24c77e57b2
Rewrite Atom generation from stream entries to use Ox instead of Nokogiri ( #1124 )
...
* Rewrite Atom generation from stream entries to use Ox instead of Nokogiri::Builder
StreamEntry is now limited to only statuses, which allows some optimization. Removed
extra queries on AccountsController#show. AtomSerializer instead of AtomBuilderHelper
used in AccountsController#show, StreamEntriesController#show, StreamEntryRenderer
and PubSubHubbub::DistributionWorker
PubSubHubbub::DistributionWorker moves n+1 DomainBlock query to PubSubHubbub::DeliveryWorker
instead.
All Salmon slaps that aren't based on StreamEntry still use AtomBuilderHelper and Nokogiri
* All Salmon slaps now use Ox instead of Nokogiri. No touch from status on account
2017-04-07 05:56:56 +02:00
70f485a9e7
Catch more errors in process_follows so it doesn't fail
2017-04-05 20:04:13 -04:00
cd68e54a7d
Split SalmonWorker into smaller parts, move profile updating into another job
2017-04-05 21:43:10 +02:00
5189ce9cd1
Replace calls to FeedManager#inline_render and #broadcast
2017-04-05 19:45:18 +02:00
54329947ab
Fix wrong variable used in publish channel
2017-04-05 18:58:32 +02:00
b9ddfff8f6
Add proper message to PushUpdateWorker, use redis directly
2017-04-05 18:48:41 +02:00
d67e1fe44d
Merge branch 'master' into broadcast-to-worker
2017-04-05 18:44:33 +02:00
8f38353c4d
Only render public payload once in FanOutOnWrite
2017-04-05 14:26:17 +02:00
192f855952
We're going to want these nice helper methods, lets share them with a parent class that matches Rails 5 practices (application level abstraction)
2017-04-04 21:14:37 -07:00
bea03b4553
Moving in the inline render
2017-04-04 20:51:18 -07:00
dd66bce930
I don't actually think we need that.
2017-04-04 20:48:22 -07:00
7d887cea4b
Moving the queue_at into the worker
2017-04-04 20:39:14 -07:00
4f2985cddc
Replacing the broadcast method with the one defined in the feed manager
2017-04-04 20:36:03 -07:00
de1bb4bd56
By pushing this into a worker we can reduce the amount of time the feed manager using workers eat up a connection
2017-04-04 20:23:40 -07:00
d2b7daf6bb
Remove PuSH subscriptions when delivery is answered with a 4xx error
2017-04-05 02:34:33 +02:00
fa2a1af90a
Spawn FeedInsertWorker to deliver status into personal feed
2017-04-04 19:21:37 +02:00
e60c473917
Reduce number of items in feeds, optimize regeneration worker slightly,
...
make regeneration worker unique, (only schedule/execute once at a time)
2017-04-04 13:58:34 +02:00
1484df2db5
Move OStatus processing back into default queue
2017-04-04 13:02:49 +02:00
7a56845060
Only call regeneration worker after first login after a 14 day break
2017-04-04 02:00:10 +02:00
8eca42b265
Separate background jobs into different queues. ATTENTION: new queue "pull"
...
must be added to the Sidekiq invokation in your systemd file
The pull queue will handle link crawling, thread resolving, and OStatus
processing. Such tasks are more likely to hang for a longer time (due to
network requests) so it is more sensible to not make the "in-house" tasks
wait for them.
2017-04-04 00:53:20 +02:00
6ebe31e716
Make default admin UI page reports. Add admin UI for creating a domain block
2017-04-03 18:55:06 +02:00
244da78105
Import feature for following/blocking lists (addresses #62 , #177 , #201 , #454 )
2017-03-30 19:42:33 +02:00
0cb5a2a0a7
Add digest e-mails
2017-03-04 00:00:48 +01:00
94a6dfab3a
Disable PuSH for blocked domains
2017-02-14 04:01:37 +01:00
8550d32165
Adding more unit tests. Fixing Salmon slaps XML
2017-02-12 17:30:15 +01:00
79f3ee31df
Take out private toot distribution out of PuSH again
2017-02-12 01:31:04 +01:00
ffd58aad9d
Make PuSH only distribute to subscriptions that match follower domains
...
Allow PuSH to distribute private toots with that condition
2017-02-11 15:41:39 +01:00
1e66536797
After FollowService, re-fetch remote account asynchronously, do nothing
...
if account lock info was up to date, otherwise re-do the FollowService
with now updated information
2017-02-11 14:25:01 +01:00
59c8c2b28a
Make follow requests federate
2017-02-11 02:58:00 +01:00
efa4065029
Preheat status cache
2017-02-05 17:24:18 +01:00
9327d05bf7
API for apps to register for push notifications
2017-01-29 01:30:32 +01:00
ea8b548ee9
Make blocks create entries and unfollows instantly, but do the clean up
...
in the background instead. Should fix delay where blocked person
can interact with blocker for a short time before background job
gets processed
2017-01-24 21:40:41 +01:00
98660a76d9
Move merging/unmerging of timelines into background. Move blocking into
...
background as well since it's a computationally expensive
2017-01-23 21:29:34 +01:00
f748a91ec7
Fix #463 - Fetch and display previews of URLs using OpenGraph tags
2017-01-20 01:00:14 +01:00
a893746e5a
Improve background jobs params and error handling
2017-01-05 03:28:21 +01:00
0471207cc3
Fix accidental distribution of hidden stream entries to PuSH subscribers
2017-01-03 00:44:23 +01:00
2bc6e7c96e
Add API for retrieving blocked accounts
2016-12-29 20:12:32 +01:00
0f4c9d1124
Adding some rescues to workers, e.g. don't fail removal job when status can't be found in the first place (already removed)
2016-12-19 09:31:12 +01:00
b2945b025f
Make unfavouriting async to prevent timeout errors from leaving orphaned records behind
2016-12-19 09:12:29 +01:00
9d4f96f440
Removing external hub completely, fix #333 fixing digit-only hashtags,
...
removing web app capability from non-webapp pages
2016-12-18 12:24:37 +01:00
b7c2c5d81d
Restoring old async behaviour of thread resolving as it proved to be more robust
2016-12-12 21:12:19 +01:00
fb36684678
Thread resolving no longer needs to be separate from ProcessFeedService,
...
since that is only ever called in the background
2016-12-11 22:23:11 +01:00
7752662f97
Add filters for suspended accounts
2016-12-06 18:03:30 +01:00
211049ae59
Fix #329 - avatar errors no longer prevent remote accounts from being saved
...
(without avatar). Also improved search position of exact matches
2016-12-02 14:14:49 +01:00
a692318382
Track successful PuSH deliveries
2016-11-30 15:24:57 +01:00
d26b8f3cce
Delete statuses asynchronously but provide instant feedback in the API
2016-11-29 15:32:25 +01:00
3c1c2b0e06
Adding rack timeout of 30sec, PuSH jobs moved to push queue so they
...
can be processed separately
2016-11-29 02:07:14 +01:00
79075e1303
Fix URLs in inline-rendered XML
2016-11-29 00:26:01 +01:00
02da8fdcbe
Fix setting of confirmed=true on successful confirmation
2016-11-28 19:24:49 +01:00
bf7c9e826d
More logging to ConfirmationWorker
2016-11-28 19:15:28 +01:00
0cc5410e03
Adding backtracing to Salmon/Processing workers
2016-11-28 19:11:36 +01:00
b5ad0eb4ea
Adding embedded PuSH server
2016-11-28 13:36:47 +01:00
f07b0dc82f
Remove unneeded indices, improve error handling in background workers, don't needlessly reload reblogged status, send Devise e-mails asynchronously
2016-11-22 17:32:51 +01:00
fbaddca49e
Move Salmon processing to background as well as PuSH
2016-11-18 23:24:57 +01:00
e71b152d89
Fix rubocop issues, introduce usage of frozen literal to improve performance
2016-11-15 16:56:29 +01:00
a7332acba3
Delegate processing of incoming PuSH data to background workers
2016-11-15 15:43:33 +01:00
f1d2c46ded
Improve feed regeneration
2016-11-08 02:08:32 +01:00
3319473b2c
Move PubSubHubbub pinging to a background worker
...
It can take as much as 0.5s if not longer to complete
2016-10-05 13:50:21 +02:00
927333f4f8
Improve code style
2016-09-29 21:28:21 +02:00
4bec613897
Fix #24 - Thread resolving for remote statuses
...
This is a big one, so let me enumerate:
Accounts as well as stream entry pages now contain Link headers that
reference the Atom feed and Webfinger URL for the former and Atom entry
for the latter. So you only need to HEAD those resources to get that
information, no need to download and parse HTML <link>s.
ProcessFeedService will now queue ThreadResolveWorker for each remote
status that it cannot find otherwise. Furthermore, entries are now
processed in reverse order (from bottom to top) in case a newer entry
references a chronologically previous one.
ThreadResolveWorker uses FetchRemoteStatusService to obtain a status
and attach the child status it was queued for to it.
FetchRemoteStatusService looks up the URL, first with a HEAD, tests
if it's an Atom feed, in which case it processes it directly. Next
for Link headers to the Atom feed, in which case that is fetched
and processed. Lastly if it's HTML, it is checked for <link>s to the Atom
feed, and if such is found, that is fetched and processed. The account for
the status is derived from author/name attribute in the XML and the hostname
in the URL (domain). FollowRemoteAccountService and ProcessFeedService
are used.
This means that potentially threads are resolved recursively until a dead-end
is encountered, however it is performed asynchronously over background jobs,
so it should be ok.
2016-09-21 01:50:31 +02:00
85b00d19b8
Moving Salmon notifications to background processing, fixing mini-profiler
...
behaviour with Turbolinks enabled, optimizing Rabl for production
2016-03-26 13:42:10 +01:00
8eeec389c1
Add link to github project to footer, move FanOutOnWriteService calls to
...
DistributionWorker. That isn't the heaviest service, yet, but gotta start
somewhere
2016-03-25 03:22:26 +01:00
Eugen