2010-10-21 Martyn Russell <martyn@lanedo.com> Release 0.9.26 Updated NEWS libtracker-miner: get_running() now ignores the ".Index" not old ".Reindex" d-bus interface This was causing a critical with tracker-status -i 2010-10-21 Aleksander Morgado <aleksander@lanedo.com> Merge branch 'miner-fs-refactor-multi-insert' libtracker-miner: Don't report processing or percentage change until crawling is finished libtracker-miner: Fix weird alignment libtracker-miner: a task is always given to the async operation libtracker-miner: Minor coding style change libtracker-miner: fix task status enumeration so that first item is -1 libtracker-miner: Try to avoid some unneeded heap allocations libtracker-miner: set processing pool logs as traces enabled at compile time libtracker-miner: Use tracker prefix in the processing pool libtracker-miner: simplify array creation as we know the size in advance libtracker-miner: Split internal PROCESS queue into two: READY and PROCESSING * When having a single queue for items in the SPARQL buffer and items being pushed to the store; as the limit of the queue was a single one, we were actually having an undesired behaviour: While an array-update of 100 items was being done, every new task pushed to the processing pool was not buffered, as the limit of the queue was reached. * So now, the READY queue holds the tasks being buffered, and the PROCESSING queue holds the items currently being pushed to the store. tracker-miner-fs: Max 10 files extracting, and 100 files processing libtracker-miner: When QUEUE_WAIT, force a processing pool buffer flush libtracker-miner: Fix deallocation of tasks when stopping the miner libtracker-miner: Fix queue looping in processing pool libtracker-miner: documented how the processing pool works libtracker-miner: Integrate the new processing pool libtracker-miner: New processing pool out of the TrackerMinerFS object 2010-10-21 Philip Van Hoof <philip@codeminded.be> tracker-extract: gstreamer: Don't extract nmm:albumArtist until we have a better solution Partly fixes NB#194908 2010-10-21 Martyn Russell <martyn@lanedo.com> Updated NEWS 2010-10-21 Aleksander Morgado <aleksander@lanedo.com> distcheck fix: distribute the .in.in files as they will need variable substitution depending on prefix during configure 2010-10-21 Philip Van Hoof <philip@codeminded.be> tracker-extract: Port QuillImageFilter usage to QImage 2010-10-21 Aleksander Morgado <aleksander@lanedo.com> libtracker-bus: Handle error properly in array_update_finish() 2010-10-20 Takayuki KUSANO <AE5T-KSN@asahi-net.or.jp> Updated Japanese translation 2010-10-20 Ivan Frade <ivan.frade@nokia.com> functional-tests: Testing addition/removal of domainIndexes in the ontology 2010-10-20 Jürg Billeter <j@bitron.ch> Use host endian byte order for updates in steroids interface Queries in steroids interface already use host endian byte order. Let's be consistent. 2010-10-20 Ivan Frade <ivan.frade@nokia.com> functional-tests: Testing removal and addition of properties in the ontology functional-test: add 'ask' method in the store helper for the tests functional-tests: rebase test ontologies to latest basic To make sure we are testing ONLY the changes we want functional-tests: add prop/subprop with cardinality n in the basic ontology Needed to test the addition removal of super/sub properties functional-tests: some doc for the ontology-changes test template ontologies: updated the modification time of NCAL ontology Forgot to do it in yesterdays commit. 2010-10-20 Jürg Billeter <j@bitron.ch> ontologies: Make sure all regular classes derive from rdfs:Resource 2010-10-19 Ivan Frade <ivan.frade@nokia.com> Fixes NB#194548, Property needed to link nco:Contact with ncal:Event ontologies: Add birthday and anniversary properties between contacts and events in SCAL and NCAL 2010-10-18 Ivan Frade <ivan.frade@nokia.com> functional-test: More relaxed assertion to check writeback results functional-tests: no need to sleeps now that the tests wait for signals functional-tests: Speed up the miner-fs tests No need to sleep, it will return from the wait_for_idle when ready functional-tests: Remove sleeps to speed up the signals tests functional-tests: Testing addition and removal of classes in the ontology functional-test: fix the 'inverse' tests using the basic ontology in the future The second ontology must be more recent than the first one in the tests. To avoid rewritting too many ontologies, included the same 'basic' ontology but in the future. So all changes 'basic'->X can be tested in the opposite direction with X->'basic-future' 2010-10-18 Deji Akingunola <dakingun@gmail.com> Fixes: Bug 629850 - Undefined reference errors during build 2010-10-18 Martyn Russell <martyn@lanedo.com> configure: Fixed warning with recent autoconf versions for AC_INIT Changed AC_INIT to not include whole bugzilla link for tracker, seems the "?" part of the URL causes this warning for some reason. Fixes: Bug 632399 - [PATCH] configure.ac: Skip `AC_CHECK_FILE` when cross compiling We have removed all checks for old installed files tracker-miner-fs: Fixed MeeGo app miner, missing include for GStrv 2010-10-18 Aleksander Morgado <aleksander@lanedo.com> libtracker-bus: plug memleak when updating, GAsyncResult should always be unref-ed 2010-10-18 Jorge González <jorgegonz@svn.gnome.org> Updated Spanish translation 2010-10-17 Mattias Põldaru <mahfiaz gmail com> [l10n] Updated Estonian translation 2010-10-16 Kjartan Maraas <kmaraas@gnome.org> Updated Norwegian bokmål translation Fix POTFILES.in 2010-10-15 Aleksander Morgado <aleksander@lanedo.com> utils, mtp-dummy: locale initialization was missing libtracker-data: Notify locales being used also for libunistring and glib collators libtracker-data: improved collation related debugging traces