Conflicts:
- `app/models/user_settings.rb`:
Upstream added a constraint on a setting textually close
to glitch-soc-only settings.
Applied upstream's change.
- `lib/sanitize_ext/sanitize_config.rb`:
Upstream added support for the `translate` attribute on a few elements,
where glitch-soc had a different set of allowed elements and attributes.
Extended glitch-soc's allowed attributes with `translate` as upstream did.
- `spec/validators/status_length_validator_spec.rb`:
Upstream refactored to use RSpec's `instance_double` instead of `double`,
but glitch-soc had changes to tests due to configurable max toot chars.
Applied upstream's changes while keeping tests against configurable max
toot chars.
Conflicts:
- `app/views/settings/profiles/show.html.haml`:
Upstream redesigned the settings page, where glitch-soc had changes because of
the ability to set some custom limits.
Went with upstream's design while keeping our custom limits.
- `yarn.lock`:
Upstream updated dependencies textually close to a glitch-soc-only dependency.
Updated the dependnencies as well.
Conflicts:
- `app/views/settings/preferences/appearance/show.html.haml`:
Upstream fixed a translation bug in the theme selector that is absent from
glitch-soc due to our different theming system.
Discarded upstream changes.
- `streaming/index.js`:
Upstream changed the signature of a function to change its return type.
This is not a real conflict, the conflict being caused by an extra
argument in glitch-soc's code.
Applied upstream's change while keeping our extra argument.
Conflicts:
- `package.json`:
Upstream changed various script definitions in lines surrounding the one for
`i18n:extract`, which had glitch-soc-specific changes.
Updated the scripts as upstream did, while keeping our changes to
`i18n:extract`.
Status reactions had an API similar to that of
announcement reactions, using PUT and DELETE at a
single endpoint. I believe that for statuses, it
makes more sense to follow the convention of the
other interactions and use separate POST endpoints
for create and destroy respectively.
Turns out the strange error where it would delete
the wrong reaction occurred because I forgot to
pass the emoji name to the query, which resulted
in the database deleting the first reaction it
found. Also, this removes the unused set_reaction
callback and includes the Authorization module for
the status reactions controller.
Erin Shepherd
