Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Remove unneeded `ActionGenerator` class. (#12691) | Patrick Cloke | 2022-05-11 | 1 | -48/+0 |
| | | | | It simply passes through to `BulkPushRuleEvaluator`, which can be called directly instead. | ||||
* | Refactor `EventContext` (#12689) | Erik Johnston | 2022-05-10 | 1 | -0/+4 |
| | | | | | | | | | | Refactor how the `EventContext` class works, with the intention of reducing the amount of state we fetch from the DB during event processing. The idea here is to get rid of the cached `current_state_ids` and `prev_state_ids` that live in the `EventContext`, and instead defer straight to the database (and its caching). One change that may have a noticeable effect is that we now no longer prefill the `get_current_state_ids` cache on a state change. However, that query is relatively light, since its just a case of reading a table from the DB (unlike fetching state at an event which is more heavyweight). For deployments with workers this cache isn't even used. Part of #12684 | ||||
* | Remove redundant "coding: utf-8" lines (#9786) | Jonathan de Jong | 2021-04-14 | 1 | -1/+0 |
| | | | | | | | Part of #9744 Removes all redundant `# -*- coding: utf-8 -*-` lines from files, as python 3 automatically reads source code as utf-8 now. `Signed-off-by: Jonathan de Jong <jonathan@automatia.nl>` | ||||
* | Import HomeServer from the proper module. (#9665) | Patrick Cloke | 2021-03-23 | 1 | -1/+1 |
| | |||||
* | Add type hints to the push module. (#8901) | Patrick Cloke | 2020-12-11 | 1 | -5/+10 |
| | |||||
* | Stop sub-classing object (#8249) | Patrick Cloke | 2020-09-04 | 1 | -1/+1 |
| | |||||
* | Convert push to async/await. (#7948) | Patrick Cloke | 2020-07-27 | 1 | -5/+2 |
| | |||||
* | Run Black. (#5482) | Amber Brown | 2019-06-20 | 1 | -3/+1 |
| | |||||
* | run isort | Amber Brown | 2018-07-09 | 1 | -3/+3 |
| | |||||
* | Remove context.push_actions | Erik Johnston | 2018-02-15 | 1 | -5/+1 |
| | |||||
* | Store ActionGenerator in HomeServer | Erik Johnston | 2017-05-19 | 1 | -1/+1 |
| | |||||
* | Faster push rule calculation via push specific cache | Erik Johnston | 2017-05-17 | 1 | -8/+4 |
| | | | | | We add a push rule specific cache that ensures that we can reuse calculated push rules appropriately when a user join/leaves. | ||||
* | Replace context.current_state with context.current_state_ids | Erik Johnston | 2016-08-25 | 1 | -2/+2 |
| | |||||
* | Move _bulk_get_push_rules_for_room to storage layer | Erik Johnston | 2016-08-19 | 1 | -1/+1 |
| | |||||
* | Make push Measure finer grained | Erik Johnston | 2016-08-17 | 1 | -4/+5 |
| | |||||
* | Use state to calculate get_users_in_room | Erik Johnston | 2016-06-01 | 1 | -1/+1 |
| | |||||
* | move filter_events_for_client out of base handler | Mark Haines | 2016-05-11 | 1 | -2/+2 |
| | |||||
* | Measure push action generator | Erik Johnston | 2016-04-14 | 1 | -11/+15 |
| | |||||
* | Fix invite pushes | David Baker | 2016-04-08 | 1 | -3/+3 |
| | | | | | | * If the event is an invite event, add the invitee to list of user we run push rules for (if they have a pusher etc) * Move invite_for_me to be higher prio than member events otherwise member events matches them * Spell override right | ||||
* | Fix relative imports so they work in both py3 and py27 | Mark Haines | 2016-03-08 | 1 | -2/+2 |
| | |||||
* | Remove dead code for setting device specific rules. | Mark Haines | 2016-02-18 | 1 | -1/+1 |
| | | | | | | It wasn't possible to hit the code from the API because of a typo in parsing the request path. Since no-one was using the feature we might as well remove the dead code. | ||||
* | Atomically persit push actions when we persist the event | Erik Johnston | 2016-02-09 | 1 | -15/+5 |
| | |||||
* | Pass in current state to push action handler | Erik Johnston | 2016-02-09 | 1 | -2/+4 |
| | |||||
* | Add regex cache. Only caculate push actions for users that have sent read ↵ | Erik Johnston | 2016-01-19 | 1 | -3/+4 |
| | | | | receipts, and are on that server | ||||
* | Re-enable urnead notifications | Erik Johnston | 2016-01-18 | 1 | -3/+0 |
| | |||||
* | Temporarily disable notification branch | Erik Johnston | 2016-01-13 | 1 | -0/+3 |
| | |||||
* | Delete notifications for redacted events | David Baker | 2016-01-06 | 1 | -0/+7 |
| | |||||
* | Add is_guest flag to users db to track whether a user is a guest user or ↵ | David Baker | 2016-01-06 | 1 | -3/+3 |
| | | | | not. Use this so we can run _filter_events_for_client when calculating event_push_actions. | ||||
* | Rename event-actions to event_push_actions as per PR request | David Baker | 2016-01-04 | 1 | -1/+1 |
| | |||||
* | Address minor PR issues | David Baker | 2016-01-04 | 1 | -4/+3 |
| | |||||
* | pep8 & unused variable | David Baker | 2015-12-22 | 1 | -2/+0 |
| | |||||
* | Insert push actions in a single db query rather than one per user/profile_tag | David Baker | 2015-12-22 | 1 | -4/+6 |
| | |||||
* | Add bulk push rule evaluator which actually still evaluates rules one by ↵ | David Baker | 2015-12-22 | 1 | -17/+11 |
| | | | | one, but does far fewer db queries to fetch the rules | ||||
* | Only run pushers for users on this hs! | David Baker | 2015-12-21 | 1 | -1/+7 |
| | |||||
* | still very WIP, but now sends unread_notifications_count in the room object ↵ | David Baker | 2015-12-16 | 1 | -1/+1 |
| | | | | on sync (only actually corrrect in a full sync: hardcoded to 0 in incremental syncs). | ||||
* | Store nothing instead of ['dont_notify'] for events with no notification ↵ | David Baker | 2015-12-10 | 1 | -4/+4 |
| | | | | required: much as it would be nice to be able to tell between the event not having been processed and there being no notification for it, this isn't worth filling up the table with ['dont_notify'] I think. Consequently treat the empty actions array as dont_notify and filter dont_notify out of the result. | ||||
* | Save event actions to the db | David Baker | 2015-12-10 | 1 | -2/+4 |
| | |||||
* | Very first cut of calculating actions for events as they come in. Doesn't ↵ | David Baker | 2015-12-10 | 1 | -0/+47 |
store them yet. Not very efficient. |