[Mahara-contributors] [Bug 2004137] [NEW] Mahara 22.10 - PHP 8.0 compatibility causing error with PHP 7.4

2023-01-29 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 22.10.0
DB: Postgres 14.3
OS: Linux 20.04
Browser: n/a

The use of "mixed" type is supported in PHP 8.0+.

It was introduced in this file:

https://reviews.mahara.org/c/mahara/+/12921/28/htdocs/lib/applicationset.php

with 2 functions:

1. current() : mixed
2. key() : mixed

With PHP 7.4, when configuring institutions with Moodle integration
(XML-RPC for example), the following error is received:

"PHP message: [WAR] 40 (lib/errors.php:529) [TypeError]: Return value of
ApplicationSet::current() must be an instance of mixed, instance of
Application returned at /var/www/site/lib/applicationset.php:45

A type should not be set for these functions until it's fully
implemented for PHP 8.0+.

I only found those 2 instances in Mahara 22.10

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/2004137

Title:
  Mahara 22.10 - PHP 8.0 compatibility causing error with PHP 7.4

Status in Mahara:
  New

Bug description:
  Mahara: 22.10.0
  DB: Postgres 14.3
  OS: Linux 20.04
  Browser: n/a

  The use of "mixed" type is supported in PHP 8.0+.

  It was introduced in this file:

  https://reviews.mahara.org/c/mahara/+/12921/28/htdocs/lib/applicationset.php

  with 2 functions:

  1. current() : mixed
  2. key() : mixed

  With PHP 7.4, when configuring institutions with Moodle integration
  (XML-RPC for example), the following error is received:

  "PHP message: [WAR] 40 (lib/errors.php:529) [TypeError]: Return value
  of ApplicationSet::current() must be an instance of mixed, instance of
  Application returned at /var/www/site/lib/applicationset.php:45

  A type should not be set for these functions until it's fully
  implemented for PHP 8.0+.

  I only found those 2 instances in Mahara 22.10

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/2004137/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 2000686] Re: LTI 1.1 misaligned auth after 22.10 upgrade

2023-01-08 Thread Ghada El-Zoghbi
hi Robert,

What about the idea of a sanity check before the upgrade and a mapping?

Thank you.
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/2000686

Title:
  LTI 1.1 misaligned auth after 22.10 upgrade

Status in Mahara:
  Triaged

Bug description:
  Mahara: 22.10.0
  OS: Linux 20.04
  DB: Postgres
  Browser: n/a

  Post an upgrade from 21.10 to 22.10, the LTI auth for an institution
  is misaligned and users can no longer log in.

  Scenario with LTI integrated Blackboard LMS:
  - In 21.10, create an institution1 with 1 LIT auth "Web services"

  - Created 2 enabled registered external apps:
  1. "Blackboard LTI" <- owner was deleted but still exists (i.e. create app 
then delete user)
  2. "Blackboard LTI Mahara"  <- genuine LTI and owner still exists

  - Crete users in instintution1 with "webservice" auth

  - Upgrde to 22.10

  - All users should be converted to "Blackboard LTI Mahara" as that is
  the valid instance.

  What happens: the users are still linked to the "webservice" auth
  which doesn't have any valid registered apps.

  
  An additional SQL during the upgrade should:
  * update all auth_remote_user records from the old auth to the new auth
  * update all usr records from the old auth to the new auth
  * the old "webservice" auth in the institution should be deleted

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/2000686/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1271808] Re: Two-factor authentication for user logins

2023-01-02 Thread Ghada El-Zoghbi
Hello Mahara Team,

Just wondering if this is something to target for this year?

I think it's an important feature considering the current security
environment and challenges.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1271808

Title:
  Two-factor authentication for user logins

Status in Mahara:
  In Progress

Bug description:
  Allow users to configure two-factor authentication for an extra layer
  of security when logging into their accounts.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1271808/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 2000686] [NEW] LTI 1.1 misaligned auth after 22.10 upgrade

2022-12-28 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 22.10.0
OS: Linux 20.04
DB: Postgres
Browser: n/a

Post an upgrade from 21.10 to 22.10, the LTI auth for an institution is
misaligned and users can no longer log in.

Scenario with LTI integrated Blackboard LMS:
- In 21.10, create an institution1 with 1 LIT auth "Web services"

- Created 2 enabled registered external apps:
1. "Blackboard LTI" <- owner was deleted but still exists (i.e. create app then 
delete user)
2. "Blackboard LTI Mahara"  <- genuine LTI and owner still exists

- Crete users in instintution1 with "webservice" auth

- Upgrde to 22.10

- All users should be converted to "Blackboard LTI Mahara" as that is
the valid instance.

What happens: the users are still linked to the "webservice" auth which
doesn't have any valid registered apps.


An additional SQL during the upgrade should:
* update all auth_remote_user records from the old auth to the new auth
* update all usr records from the old auth to the new auth
* the old "webservice" auth in the institution should be deleted

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/2000686

Title:
  LTI 1.1 misaligned auth after 22.10 upgrade

Status in Mahara:
  New

Bug description:
  Mahara: 22.10.0
  OS: Linux 20.04
  DB: Postgres
  Browser: n/a

  Post an upgrade from 21.10 to 22.10, the LTI auth for an institution
  is misaligned and users can no longer log in.

  Scenario with LTI integrated Blackboard LMS:
  - In 21.10, create an institution1 with 1 LIT auth "Web services"

  - Created 2 enabled registered external apps:
  1. "Blackboard LTI" <- owner was deleted but still exists (i.e. create app 
then delete user)
  2. "Blackboard LTI Mahara"  <- genuine LTI and owner still exists

  - Crete users in instintution1 with "webservice" auth

  - Upgrde to 22.10

  - All users should be converted to "Blackboard LTI Mahara" as that is
  the valid instance.

  What happens: the users are still linked to the "webservice" auth
  which doesn't have any valid registered apps.

  
  An additional SQL during the upgrade should:
  * update all auth_remote_user records from the old auth to the new auth
  * update all usr records from the old auth to the new auth
  * the old "webservice" auth in the institution should be deleted

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/2000686/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1968800] Re: Enhancement: User CSV upload to move users to new institution

2022-04-13 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Matthew Hilton (matthewhilton)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1968800

Title:
  Enhancement: User CSV upload to move users to new institution

Status in Mahara:
  New

Bug description:
  Mahara: master
  OS: any
  DB: any
  Browser: any

  Curtin University in Perth, Australia ( https://www.curtin.edu.au/ )
  would like to contribute a recent enhancement they applied to the User
  CSV bulk upload.

  Background: The university would like a mechanism to bulk move
  students recently graduated to an alumni institution.

  
  Modify the ‘Add people by CSV’ screen with the following:
  1. Repurpose the drop-down list of institution and authentication method:
  ▪ The screen currently has a list of authentication drop-down. But, if 
updating a user, they must belong to the institution with that authentication. 
Remove this restriction and change logic so it's to "move" user to the new 
institution/authentication.

  2. Repurpose the 'Force password' change button to send a link that requires 
students to set a password, like a first-time password link that also shows 
them their username.
  ▪ In the email, include that they have been added to the new institution.
  ▪ If possible, prevent the sending of a second email when adding the user to 
to a new institution. Otherwise, the users will receive 2 emails (password 
reset and new institution subscription).

  3. Repurpose the 'Email people about their account' switch to send the 
details off to the students.
  ▪ Currently, this switch is only applicable for new accounts.

  4. If the old user institution uses ‘remoteuser’ (i.e. LTI does) and
  new institution does not (i.e. manual), remove the ‘remoteuser’
  record.

  
  5. If the user is in more than one institution, remove them from both 
institutions and move them to the new institution.

  
  6.  All other functionality on the page will remain the same.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1968800/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1968800] [NEW] Enhancement: User CSV upload to move users to new institution

2022-04-12 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: master
OS: any
DB: any
Browser: any

Curtin University in Perth, Australia ( https://www.curtin.edu.au/ )
would like to contribute a recent enhancement they applied to the User
CSV bulk upload.

Background: The university would like a mechanism to bulk move students
recently graduated to an alumni institution.


Modify the ‘Add people by CSV’ screen with the following:
1. Repurpose the drop-down list of institution and authentication method:
▪ The screen currently has a list of authentication drop-down. But, if updating 
a user, they must belong to the institution with that authentication. Remove 
this restriction and change logic so it's to "move" user to the new 
institution/authentication.

2. Repurpose the 'Force password' change button to send a link that requires 
students to set a password, like a first-time password link that also shows 
them their username.
▪ In the email, include that they have been added to the new institution.
▪ If possible, prevent the sending of a second email when adding the user to to 
a new institution. Otherwise, the users will receive 2 emails (password reset 
and new institution subscription).

3. Repurpose the 'Email people about their account' switch to send the details 
off to the students.
▪ Currently, this switch is only applicable for new accounts.

4. If the old user institution uses ‘remoteuser’ (i.e. LTI does) and new
institution does not (i.e. manual), remove the ‘remoteuser’ record.


5. If the user is in more than one institution, remove them from both 
institutions and move them to the new institution.


6.  All other functionality on the page will remain the same.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1968800

Title:
  Enhancement: User CSV upload to move users to new institution

Status in Mahara:
  New

Bug description:
  Mahara: master
  OS: any
  DB: any
  Browser: any

  Curtin University in Perth, Australia ( https://www.curtin.edu.au/ )
  would like to contribute a recent enhancement they applied to the User
  CSV bulk upload.

  Background: The university would like a mechanism to bulk move
  students recently graduated to an alumni institution.

  
  Modify the ‘Add people by CSV’ screen with the following:
  1. Repurpose the drop-down list of institution and authentication method:
  ▪ The screen currently has a list of authentication drop-down. But, if 
updating a user, they must belong to the institution with that authentication. 
Remove this restriction and change logic so it's to "move" user to the new 
institution/authentication.

  2. Repurpose the 'Force password' change button to send a link that requires 
students to set a password, like a first-time password link that also shows 
them their username.
  ▪ In the email, include that they have been added to the new institution.
  ▪ If possible, prevent the sending of a second email when adding the user to 
to a new institution. Otherwise, the users will receive 2 emails (password 
reset and new institution subscription).

  3. Repurpose the 'Email people about their account' switch to send the 
details off to the students.
  ▪ Currently, this switch is only applicable for new accounts.

  4. If the old user institution uses ‘remoteuser’ (i.e. LTI does) and
  new institution does not (i.e. manual), remove the ‘remoteuser’
  record.

  
  5. If the user is in more than one institution, remove them from both 
institutions and move them to the new institution.

  
  6.  All other functionality on the page will remain the same.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1968800/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1963608] [NEW] Allow Comments by Default default value is not saved to the database on upgrade

2022-03-03 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 21.10
OS: Linux
DB: Postgres
Browser: N/A

When upgrading from 20.10 to 21.10, the new site wide default setting
for 'Allow page comments by default' is not set to the default of 'true'
during the upgrade script.

It is a new field that was introduced in bug 1922624. So it only affects
21.10 sites that get upgraded from previous versions.

Searching for the data in the DB:

select * from config c where c.field = 'allowcommentsbydefault';

results in an empty record.


A work-around is to:
* From the site admin > site settings: unset "Allow page comments by default" 
and save
* Then, set the field to true and save

The database field is then saved.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1963608

Title:
  Allow Comments by Default default value is not saved to the database
  on upgrade

Status in Mahara:
  New

Bug description:
  Mahara: 21.10
  OS: Linux
  DB: Postgres
  Browser: N/A

  When upgrading from 20.10 to 21.10, the new site wide default setting
  for 'Allow page comments by default' is not set to the default of
  'true' during the upgrade script.

  It is a new field that was introduced in bug 1922624. So it only
  affects 21.10 sites that get upgraded from previous versions.

  Searching for the data in the DB:

  select * from config c where c.field = 'allowcommentsbydefault';

  results in an empty record.

  
  A work-around is to:
  * From the site admin > site settings: unset "Allow page comments by default" 
and save
  * Then, set the field to true and save

  The database field is then saved.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1963608/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1922624] Re: Create a default option for comments on the 'Note' block and for pages

2022-03-03 Thread Ghada El-Zoghbi
There appears to be a bug introduced with this change.

The site wide 'allowcommentsbydefault' is not set when upgrading an
existing site. It is tuned on by default but not saved to the database.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1922624

Title:
  Create a default option for comments on the 'Note' block and for pages

Status in Mahara:
  Fix Released

Bug description:
  We can't set a default for allowing or not allowing of comments on new
  'Note' blocks. This is to be added to the 'internal' plugin type
  artefact at the bottom of the page in a new section 'Default comment
  permission'. That way, future additional default settings could be
  added there. The default comment setting for files stays in the 'file'
  plugin type artefact.

  Additionally, create a site setting that sets the default for page
  comments.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1922624/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1958499] [NEW] Upload CSV Users Bug with Failed Emails

2022-01-19 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 21.10.0
OS: Linux 20.04
DB: Postgres
Browser: n/a

While implementing a customisation for a client, we identified a very
old bug in htdocs/admin/users/uploadcsv.php.

The setting of the `$failedusers` array on line 639 occurs within the
foreach loop. Which means, the data in the $failedusers array gets reset
after each iteration.

The initialisation of $failedusers should occur outside of the foreach
loop.


https://github.com/MaharaProject/mahara/blob/main/htdocs/admin/users/uploadcsv.php#L644-L667


This is the current code at line 637:

if ($values['emailusers'] && $addedusers) {
foreach ($addedusers as $user) {
$failedusers = array();
try {
...

It should be:

if ($values['emailusers'] && $addedusers) {
$failedusers = array(); <- MOVE outside the foreach loop
foreach ($addedusers as $user) {
try {
...


Hope that makes sense.

Thank you.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1958499

Title:
  Upload CSV Users Bug with Failed Emails

Status in Mahara:
  New

Bug description:
  Mahara: 21.10.0
  OS: Linux 20.04
  DB: Postgres
  Browser: n/a

  While implementing a customisation for a client, we identified a very
  old bug in htdocs/admin/users/uploadcsv.php.

  The setting of the `$failedusers` array on line 639 occurs within the
  foreach loop. Which means, the data in the $failedusers array gets
  reset after each iteration.

  The initialisation of $failedusers should occur outside of the foreach
  loop.

  
  
https://github.com/MaharaProject/mahara/blob/main/htdocs/admin/users/uploadcsv.php#L644-L667

  
  This is the current code at line 637:

  if ($values['emailusers'] && $addedusers) {
  foreach ($addedusers as $user) {
  $failedusers = array();
  try {
  ...

  It should be:

  if ($values['emailusers'] && $addedusers) {
  $failedusers = array(); <- MOVE outside the foreach loop
  foreach ($addedusers as $user) {
  try {
  ...

  
  Hope that makes sense.

  Thank you.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1958499/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1939359] Re: LDAP Sync: Error with Adding

2021-08-09 Thread Ghada El-Zoghbi
Original commit: ea6baf488b9 from Bug # 1833495

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1939359

Title:
  LDAP Sync: Error with Adding

Status in Mahara:
  Confirmed

Bug description:
  Mahara: 20.10.1
  OS: Linux
  DB: Postgres (or MySQL)
  Browser: Any 

  
  I just discovered an error that I actually introduced!! I do beg your pardon.

  In the LDAP sync - htdocs/auth/ldap/lib.php::sync_users(), the
  following query actually has a syntax error!! It results in this call
  stack:

  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: Call stack (most recent first):
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * get_recordset_sql(string(size 
498)) at /var/www/site/auth/ldap/lib.php:1352
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * AuthLdap->sync_users() at 
/var/www/site/auth/ldap/lib.php:1692
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * 
PluginAuthLdap::auth_ldap_sync_cron() at /var/www/site/lib/mahara.php:1819
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * 
call_static_method(string(size 14), string(size 19)) at 
/var/www/site/lib/cron.php:119

  
  The query in question is the following:

  $sql = '
  SELECT
  e.extusername,
  e.firstname,
  e.lastname,
  e.email,
  e.studentid,
  e.preferredname
  FROM
  {auth_ldap_extusers_temp} e
  " . $sjoinsql . "
  WHERE u.id IS NULL
  ORDER BY e.extusername';

  The problem is that the string `$sjoinsql` doesn't exist and the
  quotes are all wrong.

  I'm so sorry.. I'll get a fix soon.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1939359/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1939359] [NEW] LDAP Sync: Error with Adding

2021-08-09 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 20.10.1
OS: Linux
DB: Postgres (or MySQL)
Browser: Any 


I just discovered an error that I actually introduced!! I do beg your pardon.

In the LDAP sync - htdocs/auth/ldap/lib.php::sync_users(), the following
query actually has a syntax error!! It results in this call stack:

Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: Call stack (most recent first):
Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * get_recordset_sql(string(size 
498)) at /var/www/site/auth/ldap/lib.php:1352
Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * AuthLdap->sync_users() at 
/var/www/site/auth/ldap/lib.php:1692
Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * 
PluginAuthLdap::auth_ldap_sync_cron() at /var/www/site/lib/mahara.php:1819
Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * call_static_method(string(size 
14), string(size 19)) at /var/www/site/lib/cron.php:119


The query in question is the following:

$sql = '
SELECT
e.extusername,
e.firstname,
e.lastname,
e.email,
e.studentid,
e.preferredname
FROM
{auth_ldap_extusers_temp} e
" . $sjoinsql . "
WHERE u.id IS NULL
ORDER BY e.extusername';

The problem is that the string `$sjoinsql` doesn't exist and the quotes
are all wrong.

I'm so sorry.. I'll get a fix soon.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1939359

Title:
  LDAP Sync: Error with Adding

Status in Mahara:
  New

Bug description:
  Mahara: 20.10.1
  OS: Linux
  DB: Postgres (or MySQL)
  Browser: Any 

  
  I just discovered an error that I actually introduced!! I do beg your pardon.

  In the LDAP sync - htdocs/auth/ldap/lib.php::sync_users(), the
  following query actually has a syntax error!! It results in this call
  stack:

  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: Call stack (most recent first):
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * get_recordset_sql(string(size 
498)) at /var/www/site/auth/ldap/lib.php:1352
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * AuthLdap->sync_users() at 
/var/www/site/auth/ldap/lib.php:1692
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * 
PluginAuthLdap::auth_ldap_sync_cron() at /var/www/site/lib/mahara.php:1819
  Aug  8 00:05:04 08a80cd1bd91 ecs_mahara_prod: * 
call_static_method(string(size 14), string(size 19)) at 
/var/www/site/lib/cron.php:119

  
  The query in question is the following:

  $sql = '
  SELECT
  e.extusername,
  e.firstname,
  e.lastname,
  e.email,
  e.studentid,
  e.preferredname
  FROM
  {auth_ldap_extusers_temp} e
  " . $sjoinsql . "
  WHERE u.id IS NULL
  ORDER BY e.extusername';

  The problem is that the string `$sjoinsql` doesn't exist and the
  quotes are all wrong.

  I'm so sorry.. I'll get a fix soon.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1939359/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1877001] Re: Skins not showing after upgrade

2021-08-09 Thread Ghada El-Zoghbi
Hi @robert,

We're applied all these fixes and we're still seeing the error in the
error logs:

Aug  9 15:35:04 cff35e4bccaa ecs_mahara_prod: 2021/08/09 15:35:04 [error] 
216#216: 
*2055 FastCGI sent in stderr: "PHP message: [WAR] 8b (skin/style.php:82) 
Undefined index: view_block_header_font
#012PHP message: Call stack (most recent first):
#012PHP message:   * log_message(string(size 39), integer, true, true, 
string(size 28), integer) at /var/www/site/lib/errors.php:515
#012PHP message:   * error(integer, string(size 39), string(size 28), integer, 
array(size 67)) at /var/www/site/skin/style.php:82
#012PHP message: 
#012PHP message: [WAR] 8b (lib/errors.php:853) Failed to get a recordset: 
postgres8 error: [-2: ERROR:  syntax error at or near "LIMIT"
#012PHP message: [WAR] 8b (lib/errors.php:853) LINE 1: SELECT * FROM 
"skin_fonts"  WHERE "name" = ?  LIMIT 2
#012PHP message: [WAR] 8b (lib/errors.php:853)  
 ^] in EXECUTE("SELECT * FROM "skin_fonts"  WHERE "name" = 
?  LIMIT 2")Command was: SELECT * FROM "skin_fonts"  WHERE "name" = ? 
#012PHP message: Call stack (most recent first):
#012PHP message:   * log_message(string(size 333), integer, true, true) at 
/var/www/site/lib/errors.php:89
#012PHP message:   * log_warn(string(size 333)) at 
/var/www/site/lib/errors.php:85
#012PHP message:   * SQLException->__construct(string(size 333)) at 
/var/www/site/lib/dml.php:541
#012PHP message:   * get_recordset_sql(string(size 45), false, integer, 
integer) at /var/www/site/lib/dml.php:360
#012PHP message:   * get_record_sql(string(size 45), array(size 0), integer) at 
/var/www/site/lib/dml.php:329
#012PHP message:   * get_record(string(size 10), string(size 4), null) at 
/var/www/site/lib/skin.php:606
#012PHP message:   * Skin::get_css_font_family_from_font_name(null) at 
/var/www/site/skin/style.php:82
#012PHP message: 
#012PHP message: [WAR] 8b (lib/dml.php:541) Failed to get a recordset: 
postgres8 error: [-2: ERROR:  syntax error at or near "LIMIT"
#012PHP message: [WAR] 8b (lib/dml.php:541) LINE 1: SELECT * FROM "skin_fonts"  
WHERE "name" = ?  LIMIT 2
#012PHP message: [WAR] 8b (lib/dml.php:541) 
  ^] in EXECUTE("SELECT * FROM "skin_fonts"  


I think the fix needs to also apply to htdocs/lib/skins.php line 601:


if ($font === '') {

=>

if (empty($font)) {


That should catch it when it's null, '', or not set.


Thank you.
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1877001

Title:
  Skins not showing after upgrade

Status in Mahara 19.10 series:
  Fix Released
Status in Mahara 20.04 series:
  Fix Released
Status in Mahara 20.10 series:
  Fix Released

Bug description:
  When upgrading from an older site to 19.10+ site the skins stop
  showing on pages

  and there are error messages in the logs regarding this

  * error(8, "Undefined index: view_block_header_font",
  "/var/www/skin/style.php", 82, array(size 68)) at
  /var/www/skin/style.php:82

  We need an upgrade step to add in blank values for the 2 new fields

  view_block_header_font
  view_block_header_font_color

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/19.10/+bug/1877001/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1938231] [NEW] Manage groups page continues eventhough group does not exist

2021-07-27 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 20.10.1
OS: Linux
DB: Postgres
Browser: FF and Chrome

A user can bookmark a page for a group and come back to it at a later
time after that group has been deleted.

The manage groups page should check that the group still exists and
notify the user if doesn't find it.


To replicate:
1. Create a group 'AAA Test' from the page /group/edit
2. As a site admin, go to Administer groups (Admin > Groups > Administer groups)
3. Edit the group 'AAA Test' by clicking on the cog
4. Bookmark the page (for me it's: 
http://localhost/admin/groups/manage.php?id=283).
5. go back to the Administer groups page and delete the group 'AAA Test'
6. go to the bookmarked page for the deleted group

What should happen:
User should be notified that the group no longer exits

What currently happens:
Warnings appear that group could not be found with a call stack:

Call stack (most recent first):
log_message("Trying to get property 'quotaused' of non-object", 8, true, true, 
"/siteroot/htdocs/admin/groups/manage.php", 22) at 
/siteroot/htdocs/lib/errors.php:515
error(8, "Trying to get property 'quotaused' of non-object", 
"/siteroot/htdocs/admin/groups/manage.php", 22, array(size 58)) at 
/siteroot/htdocs/admin/groups/manage.php:22


and many more errors as the page continues to display.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1938231

Title:
  Manage groups page continues eventhough group does not exist

Status in Mahara:
  New

Bug description:
  Mahara: 20.10.1
  OS: Linux
  DB: Postgres
  Browser: FF and Chrome

  A user can bookmark a page for a group and come back to it at a later
  time after that group has been deleted.

  The manage groups page should check that the group still exists and
  notify the user if doesn't find it.

  
  To replicate:
  1. Create a group 'AAA Test' from the page /group/edit
  2. As a site admin, go to Administer groups (Admin > Groups > Administer 
groups)
  3. Edit the group 'AAA Test' by clicking on the cog
  4. Bookmark the page (for me it's: 
http://localhost/admin/groups/manage.php?id=283).
  5. go back to the Administer groups page and delete the group 'AAA Test'
  6. go to the bookmarked page for the deleted group

  What should happen:
  User should be notified that the group no longer exits

  What currently happens:
  Warnings appear that group could not be found with a call stack:

  Call stack (most recent first):
  log_message("Trying to get property 'quotaused' of non-object", 8, true, 
true, "/siteroot/htdocs/admin/groups/manage.php", 22) at 
/siteroot/htdocs/lib/errors.php:515
  error(8, "Trying to get property 'quotaused' of non-object", 
"/siteroot/htdocs/admin/groups/manage.php", 22, array(size 58)) at 
/siteroot/htdocs/admin/groups/manage.php:22

  
  and many more errors as the page continues to display.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1938231/+subscriptions


___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1928999] Re: Error while deleting an External App if LTI plugin was never installed

2021-05-19 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1928999

Title:
  Error while deleting an External App if LTI plugin was never installed

Status in Mahara:
  New

Bug description:
  Mahara: 20.10.1
  OS: Linux 18.04
  DB: Postgres
  Browser: any

  
  An error is generated when a user deletes an External App if the LTI plugin 
was never installed on the site.

  To replicate:
  - Create a new Mahara instance
  - DO NOT enable the LTI plugin
  - Add an External App ( Site admin > Web services > External apps ) 
  - Delete the External App - confirm yes,you want to delete.
  - The following error is generated:

  
  [WAR] bd (lib/errors.php:853) Failed to get a recordset: postgres8 error: 
[-18: ERROR:  relation "lti_assessment" does not exist
  LINE 1: DELETE FROM "lti_assessment"
  ^] in EXECUTE("
  DELETE FROM "lti_assessment"
  WHERE oauthserver = ?
  ")Command was:
  DELETE FROM "lti_assessment"
  WHERE oauthserver = ?
  and values was (0:1)

  
  Call stack (most recent first):
  log_message("Failed to get a recordset: postgres8 error: [-18: ...", 8, true, 
true) at /siteroot/htdocs/lib/errors.php:89
  log_warn("Failed to get a recordset: postgres8 error: [-18: ...") at 
/siteroot/htdocs/lib/errors.php:853
  SQLException->__construct("Failed to get a recordset: postgres8 error: [-18: 
...") at /siteroot/htdocs/lib/dml.php:1156
  delete_records_sql(" DELETE FROM "lti_...", array(size 1)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:143
  webservices_server_submit(object(Pieform), array(size 4)) at 
/siteroot/htdocs/lib/pieforms/pieform.php:568
  Pieform->__construct(array(size 7)) at 
/siteroot/htdocs/lib/pieforms/pieform.php:168
  Pieform::process(array(size 7)) at /siteroot/htdocs/lib/mahara.php:5499
  pieform(array(size 7)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:558
  webservice_server_list_form(array(size 7), array(size 2)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:68

  
  The reason is because the table from the LTI plugin was never installed. A 
check that the table exists needs to be applied first. Please note - check not 
that the plugin is enabled but if the table exists.

  A situation can arise where the plugin was installed, some data is in
  the table, then the plugin is disabled. Disabling a plugin does not
  remove its existing data/tables.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1928999/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1928999] [NEW] Error while deleting an External App if LTI plugin was never installed

2021-05-19 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 20.10.1
OS: Linux 18.04
DB: Postgres
Browser: any


An error is generated when a user deletes an External App if the LTI plugin was 
never installed on the site.

To replicate:
- Create a new Mahara instance
- DO NOT enable the LTI plugin
- Add an External App ( Site admin > Web services > External apps ) 
- Delete the External App - confirm yes,you want to delete.
- The following error is generated:


[WAR] bd (lib/errors.php:853) Failed to get a recordset: postgres8 error: [-18: 
ERROR:  relation "lti_assessment" does not exist
LINE 1: DELETE FROM "lti_assessment"
^] in EXECUTE("
DELETE FROM "lti_assessment"
WHERE oauthserver = ?
")Command was:
DELETE FROM "lti_assessment"
WHERE oauthserver = ?
and values was (0:1)


Call stack (most recent first):
log_message("Failed to get a recordset: postgres8 error: [-18: ...", 8, true, 
true) at /siteroot/htdocs/lib/errors.php:89
log_warn("Failed to get a recordset: postgres8 error: [-18: ...") at 
/siteroot/htdocs/lib/errors.php:853
SQLException->__construct("Failed to get a recordset: postgres8 error: [-18: 
...") at /siteroot/htdocs/lib/dml.php:1156
delete_records_sql(" DELETE FROM "lti_...", array(size 1)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:143
webservices_server_submit(object(Pieform), array(size 4)) at 
/siteroot/htdocs/lib/pieforms/pieform.php:568
Pieform->__construct(array(size 7)) at 
/siteroot/htdocs/lib/pieforms/pieform.php:168
Pieform::process(array(size 7)) at /siteroot/htdocs/lib/mahara.php:5499
pieform(array(size 7)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:558
webservice_server_list_form(array(size 7), array(size 2)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:68


The reason is because the table from the LTI plugin was never installed. A 
check that the table exists needs to be applied first. Please note - check not 
that the plugin is enabled but if the table exists.

A situation can arise where the plugin was installed, some data is in
the table, then the plugin is disabled. Disabling a plugin does not
remove its existing data/tables.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: mahara-contributors
https://bugs.launchpad.net/bugs/1928999

Title:
  Error while deleting an External App if LTI plugin was never installed

Status in Mahara:
  New

Bug description:
  Mahara: 20.10.1
  OS: Linux 18.04
  DB: Postgres
  Browser: any

  
  An error is generated when a user deletes an External App if the LTI plugin 
was never installed on the site.

  To replicate:
  - Create a new Mahara instance
  - DO NOT enable the LTI plugin
  - Add an External App ( Site admin > Web services > External apps ) 
  - Delete the External App - confirm yes,you want to delete.
  - The following error is generated:

  
  [WAR] bd (lib/errors.php:853) Failed to get a recordset: postgres8 error: 
[-18: ERROR:  relation "lti_assessment" does not exist
  LINE 1: DELETE FROM "lti_assessment"
  ^] in EXECUTE("
  DELETE FROM "lti_assessment"
  WHERE oauthserver = ?
  ")Command was:
  DELETE FROM "lti_assessment"
  WHERE oauthserver = ?
  and values was (0:1)

  
  Call stack (most recent first):
  log_message("Failed to get a recordset: postgres8 error: [-18: ...", 8, true, 
true) at /siteroot/htdocs/lib/errors.php:89
  log_warn("Failed to get a recordset: postgres8 error: [-18: ...") at 
/siteroot/htdocs/lib/errors.php:853
  SQLException->__construct("Failed to get a recordset: postgres8 error: [-18: 
...") at /siteroot/htdocs/lib/dml.php:1156
  delete_records_sql(" DELETE FROM "lti_...", array(size 1)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:143
  webservices_server_submit(object(Pieform), array(size 4)) at 
/siteroot/htdocs/lib/pieforms/pieform.php:568
  Pieform->__construct(array(size 7)) at 
/siteroot/htdocs/lib/pieforms/pieform.php:168
  Pieform::process(array(size 7)) at /siteroot/htdocs/lib/mahara.php:5499
  pieform(array(size 7)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:558
  webservice_server_list_form(array(size 7), array(size 2)) at 
/siteroot/htdocs/webservice/admin/oauthv1sregister.php:68

  
  The reason is because the table from the LTI plugin was never installed. A 
check that the table exists needs to be applied first. Please note - check not 
that the plugin is enabled but if the table exists.

  A situation can arise where the plugin was installed, some data 

[Mahara-contributors] [Bug 1914715] [NEW] Export of large files timing out after 10 seconds

2021-02-04 Thread Ghada El-Zoghbi
Public bug reported:

Mahara 20.10.0
OS: Linux 16.04 (PHP 7.0)
DB: Postgres
Browser: FF and Chrome


When a user exports a collection with a very large file to write during the 
export, the process times out after 10 seconds: Maximum execution time of 10 
seconds exceeded


This is the error:
Feb  5 14:03:28 mahara_prod: 2021/02/05 14:03:28 [error] 227#227: 
*17943 FastCGI sent in stderr: 
"PHP message: PHP Fatal error:  Maximum execution time of 10 seconds exceeded 
in /var/www/site/export/lib.php on line 986" while reading response header from 
upstream, 
request: "POST /export/download.php HTTP/1.1", 
upstream: "fastcgi://unix:/var/run/php-fpm7.sock:", 


This is caused by 'set_time_limit(10);' in /htdocs/export/download.php line 63.

If the file takes longer than 10 seconds to write out to the directory,
the whole process fails.

Unfortunately, I cannot provide the file as we do not have permission to
do so.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1914715

Title:
  Export of large files timing out after 10 seconds

Status in Mahara:
  New

Bug description:
  Mahara 20.10.0
  OS: Linux 16.04 (PHP 7.0)
  DB: Postgres
  Browser: FF and Chrome

  
  When a user exports a collection with a very large file to write during the 
export, the process times out after 10 seconds: Maximum execution time of 10 
seconds exceeded

  
  This is the error:
  Feb  5 14:03:28 mahara_prod: 2021/02/05 14:03:28 [error] 227#227: 
  *17943 FastCGI sent in stderr: 
  "PHP message: PHP Fatal error:  Maximum execution time of 10 seconds exceeded 
in /var/www/site/export/lib.php on line 986" while reading response header from 
upstream, 
  request: "POST /export/download.php HTTP/1.1", 
  upstream: "fastcgi://unix:/var/run/php-fpm7.sock:", 

  
  This is caused by 'set_time_limit(10);' in /htdocs/export/download.php line 
63.

  If the file takes longer than 10 seconds to write out to the
  directory, the whole process fails.

  Unfortunately, I cannot provide the file as we do not have permission
  to do so.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1914715/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1912582] Re: Share access with account holders outside of institution is not matching users

2021-01-20 Thread Ghada El-Zoghbi
*** This bug is a duplicate of bug 1903620 ***
https://bugs.launchpad.net/bugs/1903620

This is a duplicate of this bug:

https://bugs.launchpad.net/mahara/+bug/1903620

** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

** This bug has been marked a duplicate of bug 1903620
   Access list needs to retain the sorted list of people

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1912582

Title:
  Share access with account holders outside of institution is not
  matching users

Status in Mahara:
  New

Bug description:
  Mahara: 20.10.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  The fix to Bug #1892664 seems to have introduced another bug.

  No records are matching the institution.

  This is because on line 71 in htdocs/view/access.json.php:

  array_keys($peoplein);

  it is not setting the array with the keys. It should be:

  $peoplein = array_keys($peoplein);

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1912582/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1912582] [NEW] Share access with account holders outside of institution is not matching users

2021-01-20 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 20.10.0
OS: Linux 16.04
DB: Postgres
Browser: FF and Chrome

The fix to Bug #1892664 seems to have introduced another bug.

No records are matching the institution.

This is because on line 71 in htdocs/view/access.json.php:

array_keys($peoplein);

it is not setting the array with the keys. It should be:

$peoplein = array_keys($peoplein);

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1912582

Title:
  Share access with account holders outside of institution is not
  matching users

Status in Mahara:
  New

Bug description:
  Mahara: 20.10.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  The fix to Bug #1892664 seems to have introduced another bug.

  No records are matching the institution.

  This is because on line 71 in htdocs/view/access.json.php:

  array_keys($peoplein);

  it is not setting the array with the keys. It should be:

  $peoplein = array_keys($peoplein);

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1912582/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1898650] [NEW] Duplicate Collection View Displayorder

2020-10-05 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.10.04  ( 2019093018 )
OS: Linux 16.04 (PHP 7.0)
DB: Postgres
Browser: FF / Chrome


In certain situations, the users are able to create a collection where the 
displayorder of the pages are duplicated. i.e. 2 pages with displayorder = 0 in 
the collection_view table.

I have no idea how this situation came about. I only saw it in one of
our production sites.

For 1 collection with 37 pages, there were:

* 2 pages with displayorder = 0
* 2 pages with displayorder = 1

The remaining pages were ordered correctly starting from 2 onwards.

The result was the error:

'A nonrecoverable error occurred. This probably means you have
encountered a bug in the system'

When trying to go to the Pages and Collections screen.

The offending subquery which returned multiple records for a column is
in lib/view.php::get_myviews_data() line 3804:

$collselect = '
UNION
SELECT (SELECT view FROM {collection_view} cvid WHERE 
cvid.collection = c.id AND displayorder = 0) as id,
null AS vid, c.name as title, c.name AS vtitle, c.description, null 
as type, c.ctime as vctime, c.mtime as vmtime, c.mtime as vatime,
c.owner, c.group, c.institution, null as locked, null as 
ownerformat, null as urlid, null AS vvisits,
   (SELECT COUNT(*) FROM {collection_view} cv WHERE 
cv.collection = c.id) AS numviews, c.id AS collid';


As a side note, we really should not be calling subqueries for a column. This 
has huge performance issues since that subquery will be run for every row.

I also note that there is another subquery to retrieve 'numviews'.
Again, another performance hit.


To eliminate the performance hit, the following should be changed:

$collselect = '
UNION
SELECT cvid.view as id,
null AS vid, c.name as title, c.name AS vtitle, c.description, null 
as type, c.ctime as vctime, c.mtime as vmtime, c.mtime as vatime,
c.owner, c.group, c.institution, null as locked, null as 
ownerformat, null as urlid, null AS vvisits,
numviews.numviews AS numviews, c.id AS collid';


$collfrom = '
FROM {view} v
LEFT OUTER JOIN {collection_view} cv ON cv.view = v.id
LEFT OUTER JOIN {collection} c ON cv.collection = c.id
INNER join {collection_view} cvid on cvid.collection = c.id AND 
cvid.displayorder = 0
INNER JOIN (SELECT COUNT(*) as numviews, numcv.collection 
FROM {collection_view} numcv
INNER JOIN {collection} numc ON numcv.collection = numc.id
GROUP BY numcv.collection) as numviews on 
numviews.collection = cv.collection';


and under 'mostcomments':

$collgroupby = ' GROUP BY cvid.view, numviews.numviews, c.id';

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1898650

Title:
  Duplicate Collection View Displayorder

Status in Mahara:
  New

Bug description:
  Mahara: 19.10.04  ( 2019093018 )
  OS: Linux 16.04 (PHP 7.0)
  DB: Postgres
  Browser: FF / Chrome

  
  In certain situations, the users are able to create a collection where the 
displayorder of the pages are duplicated. i.e. 2 pages with displayorder = 0 in 
the collection_view table.

  I have no idea how this situation came about. I only saw it in one of
  our production sites.

  For 1 collection with 37 pages, there were:

  * 2 pages with displayorder = 0
  * 2 pages with displayorder = 1

  The remaining pages were ordered correctly starting from 2 onwards.

  The result was the error:

  'A nonrecoverable error occurred. This probably means you have
  encountered a bug in the system'

  When trying to go to the Pages and Collections screen.

  The offending subquery which returned multiple records for a column is
  in lib/view.php::get_myviews_data() line 3804:

  $collselect = '
  UNION
  SELECT (SELECT view FROM {collection_view} cvid WHERE 
cvid.collection = c.id AND displayorder = 0) as id,
  null AS vid, c.name as title, c.name AS vtitle, c.description, 
null as type, c.ctime as vctime, c.mtime as vmtime, c.mtime as vatime,
  c.owner, c.group, c.institution, null as locked, null as 
ownerformat, null as urlid, null AS vvisits,
 (SELECT COUNT(*) FROM {collection_view} cv WHERE 
cv.collection = c.id) AS numviews, c.id AS collid';

  
  As a side note, we really should not be calling subqueries for a column. This 
has huge performance issues since that subquery will be run for every row.

  I also note that there is another subquery to retrieve 'numviews'.
  Again, another performance hit.

  

[Mahara-contributors] [Bug 1890258] Re: Missing files crashing 19.10.4 upgrade when updating skins

2020-08-06 Thread Ghada El-Zoghbi
Thanks Cecilia. It unfortunately didn't fix the issue when a file is
missing from disk (or storage area).

The issue are missing files. They are referenced in the skin definition.
When the thumbnail is getting generated, it fails because it's not in
the file system.

We are also using the objectfs plugin with S3. I had to ensure the file
exists when making the thumbnail.

It finally succeeded.

I've attached the patch I implemented but I'm not sure you want it in
core Mahara.



** Attachment added: "Patch applied"
   
https://bugs.launchpad.net/mahara/+bug/1890258/+attachment/5399153/+files/commit1.txt

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1890258

Title:
  Missing files crashing 19.10.4 upgrade when updating skins

Status in Mahara:
  New

Bug description:
  Mahara: 19.10.4
  OS: Linux 18.04
  DB: Postgres
  Browser: N/A (this is CLI)

  This is a follow-on to Bug #1877001

  When upgrading a Mahara (from 19.10.3 to 19.10.4), if artefact files
  are missing for some reason, the upgrade crashes.

  Also, there was a missing artefact in the DB which also crashed the
  upgrade when calling:

  $fileobj = artefact_instance_from_id($fileid);

  In Dev and Staging, we work with the DB sometimes restored from Prod.
  But, we don't always copy the site data as it can be quite large.
  Mahara has generally gracefully handled missing files.

  As for the missing artefact, I have not dug into why it is missing.

  Unfortunately, this time, the upgrade seems to not handle missing
  files for skins.

  Here is a small subset of the errors that are thrown (there were
  thousands and eventually ran out of memory).

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() 

[Mahara-contributors] [Bug 1890518] Re: Date Picker not working for Admin on first select of "Users have logged in since"

2020-08-05 Thread Ghada El-Zoghbi
** Description changed:

  Mahara: 19.04.6
  OS: Linux 19.04
  DB: Postgres
  Browser: Chrome and FF
  
  This follows on from bug# 1876625
  
  There is still a bug when the user first selects the "Users have logged
  in since". It does not filter the users based on the default date that
  first appears (i.e. 6/8/2020 12:00 AM).
  
  The user must change the date to something else, then change back to the
  default date (6/8/2020 12:00 AM). It then filters the users.
  
  The same issue for "Users have not logged in since" selection.
  
- 
  To reproduce:
  
  1. Login as admin
  2. Go to Admin > Users > User search
  3. Click Advanced options
  4. Select 'Users have logged in since'
  
  Expected result:
  The screen should filter based on the default date.
  
  What happens:
  The filter returns 0 users.
  You have to:
- - Select a date other than today (this filters)
- - Select a today at midnight (which is the default) - then, it applies the 
filter.
+ - Select a date other than today (this is applied and filters the users)
+ - Select today at midnight (which is the default) - then, it applies the 
filter.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1890518

Title:
  Date Picker not working for Admin on first select of "Users have
  logged in since"

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.6
  OS: Linux 19.04
  DB: Postgres
  Browser: Chrome and FF

  This follows on from bug# 1876625

  There is still a bug when the user first selects the "Users have
  logged in since". It does not filter the users based on the default
  date that first appears (i.e. 6/8/2020 12:00 AM).

  The user must change the date to something else, then change back to
  the default date (6/8/2020 12:00 AM). It then filters the users.

  The same issue for "Users have not logged in since" selection.

  To reproduce:

  1. Login as admin
  2. Go to Admin > Users > User search
  3. Click Advanced options
  4. Select 'Users have logged in since'

  Expected result:
  The screen should filter based on the default date.

  What happens:
  The filter returns 0 users.
  You have to:
  - Select a date other than today (this is applied and filters the users)
  - Select today at midnight (which is the default) - then, it applies the 
filter.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1890518/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1876625] Re: Date picker not working for Admin -> People when selecting a date to filter users

2020-08-05 Thread Ghada El-Zoghbi
I've raised this bug:

https://bugs.launchpad.net/mahara/+bug/1890518

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1876625

Title:
  Date picker not working for Admin -> People when selecting a date to
  filter users

Status in Mahara:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Released
Status in Mahara 19.10 series:
  Fix Released
Status in Mahara 20.04 series:
  Fix Released
Status in Mahara 20.10 series:
  Fix Committed

Bug description:
  To test:

  Go to Admin -> People -> Search Users
  Click on Advanced options and select 'People have not logged since' from the 
'Last login' dropdown

  The results should change

  Note: for good testing you will need to update some users to have varying 
'lastlogin' values 
  - you can do this by going, eg:
   UPDATE usr SET lastlogin = '2020-01-01 12:12:12' WHERE id = 2;
   UPDATE usr SET lastlogin = '2020-02-02 12:12:12' WHERE id = 3;
  etc

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1876625/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1890518] [NEW] Date Picker not working for Admin on first select of "Users have logged in since"

2020-08-05 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.6
OS: Linux 19.04
DB: Postgres
Browser: Chrome and FF

This follows on from bug# 1876625

There is still a bug when the user first selects the "Users have logged
in since". It does not filter the users based on the default date that
first appears (i.e. 6/8/2020 12:00 AM).

The user must change the date to something else, then change back to the
default date (6/8/2020 12:00 AM). It then filters the users.

The same issue for "Users have not logged in since" selection.


To reproduce:

1. Login as admin
2. Go to Admin > Users > User search
3. Click Advanced options
4. Select 'Users have logged in since'

Expected result:
The screen should filter based on the default date.

What happens:
The filter returns 0 users.
You have to:
- Select a date other than today (this filters)
- Select a today at midnight (which is the default) - then, it applies the 
filter.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1890518

Title:
  Date Picker not working for Admin on first select of "Users have
  logged in since"

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.6
  OS: Linux 19.04
  DB: Postgres
  Browser: Chrome and FF

  This follows on from bug# 1876625

  There is still a bug when the user first selects the "Users have
  logged in since". It does not filter the users based on the default
  date that first appears (i.e. 6/8/2020 12:00 AM).

  The user must change the date to something else, then change back to
  the default date (6/8/2020 12:00 AM). It then filters the users.

  The same issue for "Users have not logged in since" selection.

  
  To reproduce:

  1. Login as admin
  2. Go to Admin > Users > User search
  3. Click Advanced options
  4. Select 'Users have logged in since'

  Expected result:
  The screen should filter based on the default date.

  What happens:
  The filter returns 0 users.
  You have to:
  - Select a date other than today (this filters)
  - Select a today at midnight (which is the default) - then, it applies the 
filter.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1890518/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1876625] Re: Date picker not working for Admin -> People when selecting a date to filter users

2020-08-05 Thread Ghada El-Zoghbi
hi robertl-9

There is still a bug when the user first selects the "Users have logged
din since". It does not filter the users based on the default date that
first appears (i.e. 6/8/2020 12:00 AM).

The user must change the date to something else, then change back to the
default date (6/8/2020 12:00 AM). It then filters the users.

Is there a bug already raised for this?

Thanks,
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1876625

Title:
  Date picker not working for Admin -> People when selecting a date to
  filter users

Status in Mahara:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Released
Status in Mahara 19.10 series:
  Fix Released
Status in Mahara 20.04 series:
  Fix Released
Status in Mahara 20.10 series:
  Fix Committed

Bug description:
  To test:

  Go to Admin -> People -> Search Users
  Click on Advanced options and select 'People have not logged since' from the 
'Last login' dropdown

  The results should change

  Note: for good testing you will need to update some users to have varying 
'lastlogin' values 
  - you can do this by going, eg:
   UPDATE usr SET lastlogin = '2020-01-01 12:12:12' WHERE id = 2;
   UPDATE usr SET lastlogin = '2020-02-02 12:12:12' WHERE id = 3;
  etc

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1876625/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1876625] Re: Date picker not working for Admin -> People when selecting a date to filter users

2020-08-05 Thread Ghada El-Zoghbi
Please note, the same issue for "Users have not logged in since"
selection.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1876625

Title:
  Date picker not working for Admin -> People when selecting a date to
  filter users

Status in Mahara:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Released
Status in Mahara 19.10 series:
  Fix Released
Status in Mahara 20.04 series:
  Fix Released
Status in Mahara 20.10 series:
  Fix Committed

Bug description:
  To test:

  Go to Admin -> People -> Search Users
  Click on Advanced options and select 'People have not logged since' from the 
'Last login' dropdown

  The results should change

  Note: for good testing you will need to update some users to have varying 
'lastlogin' values 
  - you can do this by going, eg:
   UPDATE usr SET lastlogin = '2020-01-01 12:12:12' WHERE id = 2;
   UPDATE usr SET lastlogin = '2020-02-02 12:12:12' WHERE id = 3;
  etc

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1876625/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1890258] Re: Missing files crashing 19.10.4 upgrade when updating skins

2020-08-05 Thread Ghada El-Zoghbi
The steps in #2 above is for the missing artefact (i.e. the background
image with artefact id 10).

So, when a user is deleted, the skins they created are not getting
deleted (if private or not shared) or cleaned up to cater for deleted
artefacts linked to them (if public).

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1890258

Title:
  Missing files crashing 19.10.4 upgrade when updating skins

Status in Mahara:
  New

Bug description:
  Mahara: 19.10.4
  OS: Linux 18.04
  DB: Postgres
  Browser: N/A (this is CLI)

  This is a follow-on to Bug #1877001

  When upgrading a Mahara (from 19.10.3 to 19.10.4), if artefact files
  are missing for some reason, the upgrade crashes.

  Also, there was a missing artefact in the DB which also crashed the
  upgrade when calling:

  $fileobj = artefact_instance_from_id($fileid);

  In Dev and Staging, we work with the DB sometimes restored from Prod.
  But, we don't always copy the site data as it can be quite large.
  Mahara has generally gracefully handled missing files.

  As for the missing artefact, I have not dug into why it is missing.

  Unfortunately, this time, the upgrade seems to not handle missing
  files for skins.

  Here is a small subset of the errors that are thrown (there were
  thousands and eventually ran out of memory).

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • 

[Mahara-contributors] [Bug 1890258] Re: Missing files crashing 19.10.4 upgrade when updating skins

2020-08-05 Thread Ghada El-Zoghbi
Steps to reproduce:

1. Create a new vanilla Mahara version 19.04.6
2. Login as Admin and create User1 
3. Login as User1
4. Create a skin with a background image.
5. Create a page that uses the skin
6. Login as admin
7. Delete User1
8. Upgrade to 19.10.4

Upgrade fails with:

[DBG] e3 (lib/db/upgrade.php:1565) Fixing skins for new format options
The page you are looking for could not be found.

Artefact with id 10 not found

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1890258

Title:
  Missing files crashing 19.10.4 upgrade when updating skins

Status in Mahara:
  New

Bug description:
  Mahara: 19.10.4
  OS: Linux 18.04
  DB: Postgres
  Browser: N/A (this is CLI)

  This is a follow-on to Bug #1877001

  When upgrading a Mahara (from 19.10.3 to 19.10.4), if artefact files
  are missing for some reason, the upgrade crashes.

  Also, there was a missing artefact in the DB which also crashed the
  upgrade when calling:

  $fileobj = artefact_instance_from_id($fileid);

  In Dev and Staging, we work with the DB sometimes restored from Prod.
  But, we don't always copy the site data as it can be quite large.
  Mahara has generally gracefully handled missing files.

  As for the missing artefact, I have not dug into why it is missing.

  Unfortunately, this time, the upgrade seems to not handle missing
  files for skins.

  Here is a small subset of the errors that are thrown (there were
  thousands and eventually ran out of memory).

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • 

[Mahara-contributors] [Bug 1890258] Re: Missing files crashing 19.10.4 upgrade when updating skins

2020-08-05 Thread Ghada El-Zoghbi
More on this. It turns out the skin is owned by user that was deleted.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1890258

Title:
  Missing files crashing 19.10.4 upgrade when updating skins

Status in Mahara:
  New

Bug description:
  Mahara: 19.10.4
  OS: Linux 18.04
  DB: Postgres
  Browser: N/A (this is CLI)

  This is a follow-on to Bug #1877001

  When upgrading a Mahara (from 19.10.3 to 19.10.4), if artefact files
  are missing for some reason, the upgrade crashes.

  Also, there was a missing artefact in the DB which also crashed the
  upgrade when calling:

  $fileobj = artefact_instance_from_id($fileid);

  In Dev and Staging, we work with the DB sometimes restored from Prod.
  But, we don't always copy the site data as it can be quite large.
  Mahara has generally gracefully handled missing files.

  As for the missing artefact, I have not dug into why it is missing.

  Unfortunately, this time, the upgrade seems to not handle missing
  files for skins.

  Here is a small subset of the errors that are thrown (there were
  thousands and eventually ran out of memory).

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesx() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesx(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

  [WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
  Call stack (most recent first):
  • log_message("imagesy() expects parameter 1 to be resource, bool...", 8, 
true, true, "/var/www/site/lib/skin.php", 915) at 
/var/www/site/lib/errors.php:521
  • error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
  • imagesy(false) at /var/www/site/lib/skin.php:915
  • Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
  • Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
  • Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
  • xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
  • upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
  • 

[Mahara-contributors] [Bug 1890258] [NEW] Missing files crashing 19.10.4 upgrade when updating skins

2020-08-04 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.10.4
OS: Linux 18.04
DB: Postgres
Browser: N/A (this is CLI)

This is a follow-on to Bug #1877001

When upgrading a Mahara (from 19.10.3 to 19.10.4), if artefact files are
missing for some reason, the upgrade crashes.

Also, there was a missing artefact in the DB which also crashed the
upgrade when calling:

$fileobj = artefact_instance_from_id($fileid);

In Dev and Staging, we work with the DB sometimes restored from Prod.
But, we don't always copy the site data as it can be quite large. Mahara
has generally gracefully handled missing files.

As for the missing artefact, I have not dug into why it is missing.

Unfortunately, this time, the upgrade seems to not handle missing files
for skins.

Here is a small subset of the errors that are thrown (there were
thousands and eventually ran out of memory).

[WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
Call stack (most recent first):
• log_message("imagesx() expects parameter 1 to be resource, bool...", 8, true, 
true, "/var/www/site/lib/skin.php", 915) at /var/www/site/lib/errors.php:521
• error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
• imagesx(false) at /var/www/site/lib/skin.php:915
• Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
• Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
• Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
• xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
• upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
• upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

[WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
Call stack (most recent first):
• log_message("imagesy() expects parameter 1 to be resource, bool...", 8, true, 
true, "/var/www/site/lib/skin.php", 915) at /var/www/site/lib/errors.php:521
• error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
• imagesy(false) at /var/www/site/lib/skin.php:915
• Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
• Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
• Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
• xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
• upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
• upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

[WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
Call stack (most recent first):
• log_message("imagesx() expects parameter 1 to be resource, bool...", 8, true, 
true, "/var/www/site/lib/skin.php", 915) at /var/www/site/lib/errors.php:521
• error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
• imagesx(false) at /var/www/site/lib/skin.php:915
• Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
• Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
• Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
• xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
• upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
• upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

[WAR] b4 (lib/skin.php:915) imagesy() expects parameter 1 to be resource, 
boolean given
Call stack (most recent first):
• log_message("imagesy() expects parameter 1 to be resource, bool...", 8, true, 
true, "/var/www/site/lib/skin.php", 915) at /var/www/site/lib/errors.php:521
• error(2, "imagesy() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
• imagesy(false) at /var/www/site/lib/skin.php:915
• Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) at 
/var/www/site/lib/skin.php:975
• Skin::generate_thumbnail("1698") at /var/www/site/lib/skin.php:261
• Skin->commit() at /var/www/site/lib/db/upgrade.php:1579
• xmldb_core_upgrade("2019093015") at /var/www/site/lib/upgrade.php:376
• upgrade_core(object(stdClass)) at /var/www/site/lib/mahara.php:274
• upgrade_mahara(array(size 1)) at /var/www/site/admin/cli/upgrade.php:74

[WAR] b4 (lib/skin.php:915) imagesx() expects parameter 1 to be resource, 
boolean given
Call stack (most recent first):
• log_message("imagesx() expects parameter 1 to be resource, bool...", 8, true, 
true, "/var/www/site/lib/skin.php", 915) at /var/www/site/lib/errors.php:521
• error(2, "imagesx() expects parameter 1 to be resource, bool...", 
"/var/www/site/lib/skin.php", 915, array(size 11)) at Unknown:0
• imagesx(false) at /var/www/site/lib/skin.php:915
• Skin::imagebackgroundfill(resource(#3027), false, 0.125, 4, 5) 

[Mahara-contributors] [Bug 1866913] Re: Elasticsearch::add_to_queue() not found in 19.04

2020-04-02 Thread Ghada El-Zoghbi
Hi Robert,

Are you able to include the steps to reproduce the issue for testers to
verify?

Thanks,
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1866913

Title:
  Elasticsearch::add_to_queue() not found in 19.04

Status in Mahara:
  Fix Committed
Status in Mahara 18.10 series:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Committed

Bug description:
  When we fixed Bug 1840201 we backported this to 19.04 branch but that
  branch didn't have the associated function.

  So will need to fix up

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1866913/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1859118] Re: Gridstack Layout Converstion Error During New User Creation

2020-01-09 Thread Ghada El-Zoghbi
To replicate:

1. In a previous version of Mahara (18.10), create an institution named Inst1
2. Create an institution page to be copied for a user in Inst1.
3. Create the page with 3 rows and 3 columns.
4. Populate the page with artifacts in all rows and columns except for 1 column 
(i.e. the row 2 column 2 for example leave empty).
5. Upgrade Mahara to 19.10
6. Create a new user in the institution Inst1.

What happens:
- The above error will be generated

What should happen:
- The new user is created with the relevant Inst1 pages copied to their 
portfolio.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1859118

Title:
  Gridstack Layout Converstion Error During New User Creation

Status in Mahara:
  New

Bug description:
  Mahara: 19.10.1 (2019093009)
  OS: Linux 18.04
  DB: Postgres
  Browser: FF

  If an institution has pages to be copied to users upon creation and
  those pages are of the old format, an error is generated when trying
  to copy the page to the new user:

  Jan 10 14:12:34 prod: 2020/01/10 14:12:34 [error] 227
  #227: *389810 FastCGI sent in stderr: "PHP message: [WAR] 3e 
(lib/gridstacklayout.php:60) Undefined index: width
  #012PHP message: Call stack (most recent first):
  #012PHP message: * log_message(string(size 22), integer, true, true, 
string(size 37), integer) at /var/www/site/lib/errors.php:521
  #012PHP message: * error(integer, string(size 22), string(size 37), integer, 
array(size 9)) at /var/www/site/lib/gridstacklayout.php:60
  #012PHP message: * translate_to_new_layout(array(size 4)) at 
/var/www/site/lib/view.php:5929
  #012PHP message: * View->copy_contents(object(View), array(size 0)) at 
/var/www/site/lib/view.php:458
  #012PHP message: * View::create_from_template(array(size 5), string(size 5), 
string(size 5), false, false, array(size 0)) at 
/var/www/site/lib/collection.php:349
  #012PHP message: * Collection::create_from_template(array(size 2), 
string(size 5), string(size 5), false) at /var/www/site/auth/user.php:1540
  #012PHP message: * User->copy_collections(array(size 8), false) at 
/var/www/site/auth/user.php:1602
  #012PHP message: * 
User->copy_institution_views_collections_to_new_member(string(size 16)) at 
/var/www/site/lib/institution.php:360
  #012PHP message: * Institution->addUserAsMember(object(stdClass), null, null) 
at /var/www/site/lib/user.php:2622
  #012PHP message: * create_user(object(stdClass), array(size 0), 
object(Institution), false, null, array(size 27)) at 
/var/www/site/admin/users/add.php:320
  #012PHP message: * adduser_submit(object(Pieform), array(size 27)) at 
/var/www/site/lib/pieforms/pieform.php:546
  #012PHP message: * Pieform->__construct(array(size 8)) at 
/var/www/site/lib/pieforms/pieform.php:168
  #012PHP message: * Pieform::process(array(size 8)) at 
/var/www/site/lib/mahara.php:5432
  #012PHP message: * pieform(array(size 8)) at 
/var/www/site/admin/users/add.php:156
  #012PHP message:
  #012PHP message: [WAR] 3e (lib/gridstacklayout.php:73) Undefined index: width
  #012PHP message: Call stack (most recent first):
  #012PHP message: * log_message(string(size 22), integer, true, true, 
string(size 37), integer) at /var/w


  The issue is:

  (lib/gridstacklayout.php:60) Undefined index: width  <-
  translate_to_new_layout() doesn't find a relevant column in the row
  (i.e. that column is empty) and is trying to add the width of that
  column.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1859118/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1859118] [NEW] Gridstack Layout Converstion Error During New User Creation

2020-01-09 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.10.1 (2019093009)
OS: Linux 18.04
DB: Postgres
Browser: FF

If an institution has pages to be copied to users upon creation and
those pages are of the old format, an error is generated when trying to
copy the page to the new user:

Jan 10 14:12:34 prod: 2020/01/10 14:12:34 [error] 227
#227: *389810 FastCGI sent in stderr: "PHP message: [WAR] 3e 
(lib/gridstacklayout.php:60) Undefined index: width
#012PHP message: Call stack (most recent first):
#012PHP message: * log_message(string(size 22), integer, true, true, 
string(size 37), integer) at /var/www/site/lib/errors.php:521
#012PHP message: * error(integer, string(size 22), string(size 37), integer, 
array(size 9)) at /var/www/site/lib/gridstacklayout.php:60
#012PHP message: * translate_to_new_layout(array(size 4)) at 
/var/www/site/lib/view.php:5929
#012PHP message: * View->copy_contents(object(View), array(size 0)) at 
/var/www/site/lib/view.php:458
#012PHP message: * View::create_from_template(array(size 5), string(size 5), 
string(size 5), false, false, array(size 0)) at 
/var/www/site/lib/collection.php:349
#012PHP message: * Collection::create_from_template(array(size 2), string(size 
5), string(size 5), false) at /var/www/site/auth/user.php:1540
#012PHP message: * User->copy_collections(array(size 8), false) at 
/var/www/site/auth/user.php:1602
#012PHP message: * 
User->copy_institution_views_collections_to_new_member(string(size 16)) at 
/var/www/site/lib/institution.php:360
#012PHP message: * Institution->addUserAsMember(object(stdClass), null, null) 
at /var/www/site/lib/user.php:2622
#012PHP message: * create_user(object(stdClass), array(size 0), 
object(Institution), false, null, array(size 27)) at 
/var/www/site/admin/users/add.php:320
#012PHP message: * adduser_submit(object(Pieform), array(size 27)) at 
/var/www/site/lib/pieforms/pieform.php:546
#012PHP message: * Pieform->__construct(array(size 8)) at 
/var/www/site/lib/pieforms/pieform.php:168
#012PHP message: * Pieform::process(array(size 8)) at 
/var/www/site/lib/mahara.php:5432
#012PHP message: * pieform(array(size 8)) at 
/var/www/site/admin/users/add.php:156
#012PHP message:
#012PHP message: [WAR] 3e (lib/gridstacklayout.php:73) Undefined index: width
#012PHP message: Call stack (most recent first):
#012PHP message: * log_message(string(size 22), integer, true, true, 
string(size 37), integer) at /var/w


The issue is:

(lib/gridstacklayout.php:60) Undefined index: width  <-
translate_to_new_layout() doesn't find a relevant column in the row
(i.e. that column is empty) and is trying to add the width of that
column.

** Affects: mahara
 Importance: Undecided
 Status: New

** Description changed:

  Mahara: 19.10.1 (2019093009)
  OS: Linux 18.04
  DB: Postgres
- Browser: FF 
+ Browser: FF
  
  If an institution has pages to be copied to users upon creation and
  those pages are of the old format, an error is generated when trying to
  copy the page to the new user:
- 
  
  Jan 10 14:12:34 prod: 2020/01/10 14:12:34 [error] 227
  #227: *389810 FastCGI sent in stderr: "PHP message: [WAR] 3e 
(lib/gridstacklayout.php:60) Undefined index: width
  #012PHP message: Call stack (most recent first):
  #012PHP message: * log_message(string(size 22), integer, true, true, 
string(size 37), integer) at /var/www/site/lib/errors.php:521
  #012PHP message: * error(integer, string(size 22), string(size 37), integer, 
array(size 9)) at /var/www/site/lib/gridstacklayout.php:60
  #012PHP message: * translate_to_new_layout(array(size 4)) at 
/var/www/site/lib/view.php:5929
  #012PHP message: * View->copy_contents(object(View), array(size 0)) at 
/var/www/site/lib/view.php:458
  #012PHP message: * View::create_from_template(array(size 5), string(size 5), 
string(size 5), false, false, array(size 0)) at 
/var/www/site/lib/collection.php:349
  #012PHP message: * Collection::create_from_template(array(size 2), 
string(size 5), string(size 5), false) at /var/www/site/auth/user.php:1540
  #012PHP message: * User->copy_collections(array(size 8), false) at 
/var/www/site/auth/user.php:1602
  #012PHP message: * 
User->copy_institution_views_collections_to_new_member(string(size 16)) at 
/var/www/site/lib/institution.php:360
  #012PHP message: * Institution->addUserAsMember(object(stdClass), null, null) 
at /var/www/site/lib/user.php:2622
  #012PHP message: * create_user(object(stdClass), array(size 0), 
object(Institution), false, null, array(size 27)) at 
/var/www/site/admin/users/add.php:320
  #012PHP message: * adduser_submit(object(Pieform), array(size 27)) at 
/var/www/site/lib/pieforms/pieform.php:546
  #012PHP message: * Pieform->__construct(array(size 8)) at 
/var/www/site/lib/pieforms/pieform.php:168
  #012PHP message: * Pieform::process(array(size 8)) at 
/var/www/site/lib/mahara.php:5432
  #012PHP message: * pieform(array(size 8)) at 
/var/www/site/admin/users/add.php:156
  #012PHP message:
  #012PHP message: [WAR] 3e (lib/gridstacklayout.php:73) Undefined 

[Mahara-contributors] [Bug 1838220] Re: Result per page selection form on dashboard edit conflicts with other form on page

2019-07-28 Thread Ghada El-Zoghbi
** Description changed:

  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome
  
  When a block - such as My Portfolios - is on the dashboard, the 'Results
  per page' section does not work.
  
  This is due to the fact that the selection item is wrapped in a form.
- But, there is already a higher level form in play for adding block on
+ But, there is already a higher level form in play for adding blocks on
  the page.
  
  To reproduce:
  1. Create at least 11 pages for a user.
  2. For the same user, edit the user dashboard
- 3. Make sure the 'My Portfolios' block on the dashboard.
+ 3. Make sure the 'My Portfolios' block is on the dashboard.
  4. While in Edit mode, select '20' for the 'Results per page'
  
  Current Action:
  Nothing happens (only the original 10 pages continue to display)
  
  Expected Result:
  The list of portfolios for the block should display the 11 user pages.
  
- 
  This will also be the case for the Watched pages block.
- 
  
  The issue is because of this template:
  
  * htdocs/theme/raw/templates/view/blocks.tpl
  
  Which wraps the first form around the pagination form (which is created
  in lib/web.php:: build_pagination())
  
  If the outer form is closed before the other form, the theme breaks.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1838220

Title:
  Result per page selection form on dashboard edit conflicts with other
  form on page

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  When a block - such as My Portfolios - is on the dashboard, the
  'Results per page' section does not work.

  This is due to the fact that the selection item is wrapped in a form.
  But, there is already a higher level form in play for adding blocks on
  the page.

  To reproduce:
  1. Create at least 11 pages for a user.
  2. For the same user, edit the user dashboard
  3. Make sure the 'My Portfolios' block is on the dashboard.
  4. While in Edit mode, select '20' for the 'Results per page'

  Current Action:
  Nothing happens (only the original 10 pages continue to display)

  Expected Result:
  The list of portfolios for the block should display the 11 user pages.

  This will also be the case for the Watched pages block.

  The issue is because of this template:

  * htdocs/theme/raw/templates/view/blocks.tpl

  Which wraps the first form around the pagination form (which is
  created in lib/web.php:: build_pagination())

  If the outer form is closed before the other form, the theme breaks.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1838220/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1838220] [NEW] Result per page selection form on dashboard edit conflicts with other form on page

2019-07-28 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF and Chrome

When a block - such as My Portfolios - is on the dashboard, the 'Results
per page' section does not work.

This is due to the fact that the selection item is wrapped in a form.
But, there is already a higher level form in play for adding block on
the page.

To reproduce:
1. Create at least 11 pages for a user.
2. For the same user, edit the user dashboard
3. Make sure the 'My Portfolios' block on the dashboard.
4. While in Edit mode, select '20' for the 'Results per page'

Current Action:
Nothing happens (only the original 10 pages continue to display)

Expected Result:
The list of portfolios for the block should display the 11 user pages.


This will also be the case for the Watched pages block.


The issue is because of this template:

* htdocs/theme/raw/templates/view/blocks.tpl

Which wraps the first form around the pagination form (which is created
in lib/web.php:: build_pagination())

If the outer form is closed before the other form, the theme breaks.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1838220

Title:
  Result per page selection form on dashboard edit conflicts with other
  form on page

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  When a block - such as My Portfolios - is on the dashboard, the
  'Results per page' section does not work.

  This is due to the fact that the selection item is wrapped in a form.
  But, there is already a higher level form in play for adding block on
  the page.

  To reproduce:
  1. Create at least 11 pages for a user.
  2. For the same user, edit the user dashboard
  3. Make sure the 'My Portfolios' block on the dashboard.
  4. While in Edit mode, select '20' for the 'Results per page'

  Current Action:
  Nothing happens (only the original 10 pages continue to display)

  Expected Result:
  The list of portfolios for the block should display the 11 user pages.

  
  This will also be the case for the Watched pages block.

  
  The issue is because of this template:

  * htdocs/theme/raw/templates/view/blocks.tpl

  Which wraps the first form around the pagination form (which is
  created in lib/web.php:: build_pagination())

  If the outer form is closed before the other form, the theme breaks.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1838220/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1834470] Re: Profile picture generates error when uploaded to Image Gallery

2019-07-17 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Anupama Dharmajan (anupamadharmajan)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1834470

Title:
  Profile picture generates error when uploaded to Image Gallery

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  When a profile picture is uploaded to an image gallery, an error is
  generated in the console and the image gallery hangs.

  To replicate:

  • Upload a profile picture to a user account - this will add the image to the 
files collection under images
  • In a page, add an image gallery with settings:
  - Image selection = Choose individual images to display
  - Select the 'Uploaded profile picture' under the images directory
  - Save

  
  • Result: Console error

  jQuery.Deferred exception: upload_max_filesize is not defined 
http://mahara/artefact/file/js/filedropzone.js?v=8112:36:9
  l@http:/mahara/js/jquery/jquery.js?v=8112:2:29375
  a/http://mahara/js/jquery/jquery.js?v=8112:2:29677
  undefined jquery.js:2:30930

  ReferenceError: upload_max_filesize is not defined

  
  • Expected result: image is saved to gallery

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1834470/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1836657] Re: Navigation block position is not inserted in selected row

2019-07-15 Thread Ghada El-Zoghbi
** Description changed:

  Mahara: 19.04
  OS: Linux 16.04
  DB: Postgres 10.9
  Browser: FF
  
  Inserting a navigation block at the end of the middle column does not
  add the block in the same position on the other pages in the collection.
  
  Scenario:
  * Create 3 pages (Pg1, Pg2, Pg3)
  
  * Create 3 text blocks on each page across the page:
  - Pg 1:  Text 1 in row 1, col1
  - Pg 1:  Text 2 in row 1, col2
  - Pg 1:  Text 3 in row 1, col3
  - Pg 2:  Text 1 in row 1, col1
  - Pg 2:  Text 2 in row 1, col2
  - Pg 2:  Text 3 in row 1, col3
  - Pg 3:  Text 1 in row 1, col1
  - Pg 3:  Text 2 in row 1, col2
  - Pg 3:  Text 3 in row 1, col3
  
  * Edit Pg2 and add a navigation block in row2, col2
  - Block title: Nav Block from Pg 2
  - Add to all pages = Yes
  - Save
  
  * Display the collection
  - Navigation block on Pg1 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.
  - Navigation block on Pg2 is in row2, col2 => correct
  - Navigation block on Pg3 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.
  
- 
  Expected results:
  - Navigation block on Pg1 is in row2, col2
  - Navigation block on Pg2 is in row2, col2
  - Navigation block on Pg3 is in row2, col2
- 
  
  I don't actually have a solution for this. The shuffle_helper() function
  in (htdocs/lib/view.php) is doing the correct behaviour by shifting the
  existing blocks down by 1 to allow space for the newly inserted block.
  
  * Do we want to have a "force position" perhaps?
- * Or, do we want to set the 'order' of the newly added navigation block to 
something huge like  to make sure it come in last?
+ * Or, do we want to set the 'order' of the newly added navigation block to 
something huge like  to make sure it come in last? 
  
- But, I'm not sure what the implications of each will be on other blocks.
- Any thoughts?
+ The second option is not "correct" either when it's added as the first
+ item in the column...
+ 
+ If we set the order to  (i.e. make it the last block order) then, it
+ will add the block on the other pages in row 2 / col2.
+ 
+ 
+ But, I'm not sure what the implications of each will be on other blocks. Any 
thoughts?

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1836657

Title:
  Navigation block position is not inserted in selected row

Status in Mahara:
  New

Bug description:
  Mahara: 19.04
  OS: Linux 16.04
  DB: Postgres 10.9
  Browser: FF

  Inserting a navigation block at the end of the middle column does not
  add the block in the same position on the other pages in the
  collection.

  Scenario:
  * Create 3 pages (Pg1, Pg2, Pg3)

  * Create 3 text blocks on each page across the page:
  - Pg 1:  Text 1 in row 1, col1
  - Pg 1:  Text 2 in row 1, col2
  - Pg 1:  Text 3 in row 1, col3
  - Pg 2:  Text 1 in row 1, col1
  - Pg 2:  Text 2 in row 1, col2
  - Pg 2:  Text 3 in row 1, col3
  - Pg 3:  Text 1 in row 1, col1
  - Pg 3:  Text 2 in row 1, col2
  - Pg 3:  Text 3 in row 1, col3

  * Edit Pg2 and add a navigation block in row2, col2
  - Block title: Nav Block from Pg 2
  - Add to all pages = Yes
  - Save

  * Display the collection
  - Navigation block on Pg1 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.
  - Navigation block on Pg2 is in row2, col2 => correct
  - Navigation block on Pg3 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.

  Expected results:
  - Navigation block on Pg1 is in row2, col2
  - Navigation block on Pg2 is in row2, col2
  - Navigation block on Pg3 is in row2, col2

  I don't actually have a solution for this. The shuffle_helper()
  function in (htdocs/lib/view.php) is doing the correct behaviour by
  shifting the existing blocks down by 1 to allow space for the newly
  inserted block.

  * Do we want to have a "force position" perhaps?
  * Or, do we want to set the 'order' of the newly added navigation block to 
something huge like  to make sure it come in last? 

  The second option is not "correct" either when it's added as the first
  item in the column...

  If we set the order to  (i.e. make it the last block order) then,
  it will add the block on the other pages in row 2 / col2.

  
  But, I'm not sure what the implications of each will be on other blocks. Any 
thoughts?

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1836657/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1836657] [NEW] Navigation block position is not inserted in selected row

2019-07-15 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04
OS: Linux 16.04
DB: Postgres 10.9
Browser: FF

Inserting a navigation block at the end of the middle column does not
add the block in the same position on the other pages in the collection.

Scenario:
* Create 3 pages (Pg1, Pg2, Pg3)

* Create 3 text blocks on each page across the page:
- Pg 1:  Text 1 in row 1, col1
- Pg 1:  Text 2 in row 1, col2
- Pg 1:  Text 3 in row 1, col3
- Pg 2:  Text 1 in row 1, col1
- Pg 2:  Text 2 in row 1, col2
- Pg 2:  Text 3 in row 1, col3
- Pg 3:  Text 1 in row 1, col1
- Pg 3:  Text 2 in row 1, col2
- Pg 3:  Text 3 in row 1, col3

* Edit Pg2 and add a navigation block in row2, col2
- Block title: Nav Block from Pg 2
- Add to all pages = Yes
- Save

* Display the collection
- Navigation block on Pg1 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.
- Navigation block on Pg2 is in row2, col2 => correct
- Navigation block on Pg3 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.


Expected results:
- Navigation block on Pg1 is in row2, col2
- Navigation block on Pg2 is in row2, col2
- Navigation block on Pg3 is in row2, col2


I don't actually have a solution for this. The shuffle_helper() function
in (htdocs/lib/view.php) is doing the correct behaviour by shifting the
existing blocks down by 1 to allow space for the newly inserted block.

* Do we want to have a "force position" perhaps?
* Or, do we want to set the 'order' of the newly added navigation block to 
something huge like  to make sure it come in last?

But, I'm not sure what the implications of each will be on other blocks.
Any thoughts?

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1836657

Title:
  Navigation block position is not inserted in selected row

Status in Mahara:
  New

Bug description:
  Mahara: 19.04
  OS: Linux 16.04
  DB: Postgres 10.9
  Browser: FF

  Inserting a navigation block at the end of the middle column does not
  add the block in the same position on the other pages in the
  collection.

  Scenario:
  * Create 3 pages (Pg1, Pg2, Pg3)

  * Create 3 text blocks on each page across the page:
  - Pg 1:  Text 1 in row 1, col1
  - Pg 1:  Text 2 in row 1, col2
  - Pg 1:  Text 3 in row 1, col3
  - Pg 2:  Text 1 in row 1, col1
  - Pg 2:  Text 2 in row 1, col2
  - Pg 2:  Text 3 in row 1, col3
  - Pg 3:  Text 1 in row 1, col1
  - Pg 3:  Text 2 in row 1, col2
  - Pg 3:  Text 3 in row 1, col3

  * Edit Pg2 and add a navigation block in row2, col2
  - Block title: Nav Block from Pg 2
  - Add to all pages = Yes
  - Save

  * Display the collection
  - Navigation block on Pg1 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.
  - Navigation block on Pg2 is in row2, col2 => correct
  - Navigation block on Pg3 is in row1, col2 => incorrect: shuffled up because 
there was already a block in that cell.

  
  Expected results:
  - Navigation block on Pg1 is in row2, col2
  - Navigation block on Pg2 is in row2, col2
  - Navigation block on Pg3 is in row2, col2


  I don't actually have a solution for this. The shuffle_helper()
  function in (htdocs/lib/view.php) is doing the correct behaviour by
  shifting the existing blocks down by 1 to allow space for the newly
  inserted block.

  * Do we want to have a "force position" perhaps?
  * Or, do we want to set the 'order' of the newly added navigation block to 
something huge like  to make sure it come in last?

  But, I'm not sure what the implications of each will be on other
  blocks. Any thoughts?

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1836657/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1606735] Re: Accessibility - Inconsistent heading levels

2019-07-14 Thread Ghada El-Zoghbi
Hi,

This issue seems to have appeared again in 19.04. A probable regression
since the bootstrap upgrade perhaps?

Can we please reopen this?

Thanks
Ghada


** Also affects: mahara/19.04
   Importance: Undecided
   Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1606735

Title:
  Accessibility - Inconsistent heading levels

Status in Mahara:
  Fix Released
Status in Mahara 19.04 series:
  New

Bug description:
  Mahara: master

  Headings jump levels - i.e. h1 -> h3 skipping h2.

  Heading levels should be used in sequence to assist users in
  understanding the structure of the page. And, prevent screen readers
  from skipping sections.

  For additional information, please refer to:
  http://webaim.org/techniques/semanticstructure/#contentstructure

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1606735/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1836325] Re: Group search fails when Group Category is specified

2019-07-12 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1836325

Title:
  Group search fails when Group Category is specified

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  Linux: 16.04
  DB: Postgres
  Browser: FF

  
  If a group search is conducted when a group with a '.' in the name and a 
category is specified, the following error occurs:

  
  WAR] 3c (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  invalid input syntax for integer: "19.04"] in EXECUTE("SELECT COUNT(*) 
FROM "group" g
  INNER JOIN (
  SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (gm.group = g.id AND 
gm.member = ? AND gm.role = 'admin')
  UNION
  SELECT g.id, 'member' AS membershiptype, '' AS reason, 
gm.role AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (g.id = gm.group AND 
gm.member = ? AND gm.role != 'admin')
  UNION
  SELECT g.id, 'invite' AS membershiptype, gmi.reason, gmi.role
  FROM "group" g
  INNER JOIN "group_member_invite" gmi ON (gmi.group = g.id AND 
gmi.member = ?)
  UNION SELECT g.id, 'request' AS membershiptype, gmr.reason, 
'' AS role
  FROM "group" g
  INNER JOIN "group_member_request" gmr ON (gmr.group = g.id 
AND gmr.member = ?)
  ) t ON t.id = g.id WHERE g.deleted = ?
  AND (
  g.name ILIKE '%' || ? || '%'
  OR g.description ILIKE '%' || ? || '%'
  OR g.shortname ILIKE '%' || ? || '%'
  ) AND g.category = ?")Command was: SELECT COUNT(*) FROM "group" g
  INNER JOIN (
  SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (gm.group = g.id AND 
gm.member = ? AND gm.role = 'admin')
  UNION
  SELECT g.id, 'member' AS membershiptype, '' AS reason, 
gm.role AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (g.id = gm.group AND 
gm.member = ? AND gm.role != 'admin')
  UNION
  SELECT g.id, 'invite' AS membershiptype, gmi.reason, gmi.role
  FROM "group" g
  INNER JOIN "group_member_invite" gmi ON (gmi.group = g.id AND 
gmi.member = ?)
  UNION SELECT g.id, 'request' AS membershiptype, gmr.reason, 
'' AS role
  FROM "group" g
  INNER JOIN "group_member_request" gmr ON (gmr.group = g.id 
AND gmr.member = ?)
  ) t ON t.id = g.id WHERE g.deleted = ?
  AND (
  g.name ILIKE '%' || ? || '%'
  OR g.description ILIKE '%' || ? || '%'
  OR g.shortname ILIKE '%' || ? || '%'
  ) AND g.category = ? and values was 
(0:2,1:2,2:2,3:2,4:0,5:2,6:19.04,7:19.04,8:19.04)
  Call stack (most recent first):

  log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
  log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  SQLException->__construct("Failed to get a recordset: postgres8 error: 
[-1: E...") at /var/www/mahara/htdocs/lib/dml.php:517
  get_recordset_sql("SELECT COUNT(*) FROM "group" g INNER ...", array(size 
9)) at /var/www/mahara/htdocs/lib/dml.php:276
  count_records_sql("SELECT COUNT(*) FROM {group} g INNER ...", array(size 
9)) at /var/www/mahara/htdocs/lib/group.php:2288
  group_get_associated_groups("2", "all", 10, 0, 2, "19.04") at 
/var/www/mahara/htdocs/group/index.php:146


  
  The issue is not due to the '.' in the group name but rather because the SQL 
is incorrectly applying the values. i.e. $catsql and $query_where are 
concatenated in the wrong order in group_get_associated_groups().

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1836325/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1836325] [NEW] Group search fails when Group Category is specified

2019-07-12 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
Linux: 16.04
DB: Postgres
Browser: FF


If a group search is conducted when a group with a '.' in the name and a 
category is specified, the following error occurs:


WAR] 3c (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  invalid input syntax for integer: "19.04"] in EXECUTE("SELECT COUNT(*) 
FROM "group" g
INNER JOIN (
SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
FROM "group" g
INNER JOIN "group_member" gm ON (gm.group = g.id AND gm.member 
= ? AND gm.role = 'admin')
UNION
SELECT g.id, 'member' AS membershiptype, '' AS reason, gm.role 
AS role
FROM "group" g
INNER JOIN "group_member" gm ON (g.id = gm.group AND gm.member 
= ? AND gm.role != 'admin')
UNION
SELECT g.id, 'invite' AS membershiptype, gmi.reason, gmi.role
FROM "group" g
INNER JOIN "group_member_invite" gmi ON (gmi.group = g.id AND 
gmi.member = ?)
UNION SELECT g.id, 'request' AS membershiptype, gmr.reason, '' 
AS role
FROM "group" g
INNER JOIN "group_member_request" gmr ON (gmr.group = g.id AND 
gmr.member = ?)
) t ON t.id = g.id WHERE g.deleted = ?
AND (
g.name ILIKE '%' || ? || '%'
OR g.description ILIKE '%' || ? || '%'
OR g.shortname ILIKE '%' || ? || '%'
) AND g.category = ?")Command was: SELECT COUNT(*) FROM "group" g
INNER JOIN (
SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
FROM "group" g
INNER JOIN "group_member" gm ON (gm.group = g.id AND gm.member 
= ? AND gm.role = 'admin')
UNION
SELECT g.id, 'member' AS membershiptype, '' AS reason, gm.role 
AS role
FROM "group" g
INNER JOIN "group_member" gm ON (g.id = gm.group AND gm.member 
= ? AND gm.role != 'admin')
UNION
SELECT g.id, 'invite' AS membershiptype, gmi.reason, gmi.role
FROM "group" g
INNER JOIN "group_member_invite" gmi ON (gmi.group = g.id AND 
gmi.member = ?)
UNION SELECT g.id, 'request' AS membershiptype, gmr.reason, '' 
AS role
FROM "group" g
INNER JOIN "group_member_request" gmr ON (gmr.group = g.id AND 
gmr.member = ?)
) t ON t.id = g.id WHERE g.deleted = ?
AND (
g.name ILIKE '%' || ? || '%'
OR g.description ILIKE '%' || ? || '%'
OR g.shortname ILIKE '%' || ? || '%'
) AND g.category = ? and values was 
(0:2,1:2,2:2,3:2,4:0,5:2,6:19.04,7:19.04,8:19.04)
Call stack (most recent first):

log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:517
get_recordset_sql("SELECT COUNT(*) FROM "group" g INNER ...", array(size 
9)) at /var/www/mahara/htdocs/lib/dml.php:276
count_records_sql("SELECT COUNT(*) FROM {group} g INNER ...", array(size 
9)) at /var/www/mahara/htdocs/lib/group.php:2288
group_get_associated_groups("2", "all", 10, 0, 2, "19.04") at 
/var/www/mahara/htdocs/group/index.php:146


The issue is not due to the '.' in the group name but rather because the
SQL is incorrectly applying the values. i.e. $catsql and $query_where
are concatenated in the wrong order in group_get_associated_groups().

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1836325

Title:
  Group search fails when Group Category is specified

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  Linux: 16.04
  DB: Postgres
  Browser: FF

  
  If a group search is conducted when a group with a '.' in the name and a 
category is specified, the following error occurs:

  
  WAR] 3c (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  invalid input syntax for integer: "19.04"] in EXECUTE("SELECT COUNT(*) 
FROM "group" g
  INNER JOIN (
  SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (gm.group = g.id AND 
gm.member = ? AND gm.role = 'admin')

[Mahara-contributors] [Bug 1836325] Re: Group search fails when Group Category is specified

2019-07-12 Thread Ghada El-Zoghbi
To reproduce:

1. Ensure group categories are enabled (and enter a few categories: Category 1, 
Category 2, etc)
2. Create a group with name "19.04 test group"
3. Search on the group:
- search = 19.04
- category = Category 1
4. Click Search

What happens now:
You will be presented with the above error

What should happen:
The page should display a list of matching groups (or none)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1836325

Title:
  Group search fails when Group Category is specified

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  Linux: 16.04
  DB: Postgres
  Browser: FF

  
  If a group search is conducted when a group with a '.' in the name and a 
category is specified, the following error occurs:

  
  WAR] 3c (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  invalid input syntax for integer: "19.04"] in EXECUTE("SELECT COUNT(*) 
FROM "group" g
  INNER JOIN (
  SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (gm.group = g.id AND 
gm.member = ? AND gm.role = 'admin')
  UNION
  SELECT g.id, 'member' AS membershiptype, '' AS reason, 
gm.role AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (g.id = gm.group AND 
gm.member = ? AND gm.role != 'admin')
  UNION
  SELECT g.id, 'invite' AS membershiptype, gmi.reason, gmi.role
  FROM "group" g
  INNER JOIN "group_member_invite" gmi ON (gmi.group = g.id AND 
gmi.member = ?)
  UNION SELECT g.id, 'request' AS membershiptype, gmr.reason, 
'' AS role
  FROM "group" g
  INNER JOIN "group_member_request" gmr ON (gmr.group = g.id 
AND gmr.member = ?)
  ) t ON t.id = g.id WHERE g.deleted = ?
  AND (
  g.name ILIKE '%' || ? || '%'
  OR g.description ILIKE '%' || ? || '%'
  OR g.shortname ILIKE '%' || ? || '%'
  ) AND g.category = ?")Command was: SELECT COUNT(*) FROM "group" g
  INNER JOIN (
  SELECT g.id, 'admin' AS membershiptype, '' AS reason, 'admin' 
AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (gm.group = g.id AND 
gm.member = ? AND gm.role = 'admin')
  UNION
  SELECT g.id, 'member' AS membershiptype, '' AS reason, 
gm.role AS role
  FROM "group" g
  INNER JOIN "group_member" gm ON (g.id = gm.group AND 
gm.member = ? AND gm.role != 'admin')
  UNION
  SELECT g.id, 'invite' AS membershiptype, gmi.reason, gmi.role
  FROM "group" g
  INNER JOIN "group_member_invite" gmi ON (gmi.group = g.id AND 
gmi.member = ?)
  UNION SELECT g.id, 'request' AS membershiptype, gmr.reason, 
'' AS role
  FROM "group" g
  INNER JOIN "group_member_request" gmr ON (gmr.group = g.id 
AND gmr.member = ?)
  ) t ON t.id = g.id WHERE g.deleted = ?
  AND (
  g.name ILIKE '%' || ? || '%'
  OR g.description ILIKE '%' || ? || '%'
  OR g.shortname ILIKE '%' || ? || '%'
  ) AND g.category = ? and values was 
(0:2,1:2,2:2,3:2,4:0,5:2,6:19.04,7:19.04,8:19.04)
  Call stack (most recent first):

  log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
  log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  SQLException->__construct("Failed to get a recordset: postgres8 error: 
[-1: E...") at /var/www/mahara/htdocs/lib/dml.php:517
  get_recordset_sql("SELECT COUNT(*) FROM "group" g INNER ...", array(size 
9)) at /var/www/mahara/htdocs/lib/dml.php:276
  count_records_sql("SELECT COUNT(*) FROM {group} g INNER ...", array(size 
9)) at /var/www/mahara/htdocs/lib/group.php:2288
  group_get_associated_groups("2", "all", 10, 0, 2, "19.04") at 
/var/www/mahara/htdocs/group/index.php:146


  
  The issue is not due to the '.' in the group name but rather because the SQL 
is incorrectly applying the values. i.e. $catsql and $query_where are 
concatenated in the wrong order in group_get_associated_groups().

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1836325/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : 

[Mahara-contributors] [Bug 1806524] Re: edit page has duplicated "addblock" in html tags

2019-07-10 Thread Ghada El-Zoghbi
Hi,


This change has introduced a bug in Mahara 19.04.0.

When the user adds a block (by clicking on the artefact - not
drag/drop), the selection of the positions displays: 'After 1', 'After
2'

It doesn't actually display the artefact titles in the drop down as it
used to in previous Mahara versions.

The fix is to update htdocs/js/view.js - line 496

* selectbox = $('#addblock_position'); -> selectbox =
$('#newblock_position');

I don't actually have time to push this upstream. So, I leave it the
Mahara team.

Thanks,
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1806524

Title:
  edit page has duplicated "addblock" in html tags

Status in Mahara:
  Fix Released
Status in Mahara 19.04 series:
  Fix Released

Bug description:
  when editing a page, if we inspect the html we can see there are two elements 
with the id="addblock". All id's in html tags should be unique.
  We need to fix this either for the form that pop's up when adding a block to 
the page by clicking on the block menu, or fix it for the modal that contains 
it.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1806524/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1835869] [NEW] Skins are not supporting the new pageheader

2019-07-08 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgers
Browser: FF / Chrome

The skin is not respecting the new pageheader of a page.

To reproduce - Repeat only horizontally:

1. Create a new skin (Skin1):
- Add a Background image (attached)
- Background image report = Repeat only horizontally
- Background image attachment = Fixed
- Background image position = Top
- Save

2. Create a new page that uses this new skin (Skin1)
- Page title = Skin Test
- Current skin = Skin1
- Save
- Add any block to the page (i.e. text)

3. Display the page
Results:
The page header (i.e. page banner at top) is displayed across the entire page 
(from right to left) whereas the rest of the page below the banner has the 
background image displayed in the right/left margins.

Expected Result:
The page header (i.e. page banner at top) should display the background image 
in the left/right margin as the rest of the page.


=

To reproduce - Repeat Vertically:

1. Updated Skin1:
- Background image report = Repeat Vertically
- Save

2. Display (refresh) Skin Test page
Results:
The page header (i.e. page banner at top) is displayed across the entire page 
(from right to left) whereas the rest of the page below the banner has the 
background image displayed in the right/left margins.

At the bottom of the page, the image's bottom is seen in the center of
the page.


Expected Result:
The page header (i.e. page banner at top) should display the background image 
in the left/right margin as the rest of the page.

At the bottom of the page, the image's bottom is seen in the center of
the page.


=

To reproduce - Repeat both directions:

1. Updated Skin1:
- Background image report = Repeat both directions
- Save

2. Display (refresh) Skin Test page
Results:
The page header (i.e. page banner at top) is displayed across the entire page 
(from right to left) whereas the rest of the page below the banner has the 
background image repeated in the right/left margins and at the bottom of the 
page.


Expected Result:
The page header (i.e. page banner at top) should display the repeated 
background image in the left/right margin as the rest of the page and at the 
bottom of the page.

** Affects: mahara
 Importance: Undecided
 Status: New

** Attachment added: "wool.jpg"
   https://bugs.launchpad.net/bugs/1835869/+attachment/5275776/+files/wool.jpg

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1835869

Title:
  Skins are not supporting the new pageheader

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgers
  Browser: FF / Chrome

  The skin is not respecting the new pageheader of a page.

  To reproduce - Repeat only horizontally:

  1. Create a new skin (Skin1):
  - Add a Background image (attached)
  - Background image report = Repeat only horizontally
  - Background image attachment = Fixed
  - Background image position = Top
  - Save

  2. Create a new page that uses this new skin (Skin1)
  - Page title = Skin Test
  - Current skin = Skin1
  - Save
  - Add any block to the page (i.e. text)

  3. Display the page
  Results:
  The page header (i.e. page banner at top) is displayed across the entire page 
(from right to left) whereas the rest of the page below the banner has the 
background image displayed in the right/left margins.

  Expected Result:
  The page header (i.e. page banner at top) should display the background image 
in the left/right margin as the rest of the page.

  
  =

  To reproduce - Repeat Vertically:

  1. Updated Skin1:
  - Background image report = Repeat Vertically
  - Save

  2. Display (refresh) Skin Test page
  Results:
  The page header (i.e. page banner at top) is displayed across the entire page 
(from right to left) whereas the rest of the page below the banner has the 
background image displayed in the right/left margins.

  At the bottom of the page, the image's bottom is seen in the center of
  the page.

  
  Expected Result:
  The page header (i.e. page banner at top) should display the background image 
in the left/right margin as the rest of the page.

  At the bottom of the page, the image's bottom is seen in the center of
  the page.


  =

  To reproduce - Repeat both directions:

  1. Updated Skin1:
  - Background image report = Repeat both directions
  - Save

  2. Display (refresh) Skin Test page
  Results:
  The page header (i.e. page banner at top) is displayed across the entire page 
(from right to left) whereas the rest of the page below the banner has the 
background image repeated in the right/left margins and at the bottom of the 
page.

  
  Expected Result:
  The page header (i.e. page banner at top) should display the repeated 
background image in the 

[Mahara-contributors] [Bug 1835868] [NEW] Institution comparison report selections are misleading

2019-07-08 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04 (but also in 17.10 and onwards)
OS: Linux 16.04
DB: Postgres
Browser: FF / Chrome

The Institution comparison report allows the user to specify the from/to
date. But, it always displays data from the beginning of time for the
site until now.

This is confusing to the user.

The "correct" behaviour is to not allow the user to select the From/to
date when configuring the report.

i.e.  Hide From/To fields when Institution comparison is selected and
display when other reports are selected.

** Affects: mahara
 Importance: Wishlist
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1835868

Title:
  Institution comparison report selections are misleading

Status in Mahara:
  New

Bug description:
  Mahara: 19.04 (but also in 17.10 and onwards)
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome

  The Institution comparison report allows the user to specify the
  from/to date. But, it always displays data from the beginning of time
  for the site until now.

  This is confusing to the user.

  The "correct" behaviour is to not allow the user to select the From/to
  date when configuring the report.

  i.e.  Hide From/To fields when Institution comparison is selected and
  display when other reports are selected.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1835868/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1835866] [NEW] Tagged Journal blocks get added to page even when cancel is clicked on first creating the block

2019-07-08 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04 (also 18.10)
OS: Linux 16.04
DB: Postgres
Browser: FF

First noticed in 19.04 but looks to have been introduced in 18.10.

User can create an empty tagged journal block when first creating on a
page - even if there are no tags set up for any journal.

To reproduce:
* Make sure there are no tags on any of the journal entries for your user
* Create a new page
* Add a tagged journal block
* Note that the button and link at the bottom of the block are: Save and Remove
* Click the 'Save' button (without making any changes)
* An error message come up with "No tag selected. You need to add tags to your 
journal entries before being able to select them here."
* The link next to the Save button is now changed to 'Cancel' whereas when we 
first enter the block, it's 'Remove'.

Expected:
User to click the Remove button as as not to create an empty tagged journal on 
page.

Currently:
* Remove link is changed to Cancel which allows to user to create an empty 
tagged journal on the page.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1835866

Title:
  Tagged Journal blocks get added to page even when cancel is clicked on
  first creating the block

Status in Mahara:
  New

Bug description:
  Mahara: 19.04 (also 18.10)
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  First noticed in 19.04 but looks to have been introduced in 18.10.

  User can create an empty tagged journal block when first creating on a
  page - even if there are no tags set up for any journal.

  To reproduce:
  * Make sure there are no tags on any of the journal entries for your user
  * Create a new page
  * Add a tagged journal block
  * Note that the button and link at the bottom of the block are: Save and 
Remove
  * Click the 'Save' button (without making any changes)
  * An error message come up with "No tag selected. You need to add tags to 
your journal entries before being able to select them here."
  * The link next to the Save button is now changed to 'Cancel' whereas when we 
first enter the block, it's 'Remove'.

  Expected:
  User to click the Remove button as as not to create an empty tagged journal 
on page.

  Currently:
  * Remove link is changed to Cancel which allows to user to create an empty 
tagged journal on the page.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1835866/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1835851] [NEW] Accessibility - Default Menu Items break Aria rules

2019-07-08 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: Chome Version 75.0.3770.100 (Official Build) (64-bit)


* Install WCAG Accessibility Audit Developer UI Chrome extension

* Using the Default Mahara theme.

* Run the Access Audit Report. Several errors are highlighted by the
WCAG report for the drop-down menu:

* On the Dashboard page:

There are 4 fail-rules:
This implies that there were elements on the page that did not pass this audit 
rule. This is the only result you will probably be interested in.

2 Severe:
Elements with ARIA roles must use a valid, non-abstract ARIA role
https://github.com/GoogleChrome/accessibility-developer-tools/wiki/Audit-Rules#ax_aria_01
3 elements break this rule:

id("main-nav")
id("main-nav-admin")
id("main-nav-user")
Elements with ARIA roles must ensure required owned elements are present
https://github.com/GoogleChrome/accessibility-developer-tools/wiki/Audit-Rules#ax_aria_08
12 elements break this rule:

id("childmenu-1")
id("childmenu-2")
id("childmenu-3")
id("childmenu-4")
id("adminchildmenu-0")
id("adminchildmenu-1")
id("adminchildmenu-2")
id("adminchildmenu-3")
id("adminchildmenu-4")
id("adminchildmenu-6")
id("adminchildmenu-7")
id("userchildmenu-8")


* On the Pages and Collections, there are more elements:


2 Severe:
Elements with ARIA roles must use a valid, non-abstract ARIA role
https://github.com/GoogleChrome/accessibility-developer-tools/wiki/Audit-Rules#ax_aria_01
3 elements break this rule:

id("main-nav")
id("main-nav-admin")
id("main-nav-user")
Elements with ARIA roles must ensure required owned elements are present
https://github.com/GoogleChrome/accessibility-developer-tools/wiki/Audit-Rules#ax_aria_08
27 elements break this rule:

id("childmenu-1")
id("childmenu-2")
id("childmenu-3")
id("childmenu-4")
id("adminchildmenu-0")
id("adminchildmenu-1")
id("adminchildmenu-2")
id("adminchildmenu-3")
id("adminchildmenu-4")
id("adminchildmenu-6")
id("adminchildmenu-7")
id("userchildmenu-8")
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-access"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]
id("myviews")/div[@class="card-quarter card-collection"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-access"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-collection"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]
id("myviews")/div[@class="card-quarter card-collection"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="collection-list"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-access"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-access"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]
id("myviews")/div[@class="card-quarter card-collection"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-access"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-collection"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]
id("myviews")/div[@class="card-quarter card-collection"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="collection-list"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-access"]/ul[@class="dropdown-menu"]
id("myviews")/div[@class="card-quarter card-view"]/div[@class="card 
"]/div[@class="card-footer"]/div[@class="page-controls"]/ul[@class="dropdown-menu
 dropdown-menu-right"]


There are also several warnings that I'm not listing here.


* On a portfolio page with journals and several variations of headings (H4-H6): 
(see attached image)

There are 6 fail-rules:
This implies that there were elements on the page that did not pass this audit 
rule. This is the only result you will probably be interested in.

4 Severe:
ARIA state and property values must be valid
https://github.com/GoogleChrome/accessibility-developer-tools/wiki/Audit-Rules#ax_aria_04
1 element breaks this rule:


[Mahara-contributors] [Bug 1834470] [NEW] Profile picture generates error when uploaded to Image Gallery

2019-06-27 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF

When a profile picture is uploaded to an image gallery, an error is
generated in the console and the image gallery hangs.

To replicate:

• Upload a profile picture to a user account - this will add the image to the 
files collection under images
• In a page, add an image gallery with settings:
- Image selection = Choose individual images to display
- Select the 'Uploaded profile picture' under the images directory
- Save


• Result: Console error

jQuery.Deferred exception: upload_max_filesize is not defined 
http://mahara/artefact/file/js/filedropzone.js?v=8112:36:9
l@http:/mahara/js/jquery/jquery.js?v=8112:2:29375
a/http://mahara/js/jquery/jquery.js?v=8112:2:29677
undefined jquery.js:2:30930

ReferenceError: upload_max_filesize is not defined


• Expected result: image is saved to gallery

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1834470

Title:
  Profile picture generates error when uploaded to Image Gallery

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  When a profile picture is uploaded to an image gallery, an error is
  generated in the console and the image gallery hangs.

  To replicate:

  • Upload a profile picture to a user account - this will add the image to the 
files collection under images
  • In a page, add an image gallery with settings:
  - Image selection = Choose individual images to display
  - Select the 'Uploaded profile picture' under the images directory
  - Save

  
  • Result: Console error

  jQuery.Deferred exception: upload_max_filesize is not defined 
http://mahara/artefact/file/js/filedropzone.js?v=8112:36:9
  l@http:/mahara/js/jquery/jquery.js?v=8112:2:29375
  a/http://mahara/js/jquery/jquery.js?v=8112:2:29677
  undefined jquery.js:2:30930

  ReferenceError: upload_max_filesize is not defined

  
  • Expected result: image is saved to gallery

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1834470/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1834261] Re: Reports generating error about time zone note specified

2019-06-26 Thread Ghada El-Zoghbi
** Summary changed:

- Reports generating error about time zon note specified
+ Reports generating error about time zone note specified

** Summary changed:

- Reports generating error about time zone note specified
+ Reports generating error about time zone not specified

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1834261

Title:
  Reports generating error about time zone not specified

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres 10.9
  Browser: FF

  An error is generated with running several reports but I'll give the
  example of the masquerading report:

  [WAR] ac (lib/errors.php:858) Failed to get a recordset: postgres8 error: 
[-1: ERROR: invalid input syntax for type timestamp with time zone: "Not 
specified"] in EXECUTE("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e ON 
e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 AND e.event = 'loginas' AND 
e.ctime >= DATE(?) AND e.ctime <= DATE(?)")Command was: SELECT COUNT(*) FROM 
"usr" u JOIN "event_log" e ON e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 
AND e.event = 'loginas' AND e.ctime >= DATE(?) AND e.ctime <= DATE(?) and 
values was (0:Notspecified,1:Notspecified)
  Call stack (most recent first):

  log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, true, 
true) at /var/www/mahara/htdocs/lib/errors.php:95
  log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:517
  get_recordset_sql("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/dml.php:276
  count_records_sql("SELECT COUNT(*) FROM {usr} u JOIN {event_log} e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/statistics.php:3535
  masquerading_stats_table(10, 0, array(size 5), null, 
https://my_dev_mahara/admin/users/statistics.php...;) at 
/var/www/mahara/htdocs/lib/statistics.php:3504
  masquerading_statistics(10, 0, array(size 5), null) at 
/var/www/mahara/htdocs/lib/statistics.php:4285
  display_statistics("all", "users", object(stdClass)) at 
/var/www/mahara/htdocs/admin/users/statistics.php:251

  To replicate:

  1. Turn on all event logging:  Admin > Configure site > Site options > 
Logging settings > Log events = All
  2. Set up Time zone in Admin > Configure site > Site options > Site settings 
> Time zone = "Australia/Sydney"
  3. Add User1 to Mahara
  4. Masquerade User1 (ie do not log in as them bug masquerade)
  5. Return back to admin user
  65. Run the Masquerading report:  Admin > Reports > Configure report
  7. Select:
  - Report Type = Masquerading sessions
  - Submit

  What happens: Above error is generated with:

  Masquerading sessions | All institutions
  Mahara: Site unavailable
  A nonrecoverable error occurred. This probably means you have encountered a 
bug in the system

  What should happen: the report should be displayed.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1834261/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1833034] Re: Error/warning/success messages remains unless page is refreshed

2019-06-26 Thread Ghada El-Zoghbi
A bug was introduced in this fix which does not allow the user to remove
an image from the image artefact.

Steps to reproduce:

- Add an image artefact to a page.
- Select an image and save
- Edit the image artefact on the page
- Try to remove the image

Expected result: image should be removed
What actually happens: image does not get removed due to an error in the 
console.

I'll add a patch.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833034

Title:
  Error/warning/success messages remains unless page is refreshed

Status in Mahara:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Committed
Status in Mahara 19.10 series:
  Fix Committed

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome (and others)

  In the Files upload screen (and other screens that display error
  message):

  When a user accidentally clicks on the 'Create folder' icon before
  entering a folder name, a red error message appears:

  "Please provide a name for the new folder."

  
  When the user enters the folder name and clicks the button again, the folder 
is created but the red error message still appears.

  The message is only removed after the user refreshes the page.

  This can be misleading to the user.

  Can you lease clear the message when the folder is created - and
  display a success message?

  
  Another scenario:

  * The user creates a folder successfully. 
  * The user then clicks into the new folder
  * Clicks the 'Create folder' button (in the child folder) without providing a 
name
  * The error message then appears
  * Clicks the up arrow to go to the parent folder.
  * The error message still appears - even though the user has moved out of the 
child folder.

  Expected, the error message should clear.

  
  Ideally, all alert messages in danger, warning, success classes should fade 
out after perhaps 10 seconds? This should be throughout the entire site not 
just the above screens.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1833034/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832805] Re: Missing lang string when saving to timeline

2019-06-26 Thread Ghada El-Zoghbi
A bug was introduced in this fix which does not allow the user to remove
an image from the image artefact.

Steps to reproduce:

- Add an image artefact to a page.
- Select an image and save
- Edit the image artefact on the page
- Try to remove the image

Expected result: image should be removed
What actually happens: image does not get removed due to an error in the 
console.

I'll add a patch.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832805

Title:
  Missing lang string when saving to timeline

Status in Mahara:
  Fix Committed
Status in Mahara 18.10 series:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Committed
Status in Mahara 19.10 series:
  Fix Committed

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  After a portfolio assignment has been unlocked (i.e. after grading)
  and was submitted via MNET, a student tries to save the
  page/collection in timeline, it display an error message.

  Which is the correct behaviour.

  The issues is a missing language string: [[cantversionview/view]] in
  the error message that is displayed to the user.

  
  Please see attached image.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832805/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832805] Re: Missing lang string when saving to timeline

2019-06-26 Thread Ghada El-Zoghbi
Sorry... note #10 is on the wrong bug.

Should be bug #1833034

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832805

Title:
  Missing lang string when saving to timeline

Status in Mahara:
  Fix Committed
Status in Mahara 18.10 series:
  Fix Committed
Status in Mahara 19.04 series:
  Fix Committed
Status in Mahara 19.10 series:
  Fix Committed

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  After a portfolio assignment has been unlocked (i.e. after grading)
  and was submitted via MNET, a student tries to save the
  page/collection in timeline, it display an error message.

  Which is the correct behaviour.

  The issues is a missing language string: [[cantversionview/view]] in
  the error message that is displayed to the user.

  
  Please see attached image.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832805/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1834261] Re: Reports generating error about time zon note specified

2019-06-25 Thread Ghada El-Zoghbi
** Description changed:

  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres 10.9
  Browser: FF
  
  An error is generated with running several reports but I'll give the
  example of the masquerading report:
  
  [WAR] ac (lib/errors.php:858) Failed to get a recordset: postgres8 error: 
[-1: ERROR: invalid input syntax for type timestamp with time zone: "Not 
specified"] in EXECUTE("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e ON 
e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 AND e.event = 'loginas' AND 
e.ctime >= DATE(?) AND e.ctime <= DATE(?)")Command was: SELECT COUNT(*) FROM 
"usr" u JOIN "event_log" e ON e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 
AND e.event = 'loginas' AND e.ctime >= DATE(?) AND e.ctime <= DATE(?) and 
values was (0:Notspecified,1:Notspecified)
  Call stack (most recent first):
  
  log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, true, 
true) at /var/www/mahara/htdocs/lib/errors.php:95
  log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:517
  get_recordset_sql("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/dml.php:276
  count_records_sql("SELECT COUNT(*) FROM {usr} u JOIN {event_log} e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/statistics.php:3535
  masquerading_stats_table(10, 0, array(size 5), null, 
https://my_dev_mahara/admin/users/statistics.php...;) at 
/var/www/mahara/htdocs/lib/statistics.php:3504
  masquerading_statistics(10, 0, array(size 5), null) at 
/var/www/mahara/htdocs/lib/statistics.php:4285
  display_statistics("all", "users", object(stdClass)) at 
/var/www/mahara/htdocs/admin/users/statistics.php:251
  
  To replicate:
  
  1. Turn on all event logging:  Admin > Configure site > Site options > 
Logging settings > Log events = All
- 2. Add User1 to Mahara
- 3. Masquerade User1 (ie do not log in as them bug masquerade)
- 4. Return back to admin user
- 5. Run the Masquerading report:  Admin > Reports > Configure report
- 6. Select:
+ 2. Set up Time zone in Admin > Configure site > Site options > Site settings 
> Time zone = "Australia/Sydney"
+ 3. Add User1 to Mahara
+ 4. Masquerade User1 (ie do not log in as them bug masquerade)
+ 5. Return back to admin user
+ 65. Run the Masquerading report:  Admin > Reports > Configure report
+ 7. Select:
  - Report Type = Masquerading sessions
  - Submit
  
  What happens: Above error is generated with:
  
  Masquerading sessions | All institutions
  Mahara: Site unavailable
  A nonrecoverable error occurred. This probably means you have encountered a 
bug in the system
  
  What should happen: the report should be displayed.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1834261

Title:
  Reports generating error about time zon note specified

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres 10.9
  Browser: FF

  An error is generated with running several reports but I'll give the
  example of the masquerading report:

  [WAR] ac (lib/errors.php:858) Failed to get a recordset: postgres8 error: 
[-1: ERROR: invalid input syntax for type timestamp with time zone: "Not 
specified"] in EXECUTE("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e ON 
e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 AND e.event = 'loginas' AND 
e.ctime >= DATE(?) AND e.ctime <= DATE(?)")Command was: SELECT COUNT(*) FROM 
"usr" u JOIN "event_log" e ON e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 
AND e.event = 'loginas' AND e.ctime >= DATE(?) AND e.ctime <= DATE(?) and 
values was (0:Notspecified,1:Notspecified)
  Call stack (most recent first):

  log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, true, 
true) at /var/www/mahara/htdocs/lib/errors.php:95
  log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:517
  get_recordset_sql("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/dml.php:276
  count_records_sql("SELECT COUNT(*) FROM {usr} u JOIN {event_log} e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/statistics.php:3535
  masquerading_stats_table(10, 0, array(size 5), null, 
https://my_dev_mahara/admin/users/statistics.php...;) at 
/var/www/mahara/htdocs/lib/statistics.php:3504
  masquerading_statistics(10, 0, array(size 5), null) at 
/var/www/mahara/htdocs/lib/statistics.php:4285
  display_statistics("all", "users", 

[Mahara-contributors] [Bug 1834261] [NEW] Reports generating error about time zon note specified

2019-06-25 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres 10.9
Browser: FF

An error is generated with running several reports but I'll give the
example of the masquerading report:

[WAR] ac (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR: invalid input syntax for type timestamp with time zone: "Not specified"] 
in EXECUTE("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e ON e.usr = u.id 
WHERE u.id != 0 AND u.deleted = 0 AND e.event = 'loginas' AND e.ctime >= 
DATE(?) AND e.ctime <= DATE(?)")Command was: SELECT COUNT(*) FROM "usr" u JOIN 
"event_log" e ON e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 AND e.event = 
'loginas' AND e.ctime >= DATE(?) AND e.ctime <= DATE(?) and values was 
(0:Notspecified,1:Notspecified)
Call stack (most recent first):

log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, true, 
true) at /var/www/mahara/htdocs/lib/errors.php:95
log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:517
get_recordset_sql("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/dml.php:276
count_records_sql("SELECT COUNT(*) FROM {usr} u JOIN {event_log} e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/statistics.php:3535
masquerading_stats_table(10, 0, array(size 5), null, 
https://my_dev_mahara/admin/users/statistics.php...;) at 
/var/www/mahara/htdocs/lib/statistics.php:3504
masquerading_statistics(10, 0, array(size 5), null) at 
/var/www/mahara/htdocs/lib/statistics.php:4285
display_statistics("all", "users", object(stdClass)) at 
/var/www/mahara/htdocs/admin/users/statistics.php:251

To replicate:

1. Turn on all event logging:  Admin > Configure site > Site options > Logging 
settings > Log events = All
2. Add User1 to Mahara
3. Masquerade User1 (ie do not log in as them bug masquerade)
4. Return back to admin user
5. Run the Masquerading report:  Admin > Reports > Configure report
6. Select:
- Report Type = Masquerading sessions
- Submit

What happens: Above error is generated with:

Masquerading sessions | All institutions
Mahara: Site unavailable
A nonrecoverable error occurred. This probably means you have encountered a bug 
in the system

What should happen: the report should be displayed.

** Affects: mahara
 Importance: Undecided
 Status: New

** Description changed:

  Mahara: 19.04.0
  OS: Linux 16.04
- DB: Postgres 10.9 
+ DB: Postgres 10.9
  Browser: FF
  
  An error is generated with running several reports but I'll give the
  example of the masquerading report:
- 
  
  [WAR] ac (lib/errors.php:858) Failed to get a recordset: postgres8 error: 
[-1: ERROR: invalid input syntax for type timestamp with time zone: "Not 
specified"] in EXECUTE("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e ON 
e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 AND e.event = 'loginas' AND 
e.ctime >= DATE(?) AND e.ctime <= DATE(?)")Command was: SELECT COUNT(*) FROM 
"usr" u JOIN "event_log" e ON e.usr = u.id WHERE u.id != 0 AND u.deleted = 0 
AND e.event = 'loginas' AND e.ctime >= DATE(?) AND e.ctime <= DATE(?) and 
values was (0:Notspecified,1:Notspecified)
  Call stack (most recent first):
  
  log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, true, 
true) at /var/www/mahara/htdocs/lib/errors.php:95
  log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:517
  get_recordset_sql("SELECT COUNT(*) FROM "usr" u JOIN "event_log" e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/dml.php:276
  count_records_sql("SELECT COUNT(*) FROM {usr} u JOIN {event_log} e O...", 
array(size 2)) at /var/www/mahara/htdocs/lib/statistics.php:3535
  masquerading_stats_table(10, 0, array(size 5), null, 
http://d-general-mahara/admin/users/statistics.php...;) at 
/var/www/mahara/htdocs/lib/statistics.php:3504
  masquerading_statistics(10, 0, array(size 5), null) at 
/var/www/mahara/htdocs/lib/statistics.php:4285
  display_statistics("all", "users", object(stdClass)) at 
/var/www/mahara/htdocs/admin/users/statistics.php:251
  
- 
- 
  To replicate:
  
  1. Turn on all event logging:  Admin > Configure site > Site options > 
Logging settings > Log events = All
  2. Add User1 to Mahara
  3. Masquerade User1 (ie do not log in as them bug masquerade)
  4. Return back to admin user
- 5. Run the Masquerading report:  Admin > Reports > Configure report 
+ 5. Run the Masquerading report:  Admin > Reports > Configure report
  6. Select:
- - Report Type = Masquerading sessins
+ - Report Type = Masquerading sessions
  - Submit
  
  What happens: Above error is generated with:
  
  Masquerading sessions | All institutions
  Mahara: Site 

[Mahara-contributors] [Bug 1833495] Re: Syncing LDAP usernames with Mahara's internal username leads to incorrect user record matching

2019-06-24 Thread Ghada El-Zoghbi
Please note that this issue arises when LDAP is the parent of an auth
that uses the auth_remote_user table.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833495

Title:
  Syncing LDAP usernames with Mahara's internal username leads to
  incorrect user record matching

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome (really n/a)

  The LDAP sync_users() functionality is not correctly matching user
  records - i.e. LDAP username with Mahara username.

  Scenario:

  1. Create an institution with and LDAP authentication
  - Update user info on login = Yes
  - We auto-create users = Yes
  - Sync users automatically via cron job = Yes
  - Update user info in cron = Yes
  - Auto-create users in cron = Yes
  - If a user is no longer present in LDAP = Suspend user's account

  2. Create an XML-RPC authentication with a Moodle for this institution
  - Parent authority = LDAP auth created above
  - SSO direction = Yes
  - Update user info on login = Yes
  - We auto-create users = Yes
  - We import content = Yes

  * The Moodle instance needs to have LDAP set up as well.

  3. Create an internal Mahara user (i.e. not LDAP)
  - username: AAA (make sure this username exists in the LDAP directory)

  * The username 'AAA' needs to be in Moodle created via LDAP.

  4. Login to Moodle with the username AAA - the LDAP user

  5. SSO to Mahara

  This will do the following
  - a new user record is created with username AAA1
  - auth_remote_user table will link Mahara user AAA1 to Moodle user AAA

  6. As Admin, update AAA1's auth method to LDAP.

  7. Logout of Mahara and Moodle.

  8. Run cron so LDAP syncs users

  Expected Results: user AAA1 to remain active in Mahara
  Actual Results: user AAA1 is suspended because it matches on username (i.e. 
AAA)

  That is, this join is correct:

  from {usr} u
  inner join {auth_ldap_extusers_temp} e on u.username = e.extusername

  The username in auth_ldap_extusers_temp is: AAA
  The LDAP username in Mahara is: AAA1

  Hence, it will not match on username and think the user is no longer
  in the LDAP directory and suspend the user.

  The join needs to be on another field that is present in both LDAP and
  Mahara.

  We should also join to the auth_remote_user table to get the correct
  external username.

  i.e.

  select
  u.id as id,
  aru.remoteusername as username,
  u.suspendedreason as suspendedreason,
  ...
  from {usr} u
  inner join {auth_remote_user} aru on u.id = aru.localusr
  inner join {auth_ldap_extusers_temp} e on aru.remoteusername = e.extusername

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1833495/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1833495] Re: Syncing LDAP usernames with Mahara's internal username leads to incorrect user record matching

2019-06-20 Thread Ghada El-Zoghbi
** Description changed:

  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome (really n/a)
  
  The LDAP sync_users() functionality is not correctly matching user
  records - i.e. LDAP username with Mahara username.
  
  Scenario:
  
  1. Create an institution with and LDAP authentication
  - Update user info on login = Yes
  - We auto-create users = Yes
  - Sync users automatically via cron job = Yes
  - Update user info in cron = Yes
  - Auto-create users in cron = Yes
  - If a user is no longer present in LDAP = Suspend user's account
  
  2. Create an XML-RPC authentication with a Moodle for this institution
  - Parent authority = LDAP auth created above
  - SSO direction = Yes
  - Update user info on login = Yes
  - We auto-create users = Yes
  - We import content = Yes
  
  * The Moodle instance needs to have LDAP set up as well.
  
  3. Create an internal Mahara user (i.e. not LDAP)
  - username: AAA (make sure this username exists in the LDAP directory)
  
  * The username 'AAA' needs to be in Moodle created via LDAP.
  
  4. Login to Moodle with the username AAA - the LDAP user
  
  5. SSO to Mahara
  
  This will do the following
  - a new user record is created with username AAA1
  - auth_remote_user table will link Mahara user AAA1 to Moodle user AAA
  
- 6. As Admin, update AAA1's auth metho to LDAP.
+ 6. As Admin, update AAA1's auth method to LDAP.
  
  7. Logout of Mahara and Moodle.
  
  8. Run cron so LDAP syncs users
  
  Expected Results: user AAA1 to remain active in Mahara
  Actual Results: user AAA1 is suspended because it matches on username (i.e. 
AAA)
  
  That is, this join is correct:
  
  from {usr} u
  inner join {auth_ldap_extusers_temp} e on u.username = e.extusername
  
  The username in auth_ldap_extusers_temp is: AAA
  The LDAP username in Mahara is: AAA1
  
  Hence, it will not match on username and think the user is no longer in
  the LDAP directory and suspend the user.
  
  The join needs to be on another field that is present in both LDAP and
  Mahara.
  
  We should also join to the auth_remote_user table to get the correct
  external username.
  
  i.e.
  
  select
  u.id as id,
  aru.remoteusername as username,
  u.suspendedreason as suspendedreason,
  ...
  from {usr} u
  inner join {auth_remote_user} aru on u.id = aru.localusr
  inner join {auth_ldap_extusers_temp} e on aru.remoteusername = e.extusername

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833495

Title:
  Syncing LDAP usernames with Mahara's internal username leads to
  incorrect user record matching

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome (really n/a)

  The LDAP sync_users() functionality is not correctly matching user
  records - i.e. LDAP username with Mahara username.

  Scenario:

  1. Create an institution with and LDAP authentication
  - Update user info on login = Yes
  - We auto-create users = Yes
  - Sync users automatically via cron job = Yes
  - Update user info in cron = Yes
  - Auto-create users in cron = Yes
  - If a user is no longer present in LDAP = Suspend user's account

  2. Create an XML-RPC authentication with a Moodle for this institution
  - Parent authority = LDAP auth created above
  - SSO direction = Yes
  - Update user info on login = Yes
  - We auto-create users = Yes
  - We import content = Yes

  * The Moodle instance needs to have LDAP set up as well.

  3. Create an internal Mahara user (i.e. not LDAP)
  - username: AAA (make sure this username exists in the LDAP directory)

  * The username 'AAA' needs to be in Moodle created via LDAP.

  4. Login to Moodle with the username AAA - the LDAP user

  5. SSO to Mahara

  This will do the following
  - a new user record is created with username AAA1
  - auth_remote_user table will link Mahara user AAA1 to Moodle user AAA

  6. As Admin, update AAA1's auth method to LDAP.

  7. Logout of Mahara and Moodle.

  8. Run cron so LDAP syncs users

  Expected Results: user AAA1 to remain active in Mahara
  Actual Results: user AAA1 is suspended because it matches on username (i.e. 
AAA)

  That is, this join is correct:

  from {usr} u
  inner join {auth_ldap_extusers_temp} e on u.username = e.extusername

  The username in auth_ldap_extusers_temp is: AAA
  The LDAP username in Mahara is: AAA1

  Hence, it will not match on username and think the user is no longer
  in the LDAP directory and suspend the user.

  The join needs to be on another field that is present in both LDAP and
  Mahara.

  We should also join to the auth_remote_user table to get the correct
  external username.

  i.e.

  select
  u.id as id,
  aru.remoteusername as username,
  u.suspendedreason as 

[Mahara-contributors] [Bug 1833495] Re: Syncing LDAP usernames with Mahara's internal username leads to incorrect user record matching

2019-06-20 Thread Ghada El-Zoghbi
** Description changed:

  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome (really n/a)
  
  The LDAP sync_users() functionality is not correctly matching user
  records - i.e. LDAP username with Mahara username.
  
  Scenario:
  
  1. Create an institution with and LDAP authentication
  - Update user info on login = Yes
  - We auto-create users = Yes
  - Sync users automatically via cron job = Yes
  - Update user info in cron = Yes
  - Auto-create users in cron = Yes
  - If a user is no longer present in LDAP = Suspend user's account
  
  2. Create an XML-RPC authentication with a Moodle for this institution
  - Parent authority = LDAP auth created above
  - SSO direction = Yes
  - Update user info on login = Yes
  - We auto-create users = Yes
  - We import content = Yes
  
  * The Moodle instance needs to have LDAP set up as well.
  
- 
  3. Create an internal Mahara user (i.e. not LDAP)
  - username: AAA (make sure this username exists in the LDAP directory)
- 
  
  * The username 'AAA' needs to be in Moodle created via LDAP.
  
  4. Login to Moodle with the username AAA - the LDAP user
  
  5. SSO to Mahara
  
  This will do the following
  - a new user record is created with username AAA1
  - auth_remote_user table will link Mahara user AAA1 to Moodle user AAA
  
- 6. Logout of Mahara and Moodle.
+ 6. As Admin, update AAA1's auth metho to LDAP.
  
- 7. Run cron so LDAP syncs users
+ 7. Logout of Mahara and Moodle.
+ 
+ 8. Run cron so LDAP syncs users
  
  Expected Results: user AAA1 to remain active in Mahara
  Actual Results: user AAA1 is suspended because it matches on username (i.e. 
AAA)
  
  That is, this join is correct:
  
  from {usr} u
  inner join {auth_ldap_extusers_temp} e on u.username = e.extusername
- 
  
  The username in auth_ldap_extusers_temp is: AAA
  The LDAP username in Mahara is: AAA1
  
  Hence, it will not match on username and think the user is no longer in
  the LDAP directory and suspend the user.
  
  The join needs to be on another field that is present in both LDAP and
  Mahara.
  
  We should also join to the auth_remote_user table to get the correct
  external username.
  
  i.e.
  
  select
  u.id as id,
  aru.remoteusername as username,
  u.suspendedreason as suspendedreason,
  ...
  from {usr} u
  inner join {auth_remote_user} aru on u.id = aru.localusr
  inner join {auth_ldap_extusers_temp} e on aru.remoteusername = e.extusername

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833495

Title:
  Syncing LDAP usernames with Mahara's internal username leads to
  incorrect user record matching

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome (really n/a)

  The LDAP sync_users() functionality is not correctly matching user
  records - i.e. LDAP username with Mahara username.

  Scenario:

  1. Create an institution with and LDAP authentication
  - Update user info on login = Yes
  - We auto-create users = Yes
  - Sync users automatically via cron job = Yes
  - Update user info in cron = Yes
  - Auto-create users in cron = Yes
  - If a user is no longer present in LDAP = Suspend user's account

  2. Create an XML-RPC authentication with a Moodle for this institution
  - Parent authority = LDAP auth created above
  - SSO direction = Yes
  - Update user info on login = Yes
  - We auto-create users = Yes
  - We import content = Yes

  * The Moodle instance needs to have LDAP set up as well.

  3. Create an internal Mahara user (i.e. not LDAP)
  - username: AAA (make sure this username exists in the LDAP directory)

  * The username 'AAA' needs to be in Moodle created via LDAP.

  4. Login to Moodle with the username AAA - the LDAP user

  5. SSO to Mahara

  This will do the following
  - a new user record is created with username AAA1
  - auth_remote_user table will link Mahara user AAA1 to Moodle user AAA

  6. As Admin, update AAA1's auth method to LDAP.

  7. Logout of Mahara and Moodle.

  8. Run cron so LDAP syncs users

  Expected Results: user AAA1 to remain active in Mahara
  Actual Results: user AAA1 is suspended because it matches on username (i.e. 
AAA)

  That is, this join is correct:

  from {usr} u
  inner join {auth_ldap_extusers_temp} e on u.username = e.extusername

  The username in auth_ldap_extusers_temp is: AAA
  The LDAP username in Mahara is: AAA1

  Hence, it will not match on username and think the user is no longer
  in the LDAP directory and suspend the user.

  The join needs to be on another field that is present in both LDAP and
  Mahara.

  We should also join to the auth_remote_user table to get the correct
  external username.

  i.e.

  select
  u.id as id,
  aru.remoteusername as username,
  

[Mahara-contributors] [Bug 1833495] [NEW] Syncing LDAP usernames with Mahara's internal username leads to incorrect user record matching

2019-06-19 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF / Chrome (really n/a)

The LDAP sync_users() functionality is not correctly matching user
records - i.e. LDAP username with Mahara username.

Scenario:

1. Create an institution with and LDAP authentication
- Update user info on login = Yes
- We auto-create users = Yes
- Sync users automatically via cron job = Yes
- Update user info in cron = Yes
- Auto-create users in cron = Yes
- If a user is no longer present in LDAP = Suspend user's account

2. Create an XML-RPC authentication with a Moodle for this institution
- Parent authority = LDAP auth created above
- SSO direction = Yes
- Update user info on login = Yes
- We auto-create users = Yes
- We import content = Yes

* The Moodle instance needs to have LDAP set up as well.


3. Create an internal Mahara user (i.e. not LDAP)
- username: AAA (make sure this username exists in the LDAP directory)


* The username 'AAA' needs to be in Moodle created via LDAP.

4. Login to Moodle with the username AAA - the LDAP user

5. SSO to Mahara

This will do the following
- a new user record is created with username AAA1
- auth_remote_user table will link Mahara user AAA1 to Moodle user AAA

6. Logout of Mahara and Moodle.

7. Run cron so LDAP syncs users

Expected Results: user AAA1 to remain active in Mahara
Actual Results: user AAA1 is suspended because it matches on username (i.e. AAA)

That is, this join is correct:

from {usr} u
inner join {auth_ldap_extusers_temp} e on u.username = e.extusername


The username in auth_ldap_extusers_temp is: AAA
The LDAP username in Mahara is: AAA1

Hence, it will not match on username and think the user is no longer in
the LDAP directory and suspend the user.

The join needs to be on another field that is present in both LDAP and
Mahara.

We should also join to the auth_remote_user table to get the correct
external username.

i.e.

select
u.id as id,
aru.remoteusername as username,
u.suspendedreason as suspendedreason,
...
from {usr} u
inner join {auth_remote_user} aru on u.id = aru.localusr
inner join {auth_ldap_extusers_temp} e on aru.remoteusername = e.extusername

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833495

Title:
  Syncing LDAP usernames with Mahara's internal username leads to
  incorrect user record matching

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF / Chrome (really n/a)

  The LDAP sync_users() functionality is not correctly matching user
  records - i.e. LDAP username with Mahara username.

  Scenario:

  1. Create an institution with and LDAP authentication
  - Update user info on login = Yes
  - We auto-create users = Yes
  - Sync users automatically via cron job = Yes
  - Update user info in cron = Yes
  - Auto-create users in cron = Yes
  - If a user is no longer present in LDAP = Suspend user's account

  2. Create an XML-RPC authentication with a Moodle for this institution
  - Parent authority = LDAP auth created above
  - SSO direction = Yes
  - Update user info on login = Yes
  - We auto-create users = Yes
  - We import content = Yes

  * The Moodle instance needs to have LDAP set up as well.

  
  3. Create an internal Mahara user (i.e. not LDAP)
  - username: AAA (make sure this username exists in the LDAP directory)

  
  * The username 'AAA' needs to be in Moodle created via LDAP.

  4. Login to Moodle with the username AAA - the LDAP user

  5. SSO to Mahara

  This will do the following
  - a new user record is created with username AAA1
  - auth_remote_user table will link Mahara user AAA1 to Moodle user AAA

  6. Logout of Mahara and Moodle.

  7. Run cron so LDAP syncs users

  Expected Results: user AAA1 to remain active in Mahara
  Actual Results: user AAA1 is suspended because it matches on username (i.e. 
AAA)

  That is, this join is correct:

  from {usr} u
  inner join {auth_ldap_extusers_temp} e on u.username = e.extusername

  
  The username in auth_ldap_extusers_temp is: AAA
  The LDAP username in Mahara is: AAA1

  Hence, it will not match on username and think the user is no longer
  in the LDAP directory and suspend the user.

  The join needs to be on another field that is present in both LDAP and
  Mahara.

  We should also join to the auth_remote_user table to get the correct
  external username.

  i.e.

  select
  u.id as id,
  aru.remoteusername as username,
  u.suspendedreason as suspendedreason,
  ...
  from {usr} u
  inner join {auth_remote_user} aru on u.id = aru.localusr
  inner join {auth_ldap_extusers_temp} e on aru.remoteusername = e.extusername

To manage notifications about this bug go to:

[Mahara-contributors] [Bug 1833339] [NEW] Bootstap datepicker supported formats

2019-06-18 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF and Chrome (and others)

The bootstrap date picker supports the following formats:

* https://bootstrap-
datepicker.readthedocs.io/en/latest/options.html#format

The date format, combination of d, dd, D, DD, m, mm, M, MM, yy, .

d, dd: Numeric date, no leading zero and leading zero, respectively. Eg, 5, 05.
D, DD: Abbreviated and full weekday names, respectively. Eg, Mon, Monday.
m, mm: Numeric month, no leading zero and leading zero, respectively. Eg, 7, 07.
M, MM: Abbreviated and full month names, respectively. Eg, Jan, January
yy, : 2- and 4-digit years, respectively. Eg, 12, 2012.

The Mahara lib/pieforms/pieform/elements/calendar.php file is not
supporting all these formats and is still supporting the jQuery UI
calendar.

The functions:

* pieform_element_calendar_convert_dateformat() and
* pieform_element_calendar_convert_timeformat()

Need to be updated to support the new date picker.

For example, this date format should be valid:

$string['pieform_calendar_dateformat'] = '%%d/%%mm/%%Y';

Instead, when updating the view access date/times, we get the error:

* Use the format DD/MMm/


Please note... this is not the only format to test for. Please test for other 
valid formats.

** Affects: mahara
 Importance: Undecided
 Status: New

** Description changed:

  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome (and others)
  
  The bootstrap date picker supports the following formats:
  
  * https://bootstrap-
  datepicker.readthedocs.io/en/latest/options.html#format
  
  The date format, combination of d, dd, D, DD, m, mm, M, MM, yy, .
  
  d, dd: Numeric date, no leading zero and leading zero, respectively. Eg, 5, 
05.
  D, DD: Abbreviated and full weekday names, respectively. Eg, Mon, Monday.
  m, mm: Numeric month, no leading zero and leading zero, respectively. Eg, 7, 
07.
  M, MM: Abbreviated and full month names, respectively. Eg, Jan, January
  yy, : 2- and 4-digit years, respectively. Eg, 12, 2012.
  
- 
- The Mahara lib/pieforms/pieform/elements/calendar.php file is not supporting 
all these formats and is still supporting the jQuery UI calendar.
+ The Mahara lib/pieforms/pieform/elements/calendar.php file is not
+ supporting all these formats and is still supporting the jQuery UI
+ calendar.
  
  The functions:
  
- * pieform_element_calendar_convert_dateformat() and 
+ * pieform_element_calendar_convert_dateformat() and
  * pieform_element_calendar_convert_timeformat()
  
  Need to be updated to support the new date picker.
  
  For example, this date format should be valid:
  
  $string['pieform_calendar_dateformat'] = '%%d/%%mm/%%Y';
  
  Instead, when updating the view access date/times, we get the error:
  
  * Use the format DD/MMm/
+ 
+ 
+ Please note... this is not the only format to test for. Please test for other 
valid formats.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/189

Title:
  Bootstap datepicker supported formats

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome (and others)

  The bootstrap date picker supports the following formats:

  * https://bootstrap-
  datepicker.readthedocs.io/en/latest/options.html#format

  The date format, combination of d, dd, D, DD, m, mm, M, MM, yy, .

  d, dd: Numeric date, no leading zero and leading zero, respectively. Eg, 5, 
05.
  D, DD: Abbreviated and full weekday names, respectively. Eg, Mon, Monday.
  m, mm: Numeric month, no leading zero and leading zero, respectively. Eg, 7, 
07.
  M, MM: Abbreviated and full month names, respectively. Eg, Jan, January
  yy, : 2- and 4-digit years, respectively. Eg, 12, 2012.

  The Mahara lib/pieforms/pieform/elements/calendar.php file is not
  supporting all these formats and is still supporting the jQuery UI
  calendar.

  The functions:

  * pieform_element_calendar_convert_dateformat() and
  * pieform_element_calendar_convert_timeformat()

  Need to be updated to support the new date picker.

  For example, this date format should be valid:

  $string['pieform_calendar_dateformat'] = '%%d/%%mm/%%Y';

  Instead, when updating the view access date/times, we get the error:

  * Use the format DD/MMm/

  
  Please note... this is not the only format to test for. Please test for other 
valid formats.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/189/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors

[Mahara-contributors] [Bug 1832655] Re: Click on datepicker icon does not open calendar window

2019-06-18 Thread Ghada El-Zoghbi
*** This bug is a duplicate of bug 1828150 ***
https://bugs.launchpad.net/bugs/1828150

** This bug has been marked a duplicate of bug 1828150
   clicking on calendar icon doesnt bring up the datepicker on calendar inputs

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832655

Title:
  Click on datepicker icon does not open calendar window

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  
  When the user clicks on the datepicker icon next to the text field, the 
calendar window does not open.

  It is the case in the following fields:

  • Plan: Completion date
  • Group Edit settings page for the Start and End dates for Editability.
  • Resume date of birth field
  • Admin: Institution settings - the Institution expiry date
  • Admin: Report configuration - From/To date

  I believe there may be other areas as well but these are what I have
  found so far.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832655/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1833325] [NEW] Navigation block title is not set for other pages in collection

2019-06-18 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF


The title of a navigation block added to a page in a collection does not set 
the navigation title on the other pages in the collection.

To reproduce:
1. Create 4 pages (Page 1, Page 2, Page 3, Page 4) 
2. Create a new collection with Name = "Collection of Pages"
3. Add the 4 pages to the collection.
4. On Page 3, add a navigation block with the following settings:
 - Block title = "My title for Page 3"
 - Collection = "Collection of Pages"
 - Add to all pages = Yes
 - Save
5. Display the collection.

Incorrect result: 
Pages 1, 2 and 4's navigation block title is: "Collection of Pages"
Page 3 navigation block title is: "My title for Page 3"

Expected result:
All navigation blocks in the collection should be titled: "My title for Page 3"

** Affects: mahara
 Importance: Undecided
 Assignee: Ghada El-Zoghbi (ghada-z)
 Status: New

** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833325

Title:
  Navigation block title is not set for other pages in collection

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  
  The title of a navigation block added to a page in a collection does not set 
the navigation title on the other pages in the collection.

  To reproduce:
  1. Create 4 pages (Page 1, Page 2, Page 3, Page 4) 
  2. Create a new collection with Name = "Collection of Pages"
  3. Add the 4 pages to the collection.
  4. On Page 3, add a navigation block with the following settings:
   - Block title = "My title for Page 3"
   - Collection = "Collection of Pages"
   - Add to all pages = Yes
   - Save
  5. Display the collection.

  Incorrect result: 
  Pages 1, 2 and 4's navigation block title is: "Collection of Pages"
  Page 3 navigation block title is: "My title for Page 3"

  Expected result:
  All navigation blocks in the collection should be titled: "My title for Page 
3"

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1833325/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1833034] [NEW] Error/warning/success messages remains unless page is refreshed

2019-06-17 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF and Chrome (and others)

In the Files upload screen (and other screens that display error
message):

When a user accidentally clicks on the 'Create folder' icon before
entering a folder name, a red error message appears:

"Please provide a name for the new folder."


When the user enters the folder name and clicks the button again, the folder is 
created but the red error message still appears.

The message is only removed after the user refreshes the page.

This can be misleading to the user.

Can you lease clear the message when the folder is created - and display
a success message?


Another scenario:

* The user creates a folder successfully. 
* The user then clicks into the new folder
* Clicks the 'Create folder' button (in the child folder) without providing a 
name
* The error message then appears
* Clicks the up arrow to go to the parent folder.
* The error message still appears - even though the user has moved out of the 
child folder.

Expected, the error message should clear.


Ideally, all alert messages in danger, warning, success classes should fade out 
after perhaps 10 seconds? This should be throughout the entire site not just 
the above screens.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1833034

Title:
  Error/warning/success messages remains unless page is refreshed

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome (and others)

  In the Files upload screen (and other screens that display error
  message):

  When a user accidentally clicks on the 'Create folder' icon before
  entering a folder name, a red error message appears:

  "Please provide a name for the new folder."

  
  When the user enters the folder name and clicks the button again, the folder 
is created but the red error message still appears.

  The message is only removed after the user refreshes the page.

  This can be misleading to the user.

  Can you lease clear the message when the folder is created - and
  display a success message?

  
  Another scenario:

  * The user creates a folder successfully. 
  * The user then clicks into the new folder
  * Clicks the 'Create folder' button (in the child folder) without providing a 
name
  * The error message then appears
  * Clicks the up arrow to go to the parent folder.
  * The error message still appears - even though the user has moved out of the 
child folder.

  Expected, the error message should clear.

  
  Ideally, all alert messages in danger, warning, success classes should fade 
out after perhaps 10 seconds? This should be throughout the entire site not 
just the above screens.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1833034/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832805] Re: Missing lang string when saving to timeline

2019-06-16 Thread Ghada El-Zoghbi
There are 4 different scenarios:

1. The page id passed to the form is invalid

2. The user does not have permission to edit the page (i.e. they are not
the owner for example or it's an institution view and they are not an
institution admin/staff).

3. The page has been submitted for assessment.

4. This page is for a group and the editable window dates has passed.

I will create 4 separate lang strings for each scenario.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832805

Title:
  Missing lang string when saving to timeline

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  After a portfolio assignment has been unlocked (i.e. after grading)
  and was submitted via MNET, a student tries to save the
  page/collection in timeline, it display an error message.

  Which is the correct behaviour.

  The issues is a missing language string: [[cantversionview/view]] in
  the error message that is displayed to the user.

  
  Please see attached image.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832805/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832805] Re: Missing lang string when saving to timeline

2019-06-14 Thread Ghada El-Zoghbi
Please let me know what the message should say and I can upload a patch.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832805

Title:
  Missing lang string when saving to timeline

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  After a portfolio assignment has been unlocked (i.e. after grading)
  and was submitted via MNET, a student tries to save the
  page/collection in timeline, it display an error message.

  Which is the correct behaviour.

  The issues is a missing language string: [[cantversionview/view]] in
  the error message that is displayed to the user.

  
  Please see attached image.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832805/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832805] [NEW] Missing lang string when saving to timeline

2019-06-13 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF

After a portfolio assignment has been unlocked (i.e. after grading) and
was submitted via MNET, a student tries to save the page/collection in
timeline, it display an error message.

Which is the correct behaviour.

The issues is a missing language string: [[cantversionview/view]] in the
error message that is displayed to the user.


Please see attached image.

** Affects: mahara
 Importance: Undecided
 Status: New

** Attachment added: "saving to timeline after locking.jpg"
   
https://bugs.launchpad.net/bugs/1832805/+attachment/5270706/+files/saving%20to%20timeline%20after%20locking.jpg

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832805

Title:
  Missing lang string when saving to timeline

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  After a portfolio assignment has been unlocked (i.e. after grading)
  and was submitted via MNET, a student tries to save the
  page/collection in timeline, it display an error message.

  Which is the correct behaviour.

  The issues is a missing language string: [[cantversionview/view]] in
  the error message that is displayed to the user.

  
  Please see attached image.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832805/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832796] Re: Adding Journal - message displayed to user is misleading

2019-06-13 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832796

Title:
  Adding Journal - message displayed to user is misleading

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  The message displayed to a group member who is not allowed to enter a
  journal is misleading. They are given a link to 'Add one' but when
  they click on it, it displays they don't have permission to do so.

  To reproduce:
  * Create an institution: Inst1

  * Add users to Inst1 - user1, admin1

  * Create a group for Inst1: 
   Associated with institution: Inst1
   Open = No
   Controlled = Yes
   Request = No
   Friend Invitations = Yes
   Create and edit: everyone except ordinary members
   Publicly viewable group = No
   Hide group = Yes
   Hide membership = Hide members
   Hide membership from members = No
   Auto-add users = No
   Shared page notifications = All group members
   Comment notifications = All group members

  * Add user1 with role = member to the group

  * Add admin1 with role admin

  * Log in as user1 and go to group > Journals

  * Display message says: There are no journals in this group. Add one.

  * Click on 'Add one' link

  * User is presented with message they don't have permission.

  * Expected: They should not be presented with the link to click 'Add
  one'.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832796/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832803] [NEW] Adding an empty string as a comment generates a warning

2019-06-13 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF


When entering a comment which does not have any text but includes an uploaded 
file, the following warning is generated:

[WAR] bd (artefact/comment/lib.php:1184) DOMDocument::loadHTML(): Empty string 
supplied as input
Call stack (most recent first):

log_message("DOMDocument::loadHTML(): Empty string supplied as ...", 8, 
true, true, "/var/www/mahara/htdocs/artefact/commen...", 1184) at 
/var/www/mahara/htdocs/lib/errors.php:520
error(2, "DOMDocument::loadHTML(): Empty string supplied as ...", 
"/var/www/mahara/htdocs/artefact/commen...", 1184, array(size 2)) at Unknown:0
DOMDocument->loadHTML("", 8196) at 
/var/www/mahara/htdocs/artefact/comment/lib.php:1184
ArtefactTypeComment::remove_comments_classes("") at 
/var/www/mahara/htdocs/artefact/comment/lib.php:1637
add_feedback_form_submit(object(Pieform), array(size 8)) at 
/var/www/mahara/htdocs/lib/pieforms/pieform.php:544
Pieform->__construct(array(size 9)) at 
/var/www/mahara/htdocs/lib/pieforms/pieform.php:166
Pieform::process(array(size 9)) at 
/var/www/mahara/htdocs/lib/mahara.php:5400
pieform(array(size 9)) at /var/www/mahara/htdocs/view/view.php:266


To reproduce:
* Create a view with a comment section
* View the page and enter a comment
* Upload a file for the comment and leave the content empty.
* Save
* Refresh page
* The above warning appears

Expected: no warning should appear.

** Affects: mahara
 Importance: Undecided
 Assignee: Ghada El-Zoghbi (ghada-z)
 Status: New

** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832803

Title:
  Adding an empty string as a comment generates a warning

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF

  
  When entering a comment which does not have any text but includes an uploaded 
file, the following warning is generated:

  [WAR] bd (artefact/comment/lib.php:1184) DOMDocument::loadHTML(): Empty 
string supplied as input
  Call stack (most recent first):

  log_message("DOMDocument::loadHTML(): Empty string supplied as ...", 8, 
true, true, "/var/www/mahara/htdocs/artefact/commen...", 1184) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(2, "DOMDocument::loadHTML(): Empty string supplied as ...", 
"/var/www/mahara/htdocs/artefact/commen...", 1184, array(size 2)) at Unknown:0
  DOMDocument->loadHTML("", 8196) at 
/var/www/mahara/htdocs/artefact/comment/lib.php:1184
  ArtefactTypeComment::remove_comments_classes("") at 
/var/www/mahara/htdocs/artefact/comment/lib.php:1637
  add_feedback_form_submit(object(Pieform), array(size 8)) at 
/var/www/mahara/htdocs/lib/pieforms/pieform.php:544
  Pieform->__construct(array(size 9)) at 
/var/www/mahara/htdocs/lib/pieforms/pieform.php:166
  Pieform::process(array(size 9)) at 
/var/www/mahara/htdocs/lib/mahara.php:5400
  pieform(array(size 9)) at /var/www/mahara/htdocs/view/view.php:266

  
  To reproduce:
  * Create a view with a comment section
  * View the page and enter a comment
  * Upload a file for the comment and leave the content empty.
  * Save
  * Refresh page
  * The above warning appears

  Expected: no warning should appear.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832803/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832796] [NEW] Adding Journal - message displayed to user is misleading

2019-06-13 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF and Chrome

The message displayed to a group member who is not allowed to enter a
journal is misleading. They are given a link to 'Add one' but when they
click on it, it displays they don't have permission to do so.

To reproduce:
* Create an institution: Inst1

* Add users to Inst1 - user1, admin1

* Create a group for Inst1: 
 Associated with institution: Inst1
 Open = No
 Controlled = Yes
 Request = No
 Friend Invitations = Yes
 Create and edit: everyone except ordinary members
 Publicly viewable group = No
 Hide group = Yes
 Hide membership = Hide members
 Hide membership from members = No
 Auto-add users = No
 Shared page notifications = All group members
 Comment notifications = All group members

* Add user1 with role = member to the group

* Add admin1 with role admin

* Log in as user1 and go to group > Journals

* Display message says: There are no journals in this group. Add one.

* Click on 'Add one' link

* User is presented with message they don't have permission.

* Expected: They should not be presented with the link to click 'Add
one'.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832796

Title:
  Adding Journal - message displayed to user is misleading

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  The message displayed to a group member who is not allowed to enter a
  journal is misleading. They are given a link to 'Add one' but when
  they click on it, it displays they don't have permission to do so.

  To reproduce:
  * Create an institution: Inst1

  * Add users to Inst1 - user1, admin1

  * Create a group for Inst1: 
   Associated with institution: Inst1
   Open = No
   Controlled = Yes
   Request = No
   Friend Invitations = Yes
   Create and edit: everyone except ordinary members
   Publicly viewable group = No
   Hide group = Yes
   Hide membership = Hide members
   Hide membership from members = No
   Auto-add users = No
   Shared page notifications = All group members
   Comment notifications = All group members

  * Add user1 with role = member to the group

  * Add admin1 with role admin

  * Log in as user1 and go to group > Journals

  * Display message says: There are no journals in this group. Add one.

  * Click on 'Add one' link

  * User is presented with message they don't have permission.

  * Expected: They should not be presented with the link to click 'Add
  one'.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832796/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832655] [NEW] Click on datepicker icon does not open calendar window

2019-06-12 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: FF and Chrome


When the user clicks on the datepicker icon next to the text field, the 
calendar window does not open.

It is the case in the following fields:

• Plan: Completion date
• Group Edit settings page for the Start and End dates for Editability.
• Resume date of birth field
• Admin: Institution settings - the Institution expiry date
• Admin: Report configuration - From/To date

I believe there may be other areas as well but these are what I have
found so far.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832655

Title:
  Click on datepicker icon does not open calendar window

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: FF and Chrome

  
  When the user clicks on the datepicker icon next to the text field, the 
calendar window does not open.

  It is the case in the following fields:

  • Plan: Completion date
  • Group Edit settings page for the Start and End dates for Editability.
  • Resume date of birth field
  • Admin: Institution settings - the Institution expiry date
  • Admin: Report configuration - From/To date

  I believe there may be other areas as well but these are what I have
  found so far.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832655/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1832308] [NEW] Cannot toggle Group's Open button

2019-06-11 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux 16.04
DB: Postgres
Browser: Firefox


* Prerequiste: create an institution (Inst1) other than Mahara.

* Created a group with the following settings:

Group name = 1904 Upgrade Test G2
Associate with Institution = Inst1
Group homepage URL = 1904-upgrade-test-g2
Open = No
Controlled = Yes
Request = Yes
Roles: Course: Member, Tutor, Admin
Friend Invitations: No
Recommendations = No
Create and edit = All group members
Allow submissions = Yes
Allow archiving of submissions = Yes
Publicly viewable group = Yes
Hide group = No
Hide membership = Hide members
Hide membership from members = No
Participation report = Yes
Auto-add users = No
Shared page notifications = All group members
Comment notifications = All group members
Send forum posts immediately = Yes


* Save the group settings. 

* Come back into the group settings page, the Open button is disabled -
i.e. click on Controlled = No - should toggle the Open button on.

Current Behaviour: Open button remains disabled
Expected Behavious: Open button should be enabled.

** Affects: mahara
 Importance: Undecided
 Assignee: Ghada El-Zoghbi (ghada-z)
 Status: New

** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1832308

Title:
  Cannot toggle Group's Open button

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux 16.04
  DB: Postgres
  Browser: Firefox

  
  * Prerequiste: create an institution (Inst1) other than Mahara.

  * Created a group with the following settings:

  Group name = 1904 Upgrade Test G2
  Associate with Institution = Inst1
  Group homepage URL = 1904-upgrade-test-g2
  Open = No
  Controlled = Yes
  Request = Yes
  Roles: Course: Member, Tutor, Admin
  Friend Invitations: No
  Recommendations = No
  Create and edit = All group members
  Allow submissions = Yes
  Allow archiving of submissions = Yes
  Publicly viewable group = Yes
  Hide group = No
  Hide membership = Hide members
  Hide membership from members = No
  Participation report = Yes
  Auto-add users = No
  Shared page notifications = All group members
  Comment notifications = All group members
  Send forum posts immediately = Yes

  
  * Save the group settings. 

  * Come back into the group settings page, the Open button is disabled
  - i.e. click on Controlled = No - should toggle the Open button on.

  Current Behaviour: Open button remains disabled
  Expected Behavious: Open button should be enabled.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1832308/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831835] [NEW] Site Admin should be able to send messages to anyone

2019-06-05 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux
DB: Postgres 10.8
Browser: n/a


Since the beginning of Mahara, site admins have always been able to send a 
message to any user in the system regardless of the user's notification 
settings.

With the introduction of isolated institutions (in 19.04), this
capability has been removed.

It should be re-implemented.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831835

Title:
  Site Admin should be able to send messages to anyone

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux
  DB: Postgres 10.8
  Browser: n/a

  
  Since the beginning of Mahara, site admins have always been able to send a 
message to any user in the system regardless of the user's notification 
settings.

  With the introduction of isolated institutions (in 19.04), this
  capability has been removed.

  It should be re-implemented.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831835/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831419] Re: Notification when Importing Leap2A

2019-06-03 Thread Ghada El-Zoghbi
The other items to confirm:

1. When a page is shared with a group and a comment is inserted on the
page, the event log has recorded this fact: event =
sharedcommenttogroup.

2. When entering a comment on a standard page, an event is generated in
the event log.

3. When entering a comment on a group page (not shared), an event is
generated in the event log.

4. And of course, the warning message no longer appears when importing a
page with a comment.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831419

Title:
  Notification when Importing Leap2A

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.04
  DB: Postgres
  OS: Linux
  Browser: FF

  When importing a view with a comment block, the following warning is
  generated:

  
  [WAR] ed (lib/mahara.php:2056) Undefined index: onview
  Call stack (most recent first):

  
  log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
  handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
  BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
  View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
  View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
  PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
  PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
  PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
  do_import() at /var/www/mahara/htdocs/import/index.php:71

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831419/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831419] Re: Notification when Importing Leap2A

2019-06-03 Thread Ghada El-Zoghbi
This bug was introduced in Bug 1692385 - commit:
a3d904e16bb0dbf8621842dafdd0feb9c3c0f9b8

and a small fix in bug 1724113 - commit
a8f8a82d8250d7e3f447862c3e48371765f22a9f

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831419

Title:
  Notification when Importing Leap2A

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.04
  DB: Postgres
  OS: Linux
  Browser: FF

  When importing a view with a comment block, the following warning is
  generated:

  
  [WAR] ed (lib/mahara.php:2056) Undefined index: onview
  Call stack (most recent first):

  
  log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
  handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
  BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
  View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
  View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
  PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
  PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
  PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
  do_import() at /var/www/mahara/htdocs/import/index.php:71

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831419/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831419] Re: Notification when Importing Leap2A

2019-06-03 Thread Ghada El-Zoghbi
Please ignore the steps above. These are the steps to reproduce (logging
settings must be set to All).


Steps to reproduce:

1. Enable logging level to 'All' (admin > configure site > site options > 
Logging settings = 'All')
2. Create a view with a text block and a comment block
3. Export that one view to a Leap2A file
4. Import just this view to Mahara.


** Changed in: mahara
   Status: Invalid => New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831419

Title:
  Notification when Importing Leap2A

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.04
  DB: Postgres
  OS: Linux
  Browser: FF

  When importing a view with a comment block, the following warning is
  generated:

  
  [WAR] ed (lib/mahara.php:2056) Undefined index: onview
  Call stack (most recent first):

  
  log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
  handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
  BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
  View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
  View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
  PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
  PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
  PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
  do_import() at /var/www/mahara/htdocs/import/index.php:71

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831419/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831419] Re: Notification when Importing Leap2A

2019-06-03 Thread Ghada El-Zoghbi
** Changed in: mahara
   Status: New => Invalid

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831419

Title:
  Notification when Importing Leap2A

Status in Mahara:
  Invalid

Bug description:
  Mahara: 19.04.04
  DB: Postgres
  OS: Linux
  Browser: FF

  When importing a view with a comment block, the following warning is
  generated:

  
  [WAR] ed (lib/mahara.php:2056) Undefined index: onview
  Call stack (most recent first):

  
  log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
  handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
  BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
  View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
  View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
  PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
  PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
  PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
  do_import() at /var/www/mahara/htdocs/import/index.php:71

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831419/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831419] Re: Notification when Importing Leap2A

2019-06-03 Thread Ghada El-Zoghbi
Steps to reproduce:


1. Create a view with a text block and a comment block
2. Export that one view to a Leap2A file
3. Import just this view to Mahara.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831419

Title:
  Notification when Importing Leap2A

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.04
  DB: Postgres
  OS: Linux
  Browser: FF

  When importing a view with a comment block, the following warning is
  generated:

  
  [WAR] ed (lib/mahara.php:2056) Undefined index: onview
  Call stack (most recent first):

  
  log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
  handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
  BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
  View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
  View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
  PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
  PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
  PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
  do_import() at /var/www/mahara/htdocs/import/index.php:71

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831419/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831419] [NEW] Notification when Importing Leap2A

2019-06-03 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.04
DB: Postgres
OS: Linux
Browser: FF

When importing a view with a comment block, the following warning is
generated:


[WAR] ed (lib/mahara.php:2056) Undefined index: onview
Call stack (most recent first):


log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
do_import() at /var/www/mahara/htdocs/import/index.php:71

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831419

Title:
  Notification when Importing Leap2A

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.04
  DB: Postgres
  OS: Linux
  Browser: FF

  When importing a view with a comment block, the following warning is
  generated:

  
  [WAR] ed (lib/mahara.php:2056) Undefined index: onview
  Call stack (most recent first):

  
  log_message("Undefined index: onview", 8, true, true, 
"/var/www/mahara/htdocs/lib/mahara.php", 2056) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Undefined index: onview", 
"/var/www/mahara/htdocs/lib/mahara.php", 2056, array(size 23)) at 
/var/www/mahara/htdocs/lib/mahara.php:2056
  handle_event("blockinstancecommit", object(BlockInstance)) at 
/var/www/mahara/htdocs/blocktype/lib.php:1480
  BlockInstance->commit() at /var/www/mahara/htdocs/lib/view.php:2463
  View->addblockinstance(object(BlockInstance)) at 
/var/www/mahara/htdocs/lib/view.php:3400
  View::import_from_config(array(size 12), "2", "leap") at 
/var/www/mahara/htdocs/import/leap/lib.php:1036
  PluginImportLeap->import_view_from_request(object(stdClass)) at 
/var/www/mahara/htdocs/import/leap/lib.php:1005
  PluginImportLeap->import_from_requests() at 
/var/www/mahara/htdocs/import/leap/lib.php:348
  PluginImportLeap->do_import_from_requests() at 
/var/www/mahara/htdocs/import/index.php:314
  do_import() at /var/www/mahara/htdocs/import/index.php:71

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831419/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831412] [NEW] Can't delete internal notifications

2019-06-02 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0 (and previous versions)
OS: Linux
DB: Postgres
Browser: FF (and Chrome)


When going to "Sent" area and clicking "Delete all notifications", it says 
internal notifications can't be deleted for that area. Please see attached.

If the only type of messages left are internal notifications, the option
to "Delete all notifications" should not appear since they can't be
deleted.

This is causing confusion to our users.

** Affects: mahara
 Importance: Undecided
 Status: New

** Attachment added: "delete_all_notifications.jpg"
   
https://bugs.launchpad.net/bugs/1831412/+attachment/5268461/+files/delete_all_notifications.jpg

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831412

Title:
  Can't delete internal notifications

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0 (and previous versions)
  OS: Linux
  DB: Postgres
  Browser: FF (and Chrome)

  
  When going to "Sent" area and clicking "Delete all notifications", it says 
internal notifications can't be deleted for that area. Please see attached.

  If the only type of messages left are internal notifications, the
  option to "Delete all notifications" should not appear since they
  can't be deleted.

  This is causing confusion to our users.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831412/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1831182] [NEW] Use display names in Plugin Administration page

2019-05-30 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux (n/a)
DB: Postgres (n/a)
Browser: FF (n/a)


When viewing the extensions in the Plugin administration, it would be nice to 
use the display name with the "dev" name. 

Some of the dev names are a bit obscure and it's a stretch to link them
to their display names.

i.e.

* newviews  -> Latest changes I can view
* multirecipientnotification -> Notifications

Perhaps list the plugins with the display names and the dev name in
parentheses so the technical people will also be able to identify them:

* Notifications (multirecipientnotification)
* Latest changes I can view (newviews)

** Affects: mahara
 Importance: Wishlist
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1831182

Title:
  Use display names in Plugin Administration page

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux (n/a)
  DB: Postgres (n/a)
  Browser: FF (n/a)

  
  When viewing the extensions in the Plugin administration, it would be nice to 
use the display name with the "dev" name. 

  Some of the dev names are a bit obscure and it's a stretch to link
  them to their display names.

  i.e.

  * newviews  -> Latest changes I can view
  * multirecipientnotification -> Notifications

  Perhaps list the plugins with the display names and the dev name in
  parentheses so the technical people will also be able to identify
  them:

  * Notifications (multirecipientnotification)
  * Latest changes I can view (newviews)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1831182/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1830302] Re: Messages Sending ignoring user preference

2019-05-23 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1830302

Title:
  Messages Sending ignoring user preference

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux
  DB: Postgres
  Browser: FF

  There is a bug that was introduced with the isolated institution. It
  is ignoring the user message preferences.

  Scenario:

  * User1: set User > Settings > Preferences - 'Messages from other
  users' = 'Allow people on my friends list to send me messages'

  * User2: set User > Settings > Preferences - 'Messages from other
  users' = 'Allow people on my friends list to send me messages'

  * Neither User1 nor User2 are friends of each other

  * The site is not set up as isolated institution
  ($cfg->isolatedinstitutions = false;)

  
  Log in as User1 and attempt to create a message to User1. Mahara allows the 
message to be sent.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1830302/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1830302] [NEW] Messages Sending ignoring user preference

2019-05-23 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux
DB: Postgres
Browser: FF

There is a bug that was introduced with the isolated institution. It is
ignoring the user message preferences.

Scenario:

* User1: set User > Settings > Preferences - 'Messages from other users'
= 'Allow people on my friends list to send me messages'

* User2: set User > Settings > Preferences - 'Messages from other users'
= 'Allow people on my friends list to send me messages'

* Neither User1 nor User2 are friends of each other

* The site is not set up as isolated institution
($cfg->isolatedinstitutions = false;)


Log in as User1 and attempt to create a message to User1. Mahara allows the 
message to be sent.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1830302

Title:
  Messages Sending ignoring user preference

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux
  DB: Postgres
  Browser: FF

  There is a bug that was introduced with the isolated institution. It
  is ignoring the user message preferences.

  Scenario:

  * User1: set User > Settings > Preferences - 'Messages from other
  users' = 'Allow people on my friends list to send me messages'

  * User2: set User > Settings > Preferences - 'Messages from other
  users' = 'Allow people on my friends list to send me messages'

  * Neither User1 nor User2 are friends of each other

  * The site is not set up as isolated institution
  ($cfg->isolatedinstitutions = false;)

  
  Log in as User1 and attempt to create a message to User1. Mahara allows the 
message to be sent.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1830302/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1829943] Re: Error Deleting Instution Due to Foreign Key Constraint with View Table

2019-05-21 Thread Ghada El-Zoghbi
Steps to reproduce:

1. Create an institution with xml-rpc integration with a Moodle. Allow
submissions from Mahara to Moodle.

2. Add users to the new institution in Mahara.

3. In Moodle, create a course with an activity - Mahara submission. Make
sure the Mahara user is enrolled in this course.

4. As a user of the institution, create a page in Mahara and submit the
page to Moodle.

5. As an admin of Mahara, remove all users from the institution. This
will enable the 'delete' button for the institution.

6. Try to delete the institution. You should receive the error above.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1829943

Title:
  Error Deleting Instution Due to Foreign Key Constraint with View Table

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux
  DB: Postgres
  Browser: FF

  We have an institution which no longer has any users. When trying to
  delete it, it fails due to the foreign key constraint with the view
  table - which now has the institution field.


  Call stack (most recent first):

  log_message("Array to string conversion", 8, true, true, 
"/var/www/mahara/htdocs/lib/adodb/adodb...", 45) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Array to string conversion", 
"/var/www/mahara/htdocs/lib/adodb/adodb...", 45, array(size 7)) at 
/var/www/mahara/htdocs/lib/adodb/adodb-exceptions.inc.php:45
  ADODB_Exception->__construct("postgres8", "adodb_throw", -1, "ERROR: 
update or delete on table "host" violates ...", "DELETE FROM "host" WHERE 
"institution" = ? ", array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb-exceptions.inc.php:80
  adodb_throw("postgres8", "adodb_throw", -1, "ERROR: update or delete on 
table "host" violates ...", "DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb.inc.php:310
  ADODB_TransMonitor("postgres8", "adodb_throw", -1, "ERROR: update or 
delete on table "host" violates ...", "DELETE FROM "host" WHERE "institution" = 
? ", array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb.inc.php:1281
  ADOConnection->_Execute("DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1)) at /var/www/mahara/htdocs/lib/adodb/adodb.inc.php:1241
  ADOConnection->Execute("DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:981
  delete_records("host", "institution", "mytest") at 
/var/www/mahara/htdocs/admin/users/institutions.php:175
  delete_submit(object(Pieform), array(size 4)) at 
/var/www/mahara/htdocs/lib/pieforms/pieform.php:544
  Pieform->__construct(array(size 2)) at 
/var/www/clients/htdocs/lib/pieforms/pieform.php:166
  Pieform::process(array(size 2)) at 
/var/www/clients/htdocs/lib/mahara.php:5379
  pieform(array(size 2)) at 
/var/www/mahara/htdocs/admin/users/institutions.php:217

  [WAR] 55 (lib/errors.php:858) Failed to get a recordset: postgres8 error: 
[-1: ERROR:  update or delete on table "host" violates foreign key constraint 
"view_sub2_fk" on table "view"
  DETAIL:  Key (wwwroot)=(https://mytest.com.au) is still referenced from table 
"view".] in adodb_throw(DELETE FROM "host"  WHERE "institution" = ? , 
Array)Command was: DELETE FROM "host"  WHERE "institution" = ?  and values was 
(0:mytest)


  There are records in the view table that have submitted pages to this
  institution. It's setup with XML-RPC with a Moodle.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1829943/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1829943] [NEW] Error Deleting Instution Due to Foreign Key Constraint with View Table

2019-05-21 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 19.04.0
OS: Linux
DB: Postgres
Browser: FF

We have an institution which no longer has any users. When trying to
delete it, it fails due to the foreign key constraint with the view
table - which now has the institution field.


Call stack (most recent first):

log_message("Array to string conversion", 8, true, true, 
"/var/www/mahara/htdocs/lib/adodb/adodb...", 45) at 
/var/www/mahara/htdocs/lib/errors.php:520
error(8, "Array to string conversion", 
"/var/www/mahara/htdocs/lib/adodb/adodb...", 45, array(size 7)) at 
/var/www/mahara/htdocs/lib/adodb/adodb-exceptions.inc.php:45
ADODB_Exception->__construct("postgres8", "adodb_throw", -1, "ERROR: update 
or delete on table "host" violates ...", "DELETE FROM "host" WHERE 
"institution" = ? ", array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb-exceptions.inc.php:80
adodb_throw("postgres8", "adodb_throw", -1, "ERROR: update or delete on 
table "host" violates ...", "DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb.inc.php:310
ADODB_TransMonitor("postgres8", "adodb_throw", -1, "ERROR: update or delete 
on table "host" violates ...", "DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb.inc.php:1281
ADOConnection->_Execute("DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1)) at /var/www/mahara/htdocs/lib/adodb/adodb.inc.php:1241
ADOConnection->Execute("DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:981
delete_records("host", "institution", "mytest") at 
/var/www/mahara/htdocs/admin/users/institutions.php:175
delete_submit(object(Pieform), array(size 4)) at 
/var/www/mahara/htdocs/lib/pieforms/pieform.php:544
Pieform->__construct(array(size 2)) at 
/var/www/clients/htdocs/lib/pieforms/pieform.php:166
Pieform::process(array(size 2)) at 
/var/www/clients/htdocs/lib/mahara.php:5379
pieform(array(size 2)) at 
/var/www/mahara/htdocs/admin/users/institutions.php:217

[WAR] 55 (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  update or delete on table "host" violates foreign key constraint 
"view_sub2_fk" on table "view"
DETAIL:  Key (wwwroot)=(https://mytest.com.au) is still referenced from table 
"view".] in adodb_throw(DELETE FROM "host"  WHERE "institution" = ? , 
Array)Command was: DELETE FROM "host"  WHERE "institution" = ?  and values was 
(0:mytest)


There are records in the view table that have submitted pages to this
institution. It's setup with XML-RPC with a Moodle.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1829943

Title:
  Error Deleting Instution Due to Foreign Key Constraint with View Table

Status in Mahara:
  New

Bug description:
  Mahara: 19.04.0
  OS: Linux
  DB: Postgres
  Browser: FF

  We have an institution which no longer has any users. When trying to
  delete it, it fails due to the foreign key constraint with the view
  table - which now has the institution field.


  Call stack (most recent first):

  log_message("Array to string conversion", 8, true, true, 
"/var/www/mahara/htdocs/lib/adodb/adodb...", 45) at 
/var/www/mahara/htdocs/lib/errors.php:520
  error(8, "Array to string conversion", 
"/var/www/mahara/htdocs/lib/adodb/adodb...", 45, array(size 7)) at 
/var/www/mahara/htdocs/lib/adodb/adodb-exceptions.inc.php:45
  ADODB_Exception->__construct("postgres8", "adodb_throw", -1, "ERROR: 
update or delete on table "host" violates ...", "DELETE FROM "host" WHERE 
"institution" = ? ", array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb-exceptions.inc.php:80
  adodb_throw("postgres8", "adodb_throw", -1, "ERROR: update or delete on 
table "host" violates ...", "DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb.inc.php:310
  ADODB_TransMonitor("postgres8", "adodb_throw", -1, "ERROR: update or 
delete on table "host" violates ...", "DELETE FROM "host" WHERE "institution" = 
? ", array(size 1), object(ADODB_postgres8)) at 
/var/www/mahara/htdocs/lib/adodb/adodb.inc.php:1281
  ADOConnection->_Execute("DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1)) at /var/www/mahara/htdocs/lib/adodb/adodb.inc.php:1241
  ADOConnection->Execute("DELETE FROM "host" WHERE "institution" = ? ", 
array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:981
  delete_records("host", "institution", "mytest") at 

[Mahara-contributors] [Bug 1828331] Re: Inconsistent time formats across Mahara

2019-05-20 Thread Ghada El-Zoghbi
Affected locations:

1. Group Forums: delete post, delete topic, edit topic, edit post,
report post, view

2. Annotation artefact type - entering and updating feedback.

3. Entering / updating a comment on a page.

4. Entering / updating peer assessment on a page.

5. Group view - list of portfolios of their last updated date/time.

6. Group view - submitted page/collection to a group for assessment

7. Watchlist - follow activity and follow owner

8. Users list - last logged in

9. Views/Collections page - when created / updated

10. Journals - entries - posted / updated

11. Wall posts

12. Export - in export file - the date the export occurred

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1828331

Title:
  Inconsistent time formats across Mahara

Status in Mahara:
  Confirmed

Bug description:
  Mahara: 17.10.9+
  OS: Linux
  DB: Postgres
  Browser: FF, Chrome, Others as well

  The time display has inconsistent display format at different areas.
  Please see the attached screenshot.

  Some of the areas include:
  - Shared with Me (24 hr format)
  - Group Forum (24 hr format)
  - Group Journal (12 hr format)
  - Group About (12 hr format)

  It would be useful to align the whole site to display only 1 format
  (e.g 24 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1828331/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1828331] Re: Inconsistent time formats across Mahara

2019-05-19 Thread Ghada El-Zoghbi
Thanks Kristina.

As I see it, we basically need to change lang strings that contain:

* "%%l" => "%%k" (and remove "%%p" from the string)


There are only 5 strings to replace.

They are located in: langconfig.php and interaction.formum.php


Cheers,
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1828331

Title:
  Inconsistent time formats across Mahara

Status in Mahara:
  Confirmed

Bug description:
  Mahara: 17.10.9+
  OS: Linux
  DB: Postgres
  Browser: FF, Chrome, Others as well

  The time display has inconsistent display format at different areas.
  Please see the attached screenshot.

  Some of the areas include:
  - Shared with Me (24 hr format)
  - Group Forum (24 hr format)
  - Group Journal (12 hr format)
  - Group About (12 hr format)

  It would be useful to align the whole site to display only 1 format
  (e.g 24 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1828331/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1825092] Re: Undefined index: institution

2019-05-12 Thread Ghada El-Zoghbi
Hi Robert ,

Is this going to be back ported to 18.10? I'm getting the warning there
as well.

Thanks,
Ghada

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1825092

Title:
  Undefined index: institution

Status in Mahara:
  Fix Released

Bug description:
  In admin menu click on Configure site-> Site options
  (/admin/site/options.php)

  When loading the page I see the error message:

  [WAR] 94 (lib/form/elements/autocomplete.php:124) Undefined index:
  institution

  
  this error comes from the changes made in the patch 
1fa6bfc59cdaa1d7895619ee3bae062cf5edfd11

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1825092/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1828331] Re: Inconsistent time formats across Mahara

2019-05-08 Thread Ghada El-Zoghbi
** Attachment added: "24-hour format on Group Forum page.png"
   
https://bugs.launchpad.net/mahara/+bug/1828331/+attachment/5262564/+files/24-hour%20format%20on%20Group%20Forum%20page.png

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1828331

Title:
  Inconsistent time formats across Mahara

Status in Mahara:
  New

Bug description:
  Mahara: 17.10.9+
  OS: Linux
  DB: Postgres
  Browser: FF, Chrome, Others as well

  The time display has inconsistent display format at different areas.
  Please see the attached screenshot.

  Some of the areas include:
  - Shared with Me (24 hr format)
  - Group Forum (24 hr format)
  - Group Journal (12 hr format)
  - Group About (12 hr format)

  It would be useful to align the whole site to display only 1 format
  (e.g 24 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1828331/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1828331] Re: Inconsistent time formats across Mahara

2019-05-08 Thread Ghada El-Zoghbi
** Attachment added: "24-hour format on Share with me page.png"
   
https://bugs.launchpad.net/mahara/+bug/1828331/+attachment/5262565/+files/24-hour%20format%20on%20Share%20with%20me%20page.png

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1828331

Title:
  Inconsistent time formats across Mahara

Status in Mahara:
  New

Bug description:
  Mahara: 17.10.9+
  OS: Linux
  DB: Postgres
  Browser: FF, Chrome, Others as well

  The time display has inconsistent display format at different areas.
  Please see the attached screenshot.

  Some of the areas include:
  - Shared with Me (24 hr format)
  - Group Forum (24 hr format)
  - Group Journal (12 hr format)
  - Group About (12 hr format)

  It would be useful to align the whole site to display only 1 format
  (e.g 24 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1828331/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1828331] Re: Inconsistent time formats across Mahara

2019-05-08 Thread Ghada El-Zoghbi
** Attachment added: "12-hour format on Group About page.png"
   
https://bugs.launchpad.net/mahara/+bug/1828331/+attachment/5262563/+files/12-hour%20format%20on%20Group%20About%20page.png

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1828331

Title:
  Inconsistent time formats across Mahara

Status in Mahara:
  New

Bug description:
  Mahara: 17.10.9+
  OS: Linux
  DB: Postgres
  Browser: FF, Chrome, Others as well

  The time display has inconsistent display format at different areas.
  Please see the attached screenshot.

  Some of the areas include:
  - Shared with Me (24 hr format)
  - Group Forum (24 hr format)
  - Group Journal (12 hr format)
  - Group About (12 hr format)

  It would be useful to align the whole site to display only 1 format
  (e.g 24 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1828331/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1828331] [NEW] Inconsistent time formats across Mahara

2019-05-08 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 17.10.9+
OS: Linux
DB: Postgres
Browser: FF, Chrome, Others as well

The time display has inconsistent display format at different areas.
Please see the attached screenshot.

Some of the areas include:
- Shared with Me (24 hr format)
- Group Forum (24 hr format)
- Group Journal (12 hr format)
- Group About (12 hr format)

It would be useful to align the whole site to display only 1 format (e.g
24 hours)

** Affects: mahara
 Importance: Undecided
 Status: New

** Attachment added: "12-hour format on Group Journal page.png"
   
https://bugs.launchpad.net/bugs/1828331/+attachment/5262562/+files/12-hour%20format%20on%20Group%20Journal%20page.png

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1828331

Title:
  Inconsistent time formats across Mahara

Status in Mahara:
  New

Bug description:
  Mahara: 17.10.9+
  OS: Linux
  DB: Postgres
  Browser: FF, Chrome, Others as well

  The time display has inconsistent display format at different areas.
  Please see the attached screenshot.

  Some of the areas include:
  - Shared with Me (24 hr format)
  - Group Forum (24 hr format)
  - Group Journal (12 hr format)
  - Group About (12 hr format)

  It would be useful to align the whole site to display only 1 format
  (e.g 24 hours)

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1828331/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1827012] Re: Security Release 17.10.8 Makefile

2019-04-30 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1827012

Title:
  Security Release 17.10.8 Makefile

Status in Mahara:
  Fix Released

Bug description:
  Mahara: 17.10.8
  OS: Linux
  DB: Postgres
  Browser: n/a

  
  The latest security release 17.10.8 has a bug in the Makefile when it's 
building ssphp.

  It needs to specify the dependency on running the composer init first.

  i.e.

  initcomposer: installcomposer

  Otherwise, the make ssphp failes because it can't find the
  composer.phar file.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1827012/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1827012] [NEW] Security Release 17.10.8 Makefile

2019-04-30 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 17.10.8
OS: Linux
DB: Postgres
Browser: n/a


The latest security release 17.10.8 has a bug in the Makefile when it's 
building ssphp.

It needs to specify the dependency on running the composer init first.

i.e.

initcomposer: installcomposer

Otherwise, the make ssphp failes because it can't find the composer.phar
file.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1827012

Title:
  Security Release 17.10.8 Makefile

Status in Mahara:
  New

Bug description:
  Mahara: 17.10.8
  OS: Linux
  DB: Postgres
  Browser: n/a

  
  The latest security release 17.10.8 has a bug in the Makefile when it's 
building ssphp.

  It needs to specify the dependency on running the composer init first.

  i.e.

  initcomposer: installcomposer

  Otherwise, the make ssphp failes because it can't find the
  composer.phar file.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1827012/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1808935] [NEW] Error Deleting External App - Foreign Key Constraint

2018-12-18 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 18.10.0
OS: Linux
DB: Postgres
Browser: FF

I created an LTI integration with a vanilla Moodle 3.5 and set it up as
an external app.

I did log in with it from Moodle with an assignment. I think decided to
delete the app. I get the following error:

[WAR] ff (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  update or delete on table "oauth_server_registry" violates foreign key 
constraint "ltiasse_oau_fk" on table "lti_assessment"
DETAIL:  Key (id)=(1) is still referenced from table "lti_assessment".] in 
EXECUTE("
DELETE FROM "oauth_server_registry"
WHERE consumer_key = ?
  AND (userid = ? OR userid IS NULL)
")Command was:
DELETE FROM "oauth_server_registry"
WHERE consumer_key = ?
  AND (userid = ? OR userid IS NULL)
and values was 
(0:96a91f9a865716feca5030a456f7154505c131e2d,1:45349)
Call stack (most recent first):

log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, 
true, true) at /var/www/var/www/mahara/htdocs/lib/errors.php:95
log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/var/www/mahara/htdocs/lib/errors.php:858
SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/var/www/mahara/htdocs/lib/dml.php:1025
delete_records_sql(" DELETE FROM "oauth_server_reg...", array(size 2)) at 
/var/www/var/www/mahara/htdocs/webservice/libs/oauth-php/store/OAuthStoreMahara.php:156
OAuthStoreMahara->deleteServer("96a91f9a865716feca5030a456f7154505c131e2d", 
"45349", true) at 
/var/www/var/www/mahara/htdocs/webservice/admin/oauthv1sregister.php:120
webservices_server_submit(object(Pieform), array(size 4)) at 
/var/www/var/www/mahara/htdocs/lib/pieforms/pieform.php:542
Pieform->__construct(array(size 7)) at 
/var/www/var/www/mahara/htdocs/lib/pieforms/pieform.php:164
Pieform::process(array(size 7)) at 
/var/www/var/www/mahara/htdocs/lib/mahara.php:5258
pieform(array(size 7)) at 
/var/www/var/www/mahara/htdocs/webservice/admin/oauthv1sregister.php:469
webservice_server_list_form(array(size 9), array(size 3)) at 
/var/www/var/www/mahara/htdocs/webservice/admin/oauthv1sregister.php:58

[WAR] ff (lib/dml.php:1025) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  update or delete on table "oauth_server_registry" violates foreign key 
constraint "ltiasse_oau_fk" on table "lti_assessment"
DETAIL:  Key (id)=(1) is still referenced from table "lti_assessment".] in 
EXECUTE("
DELETE FROM "oauth_server_registry"
WHERE consumer_key = ?
  AND (userid = ? OR userid IS NULL)
")Command was:
DELETE FROM "oauth_server_registry"
WHERE consumer_key = ?
  AND (userid = ? OR userid IS NULL)
and values was 
(0:96a91f9a865716feca5030a456f7154505c131e2d,1:45349)
Call stack (most recent first):

delete_records_sql(" DELETE FROM "oauth_server_reg...", array(size 2)) at 
/var/www/var/www/mahara/htdocs/webservice/libs/oauth-php/store/OAuthStoreMahara.php:156
OAuthStoreMahara->deleteServer("96a91f9a865716feca5030a456f7154505c131e2d", 
"45349", true) at 
/var/www/var/www/mahara/htdocs/webservice/admin/oauthv1sregister.php:120
webservices_server_submit(object(Pieform), array(size 4)) at 
/var/www/var/www/mahara/htdocs/lib/pieforms/pieform.php:542
Pieform->__construct(array(size 7)) at 
/var/www/var/www/mahara/htdocs/lib/pieforms/pieform.php:164
Pieform::process(array(size 7)) at 
/var/www/var/www/mahara/htdocs/lib/mahara.php:5258
pieform(array(size 7)) at 
/var/www/var/www/mahara/htdocs/webservice/admin/oauthv1sregister.php:469
webservice_server_list_form(array(size 9), array(size 3)) at 
/var/www/var/www/mahara/htdocs/webservice/admin/oauthv1sregister.php:58

The delete function needs to also remove records from the lti_assessment
table due to the foreign key constraint.

But, I'm not sure if that is the correct way to go. If there are
assignments, do we want to keep them and perhaps disable the external
app?

To delete, add the following to
/htdocs/webservice/admin.oauthv1sregister.php at line 120:

delete_records_sql('DELETE FROM {lti_assessment} WHERE oauthserver IN
(SELECT id FROM {oauth_server_registry} WHERE consumer_key = ?)',
array($dbserver->consumer_key));

** Affects: mahara
 Importance: Undecided
 Status: New

** Description changed:

  Mahara: 18.10.0
  OS: Linux
  DB: Postgres
  Browser: FF
  
  I created an LTI integration with a vanilla Moodle 3.5 and set it up as
  an external app.
  
  I did log in with it from Moodle with an assignment. I think decided to
  delete the app. I get the following error:
  
- 
  [WAR] ff (lib/errors.php:858) Failed to get a recordset: postgres8 

[Mahara-contributors] [Bug 1802046] [NEW] Upgrade to 18.10 error on oAuth clean up

2018-11-06 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 18.10.0
DB: Postgres
OS: Linux
Brower: N/A

During an upgrade from 17.04 to 18.10, the following error occurs:

[INF] 79 (lib/mahara.php:238) Upgrading auth.webservice
[WAR] 79 (lib/errors.php:858) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  update or delete on table "oauth_server_registry" violates foreign key 
constraint "oautservconf_oau_fk" on table "oauth_server_config"
[WAR] 79 (lib/errors.php:858) DETAIL:  Key (id)=(7) is still referenced from 
table "oauth_server_config".] in adodb_throw(DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?), Array)Command was: DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?) and values was (0:module/lti)
Call stack (most recent first):
  * log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
  * log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
  * SQLException->__construct("Failed to get a recordset: postgres8 error: [-1: 
E...") at /var/www/mahara/htdocs/lib/dml.php:1025
  * delete_records_sql("DELETE FROM "oauth_server_registry" WHERE external...", 
array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:1002
  * delete_records_select("oauth_server_registry", "WHERE externalserviceid IN 
(SELECT id FROM {extern...", array(size 1)) at 
/var/www/mahara/htdocs/webservice/lib.php:1529
  * external_delete_descriptions("module/lti") at 
/var/www/mahara/htdocs/webservice/lib.php:1570
  * external_reload_webservices() at 
/var/www/mahara/htdocs/auth/webservice/db/upgrade.php:594
  * xmldb_auth_webservice_upgrade("2017030600") at 
/var/www/mahara/htdocs/lib/upgrade.php:450
  * upgrade_plugin(object(stdClass)) at 
/var/www/mahara/htdocs/lib/mahara.php:272
  * upgrade_mahara(array(size 1)) at 
/var/www/mahara/htdocs/admin/cli/upgrade.php:56

[WAR] 79 (lib/dml.php:1025) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  update or delete on table "oauth_server_registry" violates foreign key 
constraint "oautservconf_oau_fk" on table "oauth_server_config"
[WAR] 79 (lib/dml.php:1025) DETAIL:  Key (id)=(7) is still referenced from 
table "oauth_server_config".] in adodb_throw(DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?), Array)Command was: DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?) and values was (0:module/lti)
Call stack (most recent first):
  * delete_records_sql("DELETE FROM "oauth_server_registry" WHERE external...", 
array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:1002
  * delete_records_select("oauth_server_registry", "WHERE externalserviceid IN 
(SELECT id FROM {extern...", array(size 1)) at 
/var/www/mahara/htdocs/webservice/lib.php:1529
  * external_delete_descriptions("module/lti") at 
/var/www/mahara/htdocs/webservice/lib.php:1570
  * external_reload_webservices() at 
/var/www/mahara/htdocs/auth/webservice/db/upgrade.php:594
  * xmldb_auth_webservice_upgrade("2017030600") at 
/var/www/mahara/htdocs/lib/upgrade.php:450
  * upgrade_plugin(object(stdClass)) at 
/var/www/mahara/htdocs/lib/mahara.php:272
  * upgrade_mahara(array(size 1)) at 
/var/www/mahara/htdocs/admin/cli/upgrade.php:56


The lti service had been configured in the past but is now disabled.
During the upgrade, it tries to clean up the tables for unused modules
and the lti comes up in the list.

But, it's failing due to a foreign key constraint on the
oauth_server_config table.

The external_delete_descriptions() needs to also remove records from the
oauth_server_config table before removing records from the
oauth_server_registry table.

** Affects: mahara
 Importance: Undecided
 Assignee: Ghada El-Zoghbi (ghada-z)
 Status: New

** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1802046

Title:
  Upgrade to 18.10 error on oAuth clean up

Status in Mahara:
  New

Bug description:
  Mahara: 18.10.0
  DB: Postgres
  OS: Linux
  Brower: N/A

  During an upgrade from 17.04 to 18.10, the following error occurs:

  [INF] 79 (lib/mahara.php:238) Upgrading auth.webser

[Mahara-contributors] [Bug 1802046] Re: Upgrade to 18.10 error on oAuth clean up

2018-11-06 Thread Ghada El-Zoghbi
To replicate:

* In previous version 17.04, create an lti configuration
* disable the lti configuration
* upgrade to 18.10

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1802046

Title:
  Upgrade to 18.10 error on oAuth clean up

Status in Mahara:
  New

Bug description:
  Mahara: 18.10.0
  DB: Postgres
  OS: Linux
  Brower: N/A

  During an upgrade from 17.04 to 18.10, the following error occurs:

  [INF] 79 (lib/mahara.php:238) Upgrading auth.webservice
  [WAR] 79 (lib/errors.php:858) Failed to get a recordset: postgres8 error: 
[-1: ERROR:  update or delete on table "oauth_server_registry" violates foreign 
key constraint "oautservconf_oau_fk" on table "oauth_server_config"
  [WAR] 79 (lib/errors.php:858) DETAIL:  Key (id)=(7) is still referenced from 
table "oauth_server_config".] in adodb_throw(DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?), Array)Command was: DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?) and values was (0:module/lti)
  Call stack (most recent first):
* log_message("Failed to get a recordset: postgres8 error: [-1: E...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
* log_warn("Failed to get a recordset: postgres8 error: [-1: E...") at 
/var/www/mahara/htdocs/lib/errors.php:858
* SQLException->__construct("Failed to get a recordset: postgres8 error: 
[-1: E...") at /var/www/mahara/htdocs/lib/dml.php:1025
* delete_records_sql("DELETE FROM "oauth_server_registry" WHERE 
external...", array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:1002
* delete_records_select("oauth_server_registry", "WHERE externalserviceid 
IN (SELECT id FROM {extern...", array(size 1)) at 
/var/www/mahara/htdocs/webservice/lib.php:1529
* external_delete_descriptions("module/lti") at 
/var/www/mahara/htdocs/webservice/lib.php:1570
* external_reload_webservices() at 
/var/www/mahara/htdocs/auth/webservice/db/upgrade.php:594
* xmldb_auth_webservice_upgrade("2017030600") at 
/var/www/mahara/htdocs/lib/upgrade.php:450
* upgrade_plugin(object(stdClass)) at 
/var/www/mahara/htdocs/lib/mahara.php:272
* upgrade_mahara(array(size 1)) at 
/var/www/mahara/htdocs/admin/cli/upgrade.php:56

  [WAR] 79 (lib/dml.php:1025) Failed to get a recordset: postgres8 error: [-1: 
ERROR:  update or delete on table "oauth_server_registry" violates foreign key 
constraint "oautservconf_oau_fk" on table "oauth_server_config"
  [WAR] 79 (lib/dml.php:1025) DETAIL:  Key (id)=(7) is still referenced from 
table "oauth_server_config".] in adodb_throw(DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?), Array)Command was: DELETE FROM 
"oauth_server_registry" WHERE externalserviceid IN (SELECT id FROM 
"external_services" WHERE component = ?) and values was (0:module/lti)
  Call stack (most recent first):
* delete_records_sql("DELETE FROM "oauth_server_registry" WHERE 
external...", array(size 1)) at /var/www/mahara/htdocs/lib/dml.php:1002
* delete_records_select("oauth_server_registry", "WHERE externalserviceid 
IN (SELECT id FROM {extern...", array(size 1)) at 
/var/www/mahara/htdocs/webservice/lib.php:1529
* external_delete_descriptions("module/lti") at 
/var/www/mahara/htdocs/webservice/lib.php:1570
* external_reload_webservices() at 
/var/www/mahara/htdocs/auth/webservice/db/upgrade.php:594
* xmldb_auth_webservice_upgrade("2017030600") at 
/var/www/mahara/htdocs/lib/upgrade.php:450
* upgrade_plugin(object(stdClass)) at 
/var/www/mahara/htdocs/lib/mahara.php:272
* upgrade_mahara(array(size 1)) at 
/var/www/mahara/htdocs/admin/cli/upgrade.php:56


  The lti service had been configured in the past but is now disabled.
  During the upgrade, it tries to clean up the tables for unused modules
  and the lti comes up in the list.

  But, it's failing due to a foreign key constraint on the
  oauth_server_config table.

  The external_delete_descriptions() needs to also remove records from
  the oauth_server_config table before removing records from the
  oauth_server_registry table.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1802046/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1785985] [NEW] Remove triggers from Mahara database

2018-08-08 Thread Ghada El-Zoghbi
Public bug reported:

I've created a forum post to discuss this:

* https://mahara.org/interaction/forum/topic.php?id=8324

Basically, my reasoning is:


1. Mahara supports two DB types - MySql and Postgres. Each DB has different 
syntax when it comes to creating/updating/deleting triggers. This makes it 
complicated to develop and maintain.

2. Mahara already has the concept of events in place. We can move the
functionality of each trigger into an event. We would then have more
control over the event. For example, don't action an event during an
upgrade.

3. Triggers make it difficult to see what is actually happening. They
are not transparent. If you weren't aware that there is a trigger on the
table, it can make it difficult to track down why something is behaving
in a certain way.


We can start slowly by removing one at a time. It doesn't have to happen all at 
once.

Just my thoughts.

Cheers.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1785985

Title:
  Remove triggers from Mahara database

Status in Mahara:
  New

Bug description:
  I've created a forum post to discuss this:

  * https://mahara.org/interaction/forum/topic.php?id=8324

  Basically, my reasoning is:

  
  1. Mahara supports two DB types - MySql and Postgres. Each DB has different 
syntax when it comes to creating/updating/deleting triggers. This makes it 
complicated to develop and maintain.

  2. Mahara already has the concept of events in place. We can move the
  functionality of each trigger into an event. We would then have more
  control over the event. For example, don't action an event during an
  upgrade.

  3. Triggers make it difficult to see what is actually happening. They
  are not transparent. If you weren't aware that there is a trigger on
  the table, it can make it difficult to track down why something is
  behaving in a certain way.

  
  We can start slowly by removing one at a time. It doesn't have to happen all 
at once.

  Just my thoughts.

  Cheers.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1785985/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1773864] Re: ExternalFeed block cron task fails when SSL cert is not valid for a feed

2018-05-28 Thread Ghada El-Zoghbi
** Changed in: mahara
 Assignee: (unassigned) => Ghada El-Zoghbi (ghada-z)

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1773864

Title:
  ExternalFeed block cron task fails when SSL cert is not valid for a
  feed

Status in Mahara:
  New

Bug description:
  Mahara: 18.04
  OS: Linux
  DB: Postgres
  Browser: n/a

  
  Cron fails (and is therefore locked) when one of the rss feeds has an invalid 
ssl certificate. 

  The cron job calls lib/web.php::mahara_http_request() from
  blocktype/externalfeed/lib.php::parse_feed()

  
  The following error results:

  [WAR] 63 (lib/web.php:4425) Curl error: 60: SSL certificate problem: unable 
to get local issuer certificate
  Call stack (most recent first):
* log_message("Curl error: 60: SSL certificate problem: unable to...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
* log_warn("Curl error: 60: SSL certificate problem: unable to...") at 
/var/www/mahara/htdocs/lib/web.php:4425
* mahara_http_request(array(size 3), false) at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:421
* 
PluginBlocktypeExternalfeed::parse_feed("http://www.safeworkaustralia.gov.au/sites/SWA/medi...;,
 "0", "", "") at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:336
* PluginBlocktypeExternalfeed::refresh_feeds() at 
/var/www/mahara/htdocs/lib/mahara.php:1809
* call_static_method("PluginBlocktypeExternalfeed", "refresh_feeds") at 
/var/www/mahara/htdocs/lib/cron.php:115

  [WAR] 63 (blocktype/externalfeed/lib.php:336) error in 
/blocktype/externalfeed/lib.php line 428. error:SSL certificate problem: unable 
to get local issuer certificate
  [WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for 
MaharaException([string $message [, long $code [, Throwable $previous = NULL]]])
  Call stack (most recent first):
* exception(object(Error)) at Unknown:0


  I had added extra debugging statements to see the error:

  [WAR] 63 (blocktype/externalfeed/lib.php:336) error in
  /blocktype/externalfeed/lib.php line 428. error:SSL certificate
  problem: unable to get local issuer certificate

  
  Mahara then raises a MaharaException with the error (SSL certificate problem: 
unable to get local issuer certificate) as the code. But, it's not an integer 
so it fails with:

  [WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for
  MaharaException([string $message [, long $code [, Throwable $previous
  = NULL]]])

  
  We need to check if the error coming back is a number or string. If number, 
pass it as the code in the MaharaException. Otherwise, append it to the text of 
the error message so the data is not lost.

  
  We found this when upgrading from 17.04 to 18.04. This issue is present in 
17.04 as well.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1773864/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1773864] Re: ExternalFeed block cron task fails when SSL cert is not valid for a feed

2018-05-28 Thread Ghada El-Zoghbi
To reproduce:

* Create a page
* Add External Feed: 
** Feed Location = http://www.safeworkaustralia.gov.au/sites/SWA/mediafeed   
(at time of post) 
** Insecure SSL mode = No
** Save


These other feeds were causing issues at time of this post:

* http://www.safeworkaustralia.gov.au/sites/SWA/publicationfeed
* http://www.safeworkaustralia.gov.au/sites/SWA/newsfeed


In the cron log, error should appear and externalfeed cron task should now be 
locked (when it runs and it's time to refresh the feed).


To speed up testing:

* in blocltype/externalfeed/lib.php::refresh_feeds - set in the sql
query to refresh every 0 minutes (instead of 30 minutes).

* in blocltype/externalfeed/lib.php::parse_feeds - comment out the
'throw $cache[$source]' and the 'return $cache[$source]' at the top of
the function.

* change cron time for the task to run every minute:
update blocktype_cron set minute = '*', nextrun = '2018-05-29 11:54' where 
callfunction = 'refresh_feeds';


If feeds are now valid, can try to set $result->error after the 
mahara_http_request() call in the parse_feed() function.

i.e. $result->error = "SSL certificate problem: unable to get local
issuer certificate";

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1773864

Title:
  ExternalFeed block cron task fails when SSL cert is not valid for a
  feed

Status in Mahara:
  New

Bug description:
  Mahara: 18.04
  OS: Linux
  DB: Postgres
  Browser: n/a

  
  Cron fails (and is therefore locked) when one of the rss feeds has an invalid 
ssl certificate. 

  The cron job calls lib/web.php::mahara_http_request() from
  blocktype/externalfeed/lib.php::parse_feed()

  
  The following error results:

  [WAR] 63 (lib/web.php:4425) Curl error: 60: SSL certificate problem: unable 
to get local issuer certificate
  Call stack (most recent first):
* log_message("Curl error: 60: SSL certificate problem: unable to...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
* log_warn("Curl error: 60: SSL certificate problem: unable to...") at 
/var/www/mahara/htdocs/lib/web.php:4425
* mahara_http_request(array(size 3), false) at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:421
* 
PluginBlocktypeExternalfeed::parse_feed("http://www.safeworkaustralia.gov.au/sites/SWA/medi...;,
 "0", "", "") at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:336
* PluginBlocktypeExternalfeed::refresh_feeds() at 
/var/www/mahara/htdocs/lib/mahara.php:1809
* call_static_method("PluginBlocktypeExternalfeed", "refresh_feeds") at 
/var/www/mahara/htdocs/lib/cron.php:115

  [WAR] 63 (blocktype/externalfeed/lib.php:336) error in 
/blocktype/externalfeed/lib.php line 428. error:SSL certificate problem: unable 
to get local issuer certificate
  [WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for 
MaharaException([string $message [, long $code [, Throwable $previous = NULL]]])
  Call stack (most recent first):
* exception(object(Error)) at Unknown:0


  I had added extra debugging statements to see the error:

  [WAR] 63 (blocktype/externalfeed/lib.php:336) error in
  /blocktype/externalfeed/lib.php line 428. error:SSL certificate
  problem: unable to get local issuer certificate

  
  Mahara then raises a MaharaException with the error (SSL certificate problem: 
unable to get local issuer certificate) as the code. But, it's not an integer 
so it fails with:

  [WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for
  MaharaException([string $message [, long $code [, Throwable $previous
  = NULL]]])

  
  We need to check if the error coming back is a number or string. If number, 
pass it as the code in the MaharaException. Otherwise, append it to the text of 
the error message so the data is not lost.

  
  We found this when upgrading from 17.04 to 18.04. This issue is present in 
17.04 as well.

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1773864/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1773864] [NEW] ExternalFeed block cron task fails when SSL cert is not valid for a feed

2018-05-28 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 18.04
OS: Linux
DB: Postgres
Browser: n/a


Cron fails (and is therefore locked) when one of the rss feeds has an invalid 
ssl certificate. 

The cron job calls lib/web.php::mahara_http_request() from
blocktype/externalfeed/lib.php::parse_feed()


The following error results:

[WAR] 63 (lib/web.php:4425) Curl error: 60: SSL certificate problem: unable to 
get local issuer certificate
Call stack (most recent first):
  * log_message("Curl error: 60: SSL certificate problem: unable to...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
  * log_warn("Curl error: 60: SSL certificate problem: unable to...") at 
/var/www/mahara/htdocs/lib/web.php:4425
  * mahara_http_request(array(size 3), false) at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:421
  * 
PluginBlocktypeExternalfeed::parse_feed("http://www.safeworkaustralia.gov.au/sites/SWA/medi...;,
 "0", "", "") at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:336
  * PluginBlocktypeExternalfeed::refresh_feeds() at 
/var/www/mahara/htdocs/lib/mahara.php:1809
  * call_static_method("PluginBlocktypeExternalfeed", "refresh_feeds") at 
/var/www/mahara/htdocs/lib/cron.php:115

[WAR] 63 (blocktype/externalfeed/lib.php:336) error in 
/blocktype/externalfeed/lib.php line 428. error:SSL certificate problem: unable 
to get local issuer certificate
[WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for 
MaharaException([string $message [, long $code [, Throwable $previous = NULL]]])
Call stack (most recent first):
  * exception(object(Error)) at Unknown:0


I had added extra debugging statements to see the error:

[WAR] 63 (blocktype/externalfeed/lib.php:336) error in
/blocktype/externalfeed/lib.php line 428. error:SSL certificate problem:
unable to get local issuer certificate


Mahara then raises a MaharaException with the error (SSL certificate problem: 
unable to get local issuer certificate) as the code. But, it's not an integer 
so it fails with:

[WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for
MaharaException([string $message [, long $code [, Throwable $previous =
NULL]]])


We need to check if the error coming back is a number or string. If number, 
pass it as the code in the MaharaException. Otherwise, append it to the text of 
the error message so the data is not lost.


We found this when upgrading from 17.04 to 18.04. This issue is present in 
17.04 as well.

** Affects: mahara
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1773864

Title:
  ExternalFeed block cron task fails when SSL cert is not valid for a
  feed

Status in Mahara:
  New

Bug description:
  Mahara: 18.04
  OS: Linux
  DB: Postgres
  Browser: n/a

  
  Cron fails (and is therefore locked) when one of the rss feeds has an invalid 
ssl certificate. 

  The cron job calls lib/web.php::mahara_http_request() from
  blocktype/externalfeed/lib.php::parse_feed()

  
  The following error results:

  [WAR] 63 (lib/web.php:4425) Curl error: 60: SSL certificate problem: unable 
to get local issuer certificate
  Call stack (most recent first):
* log_message("Curl error: 60: SSL certificate problem: unable to...", 8, 
true, true) at /var/www/mahara/htdocs/lib/errors.php:95
* log_warn("Curl error: 60: SSL certificate problem: unable to...") at 
/var/www/mahara/htdocs/lib/web.php:4425
* mahara_http_request(array(size 3), false) at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:421
* 
PluginBlocktypeExternalfeed::parse_feed("http://www.safeworkaustralia.gov.au/sites/SWA/medi...;,
 "0", "", "") at 
/var/www/mahara/htdocs/blocktype/externalfeed/lib.php:336
* PluginBlocktypeExternalfeed::refresh_feeds() at 
/var/www/mahara/htdocs/lib/mahara.php:1809
* call_static_method("PluginBlocktypeExternalfeed", "refresh_feeds") at 
/var/www/mahara/htdocs/lib/cron.php:115

  [WAR] 63 (blocktype/externalfeed/lib.php:336) error in 
/blocktype/externalfeed/lib.php line 428. error:SSL certificate problem: unable 
to get local issuer certificate
  [WAR] 63 (lib/errors.php:536) [Error]: Wrong parameters for 
MaharaException([string $message [, long $code [, Throwable $previous = NULL]]])
  Call stack (most recent first):
* exception(object(Error)) at Unknown:0


  I had added extra debugging statements to see the error:

  [WAR] 63 (blocktype/externalfeed/lib.php:336) error in
  /blocktype/externalfeed/lib.php line 428. error:SSL certificate
  problem: unable to get local issuer certificate

  
  Mahara then raises a MaharaException with the error (SSL certificate problem: 
unable to get local issuer certificate) as the code. But, it's not an integer 
so it fails with:

  [WAR] 63 (lib/errors.php:536) 

[Mahara-contributors] [Bug 1754237] Re: SimpleSAML_Logger is deprecated warning

2018-05-23 Thread Ghada El-Zoghbi
There are also warnings for the store:

WARNING [TRe51b00d3] The class or interface 'SimpleSAML_Store' is now
using namespaces, please use 'SimpleSAML\Store'.

-- 
You received this bug notification because you are a member of Mahara
Contributors, which is subscribed to Mahara.
Matching subscriptions: Subscription for all Mahara Contributors -- please ask 
on #mahara-dev or mahara.org forum before editing or unsubscribing it!
https://bugs.launchpad.net/bugs/1754237

Title:
  SimpleSAML_Logger is deprecated warning

Status in Mahara:
  In Progress

Bug description:
  There is a warning about:
  ssphp- WARNING [CL05234431] The class or interface 'SimpleSAML_Logger' is now 
using namespaces, please use 'SimpleSAML\Logger'.

  We need to update the config etc to use the new style of calling the
  logger

To manage notifications about this bug go to:
https://bugs.launchpad.net/mahara/+bug/1754237/+subscriptions

___
Mailing list: https://launchpad.net/~mahara-contributors
Post to : mahara-contributors@lists.launchpad.net
Unsubscribe : https://launchpad.net/~mahara-contributors
More help   : https://help.launchpad.net/ListHelp


[Mahara-contributors] [Bug 1755046] [NEW] File name stripped of unicode chars in file transfer from Moodle to Mahara

2018-03-11 Thread Ghada El-Zoghbi
Public bug reported:

Mahara: 17.10.2
OS: Ubuntu 16.04
PHP: 7.0
DB: Postgres
Browser: n/a

During a XML-RPC 'File' transfer (not 'Leap2A' transfer) from Moodle to
Mahara, a file name with unicode characters gets stripped out. Clamdscan
then can't find the file when it tries to scan it and reports an error.


The error message coming from Clam is:

ERROR: Can't access file
/var/lib/sitedata/temp/import/339/extract/uploadxxxassignment1.txt

The file name should be: upload(xxx)assignment1.txt

Please note that the brackets are actually unicode:  %uFF08 and %uFF09


This PHP core function strips any unicode characters: escapeshellarg($fullpath)

This is in htdocs/lig/uploadmanager.php::mahara_clam_scan_file() line
337


We were able to reproduce by:

1. Set up a Moodle / Mahara xml-rpc integration
2. In Mahara, set up clamAV (or clamdscan) virus checking
3. In Moodle, create a course and set up an assignment to upload a file
4. In Moodle, Enrol a test user in the course.
5. In Moodle, Log in as the test user and go to course.
6. In Moodle, Upload a file with unicode chars %uFF08 and %uFF09 in the name 
(I've attached a sample).
7. In Moodle, export the file to Mahara and select 'File'
8. In Moodle, the following error will appear to the user:

Exception: Failed to start communication with remote server: remote server 
error: code: , message: Your administrator has enabled virus checking for file 
uploads but has misconfigured something. Your file upload was NOT successful. 
Your administrator was notified so they can fix it. Maybe try uploading this 
file later.ERROR 4:
4: remote server error: code: , message: Your administrator has enabled virus 
checking for file uploads but has misconfigured something. Your file upload was 
NOT successful. Your administrator was notified so they can fix it. Maybe try 
uploading this file later.
in ./portfolio/mahara/lib.php on line 212.
Stack trace:

line 212 of ./portfolio/mahara/lib.php: portfolio_export_exception thrown
line 510 of ./lib/portfolio/exporter.php: call to 
portfolio_plugin_mahara->send_package()
line 227 of ./lib/portfolio/exporter.php: call to 
portfolio_exporter->process_stage_send()
line 233 of ./lib/portfolio/exporter.php: call to 
portfolio_exporter->process_stage()
line 233 of ./lib/portfolio/exporter.php: call to 
portfolio_exporter->process_stage()
line 265 of ./portfolio/add.php: call to portfolio_exporter->process_stage()


9. In Mahara, an error is generated in the error log:

*14997 FastCGI sent in stderr: "PHP message: [DBG] 75 
(lib/uploadmanager.php:452) Clam AV has found a file that is infected with a 
virus. It was uploaded by  . The original file path of the infected file was 
/var/lib/sitedata/temp/import/341/extract/upload<88>xxx<89>assignment1.txt.
 The file has been moved to a quarantine directory and the new path is 
/var/lib/sitedata/quarantine/20180312093305-user-0-infected#012PHP message: 
[WAR] 75 (import/file/lib.php:58) Your administrator has enabled virus checking 
for file uploads but has misconfigured something. Your file upload was NOT 
successful. Your administrator was notified so they can fix it. Maybe try 
uploading this file later.#012PHP message: Call stack (most recent 
first):#012PHP message:   
* PluginImportFile->verify_file_contents() at 
/var/www/site/import/file/lib.php:37#012PHP message:   
* PluginImportFile->process() at /var/www/site/api/xmlrpc/lib.php:514#012PHP 
message:   
* send_content_ready("f3b1200695be08bdfeeea32918048b1be2ae89e3", "", 
"file", array(size 4), 1) at /var/www/site/api/xmlrpc/lib.php:115#012PHP 
message:   
* api_dummy_method("portfolio/mahara/lib.php/send_content_ready", array(size 
5), "send_content_ready") at Unknown:0#012PHP message:   
* xmlrpc_server_call_method(resource(#23), "#012PHP message: __construct("#012PHP 
message: 

  1   2   3   >