We now sweep all active messages for avatar changes and update
the message items and re-render, rather than patching the
DOM. This avoids some quirks that happen when subsequent messages
get sent and we re-render previous messages out of the message
store.
Our approach here is similar to how we do full-name updates.
The comments explain why this change is correct. This change is
useful because it's better to not have dead code paths, both because
it makes our life easier for coverage analysis, and because the else
statement provided the illusion that it could actually happen.
If the analysis in that comment is wrong, we'd rather have a 500 error
so we fix the bug than things silently sorta working.
This arguably regresses the Zephyr experience, in that we no longer
consider 'foo.d.d.d.d.d' to be something that gets narrowed in with
the rest, but that's a pretty rare use case anyway.
In practice, using that many '.d's anyway only happens a few times a
year.
In this commit we are modifying pretty print tool to support
Django and html singleton tags. For Addition of html singleton
tags template parser was modified to emit psudeo
html singleton end tags to accompany html singleton tags and
token class was updated to have line_span field.
In this commit we improve the way errors are handled in our
template parser and thus improving the displayed messages in
case of errors. Eg. Errors in case of unbalanced quotes now
makes more sense displaying line and column information
including line where error might be sourced.
Our client code will now receive avatar_url in
page_params.people_list during page load, so it will be
able to use more current urls for old messages (the client
already had some logic for that and was just missing the
data).
We also add avatar_url to the realm_user/add event.
When we change the avatar, we make sure to always send a
realm_user/update event (even for bots).
We also needed to add avatar_version and
avatar_source to our active users cache.
In f75af94984 I added some
lines of code that made it so that live updates for avatar
urls would affect messages currently in the browser.
This change worked well when the live update actually happened,
but then the next time the user would reload, the avatar in
the message pane would regress back to showing the avatar urls
from the server (which could have caching issues of their own).
This fix removes a couple lines of code that had the intended
effect of making all of your messages from any given sender
show the same url (good) but which generally grabbed
the url from an old message (bad).
After this fix, we go back to having old messages possibly
showing the old avatar urls, but new messages will display the
new avatar.
(There are lots of moving parts in the avatar system, because
not only do browsers cache image urls, but our server caches
messages and recipient info, so there have been "fixes" to
avatars since this change that are valid fixes in their own
right but not directly relevant to this commit.)
This provides a fairly intense highlighting of when you're hovering
over a given emoji reaction element.
We may want to tone down the color a bit; I'm hoping for some feedback on this.
This makes life a lot easier for people inviting users to a new Zulip
organization, since they can give some form of context now.
Modified by tabbott to clean up CSS, backend code flow, and improve
the formatting of the emails.
Fixes: #1409.
The current logic that we have is as follows:
* If a message is locally echoed, the draft is stored via the locally
rendered message, and that system takes care of it. So no need to
store it here.
* If the message isn't locally echoed, we don't close the compose box
until, so the content is safe here as well. It'll be saved as a draft
if the compose box is later closed due to a failure sending.
We now make tests that call EventsRegisterTest.do_test()
explicitly specify whether calls to apply_events() would
change the state of initially fetched data. Generally
these tests exist to test that logic (as well as verifying
schemas of events), so if they stop testing that logic, it
is usually a broken test.
Some tests are exempted from the check here, because I think
they don't really change state--such as updating messages or
notifications. You can set state_change_expected to False
for those tests.
For all the tests that deal with flipping boolean flags, I
set their value to False before calling do_test twice now.
For the authentication backends, I mock the settings so that
more backends are "supported" and therefore part of the event
and the fetched state.
Finally, for the bot tests, I make sure to use a bot the user
can access.
This replaces the settings toggle which had the same markup as the
current component toggle, but not the same JavaScript, along with
having an issue with inline-block spacing, with the new JS generated
one.
Steve Howell
lonerz
Tim Abbott
Rishi Gupta
adnrs96
Sourav Badami
hackerkid
Ayush Jain
Brock Whittaker