summary refs log tree commit diff
path: root/synapse/__init__.py (unfollow)
Commit message (Collapse)AuthorFilesLines
2015-11-17Bump changelog and versionErik Johnston2-1/+6
2015-11-17Always check guest = true in macaroonsSteven Hammerton1-1/+2
2015-11-16Run sytests on jenkinsDaniel Wagner-Hall1-1/+22
2015-11-16Output results files on jenkinsDaniel Wagner-Hall2-4/+18
Outputs: * results.xml * coverage.xml * violations.flake8.log
2015-11-13Don't complain if /make_join response lacks 'prev_state' list (SYN-517)Paul "LeoNerd" Evans1-0/+6
2015-11-13Pull out jenkins script into a checked in scriptDaniel Wagner-Hall3-1/+9
2015-11-13Allow guests to /room/:room_id/{join,leave}Daniel Wagner-Hall1-3/+14
2015-11-13Bump version and change log v0.11.0-rc2Erik Johnston2-1/+9
2015-11-13Implementation of state rollback in /syncRichard van der Hoff2-4/+69
Implementation of SPEC-254: roll back the state dictionary to how it looked at the start of the timeline. Merged PR https://github.com/matrix-org/synapse/pull/373
2015-11-13Fix a few race conditions in the state calculationRichard van der Hoff2-60/+77
Be a bit more careful about how we calculate the state to be returned by /sync. In a few places, it was possible for /sync to return slightly later state than that represented by the next_batch token and the timeline. In particular, the following cases were susceptible: * On a full state sync, for an active room * During a per-room incremental sync with a timeline gap * When the user has just joined a room. (Refactor check_joined_room to make it less magical) Also, use store.get_state_for_events() (and thus the existing stategroups) to calculate the state corresponding to a particular sync position, rather than state_handler.compute_event_context(), which recalculates from first principles (and tends to miss some state). Merged from PR https://github.com/matrix-org/synapse/pull/372
2015-11-13Make handlers.sync return a state dictionary, instead of an event list.Richard van der Hoff2-32/+40
Basically this moves the process of flattening the existing dictionary into a list up to rest.client.*, instead of doing it in handlers.sync. This simplifies a bit of the code in handlers.sync, but it is also going to be somewhat beneficial in the next stage of my hacking on SPEC-254. Merged from PR #371
2015-11-13Update some commentsRichard van der Hoff4-14/+98
Add a couple of type annotations, docstrings, and other comments, in the interest of keeping track of what types I have. Merged from pull request #370.
2015-11-12Return non-room events from guest /events callsDaniel Wagner-Hall2-3/+20
2015-11-12Fix missing profile data in federation joinsErik Johnston3-6/+13
There was a regression where we stopped including profile data in initial joins for rooms joined over federation.
2015-11-12Trailing whitespaceErik Johnston1-1/+1
2015-11-12Expand commentErik Johnston1-0/+6
2015-11-12Fix an issue with ignoring power_level changes on divergent graphsRichard van der Hoff2-5/+104
Changes to m.room.power_levels events are supposed to be handled at a high priority; however a typo meant that the relevant bit of code was never executed, so they were handled just like any other state change - which meant that a bad person could cause room state changes by forking the graph from a point in history when they were allowed to do so.
2015-11-12CommentErik Johnston1-0/+2
2015-11-12Fix SQL syntaxErik Johnston1-2/+3
2015-11-12Tweak guest access permissionsDaniel Wagner-Hall1-10/+11
* Allow world_readable rooms to be read by guests who have joined and left * Allow regular users to access world_readable rooms
2015-11-12Use a (hopefully) more efficient SQL query for doing recency based room searchErik Johnston1-4/+6
2015-11-12Fix bug where assumed dict was namedtupleErik Johnston1-1/+1
2015-11-12Allow guests to set their display namesDaniel Wagner-Hall1-1/+1
Depends: https://github.com/matrix-org/synapse/pull/363 Tests in https://github.com/matrix-org/sytest/pull/66
2015-11-12Consider joined guest users as joined usersDaniel Wagner-Hall1-12/+21
Otherwise they're inconveniently allowed to write events to the room but not to read them from the room.
2015-11-12Fix race creating directoriesDaniel Wagner-Hall1-1/+5
2015-11-11Update date v0.11.0-rc1Erik Johnston1-1/+1
2015-11-11Update CHANGESErik Johnston1-1/+1
2015-11-11Allow guests to access room context APIDaniel Wagner-Hall2-5/+11
2015-11-11Fix param style to work on both sqlite and postgresMark Haines1-2/+6
2015-11-11Allow guest access to room initialSyncDaniel Wagner-Hall2-24/+34
2015-11-11Snakes not camelsSteven Hammerton1-5/+5
2015-11-11Fix the background updateMark Haines3-17/+19
2015-11-11Minor review fixesSteven Hammerton2-17/+14
2015-11-11Share more code between macaroon validationSteven Hammerton2-57/+17
2015-11-10Return world_readable and guest_can_join in /publicRoomsDaniel Wagner-Hall4-29/+71
2015-11-10Allow guest users to join and message roomsDaniel Wagner-Hall7-12/+92
2015-11-10Use a background task to update databases to use the full text searchMark Haines2-80/+28
2015-11-10Test for background updatesMark Haines2-0/+79
2015-11-10Run the background updates when starting synapse.Mark Haines4-10/+67
2015-11-09Add background update task for reindexing event searchMark Haines1-2/+96
2015-11-09Don't fiddle with results returned by event sourcesRichard van der Hoff1-4/+12
Overwriting hashes returned by other methods is poor form. Fixes: SYN-516
2015-11-09Get display name from identity server, not clientDaniel Wagner-Hall2-7/+4
2015-11-09Add storage module for tracking background updates.Mark Haines2-0/+231
The progress for each background update is stored as a JSON blob in the database. Each background update is broken up into separate batches. The batch size is automatically tuned to try avoid blocking single threaded databases for too long.
2015-11-09Add a couple of unit tests for room/<x>/messagesRichard van der Hoff2-1/+57
... merely because I was trying to figure out how it worked, and couldn't.
2015-11-09SYN-513: Include updates for rooms that have had all their tags deletedMark Haines2-2/+2
2015-11-08fix comedy important missing comma breaking recent-ordered FTS on sqliteMatthew Hodgson1-1/+1
2015-11-06add a key existence check to tags_by_room to avoid /events 500'ing when ↵Matthew Hodgson1-1/+2
testing against vector
2015-11-06Remove accidentally added ID columnDaniel Wagner-Hall1-1/+0
2015-11-06remove references to matrix.org/betaMatthew Hodgson1-7/+7
2015-11-06Updates to fallback CAS login to do new token loginSteven Hammerton1-26/+12
2015-11-05Remove anonymous access, since its not ready yetErik Johnston1-2/+1
2015-11-05Add PRErik Johnston1-1/+1
2015-11-05Be explicit about what we're doingErik Johnston1-1/+1
2015-11-05Remove redundant testErik Johnston1-1/+0
2015-11-05Increment by one, not fiveErik Johnston1-1/+1
2015-11-05Bump changelog and version numberErik Johnston2-1/+18
2015-11-05Exchange 3pid invites for m.room.member invitesDaniel Wagner-Hall10-180/+230
2015-11-05Fix tox config after fa1cf5ef34Daniel Wagner-Hall1-3/+4
2015-11-05COMMENTSErik Johnston1-4/+26
2015-11-05Error handlingErik Johnston1-1/+8
2015-11-05Implement basic pagination for search resultsErik Johnston3-50/+86
2015-11-05Open up /events to anonymous users for room events onlyDaniel Wagner-Hall20-82/+299
Squash-merge of PR #345 from daniel/anonymousevents
2015-11-05Allow hs to do CAS login completely and issue the client with a login token ↵Steven Hammerton3-5/+218
that can be redeemed for the usual successful login response
2015-11-04Add service URL to CAS configSteven Hammerton1-0/+3
2015-11-04Implement order and group byErik Johnston2-17/+205
2015-11-04Allow guests to register and call /events?room_id=Daniel Wagner-Hall33-167/+272
This follows the same flows-based flow as regular registration, but as the only implemented flow has no requirements, it auto-succeeds. In the future, other flows (e.g. captcha) may be required, so clients should treat this like the regular registration flow choices.
2015-11-04Remove fields that are both unspecified and unused from the filter checks, ↵Mark Haines1-18/+2
check the right top level definitions in the filter
2015-11-04Remove the LockManager class because it wasn't being usedMark Haines4-189/+0
2015-11-03Don't rearrange transaction_queueErik Johnston1-12/+11
2015-11-03Fix broken cache for getting retry times. This meant we retried remote ↵Erik Johnston2-55/+40
destinations way more frequently than we should
2015-11-03That TODO was doneMark Haines1-1/+0
2015-11-03Spell "deferred" more correctlyMark Haines1-1/+1
2015-11-02Include read receipts in v2 syncMark Haines1-15/+30
2015-11-02Retry dead servers a lot less oftenErik Johnston2-4/+13
2015-11-02Include room tags in v2 /syncMark Haines3-11/+70
2015-11-02Remove more unused parametersDaniel Wagner-Hall6-13/+2
2015-11-02Store room tag content and return the content in the m.tag eventMark Haines4-22/+41
2015-11-02Remove unused arguments and codeDaniel Wagner-Hall3-29/+5
2015-10-30Support clients supplying older tokens, fix short poll testMark Haines2-3/+3
2015-10-30Include tags in v1 room initial syncMark Haines1-0/+11
2015-10-30Add missing columnDaniel Wagner-Hall1-1/+1
2015-10-30Include room tags v1 /initialSyncMark Haines1-0/+12
2015-10-30Update the other place check_joined_room is calledMark Haines3-2/+4
2015-10-29Don't mark newly joined room timelines as limited in an incremental syncMark Haines1-3/+5
2015-10-29Make search API honour limit set in filterErik Johnston1-0/+3
2015-10-29Inform the client of new room tags using v1 /eventsMark Haines6-14/+91
2015-10-28Optionally return event contexts with search resultsErik Johnston1-1/+52
2015-10-28Fix pyflakes errorsMark Haines2-3/+5
2015-10-28Add APIs for adding and removing tags from roomsMark Haines5-0/+320
2015-10-28DocsErik Johnston2-0/+39
2015-10-28Add room context apiErik Johnston4-2/+190
2015-10-26Implement full_state incremental syncRichard van der Hoff2-19/+38
A hopefully-complete implementation of the full_state incremental sync, as specced at https://github.com/matrix-org/matrix-doc/pull/133. This actually turns out to be a relatively simple modification to the initial sync implementation.
2015-10-26Fix a 500 error resulting from empty room_idsRichard van der Hoff2-3/+8
POST /_matrix/client/api/v1/rooms//send/a.b.c gave a 500 error, because we assumed that rooms always had at least one character.
2015-10-26Use correct service urlErik Johnston1-3/+1
2015-10-26Add login fallbackErik Johnston5-0/+280
2015-10-26Move static folder into synapseErik Johnston7-1/+3
This is because otherwise it won't get picked up by python packaging. This also fixes the problem where the "static" folder was found if synapse wasn't started from that directory.
2015-10-23Pull out sender when computing search resultsErik Johnston1-14/+17
2015-10-23Implement rank function for SQLite FTSErik Johnston3-2/+30
2015-10-22Don't assume that the event has a room_id or senderMark Haines1-2/+2
2015-10-22Actually filter resultsErik Johnston1-3/+5
2015-10-22Support filtering events represented as dicts.Mark Haines1-3/+22
This is useful because the emphemeral events such as presence and typing are represented as dicts inside synapse.
2015-10-22RenameErik Johnston1-3/+3
2015-10-22LESS THANErik Johnston1-1/+2
2015-10-22Fix receipts for room initial syncErik Johnston1-7/+1
2015-10-22Use namedtuple as return valueErik Johnston2-4/+18
2015-10-22Limit max number of SQL varsErik Johnston1-4/+6
2015-10-22Reject events which are too largeDaniel Wagner-Hall2-1/+30
SPEC-222
2015-10-22PEP8Erik Johnston1-1/+2
2015-10-22Move FTS to delta 25Erik Johnston2-3/+3
2015-10-22Don't keep appending report_stats to demo configErik Johnston1-2/+6
2015-10-22Add config option to disable password loginErik Johnston5-4/+46
2015-10-21Doc string for the SyncHandler.typing_by_room methodMark Haines1-0/+12
2015-10-21Include banned rooms in the archived section of v2 syncMark Haines2-8/+11
2015-10-21Cap the time to retry txns to appservices to 8.5 minutesKegan Dougal1-2/+2
There's been numerous issues with people playing around with their application service and then not receiving events from their HS for ages due to backoff timers reaching crazy heights (albeit capped at < 1 day). Reduce the max time between pokes to be 8.5 minutes (2^9 secs) which is quick enough for people to wait it out (avg wait time being 4.25 min) but long enough to actually give the AS breathing room if it needs it.
2015-10-21Filter room ids before hitting the databaseErik Johnston2-1/+21
2015-10-21Filter search resultsErik Johnston1-2/+7
2015-10-21Use 403 and message to match handlers/authKegan Dougal1-3/+1
2015-10-20