The upload option will no longer be limited to strictly S3 uploads. This
commit serves as a preliminary step for supporting LOCAL_UPLOADS_DIR as
part of the public only export feature.
We've been seeing nondeterministic failures in this test suite in CI
that we can't reproduce locally; these print statements should help
track them down.
This is the only function in TestEmailMirrorLibrary, so we rename this
class to more appropriate TestGetMissedMessageToken, clean it up a bit
and add some extra checks to finally get email_mirror.py to 100% test
coverage.
log_and_report and its helper functions were mostly old code no longer
well adapted to how email mirror works currently, as well as having no
test coverage. We rewrite this part of the email to report errors in a
similar manner, and add tests for it. We're able to get rid of the
clunky and now useless debug_info dictionary in process message, as
log_and_report only needs the recipient email in its third argument.
The only place in which process_stream_message used debug_info was to
set the 'stream' key, which would only be used if ZulipEmailForwardError
was raised after this line in the code - which is impossible, because after
that line only send_zulip (which doesnt raise this exception) and
logger.info get called, then process_stream_message successfully returns
and then process_message succesfully returns as well. So this debug_info
code wasn't doing anything. We remove it.
This reverts commit a717ac6d85.
.gitattributes affects the behavior of many Git operations like rebase
and merge and mergetool, not just the output of show and diff.
Treating yarn.lock as binary was making it too hard to deal with
yarn.lock diffs when we actually needed them. Also, the GitHub
website ignores this setting and suppresses large parts of diffs
regardless.
Signed-off-by: Anders Kaseorg <anders@zulipchat.com>
These seem to have been there since the very first version of our
markdown styling, and I can't imagine why we would the behavior of not
line-wrapping links now.
(I think the "weird bug" mentioned in the comment history might have
to do with an old animation when you hovered over a link on portico
pages).
This appears to caused by trying to reuse this bit of spacing logic
from the Help Center's CSS rules.
I'm not altogether happy with this fix, but it resolves the issue and
we can defer further work until we're ready to clean up the
portico/landing pages CSS more generally.
Mostly rewritten by Tim Abbott to ensure it correctly implements the
desired security model.
Administrators should have access to users' real email address so that
they can contact users out-of-band.
Clients won't have access to user email addresses, and thus won't be
able to compute gravatars.
The tests for this are a bit messy, in large part because our tests
for get_events call subsections of it, rather than the main function.
This provides a clean warning and 40x error, rather than a 500, for
this corner case which is very likely user error.
The test here is awkward because we have to work around
https://github.com/zulip/zulip/issues/12362.
In email hidden case (that is when `email_address_visibilty` is set to
everyone), for "non admins", this commit hides emails from:
- compose box user typeahead.
- PM user typeahead
In email hidden case, for admins, email is shown in user typeaheads.
In email hidden case (that is when `email_address_visibilty` is set to
everyone), for "non admins", this commit hides emails from:
- user popover
- custom profile popover
In email hidden case, for admins, email is shown in both user popovers and
custom profile popovers.
Along with this, we refactored settings_org.populate_auth_methods to use
HTML function after rendering all auth methods rows rather than appending
each row individually, which actually is a good practice.
Also in this commit, to compare `current_val` and `changed_val` in
`check_property_changed` function of the property
`realm_authentication_methods`, which are objects, and we found here
https://stackoverflow.com/a/1144249 that there is no easy way to do so. So
I followed this approach,
```js
JSON.stringify(obj1) === JSON.stringify(obj2)
```
but before converting them to string we want the same order of keys, so we
used `sort_object_by_key` to sort `current_val` by keys and
`get_auth_method_table_data` always return `changed_val` having keys
sorted.
Since these refactor were closely related we kept them as a single commit
here.
Fixes: #11954.
Actually, this is a preliminary commit which adds a general
`sort_obj_by_key` function to sort objects according to keys.
In this commit, we have refactored `populate_auth_methods` function by
extracting the logic for the desired `sort_obj_by_key` and used that to
sort `auth_methods`, but the main motive of this function is to sort
`realm_authentication_methods` in `check_property_changed` to sort
`current_val` in the upcoming commit.
The `LocalUploadBackend` returns a relative URL, while the `S3UploadBackend`
returns an absolute URL. This commit switches to using `urljoin` to obtain the
absolute URL, instead of simply joining strings.
This commit also adds a small functionality change where the results of
each webhook fixture message sent is now displayed to the user.
With a small tweak by tabbott to fix a styling bug.
Fixes#12122.
Note: If you're going to send fixtures which are not JSON or of the
text/plain content type, make sure you set the correct content type
in the custom headers.
E.g. For the wordpress fixtures the "Content-Type" should be set to
"application/x-www-form-urlencoded".
The `transition` property does not need prefixing. In fact, very
few properties need that nowadays. So remove it to simplify
the code. This is strictly a refactor with no style change intended.
Many of these styles shouldn't have been prefixed even if needed.
The prefixes exist exactly because the implementations might differ
from the incoming standard.
Looking at the supported browsers:
https://caniuse.com/#search=transition
We see that this property has had mainstream support from 2012 and
was supported on Firefox in 2006 !!!