We've been getting reports for a few months of folks coming back to
their Zulip window after a night's sleep and finding it scrolled to
the bottom, past dozens or hundreds of messages that they haven't
read. Oddly, the pointer is actually still located where it should be
(verifiable by hitting the Up key), but it's too late: everything
below gets marked as read because bottom_whitespace is in view.
There's only a few places in the zulip codebase where we scroll the
page down, and this is the main one of them. My best theory for what
could be happening is that the browser is, in its overnight
power-saving mode, not granting the Zulip window the resources to
actually repaint the early scrolls. This, in turn, would cause
scrolling down to happen that is not limited by the need to keep the
pointer in view.
I don't think that this fully closes the issue; ideally, we'd have a
reproducer and much more precise detection logic for this situation,
but it should mostly resolve the problem with likely no user-facing
visible harm.
This reverts commit ba8dc62132.
As best I can tell, the old configuration was correct for what Django
wanted. Further testing is required, but this at least brings
.tx/config to match the actual filenames; I think our Chinese
translations have been broken until now.
This commit combines a `tx pull` with updating the translations.json
files to change the values of those items whose key is equal to the
value. The new value is an empty string.
Previously we used to mark a key as unstranlated if its value was equal
to it in translations.json. This had an issue because it didn't allow
otherwise valid cases where key was equal to the value.
This commit solves the problem by disallowing an empty string as a valid
translation and then using the empty string as the value for all the
unstranslated keys.
Fixes#5261
This fixes the width of the call-to-action button to be auto, as it
previously was set in the #hero to be 150px which forced the words in
the button to wrap to two lines.
Currently when hovering on an emoji it will focus it, which makes
the browser by default scroll down or up to include the entirity
of the focused element. This corects the scrollTop to what it was
before the focus event adjusted the scroll position.
This is a follow-up to #6869.
Previously, you had to hover over the smaller area where the emoji
image was to select it, whereas the user expectation is that hovering
the emoji's padding should select it as well.
This commit makes the arrow key navigation and mouse hover affect the
same state such that for example if one moves the mouse over some emoji
and then hits down-arrow the cursor will move down by one from where he
left the mouse at rather than beginning from the top-left corner.
Fixes: #6827.
When the iOS keyboard is open and up, the positioning gotten by
getBoundingClientRect will display a `top` value that is short by the
height of the keyboard, which will usually end up placing things north
of the top of the screen.
By changing to jQuery $.fn.offset instead, the positioning appears to
be correct in all cases; iOS keyboard up, down, and desktop usage.
Fixes: #6366.
This will look through all users and not just ones active in the last
three weeks but only when you are searching with the right sidebar
input box.
Fixes: #5775.
The header line-height is too short when it collapses to multiple
lines so this sets the line-height back to "normal" from "30px"
which sets the text further apart.
The min-height for the error pages was not updated to reflect the
height of the new footer, so this updates the value and makes it a
non-scrolling page in most browsers again.
This sets the column width of the upload table actions and size
columns to always be 75 so that the buttons are always in the same
line and take up the least amount of space possible with that
constraint.
This tries to toggle the next item when clicking on an <h2>
in the sidebar, however we want to first check the next item is
an <ul> element, so that we are collapsing or showing a list,
instead of something like an <h2> which currently happens with
the "#guides" element.
This removes the underline on hover and changes the text to get to the
index.
This also changes it to an <h2> tag so that it will be more inline
with the styling of the rest of the sections, but without the
down chevron.
This restyles the headers to make the <h1> more prominent with a line
break below it, and the <h2> to be less prominent with smaller text and
pushed closer to the <p> tags.
This adds a hover state to just the checkmark that makes it darker
than the line hover state that allows a user to know that the
checkmark is allowed to be toggled.
The `have_scrolled_away_from_top` logic goes way back
to November 2012.
Now we unconditionally load older messages when we scroll
to the top of the feed. Before this bug, you could get
"stuck." It was a bit difficult to reproduce, but with
the right combination of render window sizes and batch
sizes, you would hit the home key quickly and hit the top
of the feed in a way that the flag got in the way of
going back in history.
Fixes#6628
We had a bug where once you scrolled back far enough
in the message view, your "window" for rendered messages
would be at the max, and `prepend` was not adjusting
the window correctly. Now we follow the example of
`append` and call `maybe_rerender`.
This partially addresses #6628, where users were
reporting that the home key stopped going up in their
feed. There was another bug at play for that issue
as well, which is fixed in the next commit.
Before this, the home key would go to the first message in our
render window. Now we go to the first message in our local
list of messages. (Note that there may still be older
messages, so it will still often take multiple uses of the
home key to truly get to the top of your feed.)
In the refactoring in 31d3b1ecc0 that
fixed live-updating of the medium-size avatar data, we started just
fetching the normal-size avatar, not the medium-size avatar. We fix
this by changing this code path to pass in the user object and
construct the URL using that.
While we're at it, we switch to using the user ID, not the email, to
construct these avatar URLs.
Previously, we relied on fetching the name of the user from the data
attributes on the individual elements, when we can get a more reliably
up-to-date value from the people.js data structure we're fetching
anyway.
"Mobile push notifications always" is now indented and a
sub-setting of "Mobile push notifications when offline".
It can be selected only when the outer setting is
selected, otherwise it is greyed out.
Fixes#6570.
We've iterated on this code incorrectly something like 3 times now, so
it's worth rewriting it with a lot of comments in a way that makes
sense.
The main actual functional change here is that modified key + enter
now is consistently the opposite of enter (in terms of whether to
provide a newline or send the message) in all cases.
Fixes#6489.
Now that we display the name and aliases of the currently focused
emoji at the bottom of the emoji picker, we don't need to display
the title text for emojis separately.
Fixes: #6111.
Emoji showcase refers to the space at the bottom of the emoji
picker we use for showing name as well as aliases of the currently
focused emoji.
Fixes: #6110.
This hack was used to fix the broken flag emojis in emoji-picker.
It was broken due to the incomplete migration to iamcal dataset.
See issue #4775 for more details.
This commit switches to use sprite sheets for rendering emojis
in all the remaining places, i.e., message bodies and composebox
typeahead. This commit also includes some changes to notifications.py
file so that the spans used for rendering emojis can be converted
to corresponding image tags so that we don't break the emoji rendering
in missed message emails since we can't use sprite sheets there.
As part of switching the bugdown system to use sprite sheets, we need
to switch the name_to_codepoint mappings to match the new sprite
sheets. This has the side effect of fixing a bunch of emoji like
numbers and flag emoji in the emoji pickers.
Fixes: #3895.
Fixes: #3972.
The sidebar selectors may not exist at a particular point on load but
we’d like to realistically cache the results once they are, so we try
to load them live until we know that a valid selector has been found.
This call to update the users scrollbar is inside a huddles update
method which should only affect the group PMs, so we can remove the
update function.
The `exports.build_user_sidebar` method already calls the resize
function, so there’s no need to call it again or wrap it in the
`actually_update_users_for_search` method.
When a `data-sort` is clicked in the body, it will trigger an attempt
to find the closest `list_render` instance, retrieve it from memory,
and then sort by the particular method specified.
This allows for someone to specify a generic sorting function which
accepts a prop to sort by, a sorting function which runs with just a
function on the whole object, and the ability to remove the sorting
function in play.
This adds the perfectScrollbar to the uploads table so that it will
function properly in the settings container since the parent node has a
perfectScrollbar.
This moves the stuff that should not scroll with the table such as the
search box and tips so it is moved out to be above the
`.progressive-table-wrapper` element.
The "View file" option will open the file in a new window if it
is a filetype that can open in the browser and if not, it will just
trigger a download or whatever the browser's settings are.
This adds a max-width constraint to the hero content so that the images
inside the hero don’t keep expanding forever and eventually outside of
the hero’s bounds.
Fixes: #6713.
This adds the perfectScrollbar to the right side and theoretically
updates it any time a piece of code interacts with the sidebar and
updates the counts of users displayed in it.
This shows the download instructions only selectively based on
whether the device has download instructions for it. This means
currently it shows the page for Windows, Mac, and Linux.
The list needs to be set to use perfectScrollbar so that it can
scroll due to the fact that it resides within another instance of
perfectScrollbar.
Fixes: #6351.
The popovers for the message down chevron and left sidebar had
strange side padding and non-uniform padding between the top and
bottom. This changes them to all have the same padding as the
nav `#gear_menu`; none on the sides and 5px on the top and bottom.
This moves the chevron arrow and the "All Streams" text over 10px
so that it will be inline with the hashes below and not be up
against the edge of the screen in some collapsed modes.
The issue before was that the left sidebar would become too tall for
the screen because the standard header that has “STREAMS” and buttons
is 20px tall, and this one is 30px tall. This makes it much shorter,
changes the text to be the same style as the “STREAMS” text (medium
grey, uppercase text).
The markup is then fixed to be significantly less verbose than before —
changing a list to just a simple link.
This combines two different selections of the
`#stream-filters-container` into one chained selection and writes a
comment on how it is possible due to the return value of `.css`.
We were having an anchor tag inside a button which is incorrect HTML.
Chrome and safari handle this case but firefox doesn't and hence the
dropup menu wasn't opening on firefox.
The progressively rendered table extends too far down, causing the page
to scroll needlessly, which then causes there to be two scrollbars — a
scrollbar for the table and a view scrollbar outside that.
Fixes: #6391.
This fixes the issue on subdomain.zulipchat.com/login/ where the
organization name will have characters such as the lowercase "g"
cut off near the bottom due to the line-height being too small and
the overflow being hidden.
This re-arranges the properties to fix that issue.
This adds a centered layout for mobile and responsive screens where the
emoji picker is guaranteed to be in the center of the screen, and the
rest of the screen darkens behind it.
Fixes: #6291.
We want to scroll the left sidebar to the top as soon as the user
zooms in on a stream, and we don't want to wait for the server,
otherwise we'll get jumpiness.
This commit is a bit complicated, because we do full redraws of
the topic list frequently, and we don't want to randomly obliterate
our "No more topics found" message, so we need to keep a bit of
extra state around.
We now use a template to render the "more topics" link.
We also remove an unnecessary conditional and an unnecessary
attribute.
Finally, our unit tests are a bit more granular now.
This will make testing a bit easier (we can stub stuff before
building the widget), and it will eventually give us more control
on redrawing the topic list.
We were parameterizing max_topics, but it made the calling sequences
unnecessarily complicated. We don't ever override the value, not
even in tests, so now we just set in build_list().
This puts build_list on the widget object, which will make it a
bit easier to unit test, and it's more consistent with the rest of
the function. This also reduces the scope of the `my_stream_name`
variable and moves the initialization of `self.topic_items` into
build_list.
If a user re-narrows to another stream before our server gives
us more topic history, or they zoom out, we can avoid drawing
the topic list. Note that our data structures will still be
updated, although the only time that really matters is for
the corner case of a low-traffic stream. For a low traffic
stream that only had 3 or 4 topics in the original message
fetch, but has longer history, the next time you open the
stream in the sidebar, even when you're zoomed out, you will
see more topics.
Despite a few warts, we are going forward with getting topic
history from the server when you click "more topics." This
commit simplifies the code by removing the feature flag
checks.
Our old optimizations to prevent re-rendering of locally echoed
messages created a lot of code complexity. This commit is an
experiment to simplify the code, which it clearly does. The
danger of re-rendering messages is flicker, but our message
view has changed since the original local echo code was written.
It's kind of confusing to have a filter function that has massive
side effects. Now we just have a simple loop where we triage
some messages into non_echo_messages and do an early-exit in the
loop function. This change also introduces the more explicit
variable name of `non_echo_messages`; before we were shadowing
`messages`.
This fixes the characters like “g” and “l” from overflowing the input
bounds while maintaining the previous height of the input box itself.
Fixes: #6665.
This removes the `box-sizing` attribute on the spectrum input that was
causing the characters to overflow the bounds of the padding, which
would cut off things like the bottom of a “g” or the top of an “f”.
Fixes: #6361.
We need a migration to clear the tutorial_status for existing users,
so that we don't show hotspots to anyone who signed up for Zulip in
the month or so since we deleted the old tutorial.
On a standard keyboard, 'q' is to the left of 'w', so it makes sense
for the hotkeys for the left and right sidebars to be `q` and `w`,
respectively, not the other way around.
Previously, if the operand was an invalid email, the site would
throw a scary-looking browser error. Now, it has the same behavior
as other search exeptions, and simply returns no messages found.
This allows user to view all group private conversation messages
with a specific user. That is, it views all the the group private
messages from groups which include the given user.
Add search suggestion for group-pm-with. Add operator name
and description in "Search operators" tab.
Add change in tab name to "Group Messages" when using this operator.
Add frontend_tests for group-pm-with search operator.
Fixes: #3882.
Update variable name in static/js/filter.js from 'message_ids'
to 'user_ids' for better understanding. As it is an array of user
recipients of a particular message.
This commit implements support for copying over static files
for all bots in the zulip_bots package to
static/generated/bots/ during provisioning. This directory
isn't tracked by Git. This allows us to have access to files
stored in an arbitrary zulip_bots package directory somewhere
on the system. For now, logo.* and doc.md files are copied over.
This commit should act as a starting point for extending our
macro-based Markdown framework to our bots/API packages'
documentation and eventually rendering these static files
alongside our webhooks' documentation.
This reverts commit c953759486.
The client side logic for dealing with server counts is actually
fine, as far as we know, but there are still some data-related
issues with cleaning up old unread counts.
The current behavior is that when you subscribe to a new stream
the scroll position moves back up to the top because the list updates,
when in reality the user shouldn't notice this, so we record the
previous scroll position and then apply it once the DOM update is
finished before the next paint cycle.
Fixes: #6606.
The home icon was too far to the right and did not have equal
padding within the <li> tab so this makes the padding an equal
10px on the left and right and none on the top and bottom.
Checking for window.bridge !== undefined is how the old desktop app
had always been tested for in the JS codebase.
Also, "Zulip Desktop" is how this should have been spelled in the first place.
Tweaked by tabbott to provide a proper commit message.
Fixes#6580.
This makes the total left sidebar real estate 40px taller and brings
it flush with the bottom of the screen, giving more room to the
streams list.
Fixes: #6549.
This adds perfectScrollbar to the default streams table
because it currently is inside another perfectScrollbar which
actually makes it impossible to scroll the table normally without
enabling the perfectScrollbar library on this.
Fixes: #6391.
Given a `message_id` and `emoji_name` this function returns the
alias of the emoji user used for reacting to this message, otherwise,
if he has not reacted returns the passed `emoji_name` as it is.
If a reactions picker is open then don't auto-hide the element over
which it is based off. Earlier we were inconsistently auto-hiding
some elements while keeping others visible.
Change the reaction popover to be based off the container elements
for the various message control icons. This will enable us to easily
control the visibility of the base element when the popover is opened
or closed. Also removes redundant `reactions_hover` class.
Bootstrap's `fixTitle()` function removes the base element's original
title attribute. This commit fixes some weird behaviors by restoring
the original title of the element on which the popover is based off.
This displays an error at the top of the screen on page load that
will inform any user with the userAgent string "ZulipDesktop" that
they should upgrade to our newer electron app
Fixes: #6551.
This now shows `blueslip.error` and other JS exceptions in development
in the alert box at the top of the page. Hopefully this will make it
a lot easier to notice newly introduced JS exceptions when working on
Zulip.
Tweaked by tabbott to handle all JS errors, not just blueslip.error.
Fixes: #6155.
This makes the JavaScript tooltips more legible by increasing
the font size and decreasing the line-height, while also increasing
the opacity of the tooltip from 0.8 => 1.
The function is confusing and added unnecessary complexity, given that it is
only called in one place, and is not a function that should be exposed to
other modules.
If you typed in more than one word for a stream with multiple words in
it's name, it would not show up in the search list. This fixes that
and adds some more tests covering the entire functionality of the
filter.
If both users haven't posted in the current topic, then
as a second order sort, check which user has posted first
in the stream as a whole.
Fixes part of #5956; we still need to sort by sending in the
organization.
compilemessages command now does all the heavy lifting by creating a
language_name_map.json file under locale directory. This file is used
by get_language_list to retrieve the require information.
Fixes: #6486
The server sends down lists of unread message ids in various
buckets, and we now use those on the client to provide more
complete counts of unread messages.
Use jQuery DOM construction methods, rather than string concatenation,
to keep things structured and to stay clear of the lint rules introduced
in ee6235d71.
Until we get search bubbles, the search text is kind of a
distracting detail for most users. This just makes the
height 2px smaller for now. This will also make more text
show up on mobile web.
If we use string concatenation to span i18n strings across multiple
lines then we end with such strings to be translated by the translators:
```
"This is the first line"\n + "This is the second line"
```
We should use variables in i18n strings to give proper context to the
translators. If the pattern is this:
```
i18n.t("Count " + count + " items")
```
Then it will be captured like this:
```
{"Count" + count + "items": ""}
```
Which is not good for the translators.
This comes from, in the dev environment:
./manage.py makemessages -l en -l ja
tx push -s -l ja
tx pull -f -l ja --mode=developer
The makemessages and `tx push` provide Transifex with the updated set
of strings and metadata. I'm not entirely sure why that's necessary,
but without it a lone `tx pull` left the server crashing with numerous
zerver.models.DoesNotExist exceptions.
(So to be precise, there was another `tx pull` at the start, just like
the final one. That shouldn't matter for the result, though.)
This responsively hides the last row of testimonials that only has a
single one it the row at certain widths in which the grid is a [4, 4,
1] layout, making it simply [4, 4].
This removes the scaling feature where a plan that a user
hovers over would increase in size and have a box shadow.
This used to be here because there was a clickable region that
would pop open an overlay that would have more plan information,
but that feature no longer exists.
This changes the second plans block label to dark grey responsively
when the width is less than 1390px because the white text does not
appear on that portion of the white background when the block collapses
down a line.
The original had two bugs in this line of code that cancelled each other
out. 4d0f304 fixed one, causing hotspots to no longer appear. This commit
fixes the second.
The one we currently have is the same as the one in the registration
emails, and also very low resolution (doesn't show well in lightbox).
Likely we should replace this with something completely different at
some point, but this is an improvement for now.
The moving label is acting like a placeholder here, and should be positioned
as such for consistency with the rest of the site. The two forms with a
moving label are /accounts/find/ and /accounts/password/reset.
This restructures organization settings and permissions to be
more accurately grouped and for the permissions page to not be too
long.
CHANGES:
PROFILE:
(this was split out)
organization-profile-admin.handlebars:
form #1:
name
description
(SUBMIT)
avatar:
(UPLOAD)
(DELETE)
SETTINGS:
organization-settings-admin.handlebars:
language (mostly untouched)
message editing:
time limit/history/retention
message feed:
mandatory-topics
preview images
preview websites
PERMISSIONS:
organization-permissions-admin.handlebars
(mostly stuff was removed)
Joining:
restrict domains
require invite
User Identity:
name changes
email changes
Streams/Emoji:
creating streams:
waiting period (ADDED)
adding emojis
(SUBMIT) for whole panel
The profile group (name, description, avatar) were split into a new
page that did not previously exist, and the permissions was stripped
of message settings (message editing, message feed), but keeping the
"waiting period" input and putting it in the "Streams & custom emoji"
section.
Fixes: #5844.
That's what the font is actually called, and should help future Zulip
developers save time trying to figure out what's up and why our font
is unrelated to the "Humbug" font on the Internet.
Previously, the font-weight for 500 did not exist in our set, so
all of those examples defaulted to 400. This fixes them to stay
as they previously were by migrating them to a font weight of 400.
The sidebar height was set to shorter than what the sidebar actually
was and so it would overflow the parent. By resetting it, it now
will not overflow.
Fixes: #6315.
This cleans up the styling of the page by adding the white box
around the text content, and by making the header the same
`.get-started` class as other headers in the portico-signup group.
This removes some of the min-height constraint on .flex that would
push the footer down futher than it should have been and would
introduce scrolling to the page.
This removes the old flex model for the footer to stay at the bottom
of short pages and instead always just positions the footer below
the portico content and makes the content generally tall enough that
issues don't occur.
This replaces the old footer that has one section with a small list of
items. This expands the footer to have multiple sections.
Actual content tweaked and tagged for i18n by tabbott.
This fixes the padding above the –– OR –– line on the register form.
This also fixes the left-alignment of the form starting at 795px
wide.
Fixes: #6265.
This adds perfectScrollbar to the `.subscriber_list_container` to
allow for the table to scroll naturally again. This was broken
because when perfectScrollbar is put on the parent element, any
naturally scrolling element within it will not scroll naturally
anymore.
Tweaked by tabbott to update the scrollbar on rerender.
Fixes: #6215.
The hack used to make the placeholders in the ::after element
work correctly is no longer needed, so we can revert the width
of 200% back to 100%.
The hack is no longer required because Vaida split these into
two tables, of which in the second table there are no columns,
which means that 100% represents the table width rather than
the width of the first column.
Fixes: #6271.
This is hacky, but I can't figure out another way to do it that
doesn't cause other problems.
Ideally, we'd add some sort of exclude rule to our HTML template
linter so we can check the rest of the file.
This refactor will facilitate making it possible to set CSS properties
on this controls span; in particular, we're hoping to disable user
selection of the whitespace in this region.
The main side effect of this refactor is that we need to add JS code
to also hide the icon-vector-pencil element, since it's now in a new
span.
When we were deleting a stream from the sidebar using the
stream/delete event, we were getting tracebacks due to this sequence
of operations:
* remove id from stream_list.stream_sidebar
* rebuild stream list
* remove sub from stream_data
This fixes the bug by calling stream_data.delete_sub() first.
Deletions are tricky if you do things out of order. We can probably
prevent tracebacks by having a deleted flag, but that can just cause
different problems.
Last commit tweaked by tabbott to fix a small bug in handling the case
where the user was not subscribed to the delete stream.
We continue to have page_params.realm_default_streams, but
now we do lookups on whether a stream is a default stream
by using a Dict indexed by stream_id.
We are also careful to update that during live updates.
This fixes a flaw that we weren't updating the list of realms
correctly for events that remove a default stream.
This is an attempt to more easily debug a traceback we've seen a few
times. The issue likely has to do with local echo, which would be
confirmed if this reports a local-echo-style message ID.
This commit extends the `compute_placement()` function in
`popovers.js` to take into account height/width of popover as well as
positioning preference. If vertical positioning is desired and the
popover fits in either 'top/bottom' positions then we don't check for
`left/right' positions. Earlier the behavior was to prefer
'left/right'positions over 'top/bottom' positions, which resulted in
the emoji picker popping incorrectly to the left.
This further improves the emoji picker by introducing two new behaviors:
1: If the cursor is at the end of the input box then pressing `right_arrow`
moves the focus down into `emoji_catalog.
2: If the currently focused emoji is the first emoji in the `emoji_catalog`
then pressing `left_arrow` moves focus back to search filter.
This never made sense to be a flag on the UserMessage table, since
it's not per-user state. And in fact it doesn't need to be in a
database at all, since it's easily computed from content anyway.
Fixes#1099.
Apparently, local rendering of previews had broken sometime in the
last few months in a refactoring that resulted in us passing a string,
rather than an object, into markdown.js.
Using weird characters when filtering options items in these various
settings pages would throw exceptions whenever they didn't form a
valid regular expression.
Previously on mobile, clicking on a message would make the compose
box open, but this is a relatively finnicky event whenever scrolling
so we realistically want to open the compose box on long-tap (with
a 750ms delay) to prevent false clicks and provide a closer-to-native
experience.
This makes the /help/ sidebar more discoverable at windows less
than 1000px in width because it makes it stick out a bit when it
is closed with the hamburger menu at the top.
Fixes: #6038.
The bot API key when selected in the "Your bots" panel would have
some whitespace due to the fact that the HTML markup would make
some space between the span and button.
Fixes: #6189.
For whatever reason, the clipboard doesn’t want to work if you use a
jQuery click trigger. Perhaps because the jQuery event trigger doesn’t
create a native event at all. By doing this however, it doesn’t appear
to affect any other code but does allow for the clipboard to work again.
Fixes: #6002.
Previously, we didn't check the organization-level settings when
rendering a message list; instead, we only checked it when putting
messages into the message_store. That resulted in the state being
stale in the event that the setting controlling whether one can edit
messages was changed.
We remove some node tests, because revidving the node test for their
new home in message_list_view would be more work than we probably want
to do with an upcoming release. We basically need to be better about
exporting functions like populate_group_from_message_container and
set_topic_edit_properties, so we can do fine grained testing.
When we get around to the node tests, rather than exporting these
functions, it might make sense to create a new module with a name
like message_container.js, which would have all of these
last-second type of data manipulations on message objects. This
would be nice to split out of message_list_view.js. MLV is our
biggest module, and it's mostly cohesive, but it's real job
should be about assembling messages into a DOM list, which is
probably 80% of the code now. The 20% that I'd want to consider
splitting out is actually closer in spirit to message_store.js.
Thanks to Steve Howell for helping with the node tests.
Variable `show_topic` was assigned in both branches of the
conditional, but the assignment in the "then" branch was useless,
since the variable wasn't subsequently read. Hence the assignment can
be dropped, leaving the "then" branch empty. The "if" statement can
then be simplified by removing the "then" branch entirely and flipping
the condition. Since `show_topic` is now only used inside the "if", it
is slightly tidier (though semantically equivalent) to move its
declaration inside.
Variable `stream` is a local variable (declared on [line
51](85c3f59292/static/js/tab_bar.js (L51))). It
is not read after this assignment, which hence becomes useless.
This fixes a confusing bug where administrators would be offered the
convenient topic-edit pencil even if message editing was actually
disabled.
This doesn't yet fix the real-time sync issues of changing the setting
without reloading.
Fixes#5946.
This refactors and fixes unicode issues where entities don't display
properly due to being a special character that seems to be rendered
incorrectly in a non-deterministic way every time.
This fixes 2 bugs:
* If you perform a search and search results are empty then if you try
to navigate using arrow keys, page-down/page-up etc. it will give a
traceback.
* Search for example 'a' and then navigate to the last of the search
results using arrow keys. Now press 'tab' to go back the search box
and restrict the search to for e.g. 'ab' and now try to navigate
using arrow keys, page-up/page-down etc you will get a traceback.
In this commit we basically do these things:
* Clear up section_head_offsets before pushing stuff in it so that
its size doesn't keep on growing indefinitely with time and users
opening emoji picker.
* Make use of popover element to find the correct element in DOM
to scan for section elements. This prevents us from filling stuff
twice into section_head_offsets because of presence of two
elements for '.emoji-popover-subheading' in DOM since popover
destroy is an async call.
* Using this popover element also helps in avoiding manuplation
of the DOM elements of the popover that was destroyed (Because
popover destroy is async it still maybe around). One instance of
this is associating scroll event with the right instance of
'.emoji-popover-emoji-map'.
Normally the "n" key skips over muted streams, but if we
are currently narrowed inside a muted stream, it will now
go to the next topics within that stream.
For me the use case was that I have a stream I check up on
about once a day, and "n" would be super useful for me to
clear out unread counts while still skimming some content,
and without having to temporarily unmute the stream.
This causes `upgrade-zulip-from-git`, as well as a no-option run of
`tools/build-release-tarball`, to produce a Zulip install running
Python 3, rather than Python 2. In particular this means that the
virtualenv we create, in which all application code runs, is Python 3.
One shebang line, on `zulip-ec2-configure-interfaces`, explicitly
keeps Python 2, and at least one external ops script, `wal-e`, also
still runs on Python 2. See discussion on the respective previous
commits that made those explicit. There may also be some other
third-party scripts we use, outside of this source tree and running
outside our virtualenv, that still run on Python 2.
In this commit we are moving the .emoji-popover-emoji.reaction
click handler to register_click_handlers() so as to have parity
with rest of the code design.
We now use similar code for A/D hotkeys as we do for the "n"
key.
The old code was using jQuery operations that got tripped up
by our splitters between active and inactive streams.
Fixes#4569
This allows us to traverse a list backwards, cycling to the
bottom as needed.
This code is going to be used for the "A" key that cycles
upward in the stream sidebar. It's probably overkill for
that use case, but it does give us O(1) behavior and avoids
the pitfall of accidentally mutating a list when reversing it.
Previously, the Zulip webapp would throw an exception if you used a
character like "+" in your search query, since we were using regular
expressions, when really we should have been just searching for
characters.
Use perfectScrollbar on settings sidebar, since the default scrollbar
makes settings menu break when not enough vertical space available.
Add perfectScrollbar to main settings section, and reset the scrollbar
position when switching between tabs.
Also delete the z-index on `.settings-list` since it makes the
perfectScrollbar covered.
Fixes#5216.
This adds the authors to the Zulip repository on GitHub from
/authors/ along with re-styling the page to fit the same
aesthetic as /for/open-source/ and other product-pages.
In the case of no subdomains, the input is too large because the JS
calculated size does not account for padding. The correct solution is
to do this in JS.
The new endpoints are:
/json/mark_stream_as_read: takes stream name
/json/mark_topic_as_read: takes stream name, topic name
The /json/flags endpoint no longer allows streams or topics
to be passed in as parameters.
This is the first part of a larger migration to convert Zulip's
reactions storage to something based on the codepoint, not the emoji
name that the user typed in, so that we don't need to worry about
changes in the names we're using breaking the emoji storage.
Here are the functions in top_left_corner:
get_global_filter_li: pure code move
update_count_in_dom: simplifed copy of similar function in stream_list.js
update_dom_with_unread_counts: pure code move, split out from function
of same name in stream_list.js
delselect_top_left_corner_items: pure code move
handle_narrow_activated: pure code move + rename
handle_narrow_deactivated: pure code move, split out from from function
of smae name in stream_list.js
This function was actually de-selecting stream sidebar items
before. Now we just explicitly de-select top-left items in it,
and we do stream-sidebar stuff in update_stream_sidebar_for_narrow().
Previously, when you switched to a stream narrow with the central
message outside the range of messages cached in the browser, we would
reset the UI for loading more messages, but not actually reset the
state for whether it should be possible.
This seems to have been an oversight in refactoring back in 2014.
Fixes#6109.
With this flag turned on, all streams will have a "more topics"
link, and clicking that link will always fetch topics from the
server to show a complete list of topics that you have had messages
for on that stream.
Note that if you only recently joined a public stream, your list
of topics won't go back to before you joined the stream, even though
that content is searchable. We may change that in the future, but
we will need to be careful about spamming folks who frequently
unsubscribe from streams.
Until we have an easy way to consistently determine whether a
stream has more topics than have been loaded already, we err
on the side of showing a "more topics" link. This in some ways
leads to a more consistent experience where you can zoom in on
any stream, even one that's really new.
This fix simplifies how we re-render topic lists when we
re-narrow or zoom out from a topic list.
* The topic_list.zoom_out() no longer gets called as
part of re-narrowing, and we eliminate the clear_topics
option.
* For all situations where we narrow to a filter that does
not have a topic, we simply call the new function
clear_topics().
* The stream_list code no longer calls remove_expanded_topics()
in cases where the new narrow has a topic. This allows us
to optimize away scroll/flicker churn a little more easily.
As part of this, we rename maybe_activate_stream_item() to
update_stream_sidebar_for_narrow(), since the function clears
stuff as well as turning stuff on.
This is mostly a pure code extraction. It makes the call
to reset_to_unnarrowed() happen later in sequence.
The order of operations here is mostly unimportant, but
there may actually be some tiny user-facing benefit
in terms of having the logic happen more sequentially.
BEFORE:
reset streams
fix top left
redraw streams
AFTER:
fix top left
reset streams
redraw streams
If you go into "more topics" for a stream with many topics,
and then scroll down, and then zoom out again with "All
streams", we make sure the active stream is still in view.
We have code that can automatically scroll an element into "view"
in its container. We use this for stream sidebar rows inside the
stream list.
Generally the stream sidebar rows are small enough to fit into
the container, and the prior algorithm worked correctly for that
scenario.
If you have lots of topics, however, and a short screen, the
algorithm was being too aggressive. For example, if the top
wasn't showing, it would scroll the top into view, but at the
cost of scrolling the bottom out of view.
This fix makes the general scrolling algorithm more tame.
Part of the user-facing problem is that the element we pass
into the scrolling code for the stream sidebar rows is bigger
than the part of the row that actually should be shown on
screen. Nevertheless, it makes sense here to make the general
algorithm more robust.
If you read a message, then got a topic edit for it, we were
adding the message to our data structure of unread stream/topic
messages.
Now we guard against this in unread.update_unread_topics. I
no longer expose an update() method in unread_topic_counter,
since we really want to do the unread check at a higher level
to keep other data structures consistent.
Category 'All' -> text 'Filter by category'; icon chevron right when
the dropdown is closed, icon chevron down when the dropdown is open
All other categories -> text CATEGORIES[state.category]; icon chevron down
Steve Howell
Brock Whittaker
Tim Abbott
Umair Khan
brockwhittaker
derAnfaenger
Harshit Bansal
Cynthia Lin
Vishnu Ks
Sarah
Nalin Bhardwaj
Anupam-dagar
Ali Rizwi
YJDave
Shubham Dhama
Rhea Parekh
Alena Volkova
Rishi Gupta
Joshua Pan
Cory Lynch
Abhijeet Kaur
Eeshan Garg
akashnimare
Tommy Ip
Sampriti Panda
Vishnu Ks
Yusei Tahara
Akash Nimare
Greg Price
Vaida Plankyte
rht
Steve Howell
David Taylor
Yago González
Max Schaefer
franziskagoltz
Aditya Bansal
Rafid Aslam
Jack Zhang