Christophe Dumez [Fri, 27 May 2011 13:30:39 +0000 (15:30 +0200)]
Bug #650952 - Remove unknown EContact field name runtime warning
Christophe Dumez [Fri, 27 May 2011 13:27:30 +0000 (15:27 +0200)]
Bug #650950 - entry_compare() should iterate over attributes with the same name
Extend entry_compare() to iterate over attributes that have
the same name (e.g. X-HOBBY) so that it can match any of
them, not just the first one in the vCard.
Matthew Barnes [Fri, 27 May 2011 12:01:18 +0000 (14:01 +0200)]
Bug #648736 - Crash when adding default vcard for newly created DB
Daniel Mustieles [Thu, 26 May 2011 18:05:01 +0000 (20:05 +0200)]
Updated Spanish translation
Matthew Barnes [Thu, 26 May 2011 15:34:28 +0000 (11:34 -0400)]
Automated Gtk-Doc updates from backend API changes.
Matthew Barnes [Thu, 26 May 2011 15:21:37 +0000 (11:21 -0400)]
Reduce diff noise with account-mgmt branch.
Mostly silly code cleanup type stuff, so I can see the important changes
more easily.
Raul Gutierrez Segales [Thu, 26 May 2011 11:50:52 +0000 (12:50 +0100)]
Vala bindings: metadata helpers for EContactAddress
Raul Gutierrez Segales [Thu, 26 May 2011 11:45:47 +0000 (12:45 +0100)]
Vala bindings: metadata helpers for EContactName
Raul Gutierrez Segales [Thu, 26 May 2011 11:34:16 +0000 (12:34 +0100)]
Vala bindings: add path to dependency GIRs
Chenthill Palanisamy [Thu, 26 May 2011 11:31:32 +0000 (17:01 +0530)]
CamelDB: Initiate a transaction before writing into db.
Chenthill Palanisamy [Thu, 26 May 2011 10:26:07 +0000 (15:56 +0530)]
Bug 640054 - CamelDB: do not read the db while a trasaction is in progress.
Fix the dead-lock caused due to transaction (DB WRITE_LOCK) and summary lock.
Chenthill Palanisamy [Thu, 26 May 2011 08:09:30 +0000 (13:39 +0530)]
CamelDB: Ensure that begin_transaction is called before adding queries to a transaction
Chenthill Palanisamy [Thu, 26 May 2011 07:55:32 +0000 (13:25 +0530)]
CamelFolderSummary: Remove undefined function camel_folder_summary_header_load from header file
Chenthill Palanisamy [Thu, 26 May 2011 07:53:49 +0000 (13:23 +0530)]
CamelDB: Use camel_db_select while retrieving the folder version
Chenthill Palanisamy [Wed, 25 May 2011 07:40:53 +0000 (13:10 +0530)]
CamelDB: use a RW lock to prevent reads while write is in progress
Milan Crha [Thu, 26 May 2011 09:11:09 +0000 (11:11 +0200)]
Bug #589495 - Search folder by Size (KB) counts bytes, not KB
Jorge González [Wed, 25 May 2011 20:46:15 +0000 (22:46 +0200)]
Updated Spanish translation
Daniel Mustieles [Wed, 25 May 2011 17:10:46 +0000 (19:10 +0200)]
Updated Spanish translation
Matthew Barnes [Mon, 23 May 2011 16:51:00 +0000 (12:51 -0400)]
camel_store_create_folder_sync(): Add a status message.
Milan Crha [Mon, 23 May 2011 15:59:26 +0000 (17:59 +0200)]
Fix issues reported by intltool-update
Milan Crha [Mon, 23 May 2011 14:49:19 +0000 (16:49 +0200)]
Bug #550414 - Corruption of mailbox and can't expunge trash
The second attempt - be more aggressive with locking
Seán de Búrca [Mon, 23 May 2011 12:07:13 +0000 (06:07 -0600)]
Updated Irish translation.
Seán de Búrca [Mon, 23 May 2011 10:43:15 +0000 (04:43 -0600)]
Remove groupwise files from POTFILES.in
Milan Crha [Mon, 23 May 2011 09:45:20 +0000 (11:45 +0200)]
New EClient/EBookClient/ECalClient API deprecating EBook/ECal
Matthew Barnes [Sun, 22 May 2011 02:59:58 +0000 (22:59 -0400)]
Bug 650524 - Use g_unix_signal_add_watch_full() for SIGTERM
We can't call anything safely except write() inside a signal handler,
g_unix_signal_add_watch_full() handles this for us nicely.
Matthew Barnes [Sun, 22 May 2011 00:54:47 +0000 (20:54 -0400)]
Handle SIGTERM instead of SIGQUIT.
SIGQUIT is supposed to produce a core dump. We don't want that.
Chenthill Palanisamy [Fri, 20 May 2011 10:29:29 +0000 (15:59 +0530)]
IMAPX: optionally download the messages in descending order.
TODO: Add advanced UI options for the same.
Document all the url parameters in a common place.
Raul Gutierrez Segales [Wed, 18 May 2011 22:00:15 +0000 (23:00 +0100)]
Vala bindings: mark metadata file as EXTRA_DIST
Raul Gutierrez Segales [Wed, 18 May 2011 20:08:02 +0000 (21:08 +0100)]
Vala bindings: use metadata to get the correct types for signals
The contacts_{added,removed,changed} signals were specified with
the wrong parameter types (void * instead of GLib.List<void *>) in the
generated VAPI file for libebook.
Raul Gutierrez Segales [Wed, 18 May 2011 20:07:06 +0000 (21:07 +0100)]
libedataserver: break down long lines of the Introspection part
Raul Gutierrez Segales [Wed, 18 May 2011 20:05:43 +0000 (21:05 +0100)]
libebook: add --pkg-export to get a package name in the generated GIR
Raul Gutierrez Segales [Wed, 18 May 2011 18:59:10 +0000 (19:59 +0100)]
libebook: annotate params for e_book_get_{book_view_async,self}
Raul Gutierrez Segales [Wed, 18 May 2011 18:28:37 +0000 (19:28 +0100)]
libebook: annotate more methods
Raul Gutierrez Segales [Wed, 18 May 2011 16:59:48 +0000 (17:59 +0100)]
Vala bindings: add missing case for xml. -> Xml. transformation
Raul Gutierrez Segales [Wed, 18 May 2011 16:43:47 +0000 (17:43 +0100)]
Vala bindings: help vapigen get the correct Xml names
Raul Gutierrez Segales [Wed, 18 May 2011 16:16:06 +0000 (17:16 +0100)]
Add the correct headers to libedataserver's GIR file
Matthew Barnes [Tue, 17 May 2011 14:09:23 +0000 (10:09 -0400)]
CamelService: Add async query_auth_types() functions.
Breaks ABI since sizeof(CamelServiceClass) changes.
I've already bumped the libcamel soname since 3.1.1.
Raul Gutierrez Segales [Wed, 18 May 2011 12:09:01 +0000 (13:09 +0100)]
Add vala/ to DIST_SUBDIRS
Raul Gutierrez Segales [Mon, 16 May 2011 13:19:26 +0000 (14:19 +0100)]
Fixed annotations in libedataserver
Milan Crha [Tue, 17 May 2011 09:30:38 +0000 (11:30 +0200)]
Bug #646898 - IMAPX should invalidated cache when uidvalidity changes
Patrick Ohly [Mon, 16 May 2011 09:21:04 +0000 (11:21 +0200)]
addressbook file backend: libdb must be initialized for concurrent read/write
Very bad performance (100% CPU load, several minutes run time) were
seen for multiple concurrent writes. gdb shows that libdb is
apparently busy polling while writing.
The libdb API docs for DB_ENV->open() imply that either DB_INIT_CDB or
DB_INIT_LOCK must be used in apps which are not read-only, like
EDS. This patch adds DB_INIT_CDB because it is simple and fixes the
performance problem.
In some rare cases, DB_INIT_LOCK might provide better performance by
allowing concurrent writes of independent data, but that seems too
complicated for not enough gain right now (must check for deadlocks).
Raul Gutierrez Segales [Sat, 14 May 2011 13:59:36 +0000 (14:59 +0100)]
Show the Introspection and Vala bindings in configure summary
Raul Gutierrez Segales [Sat, 14 May 2011 12:22:45 +0000 (13:22 +0100)]
Register EVCardAttributeParam as a boxed type
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=650169
Raul Gutierrez Segales [Sat, 14 May 2011 11:33:25 +0000 (12:33 +0100)]
Register EVCardAttribute as a boxed type
Without this EVCardAttribute can't be picked up by
g-ir-scanner when trying to generate the GIR file for
libebook.
Also, EVCardAttribute will need to be used by applications
trying to write back to libebook via an introspected binding.
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=650169
Matthew Barnes [Sat, 14 May 2011 18:37:19 +0000 (14:37 -0400)]
Camel: Fetch quota information asynchronously.
This bumps the libcamel soname again.
It turns out, camel_folder_get_quota_info() blocks. So convert it to be
asynchronous like the other blocking functions, with corresponding class
methods:
camel_folder_get_quota_info_sync()
camel_folder_get_quota_info()
camel_folder_get_quota_finish()
Also update the IMAP provider to make fetching IMAP quotas cancellable.
This is the only provider that currently implements this method.
Matthew Barnes [Sat, 14 May 2011 11:02:04 +0000 (07:02 -0400)]
Include git.mk in vala/Makefile.am.
Matej Urbančič [Sat, 14 May 2011 19:03:33 +0000 (21:03 +0200)]
Updated Slovenian translation
Raul Gutierrez Segales [Sat, 14 May 2011 13:33:55 +0000 (14:33 +0100)]
Add missing annotations in libebook
Matthew Barnes [Sat, 14 May 2011 10:53:24 +0000 (06:53 -0400)]
Bug 650105 - Deadlock in GroupWise connector
Instead of acquiring the CamelSession lock for the entire add_service()
method, only acquire it when directly accessing the shared 'services'
hash table. This allows camel_session_add_service() calls to be nested,
which is what CamelGroupwiseStore does.
Raul Gutierrez Segales [Fri, 13 May 2011 01:41:57 +0000 (21:41 -0400)]
Bug 649762 - Add support for Vala bindings
Raul Gutierrez Segales [Fri, 13 May 2011 01:30:02 +0000 (21:30 -0400)]
Bug 649761 - Add gobject-introspection build system to libebook
Extended Rob Bradford's initial commit to add libedataserver as a
dependency for g-ir-scanner when processing libebook.
Matthew Barnes [Thu, 12 May 2011 01:46:46 +0000 (21:46 -0400)]
camel_store_get_folder_info_sync(): Add a status message.
Vibha Yadav [Wed, 11 May 2011 09:59:22 +0000 (15:29 +0530)]
BNC 683083 - TO:field – autocomplete does not work correct
Calling up library for textual representaion of destination instead of
address format for text operations.
Matthew Barnes [Tue, 10 May 2011 23:34:15 +0000 (19:34 -0400)]
camel_folder_get_message(): Show folder name in status message
Matej Urbančič [Tue, 10 May 2011 19:22:40 +0000 (21:22 +0200)]
Updated Slovenian translation
Milan Crha [Mon, 9 May 2011 17:15:33 +0000 (19:15 +0200)]
Bug #648454 - Subfolders disappear when moving a folder
David Woodhouse [Mon, 9 May 2011 13:37:27 +0000 (14:37 +0100)]
Bug 649433 - e-calendar-factory SIGSEGV while reading hash_table
The root cause of this issue was modules being loaded with RTLD_GLOBAL, and
having conflicting "global" symbols.
There is no reason for modules to be exporting their own symbols to the rest
of the process; add the G_MODULE_BIND_LOCAL flag to the g_module_open() call
to prevent that.
Matthew Barnes [Sun, 8 May 2011 16:45:23 +0000 (12:45 -0400)]
Coding style cleanups.
Matthew Barnes [Sun, 8 May 2011 15:54:39 +0000 (11:54 -0400)]
Post-release version bump.
Matthew Barnes [Sun, 8 May 2011 14:33:45 +0000 (10:33 -0400)]
NEWS update for 3.1.1 release.
Matthew Barnes [Sun, 8 May 2011 14:33:58 +0000 (10:33 -0400)]
Update API docs.
Raul Gutierrez Segales [Sun, 8 May 2011 15:46:45 +0000 (11:46 -0400)]
Bug 649684 - GIR files should contain <package name=../>
Matthew Barnes [Sun, 8 May 2011 15:35:46 +0000 (11:35 -0400)]
Adapt Camel tests to recent API changes.
Matthew Barnes [Sat, 7 May 2011 16:20:18 +0000 (12:20 -0400)]
Whitespace and coding style cleanups.
Matthew Barnes [Sat, 7 May 2011 16:10:53 +0000 (12:10 -0400)]
Forgot to change property name for folder creation.
Matthew Barnes [Sat, 7 May 2011 14:49:05 +0000 (10:49 -0400)]
Rename "name" property of CamelFolder to "display-name".
To correspond with CamelFolderInfo changes.
Matthew Barnes [Sat, 7 May 2011 13:21:44 +0000 (09:21 -0400)]
Rename "name" field of CamelFolderInfo to "display_name".
Long as I'm already breaking the CamelFolderInfo struct, this helps
clarify the "name" field's intended purpose.
Matthew Barnes [Sat, 7 May 2011 12:13:44 +0000 (08:13 -0400)]
Remove "uri" field of CamelFolderInfo.
With this change, Camel no longer deals with folder URIs.
Matthew Barnes [Sat, 7 May 2011 05:01:09 +0000 (01:01 -0400)]
camel_vee_folder_new(): Add argument guards.
Matthew Barnes [Fri, 6 May 2011 17:27:27 +0000 (13:27 -0400)]
CamelVeeStore: Change handling of CHANGE_NOSELECT flag.
Add a CAMEL_FOLDER_NOSELECT flag to the CamelFolderInfo instead of
setting a "noselect" param on the URI. The URI is about to go away.
Matthew Barnes [Fri, 6 May 2011 17:22:59 +0000 (13:22 -0400)]
Remove CAMEL_STORE_SUMMARY_FRAGMENT flag.
Not used anywhere.
Matthew Barnes [Fri, 6 May 2011 17:16:16 +0000 (13:16 -0400)]
CamelSaslPOPB4SMTP: Fix a compiler warning.
Matthew Barnes [Fri, 6 May 2011 17:13:03 +0000 (13:13 -0400)]
CamelFolder: Remove "uri" property.
Also remove camel_folder_get_uri().
Too bad I just added this for 3.0. It was handy at the time, but Camel
won't be using folder URIs anymore. That's Evolution's domain now.
Matthew Barnes [Fri, 6 May 2011 17:07:16 +0000 (13:07 -0400)]
CamelFolderSearch: Hack to deal with "message-location".
This is the one place in Camel that needs to know about Evolution folder
URI formats. It's fine for now; it's marked with a FIXME comment and I'll
clean it up properly later.
Matthew Barnes [Fri, 6 May 2011 15:23:08 +0000 (11:23 -0400)]
Camel: Store UIDs in X-Evolution-Source headers.
CamelStore UIDs are easier to deal with than URLs.
For backward-compatibility, CamelFilterDriver can still handle
X-Evolution-Source headers with URLs.
Matthew Barnes [Fri, 6 May 2011 14:36:15 +0000 (10:36 -0400)]
Remove camel_filter_driver_filter_mbox()
Function is no longer needed since the switch to Maildir.
Dan Vrátil [Thu, 5 May 2011 14:01:43 +0000 (10:01 -0400)]
Bug 332497 - Add Edit -> Available Categories
Split ECategoriesDialog into smaller widgets:
ECategoriesEditor
ECategoriesSelector
ECategoryEditor
Vibha Yadav [Thu, 5 May 2011 07:16:09 +0000 (12:46 +0530)]
Bug 649365 - Creation of new repository evolution-groupwise
Removing groupwise plugin code from evolution.
Please find the evolution-groupwise repository
http://git.gnome.org/browse/evolution-groupwise
Kjartan Maraas [Wed, 4 May 2011 08:56:48 +0000 (10:56 +0200)]
Updated Norwegian bokmål translation
Kjartan Maraas [Fri, 1 Apr 2011 17:04:32 +0000 (19:04 +0200)]
Mark the param spec strings for translation correctly.
Matthew Barnes [Tue, 3 May 2011 15:22:46 +0000 (11:22 -0400)]
Remove camel_store_folder_uri_equal().
Function was not used by Camel or any of its providers. Evolution
now has an equivalent function named e_mail_folder_uri_equal().
Abduxukur Abdurixit [Mon, 2 May 2011 16:04:28 +0000 (18:04 +0200)]
Added UG translation
Abduxukur Abdurixit [Sun, 1 May 2011 21:08:00 +0000 (23:08 +0200)]
Added UG translation
Fran Diéguez [Sat, 30 Apr 2011 21:13:50 +0000 (23:13 +0200)]
Updated galician translations
Daniel Mustieles [Fri, 29 Apr 2011 17:38:10 +0000 (19:38 +0200)]
Updated Spanish translation
Matthew Barnes [Fri, 29 Apr 2011 14:14:01 +0000 (10:14 -0400)]
Fix definition of CamelProvider.url_equal().
Functions of type GCompareFunc are supposed to return -1/0/1 as in
strcmp(). CamelProvider.url_equal() should instead be a GEqualFunc
since it returns TRUE or FALSE.
Function signatures are identical except for the return type, but
gboolean is a type of gint so no changes required for providers.
Matthew Barnes [Fri, 29 Apr 2011 13:11:38 +0000 (09:11 -0400)]
Revise camel_session_get_service_by_url().
Some groupware providers use the same URL for their mail store and mail
transport services, so add a CamelProviderType argument so callers have
to explicitly say which service type they want.
The need for this was discovered in bug #648841 for evolution-mapi.
Milan Crha [Thu, 28 Apr 2011 08:49:24 +0000 (10:49 +0200)]
Bug #648844 - IMAP Error: LIST failed: wildcards not permitted in username
David Woodhouse [Wed, 27 Apr 2011 23:35:56 +0000 (00:35 +0100)]
Bug 628142 - Fix handling of simultaneous get_message requests
Drop the hash table of EFlags completely. It's broken, because the UID
we use as the hash key isn't actually unique; the same UID can exist in
multiple folders. And the lifetime issues on the EFlag weren't cleanly
solvable (yeah, we can add a refcounting wrapper, but ick).
We were *already* using imapx_is_job_in_queue() to check *properly* if
there was an existing fetch. So just implement a simple 'fetch counter'
with a GCond and a corresponding GMutex, bump that count by one each
time any fetch completes, and use the GCond when waiting for a *specific*
fetch to complete, inside a while (imapx_is_job_in_queue()) loop.
Andrej Žnidaršič [Wed, 27 Apr 2011 17:46:52 +0000 (19:46 +0200)]
Updated Slovenian translation
Matthew Barnes [Sat, 23 Apr 2011 13:45:35 +0000 (09:45 -0400)]
Replace CamelSessionThreadMsg with a more modern API.
This introduces camel_session_submit_job() to replace the older
CamelSessionThreadMsg API, providing a simpler mechanism for providers
to initiate low-priority background jobs. Jobs can be submitted from
any thread, but execution of the job is always as follows:
1) A "job-started" signal is emitted from the thread in which the
CamelSession was created. This is typically the same thread that
hosts the global default GMainContext, or "main" thread.
2) The callback function passed to camel_session_submit_job() is invoked
from a different thread where it's safe to call synchronous functions.
3) Once the callback has returned, a "job-finished" signal is emitted
from the same thread as "job-started" was emitted.
4) If a GDestroyNotify callback was passed to camel_session_submit_job(),
it is invoked and passed the callback closure so it can be freed.
The signature of the callback function is easier to work with:
void (*CamelSessionCallback) (CamelSession *session,
GCancellable *cancellable,
gpointer user_data,
GError **error)
The cancellable is actually a CamelOperation, so status messages can be
pushed and popped and progress reported through the CamelOperation API.
The GError pointer is never NULL, so you can safely dereference it in
the callback function to check for errors.
Matthew Barnes [Sat, 23 Apr 2011 00:22:34 +0000 (20:22 -0400)]
Remove unused camel_session_thread_wait().
Matthew Barnes [Fri, 22 Apr 2011 21:10:23 +0000 (17:10 -0400)]
Remove unused CamelTimeoutCallback typedef.
Matthew Barnes [Fri, 22 Apr 2011 13:04:43 +0000 (09:04 -0400)]
CamelStore: Remove a dead assignment.
Matthew Barnes [Fri, 22 Apr 2011 12:49:51 +0000 (08:49 -0400)]
CamelSession: Remove some leftover debug messages.
Matthew Barnes [Fri, 22 Apr 2011 12:48:51 +0000 (08:48 -0400)]
CamelStore: Chain up in the constructed() method
Forgot to add that part yesterday. Nothing works without it.
Matthew Barnes [Tue, 19 Apr 2011 19:32:47 +0000 (15:32 -0400)]
Rework how CamelServices are added to CamelSession.
* Give CamelServices a simple unique ID string. That will be its
identity from now on, not its URL.
* Split adding a CamelService and retrieving a CamelService into two
separate operations. Adding a CamelService requires both a UID and
CamelURL (for now), retrieving a CamelService just requires the UID.
* CamelService now implements the GInitable interface, replacing its
construct() method.
Matthew Barnes [Sun, 17 Apr 2011 23:59:15 +0000 (19:59 -0400)]
Define a boxed GType for CamelURL.
And clean up the header file.
Milan Crha [Wed, 20 Apr 2011 15:24:34 +0000 (17:24 +0200)]
Bug #645498 - [maildir] Fails to show already read messages again
Wouter Bolsterlee [Tue, 19 Apr 2011 19:14:13 +0000 (21:14 +0200)]
Update the translations now that bug #646619 is fixed