For filters that cannot be applied locally, we don't know if the
messages are still a part of the filter. So, we remove the
existing message that were updated and let `maybe_add_narrowed_messages`
treat them all as new messages to the filter.
This fixes the bug where existing messages that were present in
the narrow were not updated.
Putting all of the logic in a `finally` block is equivalent to a bare
`except` block, which silently consumes all exceptions.
Move only the most-necessary parts into the except; this lets
`BadImageError` exceptions from `zerver/lib/upload.py` to escape,
allowing better the generic "Image file upload failed" to be replaced
with a more specific message.
It also allows unexpected exceptions, as the previous commit resolved,
to escape and 500. This lets them be detected and resolved, rather
than give users a silently bad experience.
5dab6e9d31 began honoring the list of disposals for every frame.
Unfortunately, passing a list of disposals for a non-animated image
raises an exception:
```
File "zerver/lib/upload.py", line 212, in resize_emoji
image_data = resize_gif(im, size)
File "zerver/lib/upload.py", line 165, in resize_gif
frames[0].save(
File "[...]/PIL/Image.py", line 2212, in save
save_handler(self, fp, filename)
File "[...]/PIL/GifImagePlugin.py", line 605, in _save
_write_single_frame(im, fp, palette)
File "[...]/PIL/GifImagePlugin.py", line 506, in _write_single_frame
_write_local_header(fp, im, (0, 0), flags)
File "[...]/PIL/GifImagePlugin.py", line 647, in _write_local_header
disposal = int(im.encoderinfo.get("disposal", 0))
TypeError: int() argument must be a string, a bytes-like object or a
number, not 'list'
```
`check_add_realm_emoji` calls this as:
```
try:
is_animated = upload_emoji_image(image_file, emoji_file_name, a
uthor)
emoji_uploaded_successfully = True
finally:
if not emoji_uploaded_successfully:
realm_emoji.delete()
return None
# ...
```
This is equivalent to dropping _all_ exceptions silently. As such,
Zulip has silently rejected all non-animated images larger than 64x64
since 5dab6e9d31.
Adjust to only pass a single disposal if there are no additional
frames. Add a test for non-animated images, which requires also
fixing the incidental bug that all GIF images were being recorded as
animated, regardless of if they had more than 1 frame or not.
In the dark theme, disable the hover behavior to display sorting options
for actions column in tables (Which doesn't support this).
Matches existing behavior in light theme.
Fixes: #21137
Added a property `mouse_moved_since_typeahead` to the typeahead class
which tracks whether the mouse has been moved since the typeahead
menu appeared.
The hovered over menu item is highlighted on `mouseenter` only if
`mouseMoved` is true. Otherwise, the cursor is hidden temporarily.
Code substantially reorganized by tabbott.
Fixes: #21018.
For our Subscribers tab we want to enclose our
add-subscriber widget in a form for now, and we
continue to do so, but now we have the form tag
getting created in the parent template, not the
child.
Here is why:
We want to re-use our add-subscriber widget in an
upcoming commit for when you create streams.
In our create-stream UI, there is already an outer
<form>...</form> section of the HTML for the entire
process of creating a stream.
HTML does not yet you nest forms, and even though
browsers just silently ignore an inner form, we won't
want to create semantically incorrect HTML.
Therefore, we want the child template not to have
the form tags around them.
It's possible that we don't really even want to
enclose the input-pill widget and Add button inside
a form for the Subscribers tab, but tweaking that
for now is too risky. (We don't really take advantage
of the form tag in any meaningful way, since we
don't directly submit form data to the server, and
we can't use a single submit handler for the Enter
key and Add button due to some magic in our input
pills.)
We want to avoid submit handlers here, because we may
have embedded widgets that have their own forms or
buttons.
We use "finalize" here to distinguish the two Create
buttons related to streams. You hit one button to
start the UI and then the second button to finalize
the process.
I also fix the bad test idiom of clicking on the
sea-green button.
You don't start auto-completing user groups with an @,
so the previous placeholder text made no sense.
There might be a better way to do autocomplete in
these add-subscriber input pills, but that is beyond
the scope of my current efforts.
It would confuse a future Debian 15.10 release with Ubuntu 15.10, it
relies on the legacy fact $::operatingsystemrelease, the modern fact
$::os provides this information without extra logic, and it’s unused
as of commit 03bffd3938.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The deferred_work events can't be processed in the test suite
environment, since it tries to make requests to the host <realm.uri>
which is "zulip.testserver" and obviously not going to work. Since the
test suite base data set has no animated emoji, there's nothing to do,
and we can skip this step.
This code only runs when applying the migration to an already
provisioned test db - because during an initial db set up there are no
realms yet, so no events get pushed by the migration.
This adds tabindex='0' in anchor tags for allowing the message menu
popover box to be focusable using hotkey 'i' through keyboard.
This is a necessary follow up to
729c09074a, which removed the href
attribute from anchor tags to which click-handlers were attached,
which resulted in making them non-focusable as jQuery doesn't allow
anchor tags to be focusable without tab-index or href attribute.
Fixes#21125.
This seems to have no negative effect, and substantially improves the
visuals in situations where we're displaying a user's name.
While we're at it, switch to a relative font size.
When viewing a narrow such as a search or `sender:` view, where
consecutive messages in the view may not be consecutive in their
original stream/topic context, we should avoid displaying the messages
with a shared sender/recipient bar header, as that creates the
incorrect perception that they are consecutive.
Back in 2013 (bc8bc8567b), we
implemented this via the collapse_messages flag, but it appears more
recent refactoring (no more recent than
dbffb2a614) made it always true.
The original logic was incorrect, in that it only considered full-text
search views, and not other views with this property.
I originally planned to use the existing logic for
can_mark_message_read designed for this purpose, but I think there
might be product reasons why might want the logic to be independent.
Removes the 5px `margin-right` on images, replacing it
with a 5px `margin-left`. This change aligns the images
with the message content while making sure they do not
stick stick together in other layouts.
Aman Agrawal
Alya Abbott
Tim Abbott
Alex Vandiver
Anders Kaseorg
Jai soni
Yogesh Sirsat
N-Shar-ma
Sahil Batra
Steve Howell
Mateusz Mandera
jai2201
ditsuke
kraktus