Fix a bug where the compose box cut the message without warning the user
the message pasted was longer than the allowed. It was fixed by stopping
cutting the message off and showing an indicator whenever the limit exceeds
and removing the indicator as soon as message gets less than that.
The cut off for showing the indicator is set as 90% of the limit.
Fixes#15909
Co-authored-by: João Maurício <carvalho.joaomauricio@gmail.com>
Since do_create_realm also creates general and core team streams,
we rename general to verona right after the realm is created. Mostly
because we dont really want two additional streams and this might
probably make it easy to review things.
There are puppeteer test changes because, we have a new "core team"
stream in tests as well as there is a new default notification stream
"Verona". Because of this tests in message-basics for example have
to be changed since the newly added core team affects the order in
which we navigate through the streams using arrow keys.
The extra await for selector was added in subscriptions test to make
the tests wait. Without the await the tests were passing ocassionally
and failing in some other times.
Fixes#6967
Right now, on clicking `quote and reply` on any message, the quoted
message is always inserted at the top of compose-box irrespective of
the current cursor position. Also, after insertion of the quoted
message, the cursor is shifted at the end of the compose box.
This commit changes this behaviour to insertion of quoted message
at the current cursor position with a newline at the end of quote
and moving the cursor position to that newline after insertion.
A newline is added at the beginning of quoted message only if there
was some content already present in compose box before the previous
cursor position.
Tested on Google Chrome and Firefox browsers on Ubuntu dev environment.
Fixes: #16836
This commit renames confirm_dialog_yes_button class to
confirm_dialog_submit_button. This will help in keeping
a general class name when deduplicating the code for
confirm_dialog and edit_fields_modal.
This commit first moves the compose.validate() function out
with the functions that are needed by it. Then one by one
checked for which function is now not needed in compose.js.
This moves all validation related functions out of "compose.js"
to "compose_validate.js".
Splitting compose announce variables out of compose.js.
This commit moves the "user_acknowledged_all_everyone" and
"user_acknowledged_announce" out of compose.js to reduce
cyclic dependency of compose_validate on compose.js.
Moving wildcard mentions to compose_validate.
The wildcard mention settings are mostly used while validating.
Also to reduce the cyclic dependence of compose in
compose_validate, the related wildcard mentions are moved out to
compose_vaidate.js.
This also converts reset_acknowledged functions to set values
by passing values.
We currently have the resolved topics prefix shown as part of
the topic name in the left sidebar. However this causes inconsistency
while showing topic names. Hence this adds support for showing the
prefix in the cutter to the left of the topic name.
Fixes#18989.
The current system to change stream information like Stream name and
description isn't consistent with what we use everywhere else. It's
also slightly difficult to maintain.
Co-authored-by: Pragati Agrawal <pragati22066@gmail.com>
This commit adds a new trigger for compose.start that is
"new private message". It will clear the message recipients
whenever compose.start is called with this trigger.
This solves the bug, when a person is in a PM narrow and
clicks the new private message button, it opens the
composebox with the recipients filled out with whoever
you're narrowed to, rather than opening a new, blank PM.
CZO link for the issue
https://chat.zulip.org/#narrow/stream/9-issues/topic/.22New.20private.20message.22.20isn't/near/1222712
This shortcut now renarrows to an empty topic, rather than to the
stream, as long as the topic input is nonempty.
This was debated in the original implementation of this feature:
https://github.com/zulip/zulip/pull/9511#discussion_r190323319
Having seen the experience in practice, the current behavior is
surprising and we should migrate to the other option originally
considered. In particular, the argument that "narrowing to the stream
would still tell you the topic is empty" turns out to be inaccurate,
since the history within the stream may not be in the most recent
messages in the stream.
Fixes#19122.
I also remove a conditional here that's not necessary. For
anything that's obviously a template in static/templates,
we now create a stub implementation that behaves according
to whether we're stubbed or being included as a partial.
d779a1c tweaked `get_full_datetime` to return a string instead
of a {date, time} object. This function is used for recent topics
too but wasn't fixed to use the string.
This resulted in showing 'undefined undefined' in tooltip.
Previously, we showed a `Copied!` alert on copying link to a message
irrespective of the fact that the link was copied or not.
Hence add an event listener that shows the `Copied!` tooltip only
if the action was successful.
Fixes#19019.
We separate "Your account" section to two different sections -
"Profile" section for user name, custom profile fields, and avatar
and "Account & Security" section for email, password, role, api-key
and deactivating button.
Another important change here is that the modal for changing name
is removed and now the name has a simple input text box and it
behaves similar to inputs for custom-profile-fields.
Fixes#18848.
We rename `stream_subscription_info.hbs` and its related
variables and functions to `stream_subscription_request_info.hbs`.
So that its clear that it is used for displaying info about
requests sent for stream subscriptions.
This is a follow-up for #18957.
We add disabled prop on buttons only and we add the css for
buttons (and some specific class elements) when disabled as
'cursor: not-allowed' or 'pointer-events: none' which means
the user cannot use these buttons when disabled.
This is not the case for the avatar widget as we use a div
element there and not button and neither those specific
classes which has appropriate styles defined.
We use the avatar image block in two ways - for showing image
and as a clickable target (which is actually a div element)
for uploading, so instead of adding 'disabled' prop we are
hiding (adding display: none) the div such that it is not
visible on hover and also cannot be clicked.
This commit extracts the logic in can_edit_topic_of_any_message test
in a separate function test_message_policy. This change will help in
avoiding duplication while adding test for can_delete_own_message.
We ignore deactivated users when they are being added to streams.
To give current user information about the ignored user ids
we show it along with other details about subscribed and
unsubscribed users.
This partially addresses: #18949.
When the hash changes to `#reload...` before a reload, the app tries
to show default_view since there is no `case` handled for it. What we
want to do is ignore this hashchange, since we're about to be reloaded
(so it's not helpful) and if the timing is wrong, it can cause the
browser to reload back to "Recent topics" rather than saving the
user's view.
This fixes a bug introduced in 0e65225d06.
With tests from Riken Shan.
While writing a long message in compose-box, the last few messages of
the current stream gets covered by the compose-box and it gets pretty
annoying sometimes trying to figure out a way to read the last message
of the stream while writing. Right now, the only way to get past this
is to resize `compose-textarea` by using the resize tool at the
bottom-right corner of the `compose-textarea`. But, that small resize
tool is not always readily visible to the user.
The proposed solution in this commit is to reset the `max-height`
property of `#compose-textarea` everytime `bottom_whitespace_height`
is resized such that the total height of `#compose` is always less
than or equal to the height of `bottom_whitespace_height`. Doing
this, the compose-box never covers the last message of the current
stream.
The only problem with this is that if the compose-box is closed at the
time of bottom-whitespace resize, we cannot find the
`compose_non_textarea_height` and so, we cannot reset the max-height
of `#compose-textarea`. To solve this, max-height of
`compose-textarea` is also reset everytime a new compose-box is opened
according to the value of `bottom_whitespace_height` at that time.
Thus, if the compose-box is already open at the time of
bottom-whitespace resize, the max-height of `#compose-textarea` will
also get reset at the same time, whereas, if the compose-box is closed
at the time of bottom-whitespace resize, the max-height of
`#compose-textarea` won't get reset at that time, but it will surely
get reset whenever the user will open the compose-box.
Tested on my Ubuntu Development Environment on Chrome and Firefox browsers.
Fixes: #16038.
We've had for years a subtle bug, where after editing a topic in the
left sidebar that had previously had unread messages (but doesn't
anymore), the old topic might still appear in the sidebar.
The bug was hard to notice except for new organizations or in the
development environment, because the pre-edit topic appeared with a
sort key of -Infinity (that being the max ID in an empty list of
message IDs). But this is an important onboarding bug in reducing
faith in Zulip's topic editing just working, so I'm glad to have it
fixed.
Fixes#11901.
This commit fixes a bug of not enabling/disabling
the email-change button when email_changes_disabled
setting is changed. Bug was because of using wrong
selector.
If a user sets themselves to unavailable, or otherwise
drops out of our presence data, we should still show
them at the top of their own buddy list.
See https://chat.zulip.org/#narrow/stream/137-feedback/topic/Users.20Sidebar/near/1220135
for more context.
I believe this change makes sense as a defensive
fallback, but it's quite possible that we may
want to change the server to record presence info
about users who are "unavailable" and then only
send that info to them (and not their peers).
This was rebased and significantly modified by @showell:
* introduce vars for textarea_val and textarea_caret_pos
* simplify mock_esm call
* use override for hash_util
* minor things like whitespace cleanup
(see #18849 for the original PR)
We now only expose mock_template as a helper in run_test.
This has the following advantages:
* less boilerplate at the top of the file
* more surgical control with setting exercise_templates
* no more "f" hack (or render_foo consts)
* we force devs to explicitly mock the template
See frontend_tests/zjsunit for the substantive changes.
All the changes to the tests are very mechanical in nature.
This appears to be a silly case of using mock_template,
since the tests directly "require" the template, but
we are using it here to set up a generalization.
13 choose 1 is obviously 13, but it's also 13! divided
by the product of 12! and 1!.
The cruftiest part of this change gets fixed by the
subsequent commit (no need to call mock_template at
the top level).
This keeps it consistent with other widgets in message body area.
Set the display position to top to be consistent with
compose control buttons.
Changed the tooltip content to be more readable like
Thursday, May 18, 2017
7:12:53 AM India Standard Time
Also changed timerender.get_full_datetime() to consider
users' 24 hour format preference.
When a user tries to send a (group) private message to a deactivated
user, a compose error is displayed.
Fixes#13766.
Co-authored-by: Signior-X <b19188@students.iitmandi.ac.in>
This modifies the appearance of pills for deactivated users
in the following ways:
* Adds `(deactivated)` with the deactivated user's name.
* Add a tooltip to the deactivated user's pill stating
that one can't send a message to this user.
* Color the deactivated user's pill reddish.
This now also adds tests for user_pills.
Part of #13766.
Co-authored-by: Signior-X <b19188@students.iitmandi.ac.in>
This commit treats deactivated users as valid users
for compose box which prevents the PM list to collapse
when a deactivated user in the PM list is clicked.
Part of #13766
Co-authored-by: Signior-X <b19188@students.iitmandi.ac.in>
This commit modifies the linkifier-edit modal to use newly added
edit_fields_modal framework.
One important change is that we remove the "edit-linkifier-status"
element as the corresponding "edit-fields-modal-status" element
is added in edit_fields_modal.hbs and we also modify the css
accordingly. This "edit-fields-modal-status" is used only for
this modal and remains empty for others, so this change does not
cause problems with other modals.
There is another element which uses "edit-linkifier-status" as a
class, but the css we defined was for "edit-linkifier-status" as
id, so the css change is also safe.
This commit moves the compose_error function to its own module, which
will be useful for future work splitting the compose module.
We also simplify compose_not_subscribed_error to call the
compose_error show function.
We show a spinner inside the button instead of hiding the button
and then showing the spinner in the bottom area. We also disable
the button to avoid repetitive clicking by user.
There is still no need for mock_template to reach into the internals
of mock_cjs. Make it a normal caller of mock_cjs.
This is basically the same as 60f5a00c09
(#18804), because the same abstraction violation was since
reintroduced in the same way.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The language_list_dbl_col parameter in the page_params
is used by only the web client frontend. The value is
calculated in the backend and then passed as a page_param
which is unnecessary considering that the whole process
is beneficial for the front_end only. Hence move the entire
calculation code to the frontend.
Fixes part of #18673.
default_language_name was a part of page_params which is actually
redundant considering that we already have language_list and
default_language available to frontend which can be used to
get the default_language_name and hence prevents the backend
from sending an additional parameter.
Fixes part of #18673.
This avoids the issue of all the topics in recent topics marked
as yesterday after mid-night.
This change also affects other pieces of UI using this function
like buddy list in a similar way.
We now validate the linkifier urls and patterns together, and add
the following additional checks:
1. All groups in the pattern must be used in the URL format string.
2. All groups in the URL format string must be declared in the pattern.
Linkifier pattern is now validated inside the `clean` method.
`filter_pattern_validator` is moved from `clean_fields` to `clean`
method as a safe check. As a result of this, a Puppeteer test case
is updated.
NOTE: The changes here are IN ADDITION to the existing validations.
Fixes#16482.
Co-authored-by: akshatdalton <akshat.dak@students.iiit.ac.in>
This commit replaces the allow_community_topic_editing boolean with
integer field edit_topic_policy and includes both frontend and
backend changes.
We also update settings_ui.disable_sub_settings_onchange to not
change the color of label as we did previously when the setting
was a checkbox. But now as the setting is dropdown we keep the
label as it is and we don't do anything with label when disabling
dropdowns. Also, this function was used only here so we can safely
change this.
If you call mock_template(fn, true), we will
call the actual template code and pass it to your
stub for verification.
We make this opt-in to prevent false positives
on template line coverage.
We special-case our handling of static/js/templates.js,
since it's important that all of our tests have
the Zulip-specific handlers for Handlebars pre-registered.
This runs in roughly the same amount of time as the
previous commit.
We now ignore question edits if the sender of the submessage
is not the message author.
The webapp UI prevents folks from editing the question for
somebody else's poll, but a determined person could use our
low level API to do it. We will add safeguards on the server
side for this, but this change is sufficient to protect the
webapp (and mobile when they upgrade the library).
This is necessary for the new `mock_template` helper added in
97ad6b6b62 to work correctly without
leaks, since it uses `mock_cjs` under the hood.
This logic was added in 216493aae8.
Functions for input pill typeahead in pill_typeahead.js
were dependent on other modules for testing, and most of
those only tested the part required to ensure coverage.
As part of changes done to pill_typeahead for code clarity we
split its functions logic to clearly handle three types of
pills instead of hacky checks. So to completely test the logic
for various possible combinations of types, we add separate tests
for these functions.
Functions in pill_typeahead.js assumed the type of required pills
to be user by default. But this created several readability
issues with its code, as a result, this codebase appeared
hard to change because of hacky checks involved due to this
assumption.
To improve these we use `include_users` check wherever possible
and simplify the code to clarify the logic involved in typeahead
functions in this module.
Also updated existing stream_edit tests to adjust them with changed
code.
In this 90041ff453,
we remove the `info` class from the message action
menu, that's why it was failing.
This commit replaces it with the correct selector.
There is no need for mock_template to reach into the internals of
mock_cjs. Make it a normal caller of mock_cjs.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
We split recent_topics module into recent_topics_(ui + data + util).
This allows us to reduce cyclical dependencies which were
created due to large list of imports in recent topics. Also, this
refactor on its own makes sense.
`user-profile-modal` is shown using `overlays.open_overlay` which
disables mouse pointer events. The user can't click anywhere while a
modal is present, except to close it.
We use `hide_all_except_sidebars` and `hide_all` to hide popovers.
But since the user can't interact while a modal is present,
closing it manually is redundant.
This commit fixes the invalid assert call in dispatch.js and also fixes the
invalid settings parameter of update_display_settings__default_view_all_messages
fixture in event.js, it should be all_messages instead of 1.
Fixes#18685.
We use css() pretty rarely in our codebase, and
it can sometimes be used mistakenly, when a better
alternative is to toggle a class for external css.
It's hard to support the full API in zjquery, so
we just punt and tell folks to create their own
stubs.
Most of the existing tests that were "fixed" here
weren't actually verifying the behavior of the css()
calls, and for those I just create no-op stubs.
In a few places I verify that css() was called as
expected.
This commit renames `delete_draft_after_send` to
`delete_active_draft` for two main reasons:
1) This function does the job of deleting the draft that we
are currently using in compose box and it is independent
of the message sending functionality.
2) This is prep refactoring for new drafts behavior where
we delete the draft if the compose box is cleared and
closed.
This commit divides the user_invite_restriction setting dropdown to
a checkbox and a dropdown.
The checkbox is used for 'realm_invite_required' setting and dropdown
for 'realm_invite_to_realm_policy'.
This separation of UI elements is fine as these two settings are
separate in database also and also helps in removing excess if-else
conditions and switch cases.
There was a lot of duplicate code in test_sync_realm_settings where
each value in common_policy_values was being tested for different
policies (create_stream_policy and invite_to_stream_policy).
This commit deduplicates is by using a for-loop for testing all
common_policy values and extracting the code as a function to test
different policies.
Previously, we had this complicated layering where the right sidebar
logic would display "Last active: foo" but the user popovers would
just display "foo", which doens't make any sense, since the two
settings have equal context about the string.
We deduplicate that and also arrange that the "Last active:" prefix is
used when it's not clear what we're talking about; i.e. all the values
except for "Active now".
There is no clear reason to not use a button element here. According
to the spec pharasing content, which includes the <span> element,
are allowed in the button element.
Manually tested both buttons to make sure it works and made sure all
the selectors are updated by grepping all the selector classes/id in
the handlebars templates that are parents of the button or are
present on the button.
(One of the jQuery handler code got reformatted due to it fitting
the line limit due to one character deletion for the selector)
This commit fixes a rare flake where it accidentally clicks
on "Subscribe to more streams" instead of "Verona"
(stream). This happens because `#stream_filters` list
was not completely updated, i.e. It still had "Venice" and
"Verona" on the list, when it searched for "ver".
When it decided to click on the "Verona", "Venice"
disappeared, which causes "Verona" to move up and
causing "Subscribe to more streams" to click.
Tested by the adding a console log to the handlers and typing Tab
and Enter in the group PM and stream compose box. This two event
handlers are attached to form#send_message_form which is the compose
box form.
The tests set keyCode to 42 previously, which does not represent a
valid key, so I assume they were trying to test a random key, and
replace it with "a".
There is no functionality change caused by this change. We turn the
event key to lowercase so hotkeys work even when Caps Lock is on as
it did before.
Tested by using the keybinds with and without Caps Lock in compose
box:
- "Ctrl + B" and "Ctrl + Shift + B",
- "Ctrl + I", and
- "Ctrl + Shift + L".
(Also tested that "Ctrl + Shift + I" and "Ctrl + L" do not work.)
Tested by doing the following in the compose box input for PMs:
- "Enter" works as expcected with and without input
- "Backspace" when there is no input deletes the last pill
- Left arrow in the input focuses last pill
- Comma works as expected for correct and incorrect input
- Left arrow, Right arrow and Backspace work correctly for pills.
The Event.which and Event.keyCode are deprecated as pointed out by
TypeScript intellisense based on the jQuery types. We use Event.key
instead which behaves similarly to Event.which & Event.keyCode for
our use case.
The only difference in functionality by this change is that the vim
keys won't work when Caps Lock is on. This is because, in this case,
the key property will be "J" instead of 'j'. We can fix this by
adding a mapping for this, however, I think we don't want to handle
this case so I left this change out. Tested by trying out the
everywhere keydown_util is used.
Finally, we also turn off the new-cap rule for tests since I think
it fine to only enforce it on real code and exempting test code is
fine.
In options that we pass to pill_typeahead.set_up we
specify if we want typeahead to support stream or
user_group pills, and use users as source by default.
Using users for source by default, can have unnecessary
suggestions in typeaheads where only user_groups or streams
are needed.
So to solve that, we specify if we want users pill in the input.
This is then utilized in further commits, to clean up hacky code
that deals with intializing source for typeahead.
We make the typeahead a little more nicer but only showing one alias
per language. For example if the user searches for prefix "j", then
the typeahead list should contain "javascript" only, and not "js" and
"javascript".
We now organize the pygment language codes into meaningful categories
- default, custom and aliases.
Further the `lang.json` list now contains a dataset extracted from the
"language" section of https://insights.stackoverflow.com/survey/2020
and is prioritized based on current language trends.
On slower systems it may take more than 4 seconds for the
stataus message to dissapear from the DOM, so it's better
to wait for the default duration of 30 seconds before
proceeding.
Thanks to @Riken-Shah for suggesting the fix.
Since this is currently only useful to interpret presence data, we
send this only if presence is requested.
I'm not sure that server_timestamp is the right name for this field,
but ultimately it should match the main presence API format.
When quoting a reply or mentioning a person having full name matching
wildcard mention, in such case, `get_mention_syntax` doesn't return
mention syntax of format: **full_name|user_id**.
As a result, a normal user can't mention such users and users who
can mention them may unwillingly trigger wildcard mention.
This commit fixes such issue.
Prior to this, we restricted to show any user group suggestions
if silent mention syntax is used. But with the addition of
user group silent mentions, there are cases where one may want to
refer to some user groups without actually notifying them.
So, we add typeahead logic for such cases.
Test cases for silent syntax (@_*) and non-silent syntax (@*)
are added.
Fixes this KaTeX warning:
Warning: KaTeX doesn't work in quirks mode. Make sure your website has
a suitable doctype.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This is pretty much a verbatim code move,
except that I inlined noop and reset_jquery
in the new file.
Also, in the last test, I no longer reset
the compose state.
We split them in the commit: c50dbf8297
because at that time we thought markdown_katex was gonna
require rewireproxy magic in the later commits.
But we later removed the rewiremock dependency in the
commit: 30c7108955.
So now, we can safely merge `markdown_katex.js` test file
with `markdown.js` test file.
The extracted logic is in linkifier.js.
We have decided to name it linkifier.js instead of realm_linkifier.js
because in future when we will add stream-level linkifiers, we'll
likely want them to be managed by this same file.
I'm not sure that settings UI is particularly worth having puppeteer
tests for, so it's possible we shouldn't bring these back at all. But
in any case, it's worth disabling them as they've been failing for
some time.
This commit solves a rare flake, where the `realm
_linkifier.ts` test was failing because there was no
appropriate wait call for the table
(`#admin_linkifiers_table`) to get updated after editing
the pattern.
The command:
codespell --skip='./locale,*.svg,./docs/translating,postgresql.conf.template.erb,.*fixtures,./yarn.lock,./docs/THIRDPARTY,./tools/setup/emoji/emoji_names.py,./tools/setup/emoji/emoji_map.json,./zerver/management/data/unified_reactions.json' --ignore-words=codespell_ignore_words.txt .
The content of codespell_ignore_words:
```
te
ans
pullrequest
ist
cros
wit
nwe
circularly
ned
ba
ressemble
ser
sur
hel
fpr
alls
nd
ot
```
Priyam Seth
Vishnu KS
Priyansh Garg
sahil839
Dinesh
Gaurav Pandey
Priyank Patel
Anders Kaseorg
m-e-l-u-h-a-n
Tim Abbott
Steve Howell
Aman Agrawal
YashRE42
Hashir Sarwar
Ganesh Pawar
Riken Shah
Rohitt Vashishtha
Abhijeet Prasad Bodas
aryanshridhar
Palash Raghuwanshi
Sumanth V Rao
akshatdalton
rht