Re: [Evolution-hackers] e_cal/book_open() + only_if_exists

2012-02-14 Thread Patrick Ohly
On Tue, 2012-02-14 at 08:45 +0100, Milan Crha wrote:
 On Mon, 2012-02-13 at 21:52 +0100, Patrick Ohly wrote:
* A second invocation of SyncEvolution finds the definition of the
  system address book via e_book_get_addressbooks() and tries to
  open it with e_book_open(only_if_exists=false), which then fails
  with various errors, depending on the EDS version (Cannot open
  book: db error 0x2 (No such file or directory) in 3.2,
  DB_RUNRECOVERY in 3.3).
  
  The problem can be avoided by always passing only_if_exists=true to
  the open calls. The way how the API is defined and/or implemented,
  creating an database definition and creating the actual database files
  are two different steps.
 
   Hi,
 that's rather a bug in the backend. Calling with only_if_exists=false
 might mean that the backend should create a new book/calendar, in case
 it's not available yet. Otherwise, if you call with only_if_exists=true,
 the backend should claim an error when the requested book/calendar
 cannot be opened.

Sorry, typo in my email. I meant the problem can be avoided by always
passing only_if_exist=false. In other words, never use
only_if_exists=true because it serves no useful purpose. Is that correct
or is there some legitimate usage for only_if_exists=true?


-- 
Bye, Patrick Ohly
--  
patrick.o...@gmx.de
http://www.estamos.de/


___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] e_cal/book_open() + only_if_exists

2012-02-14 Thread Milan Crha
On Tue, 2012-02-14 at 11:20 +0100, Patrick Ohly wrote:
 Sorry, typo in my email. I meant the problem can be avoided by always
 passing only_if_exist=false. In other words, never use
 only_if_exists=true because it serves no useful purpose. Is that correct
 or is there some legitimate usage for only_if_exists=true?

Hi,
I do not know how the idea of only_if_exists begun, it predates me.
The current behavior depends on each backend, some, like those remote,
doesn't count with it at all (like for the webcal backend, it cannot
create remote calendars, thus it always either fail or succeed, based on
the existence of the remote calendar).

I left it there when changing API due to EClient changes, because it
seemed to me that someone can find it useful, say with local, non-system
calendars. But maybe not.

It will be good to have opinion from other team members, either former
or current.
Bye,
Milan

___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] e_cal/book_open() + only_if_exists

2012-02-14 Thread Chenthill
On Tue, 2012-02-14 at 11:44 +0100, Milan Crha wrote:
 On Tue, 2012-02-14 at 11:20 +0100, Patrick Ohly wrote:
  Sorry, typo in my email. I meant the problem can be avoided by always
  passing only_if_exist=false. In other words, never use
  only_if_exists=true because it serves no useful purpose. Is that correct
  or is there some legitimate usage for only_if_exists=true?
 
   Hi,
 I do not know how the idea of only_if_exists begun, it predates me.
 The current behavior depends on each backend, some, like those remote,
 doesn't count with it at all (like for the webcal backend, it cannot
 create remote calendars, thus it always either fail or succeed, based on
 the existence of the remote calendar).
 
 I left it there when changing API due to EClient changes, because it
 seemed to me that someone can find it useful, say with local, non-system
 calendars. But maybe not.
 
 It will be good to have opinion from other team members, either former
 or current.
These are used atm for operations such as copy-to-calendar or while
changing the calendar from the event-page. My guess is that it could
have been introduced to avoid opening the calendar unless its already
opened to avoid any UI in-responsiveness. IMHO that option can be
removed and the async api's can be used to solve it unless there is some
other case.

- Chenthill.
   Bye,
   Milan
 
 ___
 evolution-hackers mailing list
 evolution-hackers@gnome.org
 To change your list options or unsubscribe, visit ...
 http://mail.gnome.org/mailman/listinfo/evolution-hackers


___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
http://mail.gnome.org/mailman/listinfo/evolution-hackers


[Evolution-hackers] Evolution community meeting at #evolution-meet channel - Feb 15 - 3:30 PM UTC

2012-02-14 Thread Chenthill
Hi,
  The meeting goes as follows,

* Project updates
* Discussion on queries/decisions
* Individual updates

http://live.gnome.org/Evolution/CommunityMeet

- Chenthill.

___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] Home news from the Evo/WebKit Universe

2012-02-14 Thread Chenthill
On Thu, 2012-02-09 at 09:09 -0600, Matthew Barnes wrote:
 On Thu, 2012-02-09 at 12:12 +0100, Dan Vratil wrote:
  just a short note - thanks to Milan who has decied to dig into WebKit,
  embedding widgets into WebKit webview works again.
 
 That's awesome news!  I think I suffered that same bug early on in the
 branch.  Nearly drove myself crazy trying to get a simple embedded icon
 to draw.
 
 Well done to both of you!
+1 :)

- Chenthill.
 
 Matt
 
 ___
 evolution-hackers mailing list
 evolution-hackers@gnome.org
 To change your list options or unsubscribe, visit ...
 http://mail.gnome.org/mailman/listinfo/evolution-hackers


___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] Evolution community meeting at #evolution-meet channel - Feb 15 - 3:30 PM UTC

2012-02-14 Thread Matthew Barnes
On Tue, 2012-02-14 at 17:03 +0530, Chenthill wrote:
 Hi,
   The meeting goes as follows,
 
 * Project updates
 * Discussion on queries/decisions
 * Individual updates
 
 http://live.gnome.org/Evolution/CommunityMeet


I won't be able to make the meeting tomorrow as I'll be driving all day.
Been on travel for the past week and a half, back on Thursday.

If someone could send me the meeting log I'll get it posted to the web
site as soon as I can.

Matt

___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
http://mail.gnome.org/mailman/listinfo/evolution-hackers