In addition to many small edits for formatting and clarity, a few more
significant changes:
* In the main instructions, refer specifically to restarting the
server and to testing that the config works.
* Add SendGrid to the recommended list, as it seems like people
give it a somewhat stronger reputation these days than Mailgun.
* Discuss EMAIL_USE_TLS and EMAIL_PORT along with host, user, and
password in the "free services" section. Though those bullets feel
kind of duplicative to me already.
Let's get right to the point of how to configure SMTP once you know
what you want. That section is pretty short anyway; and we can have
a first step direct the reader to our suggestions if they don't know
what service they want to use.
Also adjust the hierarchy of the headings: group the various
alternative email services under one heading, and group
troubleshooting together under an independent heading.
Also correct what we say about EMAIL_PORT: the Django default is
apparently 25, so if the provider *does* use the usual port 587
then we'll need the port to be set.
Add a clear heading, and use fewer words and simpler sentences. Also
explain the password thing a bit more, and put that more inline next
to the username.
Also, on checking the Django docs, the default for EMAIL_USE_TLS
is False and for EMAIL_PORT is 25. So most admins, certainly any that
are using an SMTP service on the public Internet (that is at all
decently run), will need to set those settings. Mention that.
This commit adds the ability to pass the sticky option to
in the change_display_setting function in order to have the
feedback element remain visible instead of fading out which
is the default behavior. Also passes true for that option in
two instances on the page.
This commit changes the do_settings_change function so that it
defaults to showing the loading spinner for 500ms before fading
out the the feedback element. It also adds a sticky option so you
can override the fading out of the feedback element and have it
remain visible.
This tweak to our compose on-topic-narrow logic may help make it a bit
easier to do quick replies without needing to re-open compose. I'm
not 100% confident this actually makes Zulip better, but it's worth
testing and getting some feedback.
Fixes#6473.
For a non-empty compose box, we previously considered closing the
compose box when switching topic narrows with content in the compose
box; now we leave it open unconditionally.
Part of #6473.
The margin between the first sentence/heading for the
instructions and the numbered list that follows was too small,
which made headings look very awkward.
The indentation was off because there was no CSS anywhere that
properly indented <ul> elements that were nested inside <ol>
lists with custom numbering.
Since every <li> element is prefixed by a custom list number, in
<li> elements with multiple <p> elements inside, the <p> elements
after the first one did NOT take into account the space occupied
by the custom list number, which resulted in inconsistent
indentation. Now, it does!
This one is one of the most tedious to set up and get working.
We now also rely on the Trello scripts available as part of the
`python-zulip-api/zulip` API package to make the setup process
easier.
After some thinking, I don't think there's any actual value to doing
the ../ style relative links here, whereas there is actual harm from
the links being slightly broken in the current model. We fix this by
just using /#settings as the URL.
Fixes#8978.
For certain queries where both include_history and
use_first_unread_anchor are set to True, we were excluding
historical rows. Now we only use the use_first_unread_anchor
flag to filter rows that we use to find the anchor, without
having it filter the actual search results.
The bug went unreported for a long time, because it only
affected mobile users who had newly subscribed to streams.
Note that we make a small change to the test called
test_use_first_unread_anchor_with_muted_topics, which has
a very scary comment about being "arcane" and "be
absolutely sure you know what you're doing." I think it's
fine.
Also, the new test code would fail before this fix, so it
should help prevent future regressions.
Fixes#8958
This is a bit more than a pure refactor, because we duplicate a
chunk of code to calculate a query inside of
find_first_unread_anchor(), so we're doing a bit more work
than before.
We need this refactoring to start decoupling find_first_unread_anchor
from get_messages_backend for the case where include_history is
True. This will happen in a subsequent commit.
The only test that changes here is a direct test on
find_first_unread_anchor(). All other tests pass without
modification, and we have decent coverage on get_messages_backend.
We use an array now to build up the list of search operands and
then consolidate the special search handling after the loop (which
means setting the flag, putting two more columns in the query, and
using ' '.join to build the string).
We have a debugging statement for some obscure errors we get
when narrows have search terms. We now show all the narrow
operators. This isn't really to improve debugging; it's more
to make it easier in the next commit to extract a function
that would make search_term have to be passed back in a tuple.
But it shouldn't hurt debugging either.
Refactoring in this file had resulted in the logic for
html_settings_link being duplicated and extra logic being needed to
ensure these variables were set where they were needed.
This fixes subscriptions_html not being rendered properly in the /help
and /api pages, in addition to removing duplicate code.
This is a pure refactoring and just pulls the function out
to the top level of the module. (The prior commit extracted
it inside a larger function to make a nicer diff.)
Fixes#8965.
Mark_message(s)_as_read is used in marking a message as having been
read by the browser, rename it to notify_server_message(s)_read to
avoid any confusion.
We've had a longstanding bug where the streams settings code
was getting an i18n'ed value in the middle of a callback from
the toggle component, so it would have been broken for
non-English sites. And then a recent cleanup of the toggle
code introduced a bug where the callback-in-the-callback was
getting stale state, so English sites broke too.
This fix just simplifies everything by using the key that
comes into our callback to determine whether we filter or not.
Fixes#8945
This is a recent regression where we I refactored the toggle
component. For some reason the old code was waiting until
after the callback to set some of its state, and I did the
same thing when I simplified how the state was stored.
Under the old code, this didn't manifest as a bug, although
the old code was problematic for other reasons.
This "fix" doesn't actually change anything user facing, as the
follow up commit fixes the proximal problem more directly. And
the toggle component is still prone to people writing code that
tries to inspect the state of the widget as it's being built.
For info overlays (keyboard/markdown/search help) we now let
the modal portions of the widget have focus, so that you can
page around. And then tab switching still works with the arrow
keys.
This is a pretty thin abstraction to prevent having to put
magic numbers in code, doing the which/keyCode hack, and remembering
to all preventDefault.
Hopefully we'll expand it to handle things like shift/alt keys
for components that want their own keyboard handlers (vs. going
through hotkey.js).
Greg Price
Tarun Kumar
Armaan Ahluwalia
Balaji2198
Shubham Dhama
Cynthia Lin
Tim Abbott
Priyank
Eeshan Garg
Rishi Gupta
Ben Reeves
Steve Howell
Nikhil Kumar Mishra
Shubham Padia