Ronan Le Martret [Tue, 7 Oct 2014 12:59:21 +0000 (14:59 +0200)]
use Tizen project configuration
- introspection is define in Tizen prj configuration and should be not overwritting.
- this fix do not change the package build
Change-Id: I7442abe797b80f39492e681546f39f85596bb6af
Signed-off-by: Ronan Le Martret <ronan@fridu.net>
Patrick Ohly [Tue, 16 Sep 2014 07:41:31 +0000 (09:41 +0200)]
update changelog
Patrick Ohly [Tue, 16 Sep 2014 07:36:53 +0000 (09:36 +0200)]
rebase to 3.12.2 + intel-work-3-12 patches
Previously, Tizen used the heavily modified openismus-work-3-8 branch
as upstream base. The upstream intel-work-3-12 branch contains a much
smaller set of patches not yet included in 3.12, so the rebased Tizen
package uses the official 3.12.2 as base and adds the work branch as
patches to make the difference more obvious.
Patrick Ohly [Tue, 9 Jul 2013 12:22:11 +0000 (14:22 +0200)]
update to openismus-work-3-8 branch
This introduces the cursor API for reading a subset of sorted search
results in chunks. The version gets bumped to match the one from the
master branch with a similar feature set.
Patrick Ohly [Fri, 28 Jun 2013 12:18:49 +0000 (14:18 +0200)]
minimal build
This build removes all optional dependencies and leaves only the local
file backend. Everything can still be enabled via defines.
Philippe Coval [Fri, 24 May 2013 10:37:31 +0000 (12:37 +0200)]
changes: Bump to 3.8.2
Philippe Coval [Mon, 13 May 2013 10:40:17 +0000 (12:40 +0200)]
Bump to 3.8.2
Philippe Coval [Thu, 25 Apr 2013 13:34:27 +0000 (15:34 +0200)]
changes: Bump to 3.8.1
Philippe Coval [Mon, 15 Apr 2013 14:52:51 +0000 (16:52 +0200)]
Bump to 3.8.1
Philippe Coval [Tue, 2 Apr 2013 09:32:59 +0000 (11:32 +0200)]
changes: Bump to 3.8.0
Philippe Coval [Tue, 2 Apr 2013 09:30:30 +0000 (11:30 +0200)]
spec: cleanup rpmlint=0E0W
Philippe Coval [Wed, 27 Mar 2013 13:18:32 +0000 (14:18 +0100)]
Bump to 3.8.0
Philippe Coval [Mon, 25 Mar 2013 12:39:43 +0000 (13:39 +0100)]
changes: Bump to 3.7.92
Philippe Coval [Fri, 22 Mar 2013 17:16:19 +0000 (18:16 +0100)]
Bump to 3.7.92
Philippe Coval [Thu, 14 Mar 2013 09:45:40 +0000 (10:45 +0100)]
changes: Bump to 3.7.91
Philippe Coval [Mon, 11 Mar 2013 12:08:10 +0000 (13:08 +0100)]
Bump to 3.7.91
Corentin Lecouvey [Wed, 27 Feb 2013 15:42:34 +0000 (16:42 +0100)]
changes: Bump to 3.7.90
Corentin Lecouvey [Mon, 25 Feb 2013 13:03:37 +0000 (14:03 +0100)]
spec: add libebook-contacts package
Philippe Coval [Mon, 25 Feb 2013 11:19:02 +0000 (12:19 +0100)]
Bump to 3.7.90
Philippe Coval [Wed, 13 Feb 2013 10:28:39 +0000 (11:28 +0100)]
changes: migrate to devel while it breaks folks
Philippe Coval [Wed, 13 Feb 2013 10:28:13 +0000 (11:28 +0100)]
changes: migrate to devel while it breaks folks
Jean-Benoit MARTIN [Tue, 12 Feb 2013 17:21:11 +0000 (18:21 +0100)]
Changes : Bump to 3.7.5
Jean-Benoit MARTIN [Tue, 12 Feb 2013 15:53:25 +0000 (16:53 +0100)]
Remove configure macro in build part, change group and license
Philippe Coval [Tue, 12 Feb 2013 14:25:44 +0000 (15:25 +0100)]
disable ubuntu
Philippe Coval [Fri, 8 Feb 2013 16:52:36 +0000 (17:52 +0100)]
Bump to 3.7.5
Philippe Coval [Tue, 5 Feb 2013 15:10:57 +0000 (16:10 +0100)]
changes: Initial packaging on 3.7.4 (using previous ones)
Jean-Benoit MARTIN [Tue, 5 Feb 2013 14:42:42 +0000 (15:42 +0100)]
Remove libraries not builded
Philippe Coval [Tue, 5 Feb 2013 12:49:25 +0000 (13:49 +0100)]
Initial packaging on 3.7.4 (using previous ones)
Mateusz Polrola [Wed, 9 Jul 2014 09:59:22 +0000 (11:59 +0200)]
Addressbook: Notify views after locale change
After changing locale, views of EBookBackend are not notified about
modifications made during that operation. Changing locale of
EBookBackend may result in updating EVC_X_E164 parameter of phone
numbers as its value depends on currently used locale. Calling
e_book_backend_notify_complete after successfully changing locale
notifies views about possible modifications.
Cherry picked from commit in master:
Bug 732730 - Addressbook: Notify views after locale change 0513c56
Fixes BGO: 732730
Mateusz Polrola [Mon, 2 Jun 2014 06:17:09 +0000 (08:17 +0200)]
Added support for libphonenumber >= r568
Cherry picked from commit in master:
Bug 730827 - Cannot use libphonenumber >= r568 f8e6b6f
Fixes BGO #730827
Mateusz Polrola [Fri, 30 May 2014 11:11:57 +0000 (13:11 +0200)]
Removed C++ compiler flags not supported by older GCC
Older versions of GCC don't recognize -Wnoexcept flag
Cherry picked from commit in openismus-work-3-8:
Fixed C++ compiler flags to work with various GCC versions 0c938d
Mateusz Polrola [Fri, 30 May 2014 10:45:28 +0000 (12:45 +0200)]
EBookSqlite: Added support for recording detailed changes of contacts
Cherry picked from commit in master:
EBookSqlite: Allow record detailed changes made to contacts a0d905
Fixes BGO #729251
Mateusz Polrola [Fri, 30 May 2014 10:44:35 +0000 (12:44 +0200)]
EBookSqlite: Added support for transliterated queries
Cherry picked from commits in openismus-work-3-8:
Added ETransliterator wrapper object. 0471dc
EBookQuery: Added transliteration queries. c99eba
EBookBackendSexp: Support new transliterated query types. 569d94
test-client-custom-summary.c: Added tests for transliteration queries.
ebe6d9
test-client-custom-summary.c: Added test for transliterated queries
c7be1c
test-client-cursor-operations.c: Added tests for using
transliteratedqueries f0a2d1
Tests: Added 3 new contacts and a few more query tests. 87e1d1
EBookSqlite: Not handle transliterated queries gracefully 7b6d9a
Added E_BOOK_INDEX_TRANSLIT index type 19f482
EBookSqlite: Full support for transliterated queries e59b57
Added transliterated indexes to the custom book in the test case 6b857b
Mateusz Polrola [Fri, 30 May 2014 10:03:32 +0000 (12:03 +0200)]
EDataBook: Use separate environment variable to control localed bus
Cherry picked from commits in openismus-work-3-8:
Use separate environment variable to control localed bus, instead of
EDS_TESTING 089526
Matthew Barnes [Sun, 11 May 2014 17:36:49 +0000 (13:36 -0400)]
NEWS update for 3.12.2 release.
Milan Crha [Fri, 9 May 2014 10:39:59 +0000 (12:39 +0200)]
Bug #728973 - [IMAPX] Recover after store summary version mismatch
Milan Crha [Fri, 9 May 2014 10:03:35 +0000 (12:03 +0200)]
camel_imapx_conn_manager_get_connection: Ensure only connected servers are returned
It could happen, still, that the CamelIMAPXServer is disconnected or shutdown
in the list of connections and it is being returned, which is not
correct, thus better never return such dead connections. Once
the server's state is set to shutdown it is automatically removed
from the list, but here plays a role thread interleaving or something,
which does the remove slightly later than expected.
Pau Iranzo [Wed, 7 May 2014 23:09:00 +0000 (01:09 +0200)]
[l10n] Update Catalan translation
Milan Crha [Tue, 6 May 2014 10:40:04 +0000 (12:40 +0200)]
[IMAPX] Do not hold queue_lock when calling imapx_server_ref_job()
Holding the queue_lock when calling imapx_server_ref_job() can cause
deadlock when the store has opened more than one connection and each
is searching for a duplicate job, both/all servers holding respective
queue lock prevents in a search of active jobs in them, both waiting
for a release of the queue_lock of the other server.
Milan Crha [Tue, 6 May 2014 10:13:19 +0000 (12:13 +0200)]
[IMAPX] Missing folder operation finish call at imapx_search_process_criteria()
Once the server is opened for certain folder it should also claim
the folder operation is over, which was missing here.
Milan Crha [Mon, 5 May 2014 17:08:09 +0000 (19:08 +0200)]
[IMAPx] Try to address "Empty cache file" error
This is rather a wild guess. The UI can be filled with the "Empty
cache file" error, which means that the just-downloaded message
from the server had 0-length size. This can happen when the operation
was cancelled by a user in a "correct time", probably, thus let's
try to check for cancelled operation too - just in case.
Milan Crha [Mon, 5 May 2014 16:24:13 +0000 (18:24 +0200)]
imapx_register_job: Try reconnect if the server is shutdown/disconnected
There could sometimes bubble "Not authenticated" error up to the UI,
which didn't look nicely and was not necessary, because in the case
of the server being shutdown or disconnected it's easy to reconnect.
The "Not authenticated" error message is still used, only the error
code is different, it tells the caller to reconnect.
Milan Crha [Mon, 5 May 2014 15:49:13 +0000 (17:49 +0200)]
camel_imapx_conn_manager_close_connections: Avoid possible deadlock
A deadlock could happen by calling CamelIMAPXServer functions with
the CamelIMAPXConnManager write lock locked, where there was a race
condition between the write lock and the server's queue lock. Closing
all the connections without holding the write lock should avoid
the deadlock.
Milan Crha [Fri, 2 May 2014 14:03:48 +0000 (16:03 +0200)]
Bug #728414 - Deliver ESource data changes on idle
The data D-Bus property change is received in a gdbus dedicated thread,
thus any ESource property changes, caused by reload of the data file,
are also delivered in the same gdbus dedicated thread, which is wrong
when there are still left property bindings which update gtk widgets.
Update the ESource data on idle, to workaround this issue.
Milan Crha [Fri, 2 May 2014 13:31:22 +0000 (15:31 +0200)]
[IMAPx] Set local cache to never expire
This is not like an HTTP cache, the downloaded messages might not be
discarded periodically, because, apart of other things, it breaks
the "Copy for offline usage" option.
Milan Crha [Fri, 2 May 2014 10:51:04 +0000 (12:51 +0200)]
Bug #728167 - [IMAPx] Message headers downloaded twice
Milan Crha [Wed, 30 Apr 2014 13:24:16 +0000 (15:24 +0200)]
Bug #720154 - [NNTP] Silently recover from "Connection reset by peer" error
The second, and final, part for this bug report, involving NNTP.
Milan Crha [Wed, 30 Apr 2014 11:10:08 +0000 (13:10 +0200)]
Bug #720154 - [IMAPx] Silently recover from "Connection reset by peer" error
This is a partial commit for the bug, to address "Connection reset by peer"
errors, and even the BYE response from the server, more gracefully, like
reconnecting to the server and re-issue the job. There will be at least
one more commit for this bug report, for an NNTP provider.
Matej Urbančič [Mon, 28 Apr 2014 19:05:52 +0000 (21:05 +0200)]
Updated Slovenian translation
Milan Crha [Fri, 25 Apr 2014 06:33:48 +0000 (08:33 +0200)]
Remove unused imapx_unmark_folder_subscribed()
I overlooked to remove it in one of the previous related commits.
Milan Crha [Fri, 25 Apr 2014 06:11:03 +0000 (08:11 +0200)]
imapx_connect_sync: Use connection manager to connect to the server
This function opened a new connection to the server, but immediately
dropped it, which, apart of waste of resources, can cause trouble
when other parts asked for the connection to the server, but those
using the connection manager, which is the correct way of dealing
with connections to the server now. The previous behaviour can cause
a race condition on the imapx_store->priv->connecting_server, which
led later to runtime warnings and misbehaviour in general.
Milan Crha [Fri, 25 Apr 2014 05:38:42 +0000 (07:38 +0200)]
Do not return CAMEL_AUTHENTICATION_REJECTED on runtime errors
Returning CAMEL_AUTHENTICATION_REJECTED on runtime errors also means
that a stored password is forgotten and a user is asked for it,
which is eventually not used, because the runtime error can repeat.
The correct return code is CAMEL_AUTHENTICATION_ERROR in this case.
Milan Crha [Thu, 24 Apr 2014 16:55:08 +0000 (18:55 +0200)]
camel_network_service_can_reach_sync: Forgotten local_error propagate
Milan Crha [Thu, 24 Apr 2014 10:21:46 +0000 (12:21 +0200)]
Fix two calendar-related memory leaks
Milan Crha [Wed, 23 Apr 2014 13:06:44 +0000 (15:06 +0200)]
[IMAPx] Use case-insensitive compares and hash table functions
Almost all attributes, responses and commands in general are
case-insensitive in IMAPx, thus even the code can "normalize"
the respective strings to proper case, it's safer to use
case-insensitive compares and equal-functions in hash tables
where applicable.
Milan Crha [Wed, 23 Apr 2014 13:01:48 +0000 (15:01 +0200)]
[IMAPx] Keep store summary's flags in sync with mailbox flags
This way mailbox attributes like \NoSelect gets in sync with
folder info flags, which makes it work properly, the non-selectable
folders will not be updated (tested with GMail's "[GMail]" folder).
Milan Crha [Wed, 23 Apr 2014 10:33:54 +0000 (12:33 +0200)]
Bug #728737 - [IMAPx] Namespaces not recognized in subscription list
Dirgita [Tue, 22 Apr 2014 14:36:20 +0000 (14:36 +0000)]
Updated Indonesian translation
Milan Crha [Tue, 22 Apr 2014 11:12:07 +0000 (13:12 +0200)]
[ownCloud backend] Check color format before using it
Calendar-based sources have always overwritten color, to ensure
the color defined on the server matches the one used in evolution,
but there was not checked whether the color is of the right format,
which may eventually cause reset of a user-specified color to
an invalid one, showing the calendar in an odd color.
Christian Schaarschmidt [Fri, 18 Apr 2014 17:41:19 +0000 (19:41 +0200)]
Bug #728087 - Use-after-free in contact address switch in composer
Milan Crha [Fri, 18 Apr 2014 13:20:05 +0000 (15:20 +0200)]
[IMAPx] Manage mailboxes and namespaces in CamelIMAPXStore
The mailboxes and namespaces had been managed in CamelIMAPXServer,
which was perfectly fine before the concurrent connections were added,
because each of the connection could know only on part of the two,
which is not enough. This could let to a failure of folder update
(the case I face were when an empty folder had added old messages,
where one connection new there are some messages, but the connection
serving this update thought there are no messages, thus it didn't try
to download anything).
Milan Crha [Thu, 17 Apr 2014 16:34:00 +0000 (18:34 +0200)]
[IMAPx] Automatically limit maximum number of concurrent connections
In case a user has set more connections than the server has allowed,
the IMAPx may limit the connection count automatically, instead of
panic and close already established connections.
Milan Crha [Thu, 17 Apr 2014 12:40:05 +0000 (14:40 +0200)]
Set properly children availability for newly created local folders
This is to help with a UI regression mentioned at
https://bugzilla.gnome.org/show_bug.cgi?id=722698#c13
with the local folders.
Milan Crha [Wed, 16 Apr 2014 17:39:55 +0000 (19:39 +0200)]
Bug #727598 - [IMAPX] Fails to list folders without NAMESPACE capability
Milan Crha [Wed, 16 Apr 2014 16:56:10 +0000 (18:56 +0200)]
[IMAPX] Do not expose user credentials in debug logs
And skip trailing \r\n in the io logs as well.
Daniel Mustieles [Tue, 15 Apr 2014 12:17:37 +0000 (14:17 +0200)]
Updated Spanish translation
Matthew Barnes [Sun, 13 Apr 2014 15:01:46 +0000 (11:01 -0400)]
Post-release version bump.
Matthew Barnes [Sun, 13 Apr 2014 14:42:59 +0000 (10:42 -0400)]
NEWS update for 3.12.1 release.
Milan Crha [Fri, 11 Apr 2014 10:45:11 +0000 (12:45 +0200)]
camel-imapx-conn-manager.h/c: Correct LGPL license notices
Milan Crha [Thu, 10 Apr 2014 13:56:27 +0000 (15:56 +0200)]
Always reject revoked certificates
If there is recognized a revoked certificate being used for a secure
connection, then reject the connection immediately, for security reasons.
This behaviour cannot be overwritten with a user's trust.
Milan Crha [Thu, 10 Apr 2014 12:02:59 +0000 (14:02 +0200)]
Bug #722275 - IMAPX: Incorrectly reads astring values (mailbox names)
Milan Crha [Thu, 10 Apr 2014 07:34:36 +0000 (09:34 +0200)]
[IMAPx] Do not open each folder on LIST/LSUB responses
It has sever performance issues, especially with large folders,
which load (part of) the folder summary into memory too.
Milan Crha [Thu, 10 Apr 2014 06:18:02 +0000 (08:18 +0200)]
[IMAPx] Enable concurrent connections
This returns back CamelIMAPXConnManager, and enables use of concurrent
connections with it, which helps with response time of user requests.
Milan Crha [Wed, 9 Apr 2014 10:00:30 +0000 (12:00 +0200)]
Claim missing user name when authenticating
When user's CalDAV/HTTP calendar or WebDAV address book had missing
set login user name, the UI did not show anything useful, the worse
it reasked for a password which were not used at all, because libsoup's
session on the factory side failed in authentication due to runtime
check about non-NULL user name. This makes it return "Authentication
required" error to the UI, instead of asking for the password.
Milan Crha [Wed, 2 Apr 2014 17:12:53 +0000 (19:12 +0200)]
CamelIMAPXServer not shut down properly on network outage
Even the CamelIMAPXStore had the server removed from its references,
the instance had been still alive in the background, when the network
changed and the destination host was unreachable.
Gianvito Cavasoli [Thu, 27 Mar 2014 07:42:54 +0000 (08:42 +0100)]
[l10n] Updated Italian translation.
Milan Crha [Wed, 26 Mar 2014 10:45:13 +0000 (11:45 +0100)]
CamelIMAPXSettings: Real trash/junk path properties leak
Milan Crha [Wed, 26 Mar 2014 07:23:33 +0000 (08:23 +0100)]
Bug #726060 - Password prompt initiated before D-Bus name acquired
Inaki Larranaga Murgoitio [Tue, 25 Mar 2014 19:54:55 +0000 (20:54 +0100)]
Updated Basque language
Gianvito Cavasoli [Tue, 25 Mar 2014 08:45:48 +0000 (09:45 +0100)]
[l10n] Updated Italian translation.
Jiro Matsuzawa [Mon, 24 Mar 2014 20:07:48 +0000 (05:07 +0900)]
l10n: Update Japanese translation
Milan Crha [Mon, 24 Mar 2014 18:39:45 +0000 (19:39 +0100)]
e_source_registry_dup_unique_display_name: Fix a memory leak
g_list_remove_link() doesn't delete the link itself, thus it made
it leaked here.
Philip Withnall [Mon, 24 Mar 2014 15:15:12 +0000 (16:15 +0100)]
Bug #660535 - Google contact groups/categories not showing in category list
This means they should automatically appear in Evolution’s category
selector and editor. Previously, only new groups added manually by
editing a Google contact in Evolution using the category editor were
shown in the list.
Milan Crha [Mon, 24 Mar 2014 14:55:33 +0000 (15:55 +0100)]
Bug #726767 - [imapx] Implement CamelFolder::get_message_cached()
Xavier Claessens [Mon, 24 Mar 2014 14:51:05 +0000 (15:51 +0100)]
Bug #726788 - Crash on malformed vCard
If the vcard is "END:VCARD" then the attribute is added to
the list, then freed. So next time we iterate on attributes it
will read freed memory and crash.
Matthew Barnes [Mon, 24 Mar 2014 12:01:28 +0000 (08:01 -0400)]
Add book migration database for 3.12.
(cherry picked from commit
45afa54dcffdc028256b3c2deaa1619f046a9613)
Milan Crha [Mon, 24 Mar 2014 09:32:01 +0000 (10:32 +0100)]
Update ChangeLog URL
Matthew Barnes [Sun, 23 Mar 2014 01:14:19 +0000 (21:14 -0400)]
Post-release version bump.
Matthew Barnes [Sun, 23 Mar 2014 00:08:15 +0000 (20:08 -0400)]
NEWS update for 3.12.0 release.
Christian Kirbach [Sat, 22 Mar 2014 18:57:55 +0000 (18:57 +0000)]
Updated German translation
Gábor Kelemen [Fri, 21 Mar 2014 15:56:07 +0000 (15:56 +0000)]
Updated Hungarian translation
Milan Crha [Fri, 21 Mar 2014 13:03:34 +0000 (14:03 +0100)]
Bug #726723 - Data loss on a message copy to an IMAPx folder
Ask H. Larsen [Thu, 20 Mar 2014 20:05:55 +0000 (21:05 +0100)]
Updated Danish translation
Milan Crha [Wed, 19 Mar 2014 10:26:37 +0000 (11:26 +0100)]
Bug #726681 - Check GConf XDG folder for migration too
Duarte Loreto [Tue, 18 Mar 2014 23:50:54 +0000 (23:50 +0000)]
Updated Portuguese translation
Мирослав Николић [Mon, 17 Mar 2014 22:21:48 +0000 (23:21 +0100)]
Updated Serbian translation
Matthew Barnes [Sun, 16 Mar 2014 14:06:33 +0000 (10:06 -0400)]
Post-release version bump.
Matthew Barnes [Sun, 16 Mar 2014 13:56:21 +0000 (09:56 -0400)]
NEWS update for 3.11.92 release.
Milan Crha [Fri, 14 Mar 2014 10:21:48 +0000 (11:21 +0100)]
Add missing colon into a developer-documentation comment
Milan Crha [Fri, 14 Mar 2014 09:15:37 +0000 (10:15 +0100)]
e_cal_backend_get_timezone_finish: Do not panic when the timezone cannot be found
When the specific timezone could not be found, the function threw a runtime
warning, and did not set an error, which further in the call made fail also
data_cal_complete_get_timezone_cb(), which has a check for "either have
a timezone object or an error", which finally prevented a correct finish
of the D-Bus call, thus the client side which asked for an unknown timezone
was left stuck, waiting for a response from the server which was not
interested in the call anymore.
This behaviour could lead to a UI freeze, as had been reported at:
https://bugzilla.redhat.com/show_bug.cgi?id=1075461