In narrow screens, the save/discard widget was incorrectly shown
even with no changes, due to the CSS intended to switch it from `inline-block`
to `block display.
Fixed by adding `display: none` to `.hide` class for @media(width < 575px)
to hide buttons on screen width less than 575px.
Fixes#24589.
Zulip already has integrations for server-side Sentry integration;
however, it has historically used the Zulip-specific `blueslip`
library for monitoring browser-side errors. However, the latter sends
errors to email, as well optionally to an internal `#errors` stream.
While this is sufficient for low volumes of users, and useful in that
it does not rely on outside services, at higher volumes it is very
difficult to do any analysis or filtering of the errors. Client-side
errors are exceptionally noisy, with many false positives due to
browser extensions or similar, so determining real real errors from a
stream of un-grouped emails or messages in a stream is quite
difficult.
Add a client-side Javascript sentry integration. To provide useful
backtraces, this requires extending the pre-deploy hooks to upload the
source-maps to Sentry. Additional keys are added to the non-public
API of `page_params` to control the DSN, realm identifier, and sample
rates.
Currently, when a user marks messages as unread in a stream/topic
and unsubscribes from the stream, both subscribe button and compose
banner will remain visible. This change will hide the compose banner
when the user unsubscribes from the stream and hopes to create a
better flow and reduce confusion.
Previously, when you did a "mark as unread from here" operation, we triggered
a full rerender of both the recent topics view and the message feed. This was
needlessly expensive, on and a large server with a somewhat busy CPU from
other applications, can cause a visible lag, even when the message feed that
you're looking at only has like 3 messages in it.
Improve this by passing the set of modified messages to the rerender.
There's likely further improvements to be made here -- we shouldn't need to
do more than toggle the unread markers -- but this should be good enough to
eliminate the visible lag.
Fixes#24263.
Previously the 'Support Zulip' option linked to Github Sponsors.
We now link to our own https://zulip.com/help/support-zulip-project,
which describes more options for how to support the project.
Fixes#24230.
Adds a `div` wrapper with the `rendered_markdown` class to the
content in `confirm_emoji_settings_warning.hbs`, so that the
inline code element has the intended CSS rules applied.
Because the `p` element also has the `rendered_markdown` class,
the margins for this element in the modal are slightly changed
(3px on top/bottom). Previously, the margins for this element were
being set from a bootstrap rule (10px on bottom).
This function was mistyped; it was used in practice both accepting
string[] and number[], with the implementation taking advantage of the
fact that number.parseInt(<int>, 10) = <int>.
The only string[] callers were some overly defensive typing_data tests
that don't match the actual typing_data interface, so we remove the
string[] support and adjust the function's type, as well as those
tests.
When someone hovers over the Github username in the popover
with dark theme, a background will appear. These changes
ensures that the `background-color` isn't overwritten by
other CSS.
This commit adds inline_decorated_stream_name component
which is used to show stream name along with its privacy
type icon. This component is added such that we can align
the icon and stream name properly as there are many
instances where the icon and name are not aligned in the
current UI.
This component is only used in "Archive stream" modal for
now and will be used for other UIs as well in future.
Revises descriptive text and examples at the top of the code
playgrounds tab in the organization settings overlay to be shorter,
have only one example that includes the name field.
Also, adds the `rendered_markdown` class to the HTML code elements,
via an HTML span element so that the specific CSS rules for code
elements with that class will be applied to these examples.
Revises descriptive text and examples at the top of the linkifiers
tab in the organization settings overlay to be shorter, have only
one example and updates help center link text for capitalization.
Adds a link to the help center in the form/input area for adding
a new linkifier, which is consitent with the code playgrounds tab,
and means the link would be there if the form was converted to a
modal.
Also, adds the `rendered_markdown` class to the HTML code elements,
via an HTML span element so that the specific CSS rules for code
elements with that class will be applied to these examples.
Updates `markdown.css` to remove border styling from code
elements, and instead use background-color (as well as
font-family) to visually highlight inline code elements
as distinct from regular text.
Updates code element font-family to be the same as in
`zulip.css`.
Makes padding on the left and right the symmetrical for
inline code elements. Previously there was 4px padding on
the right, but not on the left.
Maintains anchor font-color styling for code elements that
are also links.
Updates `markdown.css` and `rendered_markdown.css` for the rules
in `bootstrap.css` that were being used to style code elements and
removes the now redundant/ignored rules from `bootstrap.css`.
After b21f533af, we now update the organization bots
list on receiving realm_user events since non-admins
can also see bots not owned by them in the list.
But the functions to update bot list should be called
only for bot users and not for others, otherwise it
results in an error.
This can be reproduced by first opening the organization
bots list and then just updating the name of the user.
Calling `redraw_bots_list` on receiving "realm_user/add"
event for non bot users will not raise any error but
still we avoid redrawing the whole list when not required.
This commit fixes the code to call functions to update
bot list only when the event is received for a bot.
This added functionality will be used to compare pygment_language from
Code Playgrounds.
There is a choice of how to sort languages with popularity versus
without popularity. I chose to sort the Code Playground custom language
after other pygment languages based on the reasoning in the comments.
In preparation for adding logic for Code Playground languages, which has
no popularity score, extracting the popularity comparison part of the
logic out to evolve it in future diffs.
compare_by_popularity() actually does two things:
1. compare A and B by their popularity
2. then if needed, compare alphabetically to break ties
Someone reading the function name might not be clear about that detail.
I'm renaming this function to remove the "by_popularity" wording, which
can mislead readers. I also added comments to clarify the detail.
The goal in future diffs is to break this up and make it more self
explanatory.
The `get_pygments_typeahead_list()` function is only used by
`settings_playgrounds.js`. It isn't used by the autocomplete of the
composebox.
Renaming this function to make it clear where it's meant for.