* shared.ts first test of getReplies
* comment testing now validates reply notifications and mentions, some code comment cleanup in other functions
* comments revised
* first use of getUnreadCount in testing
* test notification of new comment replies, clarify usage of getReplies
* killall moved earlier in bash script
* api-test jest run does not need directory prefix, make consistent with other jest runs
* do not put my testing system password into script
* fix, killall exits script when no process found
* killall now moved to parent script to release locks before database create
* need to run killall a second time, before database drop
* first use of getReplies getPosts saveUserSettings
* accidental duplication of functions, removed
* try to sync shared library with main
* Nutomic feedback: Better to rename the var instead of putting a comment which can easily get outdated.
* Correct logic to meet join-lemmy requirement, don't have closed signups. Allows Open and Applications. (#3761)
Co-authored-by: Josh Bernardini <josh.bernardini@cologix.com>
* Fix fetch instance software version from nodeinfo (#3772)
Fixes#3771
* remove unused code, revert killall change
---------
Co-authored-by: Dessalines <dessalines@users.noreply.github.com>
Co-authored-by: figure-0e <133478007+figure-0e@users.noreply.github.com>
Co-authored-by: Josh Bernardini <josh.bernardini@cologix.com>
Co-authored-by: Denis Dzyubenko <denis@ddenis.info>
Co-authored-by: Felix Ableitner <me@nutomic.com>
* Only run slow CI checks for actual code changes
Its a waste of time to compile Rust and run tests when only
something like the readme was changed.
* also make cache restore/rebuild conditional
* remove unused condition
* remove n^2 part of person triggers, improve community aggregate trigger
* comment out comment_score tests since previously they only accidentally succeeded
* empty
* more robust test of unlike a comment, confirm replication to instance downstream from community home
* more robust 'delete a comment' test, confirm replication
* Far more robust "Report a comment" test. Many comments about situation, this is currently failing because gamma does not get the report
* typo and actually have Gamma comment check use gamma, not alpha
* prepare-drone-federation-test.sh has some more echo output and note about the LEMMY_DATABASE_URL format (#3651)
* Add http cache for webfingers (#3317)
* Add http cache for webfingers
* Remove the outgoing cache middleware & adjust the cache headers directive
* Use 1h & 3day cache header
* Update routes and adjust the cache headers location
* revert apub caching
---------
Co-authored-by: Dessalines <dessalines@users.noreply.github.com>
Co-authored-by: Felix Ableitner <me@nutomic.com>
* Rewrite activity lists to fix delete federation (fixes#3625)
* Revert "typo and actually have Gamma comment check use gamma, not alpha"
This reverts commit 7dfb6ee0f4.
* Revert "Far more robust "Report a comment" test. Many comments about situation, this is currently failing because gamma does not get the report"
This reverts commit 7bd3b20ae0.
* prettier TypeScript
* revised comments, as ResolveObject isn't using routine replication
* fmt
* fix api tests
* remove comment
---------
Co-authored-by: cetra3 <cetra3@hotmail.com>
Co-authored-by: Dessalines <dessalines@users.noreply.github.com>
Co-authored-by: Felix Ableitner <me@nutomic.com>
* HTML sanitization in apub code
* Sanitize API inputs
* fmt
* Dont allow html a, img tags
---------
Co-authored-by: Dessalines <dessalines@users.noreply.github.com>
* add option to only show posts from moderated communities
* rename moderated_only to moderator_view and show blocked users in moderator view
* add test for moderator view
* bump lemmy-js-client version for moderation view tests
* fix yarn lockfile
* retry build
* Delete logfile
* retry checks
* remove unused select statement from list posts
---------
Co-authored-by: Dessalines <dessalines@users.noreply.github.com>
* Added controversy rank property to posts and comments, and ability to sort by it
* Triggers instead of schedules tasks, integer -> double, TODO: comments don't seem to get updated with floats, divide SortTypes
* Created PersonSortType
* PersonSortType::MostComments case
* Removed unused PartialOrd trait
* Added new person sort type mappings
* SortType -> PersonSortType
* fixes
* cargo fmt
* fixes after merge with main
* Fixed bug in controversy rank trigger, removed TopX sorts from PersonSortType and added CommentScore instead
* Uncovered enum case
* clippy
* reset translation changes
* translations
* translations
* Added additional hot ordering on controversial posts and comments
* featured local and featured community added to controversy rank index, additional order_by removed (?), added post_score and post_count to PersonSortType
* Woodpecker rerun
* cargo fmt
* woodpecker rerun
* fixed controversy_rank order
* fix
* Readded migration as latest, removed second update statement for setting controversy rank
* Resolves issue #3685
If user isn't authenticated with resolve_object, only allow a local search instead of possibly making an http request.
* Making sure to validate auth before doing a potential remote lookup.
* Skip fragile API tests
These tests fail very often for no reason, eg in #3712 and #3696.
Better to disable them until they can be fixed.
* killall -s1
* add new flag to api
* add new ui settings for local user
* remove extraneous def
* add props to application reg.
* fix clippy updated these
* re-order db schema entries
* remove dupe
* update lemmy sdk
* update lemmy js client
---------
Co-authored-by: Nutomic <me@nutomic.com>
* detailed error message for blocked domains (#3698)
* Pass the domain as an error param
Not formatting the error message to support i18n
---------
Co-authored-by: Freek van Zee <freek.van.zee@mediamonks.com>
I noticed that stopping the Lemmy process with ctrl+c wasnt working
because the activity channel isnt properly closed. This is now fixed.
Later we should also move the channel from static into LemmyContext,
Im not doing that now to avoid conflicts with #3670.
* Denormalize community_id into post_aggregates for a 1000x speed-up when loading posts
* Remove unused index
* Add creator_id to post_aggregates
* Use post_aggregates as main table for PostQuery
* Make post_aggregates the main table for PostView
* Reformat SQL
