date:20130122

[web2py] Re: 'list:reference', contains, Mysql, instr

2013-01-22 Thread pablo . angulo



El lunes, 21 de enero de 2013 22:47:41 UTC+1, Massimo Di Pierro escribió:

 Now I understand the problem. You want to pass an expression where a value 
 is expected. I do not believe we can support this. INSTR does not solve the 
 problem because it would not be correct to looks for substrings. In fact 
 paper.authors may contain '|12|13|' and it would incorrectly return that 
 author '2' is one of authors because it is a substring.


ok, but we can use CONCAT to append and prepend '|', that's more standard 
that instr.
 


 You have to do it in two steps:

 person = db(db.person.name='').select().first()
 papers = db.paper.authors.contains(person.id).select()


But it's not a one-time query. I'm working in something like appadmin, 
tweaking smart_query and SQLFORM.grid to help the user build compund 
querys. I'd rather normalize the database.

BTW, I will come later with some ideas/questions.

Thanks again
 


 Massimo


 On Jan 21, 2013, at 3:33 PM, pablo@uam.es javascript: wrote:



 El lunes, 21 de enero de 2013 18:56:07 UTC+1, Massimo Di Pierro escribió:

 Before we attempt to fix this. It works for me as it is with sqlite. Why 
 does doesn't it wirk with mysql? MySQL has a LIKE operator. Wat error do 
 you get?


 execution of:

 db(db.paper.authors.contains(person.id))._select()

 gives:

 SELECT  person.id, paper.id, paper.title, paper.authors FROM person, 
 paper WHERE (paper.authors LIKE '%|person.id|%');
  
 which I think tries to find those papers whose authors field contains 
 the literal person.id, instead of the pairs (paper,person) such that 
 the person is one of the authors of the paper.

 With the modification I wrote before, it gets converted into:

 'SELECT  person.id, paper.id, paper.title, paper.authors FROM person, 
 paper WHERE INSTR(paper.authors, person.id);'

 with a different meaning.

 I can see why using the INSTR operator on MySQL may be better but do we 
 know it is faster?


 I haven't made any speed test, I'd go for anything that works in sqlite 
 and mysql/mariadb.

 Thanks for your attention!
  


 Massimo

 On Monday, 21 January 2013 08:55:40 UTC-6, pablo@uam.es wrote:

   Hello:
   I was getting incorrect SQL from statements like:

 db(db.paper.authors.contains(person.id)).select()

 where

 db.define_table('paper',
 Field('title'),
 Field('authors', 'list:reference person'),
 )


   so I rewrote dal.MySQLAdapter.CONTAINS from:

 def CONTAINS(self, first, second):
 if first.type in ('string', 'text'):
 key = '%'+str(second).replace('%','%%')+'%'
 elif first.type.startswith('list:'):
 key = 
 '%|'+str(second).replace('|','||').replace('%','%%')+'|%'
 return '(%s LIKE %s)' % 
 (self.expand(first),self.expand(key,'string'))

 into:

 def CONTAINS(self, first, second):
 if first.type in ('string', 'text'):
 key = '%'+str(second).replace('%','%%')+'%'
 elif first.type.startswith('list:reference'):
 return 'INSTR(%s, %s)'%(first, str(second))
 elif first.type.startswith('list:'):
 key = 
 '%|'+str(second).replace('|','||').replace('%','%%')+'|%'
 return '(%s LIKE %s)' % 
 (self.expand(first),self.expand(key,'string'))

 now the above query works, but that's in MySQL, not SQLite, which 
 apparently does not have  a INSTR operation, despite the fact that it is 
 advertised!

 https://www.sqlite.org/lang_corefunc.html

 Can anybody think of a solution that works in more databases




--

[web2py] Re: Scheduler ignoring period/repeat setting.

2013-01-22 Thread Stormcrow

Hi Niphlod. Thanks so much for the reply. I suspected I had the* 
queue_task*method in the wrong place and I did.

Is there isn't any neat way of automatically checking that these tasks are 
in the database and adding them if they're not there? I don't really want 
to have to access a controller to get them going I just want them to always 
be there when I start the server.

--

[web2py] Re: 'truncate' broken in SQLTABLE with 'headers'.

2013-01-22 Thread Mandar Vaze

Just to revive the discussion, if SQLTABLE is passed a headers dict AND 
truncate value, then truncate isn't honored.

e.g 

mytable = SQLTABLE(somequery,
   headers={'table.field1': 'Label1', 'table.field2': 'Label2'}
   truncate=25)



On Wednesday, July 20, 2011 5:45:21 AM UTC+5:30, Anthony wrote:

 On Tuesday, July 19, 2011 7:48:35 PM UTC-4, Carlos wrote: 

 So 'truncate' is not being respected when using the previous 'headers' 
 format (simple dict of strings).

  
 Oh, I see, you're right -- that should be fixed.
  
 Anthony


This continues to be broken after 12+ months. May be not enough people are 
using SQLTABLE + truncate combination

Any plans to fix this ?

-Mandar


 

  


--

[web2py] Is there a way to use SQLFORMs to generate form that can update multiple records?

2013-01-22 Thread Srinath

Hi all,

As the subject states, is there a way to update multiple records in a
single form using SQLFORM?

Thanks in advance,
Srinath

--

[web2py] Re: Is there a way to use SQLFORMs to generate form that can update multiple records?

2013-01-22 Thread Alan Etkin



 As the subject states, is there a way to update multiple records in a 
 single form using SQLFORM?


If you don't need to use specifically SQLFORM, you can try 
contrib/spreadsheet using the data option (for viewing/updating a db)
It's intended to deal with numbers and short text, but it could be improved 
to support other field types.

--

[web2py] Re: 'truncate' broken in SQLTABLE with 'headers'.

2013-01-22 Thread Mandar Vaze



 On Wednesday, July 20, 2011 5:45:21 AM UTC+5:30, Anthony wrote:

 On Tuesday, July 19, 2011 7:48:35 PM UTC-4, Carlos wrote: 

 So 'truncate' is not being respected when using the previous 'headers' 
 format (simple dict of strings).

  
 Oh, I see, you're right -- that should be fixed.
  
 Anthony

  
commented lines below are potential fix. (gluon/sqlhtml.py)

elif field.type in ['string', 'text']:
r = str(field.formatter(r))
if headers != {}:  # new implement dict
if isinstance(headers[colname], dict):
if isinstance(headers[colname]['truncate'], int
):
r = truncate_string(
r, headers[colname]['truncate'])
#elif truncate is not None:
#r = truncate_string(r, truncate)
elif not truncate is None:
r = truncate_string(r, truncate)


I have tested it in 2.3.2 (2012-12-17 15:03:30) stable it works. (and 
then commented the code so that I get the out of the box behavior)

-Mandar

--

Re: [web2py] How I can use code tag with Markmin?

2013-01-22 Thread Alan Etkin

How about

MARKMIN(text, extra={pre_with_code: lambda text: 
precode{0}/code/pre.
format(XML(text, sanitize=True))})

--

[web2py] Re: Scheduler ignoring period/repeat setting.

2013-01-22 Thread Niphlod

its more of an architectural problem than a web2py one  to be sure that 
the tasks are there you may have to check for their presence on the 
scheduler_task table. There is the uuid column that is enforced as unique, 
so you can use it to query the table safely.
Now, the problem is that you generally don't want to do a query at every 
request: you may want to look at cron's @reboot or use a little trick using 
cache

lock_tasks = cache.ram('lock_tasks', lambda: request.now, time_expire=120)
if lock_tasks == request.now:
 #do your checks here

What it does? Stores request.now in cache for 120 seconds. Next, it checks 
if the value retrieved from the cache is the same one (meaning that the 
cache value has been refreshed in the same request)  the code in the 
section will be executed at most every two minutes (always assuming that 
requests come along)
Please mind that if you put this in models it works well for web requests, 
but the scheduler executes models too and every time in a fresh process, so 
cache.ram doesn't survive... to eliminate that issue, you may have to put 
this logic too

if not global_settings.cmd_options.scheduler:
#this piece of code is not running on the scheduler



--

[web2py] records as fields in a single form

2013-01-22 Thread Annet

I defined the following tables:

db.define_table('nav',
Field('navbarID','reference 
navbar',default='',requires=IS_IN_DB(db,'navbar.id','%(name)s',zero=T('select 
a value')),notnull=True),

Field('name',length=32,default='',requires=[IS_LENGTH(32,error_message=T('length
 
exceeds 32 characters')),IS_NOT_EMPTY()],notnull=True),

Field('backendcontroller',length=32,default='',requires=IS_LENGTH(32,error_message=T('length
 
exceeds 32 characters'))),

Field('frontendcontroller',length=32,default='',requires=IS_LENGTH(32,error_message=T('length
 
exceeds 32 characters'))),

Field('function',length=32,default='',requires=[IS_LENGTH(32,error_message=T('length
 
exceeds 32 characters')),IS_NOT_EMPTY()],notnull=True),
format='%(name)s',
migrate=False)

db.define_table('nodeNav',
Field('nodeID','reference node',**isnode),
Field('navID','reference nav',default='',notnull=True'),
Field('backend',type='boolean',default=False),
Field('frontend',type='boolean',default=False),
on_define=set_requirement,
migrate=False)

Given the following records in nodeNav:

46TrueTrue
47TrueFalse
48TrueFalse
49TrueTrue

I'd like to create a form in which every record is a field, the name should 
be the navID, the label should be the name  in table nav, the field should 
be of type boolean and its value should be the value of frontend.

I thought of something like:

query=db((db.nodeNav.nodeID==4)(db.nodeNav.navID==db.nav.id)(db.nav.navbarID==3))
records=query.select(db.nodeNav.ALL,db.nav.ALL)
fields_list=[]
for r in records:

fields_list.append(Field(name=str(r.nav.id),type='boolean',default=r.nodeNav.frontend,label=r.nav.name))
form=SQLFORM.factory(*fields_list)
if form.process().accepted:
records=query.select(db.nodeNav.ALL)
for r in records:
field=r.navID
r.update_record(frontend=form.vars.field)
return dict(form=form)

Is this the way to go or is there a better way to implement this?

Kind regards,

Annet

--

[web2py] Compare row objects inside a for... in loop

2013-01-22 Thread Luciano Laporta Podazza

Hello, 

I've been searching a lot about this but couldn't find anything so sorry 
for this basic question.

How can I compare an actual row object with the next one inside a for...in 
loop?

Thanks in advance!

--

Re: [web2py] Re: new setup-web2py-nginx-uwsgi-ubuntu.sh

2013-01-22 Thread paolo.vall...@gmail.com

Hi all, I've just discovered that the regex used to retrieve the static
files doesn't work if I use languages abbreviation in urls.
How can we fix this?

2012/12/28 Richard Vézina ml.richard.vez...@gmail.com

Hello,

I publish a new script that allow deployment of Redmine beside web2py.

Here :
https://groups.google.com/forum/?fromgroups=#!searchin/web2py/redmine/web2py/ZqL7Si8Khbo/Es-wK1yXdgQJ

Notice : After some read, I choose Unicorn over Phussion Passenger.

Richard

On Thu, Dec 20, 2012 at 9:57 AM, Massimo Di Pierro
massimo.dipie...@gmail.com wrote:

please email me the patch of latest file when ready for inclusion.

On Thursday, 20 December 2012 03:18:20 UTC-6, Niphlod wrote:

perfect, Ccing Massimo on this final one.

Il giorno giovedì 20 dicembre 2012 00:16:25 UTC+1, Paolo ha scritto:

Hi
I was trying with the script that comes with the stable web2py. with
the one on dropbox I problem has gone.

Paolo

2012/12/19 Niphlod nip...@gmail.com

@Simone, an other improvement to the script could be to combine into a
single server the 80, and 443 to avoid duplicating configuration, as
stated
here :
http://nginx.org/en/docs/http/configuring_https_servers.**html**
#single_http_https_serverhttp://nginx.org/en/docs/http/configuring_https_servers.html#single_http_https_server
I don't had test this.

Also, here the code for permanent redirection
server {
server_name $hostname;
listen 80;
return 301 http*s*://$hostname$request_uri; # NOTE: I am not sure
for $hostname here, because I didn't set hostname for my VM until now, as
in the example (URL below) we can use domainName.com instead if properly
configure in nginx
...
}

People may want separate configs for http and https. The script
objective is to have a working copy of web2py. If we start to follow such
requests, we'd end up installing postgresql and redis too :P

@Paolo: try the script found at the dropbox link. If the same error
happens I think we need the log of pip and a hand by Roberto on the
specific error...

--
Paolo

Re: [web2py] Re: Is there a way to use SQLFORMs to generate form that can update multiple records?

2013-01-22 Thread Srinath

Is there any guide on how I can make use of this spreadsheet module? How I
create the form and other related guides?


On Tue, Jan 22, 2013 at 4:21 PM, Alan Etkin spame...@gmail.com wrote:

 As the subject states, is there a way to update multiple records in a
 single form using SQLFORM?


 If you don't need to use specifically SQLFORM, you can try
 contrib/spreadsheet using the data option (for viewing/updating a db)
 It's intended to deal with numbers and short text, but it could be
 improved to support other field types.

 --





--

[web2py] Re: 'list:reference', contains, Mysql, instr

2013-01-22 Thread pablo . angulo



 person = db(db.person.name='').select().first()
 papers = db.paper.authors.contains(person.id).select()


 Sorry, I didn't read carefully enough. That's not what I need, I need to 
perform a join. My real-world query is rather like:

db.define_table('person',
Field('name', unique=True),
Field('some_field'),
)

Advisor = db.person.with_alias('advisor')
GradStudent = db.person.with_alias('grad')

db.define_table('thesis_project',
Field('advisors', 'list:reference person'),
Field('candidate', db.person),
Field('title', 'text')
)



and then:

db( db.thesis_project.advisors.contains(Advisor.id)  
   (db.thesis_project.candidate==GradStudent.id)  
   (Advisor.some_field==some_value)  
   (GradStudent.some_field==some_value)).select(...)



and it works in mysql using INSTR and CONCAT, and there are other solutions 
with other functions that don't seem to work in sqlite, even though some of 
them are advertised. That's my puzzle.


--

[web2py] Re: Compare row objects inside a for... in loop

2013-01-22 Thread Anthony

I suppose you could do something like:

rows = db(query).select()
for i in range(len(rows)-1):
code comparing rows[i] to rows[i+1]

Anthony

On Tuesday, January 22, 2013 8:17:10 AM UTC-5, Luciano Laporta Podazza 
wrote:

 Hello, 

 I've been searching a lot about this but couldn't find anything so sorry 
 for this basic question.

 How can I compare an actual row object with the next one inside a for...in 
 loop?

 Thanks in advance!


--

Re: [web2py] Re: Compare row objects inside a for... in loop

2013-01-22 Thread Luciano Laporta Podazza

Hi Anthony!,

Yes, a friend of mine just gave me the same solution and I can't believe I
didn't thought about that :P.

Thanks a lot!!!


On Tue, Jan 22, 2013 at 11:10 AM, Anthony abasta...@gmail.com wrote:

 I suppose you could do something like:

 rows = db(query).select()
 for i in range(len(rows)-1):
 code comparing rows[i] to rows[i+1]

 Anthony

 On Tuesday, January 22, 2013 8:17:10 AM UTC-5, Luciano Laporta Podazza
 wrote:

 Hello,

 I've been searching a lot about this but couldn't find anything so sorry
 for this basic question.

 How can I compare an actual row object with the next one inside a
 for...in loop?

 Thanks in advance!

  --







-- 
Atte
Luciano Laporta Podazza

--

79 matches

Mail list logo