| Commit message (Expand) | Author | Files | Lines |
2015-05-18 | Comments and shuffle of functions | Erik Johnston | 1 | -38/+42 |
2015-05-18 | Remove unused metric | Erik Johnston | 1 | -1/+0 |
2015-05-18 | Remove rejected events if we don't want rejected events | Erik Johnston | 1 | -0/+3 |
2015-05-18 | Move get_events functions to storage.events | Erik Johnston | 2 | -155/+132 |
2015-05-18 | Remove pointless newline | Erik Johnston | 1 | -1/+0 |
2015-05-18 | Newline, remove debug logging | Erik Johnston | 3 | -4/+1 |
2015-05-18 | Remove debug logging | Erik Johnston | 1 | -8/+0 |
2015-05-15 | PEP8 | Erik Johnston | 2 | -2/+1 |
2015-05-15 | Revert limiting of fetching, it didn't help perf. | Erik Johnston | 1 | -9/+2 |
2015-05-15 | init j | Erik Johnston | 1 | -0/+1 |
2015-05-15 | Don't completely drain the list | Erik Johnston | 1 | -2/+9 |
2015-05-15 | Don't fetch prev_content for current_state | Erik Johnston | 1 | -1/+1 |
2015-05-15 | Err, defer.gatherResults ftw | Erik Johnston | 1 | -21/+12 |
2015-05-15 | Make store.get_current_state fetch events asyncly | Erik Johnston | 2 | -4/+4 |
2015-05-15 | Remove more debug logging | Erik Johnston | 1 | -2/+0 |
2015-05-15 | Remove debug logging | Erik Johnston | 1 | -11/+3 |
2015-05-15 | preserve log context | Erik Johnston | 1 | -1/+2 |
2015-05-15 | Add a wait | Erik Johnston | 2 | -3/+12 |
2015-05-15 | Srsly. Don't use closures. Baaaaaad | Erik Johnston | 1 | -10/+8 |
2015-05-15 | Correctly pass through params | Erik Johnston | 1 | -3/+3 |
2015-05-15 | More whitespace | Mark Haines | 1 | -0/+1 |
2015-05-15 | Sort out error handling | Erik Johnston | 1 | -25/+22 |
2015-05-15 | Fix daedlock | Erik Johnston | 7 | -84/+122 |
2015-05-14 | Add some missed argument types, cleanup the whitespace a bit | Mark Haines | 1 | -6/+17 |
2015-05-14 | More doc-strings | Mark Haines | 1 | -39/+202 |
2015-05-14 | Remove race condition | Erik Johnston | 4 | -96/+157 |
2015-05-14 | Support gzip encoding for client, client v2 and web client resources (SYN-176). | David Baker | 1 | -5/+16 |
2015-05-14 | Call from right thread | Erik Johnston | 1 | -1/+3 |
2015-05-14 | Count and loop | Erik Johnston | 2 | -37/+35 |
2015-05-14 | Don't start up more fetch_events | Erik Johnston | 1 | -0/+1 |
2015-05-14 | Awful idea for speeding up fetching of events | Erik Johnston | 3 | -40/+139 |
2015-05-14 | Remove unused arguments and doc PresenceHandler.push_update_to_clients | Mark Haines | 3 | -60/+21 |
2015-05-14 | Turn off preemptive transactions | Erik Johnston | 1 | -17/+7 |
2015-05-14 | PEP8 | Erik Johnston | 1 | -2/+1 |
2015-05-14 | Move from _base to events | Erik Johnston | 2 | -232/+247 |
2015-05-14 | Jump out early | Erik Johnston | 1 | -0/+6 |
2015-05-14 | Actually, we probably want to run this in a transaction | Erik Johnston | 1 | -15/+12 |
2015-05-14 | Treat setting your display name to the empty string as removing it (SYN-186). | David Baker | 1 | -0/+3 |
2015-05-14 | Preemptively jump into a transaction if we ask for get_prev_content | Erik Johnston | 1 | -12/+22 |
2015-05-14 | loop -> gatherResults | Erik Johnston | 1 | -10/+14 |
2015-05-14 | Err, we probably want a bigger limit | Erik Johnston | 1 | -1/+1 |
2015-05-14 | Move fetching of events into their own transactions | Erik Johnston | 4 | -53/+45 |
2015-05-14 | Refactor _get_events | Erik Johnston | 3 | -253/+123 |
2015-05-14 | Disallow whitespace in aliases here too | David Baker | 1 | -0/+5 |
2015-05-14 | Throw error when creating room if alias contains whitespace #SYN-335 | David Baker | 1 | -0/+5 |
2015-05-14 | Make shared secret registration work again | David Baker | 1 | -1/+4 |
2015-05-14 | Add ID generator for push_rules_enable to #resolve SYN-378 | David Baker | 2 | -2/+13 |
2015-05-14 | Specify python 2.7 in the virtualenv setup (SYN-319) #resolved | David Baker | 1 | -1/+1 |
2015-05-13 | Add index on events | Erik Johnston | 2 | -1/+20 |
2015-05-13 | Actually use async method | Erik Johnston | 1 | -3/+1 |
2015-05-13 | Fetch events from events_id in their own transactions | Erik Johnston | 3 | -18/+168 |
2015-05-13 | Typo | Erik Johnston | 1 | -1/+1 |
2015-05-13 | Don't insert None | Erik Johnston | 1 | -1/+4 |
2015-05-13 | Don't return None | Erik Johnston | 1 | -1/+1 |
2015-05-13 | Limit batch size | Erik Johnston | 1 | -5/+10 |
2015-05-13 | Fetch events in bulk | Erik Johnston | 2 | -34/+63 |
2015-05-13 | Don't fetch redaction and rejection stuff for each event, so we can use index... | Erik Johnston | 2 | -6/+24 |
2015-05-13 | Temp turn off checking for rejections and redactions | Erik Johnston | 1 | -5/+2 |
2015-05-13 | defer.gatherResults loop | Erik Johnston | 1 | -1/+7 |
2015-05-13 | Batch fetching of events for state groups | Erik Johnston | 1 | -1/+24 |
2015-05-13 | Load events for state group seperately | Erik Johnston | 2 | -5/+11 |
2015-05-12 | Don't talk to yourself when backfilling | Erik Johnston | 1 | -0/+2 |
2015-05-12 | You need to call contextmanager | Erik Johnston | 1 | -0/+1 |
2015-05-12 | Revert "Improvement to performance of presence event stream handling" | Mark Haines | 1 | -20/+6 |
2015-05-12 | Err, delete the right stuff | Erik Johnston | 1 | -4/+5 |
2015-05-12 | Newlines | Erik Johnston | 1 | -2/+0 |
2015-05-12 | We do actually want to delete rows out of event_backward_extremities | Erik Johnston | 1 | -4/+11 |
2015-05-12 | Fix up _handle_prev_events to not try to insert duplicate rows | Erik Johnston | 1 | -23/+13 |
2015-05-12 | Do state groups persistence /after/ checking if we have already persisted the... | Erik Johnston | 1 | -6/+9 |
2015-05-12 | s/backfil/backfill/ | Erik Johnston | 1 | -4/+4 |
2015-05-12 | Typos everywhere | Erik Johnston | 1 | -1/+1 |
2015-05-12 | And another typo. | Erik Johnston | 1 | -3/+3 |
2015-05-12 | Err, gatherResults doesn't take a dict... | Erik Johnston | 1 | -4/+7 |
2015-05-12 | Make distributer.fire work as it did | Erik Johnston | 1 | -20/+30 |
2015-05-12 | Unwrap defer.gatherResults failures | Erik Johnston | 6 | -7/+18 |
2015-05-12 | Update the end_token correctly, otherwise the token doesn't advance and the c... | Mark Haines | 1 | -1/+1 |
2015-05-12 | SYN-377: Make sure that the StreamIdGenerator.get_next.__exit__ is called fro... | Mark Haines | 2 | -23/+27 |
2015-05-12 | Handle the case where the other side is unreachable when backfilling | |