summaryrefslogtreecommitdiffstats
path: root/common
Commit message (Collapse)AuthorAge
* Fixed getUids by type filtering.Christian Mollekopf2017-09-07
| | | | | We used to simply return all uids. Requires "sinksh upgrade"
* Avoid hiding the index() functionChristian Mollekopf2017-09-02
|
* Ensure we monitor resources for status changes that have been createdChristian Mollekopf2017-09-02
| | | | | | after the query. This fixes status monitoring when creating a new account.
* Use Q_GLOBAL_STATIC for threadsafety.Christian Mollekopf2017-08-31
| | | | | This resolves the following warning on shutdown it seems: "QObject::connect: No such signal QObject::aboutToClose() in ../../include/QtCore/5.9.1/QtCore/private/../../../../../src/corelib/io/qtextstream_p.h:75"
* Avoid non threadsafe initialization.Christian Mollekopf2017-08-31
| | | | | | local static initialization is only threadsafe if initialized on construction. The other codepath is not threadsafe, but is only used in testcode.
* Avoid use after destructionChristian Mollekopf2017-08-28
|
* use Q_GLOBAL_STATICChristian Mollekopf2017-08-28
| | | | In an attempt to resolve T6890.
* Detect connection lost so we can go to offline stateChristian Mollekopf2017-08-28
| | | | kimap should really have better error codes...
* Only return connected if there is any resource connected.Christian Mollekopf2017-08-26
|
* Default to NoStatus for resources until we know more.Christian Mollekopf2017-08-25
| | | | | | This allows the aggregation to ignore resources where we don't have any status information yet, so the account doesn't always end up being offline.
* Ensure the copied enum matchesChristian Mollekopf2017-08-22
|
* Implemented thread mergingChristian Mollekopf2017-08-18
| | | | | It can happen that thread messages are not delivered in order, which means we will have to merge threads once all messages are available.
* static members can apparently still lead to crashes...Christian Mollekopf2017-08-13
|
* Silence the compiler warningChristian Mollekopf2017-08-11
|
* CleanupChristian Mollekopf2017-08-11
|
* Avoid warningChristian Mollekopf2017-08-09
| | | | | | | setFuture can emit signals directly if the future is already stopped. This does not apply to our case but it fixes the warning. "QFutureWatcher::connect: connecting after calling setFuture() is likely to produce race"
* Use imported targets instead of qt5_use_modulesHeiko Becker2017-07-29
| | | | | | | | | | | | | | | | | | | From Qt's documentation: "This macro is obsolete. Use target_link_libraries with IMPORTED targets instead." It's only recommended with cmake >=2.8.9 & < 2.8.12. Sink already requires cmake 3.0. One advantage of using the imported targets is, that cmake complains if a target isn't found before it's used, like Qt5Concurrent missing from the find_package_call here. Reviewers: #sink, cmollekopf Reviewed By: #sink, cmollekopf Subscribers: #sink Tags: #sink Differential Revision: https://phabricator.kde.org/D6361
* Adjusted docs and test.Christian Mollekopf2017-07-27
|
* Empty resource id's are valid when we search for resources.Christian Mollekopf2017-07-27
|
* Skip modifications that do nothing.Christian Mollekopf2017-07-27
| | | | | | This allows us to i.e. blindly mark mails as read in kube, with the modification automatically being dropped if it doesn't do anything useful.
* Only print modified properties if we have any.Christian Mollekopf2017-07-27
|
* Sanity check queriesChristian Mollekopf2017-07-16
|
* Simplified codeChristian Mollekopf2017-07-16
| | | | The while loop is executed at least once, so advanced is always true.
* Better error messagesChristian Mollekopf2017-07-16
|
* No return value needed here.Christian Mollekopf2017-07-16
|
* Filter resources for syncing by type.Christian Mollekopf2017-07-16
| | | | | Otherwise we end up sending sync requests for contacts to imap resources.
* Avoid regenerating the messageId on every modficationChristian Mollekopf2017-07-06
|
* Keep notifier alive for notifications of new resources.Christian Mollekopf2017-07-05
| | | | Necessary to get notifications for newly created resources.
* Notification printingChristian Mollekopf2017-07-05
|
* Ensure we have the right facadeChristian Mollekopf2017-06-27
|
* Ensure we can deal with non-null terminated strings.Christian Mollekopf2017-06-27
|
* Photo supportChristian Mollekopf2017-06-27
|
* Ensure revision updates are instantaneous.Christian Mollekopf2017-06-23
| | | | | | | We could do some event compression with a timer if necessary, but for the time being we'll just flush after every notification. This is necessary so i.e. a mail in the outbox appears immediately as it is created in the resource.
* Build with flatbuffers 1.7Christian Mollekopf2017-06-23
| | | | | | There's a new template version that expects methods that QByteArray doesn't have but breaks the implicit conversion to const char *, std::string is safer anyways (doesn't require a null terminated string).
* We can run into this on empty modelsChristian Mollekopf2017-06-21
|
* Added threading index cleanupChristian Mollekopf2017-06-15
|
* No more threading by subjectChristian Mollekopf2017-06-15
| | | | | It seems to do more harm than good, creating huge threads, and the webclient doesn't do it either.
* Fixed incremental queriesChristian Mollekopf2017-06-10
| | | | | | The incremental querying broke as soon as a revision update came in since it would nuke the base-set. This fixes it, but it's definitely not pretty.
* A stab at the undefined reference to QMap<QByteArray, int>::QMap() errorChristian Mollekopf2017-06-06
| | | | ...that I only get on copr and nowhere else.
* no need to hardcode thisChristian Mollekopf2017-05-20
|
* Ensure change-replay errors make it through to the correct errorChristian Mollekopf2017-05-20
| | | | handling and are appropriately dealt with.
* We don't need the layout just to check for existenceChristian Mollekopf2017-05-20
|
* Reread the location if we enable the test modeChristian Mollekopf2017-05-20
|
* Avoid notifcations for requests that do nothing, progress with folderidChristian Mollekopf2017-05-20
|
* Don't commit after every replayed revisionChristian Mollekopf2017-05-12
| | | | | | If we didn't actually do anything we just carry on. Failing to commit is harmless in that case and committing for every revision is rather expensive.
* Completely shortcut the stream evaluation if we're not going to use theChristian Mollekopf2017-05-12
| | | | | | | | | output Serializing the data is the expensive part, so we want to completely avoid that for the noisier part if we're not going to use it. Additionally we're now using a stringbuilder for the debugarea to try to further improve the situation with temporary memory allocations.
* Avoid redoing stuff over and over that we can easily avoid.Christian Mollekopf2017-05-12
|
* QDateTime::toString is really slowChristian Mollekopf2017-05-12
|
* We run into this when trying to index a property that is not set.Christian Mollekopf2017-05-12
| | | | i.e. in tests.
* Don't add resources that don't match the query during a livequeryChristian Mollekopf2017-05-12
|
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-28 19:45:48 +0000