Previously, emoji.json was read from
"$ZULIP_PATH/node_modules/emoji-datasource-google/emoji.json".
This path doesn't exist in production when installing from scratch from
a release tarball. And so, we ensure emoji.json exists by copying it to
`static/generated/emoji`.
With tweaks to comments by tabbott.
Fixes: #23469
In 2f547ea, the custom `has_error` logic in BaseDocumentationSpider
was removed in favor of checking whether any errors were logged.
One reference to that method was missed in UnusedImagesLinterSpider
and went unremarked as no images were added to the documentation
that were not used.
Replaces that reference with an error log.
Fixes a warning caused by using the deprecated `2.6` value
of the `REQUEST_FINGERPRINTER_IMPLEMENTATION` setting.
Upgrades `settings.py` to what would have been generated
by starting the documentation_crawler project using the
`startproject` command of scrapy v2.7.
This is still messy, in that it does the `chdir` using a different
library than the rest, but it's at least more consistent with the
codebase and it should be possible to bulk migrate the `ZULIP_PATH`
calculation, which we have in a dozen files, to pathlib later.
Setting `credential_source` is used when assuming role credentials --
that is, when running as one role, use the AssumeRole right to become
someone else.
The AWS command-line tools only do this if `role_arn`, the role to
assume, is also set -- if it is not set, it transparently falls
through to IAM role attached to the EC2 instance profile. However,
with the `aws-sdk-go` package, used by Teleport, this configuration
produces an error.
Remove the `credential_source = Ec2InstanceMetadata` line, which isn't
necessary for the AWS CLI, and interferes with Teleport operation.
This commit introduces the change of rendering private messages
section as collapsible, whose data-fetching logic came with zulip#21357.
We now have separated out `Private messages` from `top_left_corner`
section and shifted it below the `global_filters` in a different
separate section along with stream list with common scroll bar
in left-sidebar.
The new PM section will be opened by-default on loading the page
and will have a toggle-icon in its header, clicking on which makes the
section collapse/expand accordingly.
In default view, only recent 5 PM threads would be shown
and would append the active conversation as the 6th one at last
if not present in those 5, similar to how topics list work.
In PM section with unreads, a maximum of 8 conversations
would be shown and rest of them would be hidden behind
the 'more conversations' li-item, clicking on which takes
to the zoomedIn view of PM section where all the present
PM threads would be visible and rest of the sections of left-sidebar
will get collapsed.
Fixes#20870.
Co-authored-by: Aman Agrawal <amanagr@zulip.com>
Before Zulip 4.9, the Zulip install process left any already-installed
rabbitmq with whatever nodename it had previously configured. Wince
this encodes the name of the host when it was installed, this does not
function well with containers.
Leave rabbitmq-server uninstalled, which lets the Zulip installation
process set the nodename to `localhost`, which ensures that it is
usable across container restarts.
Since /communities/#all was failing the spider test due to `#all`
not being present as an `id` on the page, we exclude it from
this check saying that it is not really a documentation page.
Fixes#22984
Add an `@` icon in unread topics where user is mentioned.
We track a new set of `stream_id:topic` pairs for the unread mentions
so that recent topics instantly knows if a topic is unread and mentioned
or not.
This commit adds private messages to the Recent topics view, to make
it an all-encompassing overview of recent activity visible to the user.
We add a filter "Include PM" to toggle whether PMs should be shown in
recent topics.
Fixes#19449.
This way Puppeteer doesn’t have to re-download Chromium every time we
install new JavaScript dependencies.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This script pulls from our previously custom-written emoji strings
and fills in the rest from CLDR. It also removes 4 custom emoji which
collide with some of the new CLDR names (they will now just be called
by their CLDR name).
Fixes “E713 Test for membership should be `not in`” found by ruff (now
that I’ve fixed it not to ignore scripts lacking a .py extension).
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The implementation is simple, we just check if the
the message sender is a notification bot to decide if we
should show the read receipts list.
We also update the modal content styling to match the padding at the
top of the modal.
Fixes#22905
Transitions the frontend of the web app to no longer use the
user status `away` field for setting a user's activity status
to be 'unavailable' (which is now a deprecated way to access
a user's `presence_enabled` setting).
Instead we now directly use and update the user's `presence_enabled`
setting for this feature.
Renames frontend code related to the feature to `invisible_mode`
vs `away`.
We lose node test coverage in `user_status.js` because we are now
using `channel.patch` to send these user setting updates to the
server.
Removes the temporary updates to `server_events_dispatch.py` (and
related tests) made in a previous commit, since we no longer have
or need the `away_user_ids` set.
Renames the filename so that it accurately reflects its contents
given the changes to the "Recommended setup" page in the previous
commit, and updates all links accordingly.
As a prep-commit for updating the billing / corporate pages for
demo organizations, initialize tippy.js with a default setting
for portico pages to use in general.
Fixes#21037.
This is part of fixing #19371. To bulk-add new emoji regularly,
mobile needs to know which servers support which emoji.
`staticfiles_storage.url` generates a unique URL with a hash
based on the file content, which lets mobile know if it needs
to update its locally stored data.
Django has always expected this, but Django 4.0 added a system check
that spews warnings in production.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
We should rearrange Zulip's developer docs to make it easier to
find the documentation that new contributors need.
Name changes
Rename "Code contribution guide" section -> "Contributing to Zulip".
Rename "Contributing to Zulip" page -> "Contributing guide".
Organizational changes to the newly-named "Contributing to Zulip":
Move up "Contributing to Zulip", as the third link in sidebar index.
Move up renamed "Contributing guide" page to the top of this section.
Move up "Zulip code of Conduct", as the second link of this section.
Move down "Licensing", as the last link of this section.
Move "Accessibility" just below "HTML and CSS" in Subsystems section.
Update all links according to the changes above.
Redirects should be added as needed.
Fixes: #22517.
This is preparatory commit that does basic UI set up for
user group edit in group settings overlay. This allows us to
write proper hashchange logic for user group settings overlay
under diffrent situations.
The work in this commit will be extended in further commits
to add proper UI and group edit logic.
Add support for creation of user groups using right panel
of new user group settings overlay being developed as part
of https://github.com/zulip/zulip/issues/19526.
In further commits we will add support for editing user
groups using right panel of the overlay.
This commit also introduces a minor bug related hashchange
for #groups which would be a quick fix once we have UI
for group edit on #groups overlay.
Dedicated overlay for user group settings is added as part of
addressing zulip#19526.
The newely added overlay is currently empty and more UI
related to settings is to be added in further commits.
A preparatory commit to have legacy user group settings logic
as we move forward to redesign the user group settings.
This is done so that current user group settings are functional
while we are working on the redesign, and also to make it clear
that most of the code in this file will be deleted and developers
should avoid spending much time on it.
Adds an API endpoint for accessing read receipts for other users, as
well as a modal UI for displaying that information.
Enables the previously merged privacy settings UI for managing whether
a user makes read receipts data available to other users.
Documentation is pending, and we'll likely want to link to the
documentation with help_settings_link once it is complete.
Fixes#3618.
Co-authored-by: Tim Abbott <tabbott@zulip.com>
Now that we can assume Python 3.6+, we can use the
email.headerregistry module to replace hacky manual email address
parsing.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
`DiscoverRunner.run_tests` has a return type of `int`. While
`Runner.run_tests` has a wildly different `Tuple[bool, List[str]]`.
This refactors it so that we have the correct return type, by passing
the additional information about failed tests through a side effect to directly
write the failed tests to a file.
Note that we have to make `failed_tests_path` optional as otherwise the method
signature will not be compatible with the supertype.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Mypy does not know the acccurate return type of `get_runner` that is
determined by the `TEST_RUNNER` setting. We need to cast it to the
correct type to use methods like `get_shallow_tested_templates`. Note
that we import conditionally to avoid adding runtime dependency on
`zerver`.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Emoji that we specifically choose names or aliases for will be
stored in this new file so that we can generate emoji_names.py
from the custom names and also CLDR (for the rest).
This new file isn't used for anything yet (no user facing change);
it will be used in future commits.
distutils is deprecated in Python 3.10 and will be removed in Python
3.12. We don’t need a full-powered version parser for this anyway.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This fixes having incompatible types of different items in
the dictionary, and makes `**identity_args` type-safe.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
When we were preparing the conversion to ES modules in 2019, the
primary obstacle was that the Node tests extensively relied on the
ability to reach into modules and mutate their CommonJS exports in
order to mock things. ES module bindings are not mutable, so in
commit 173c9cee42 we added
babel-plugin-rewire-ts as a kludgy transpilation-based workaround for
this to unblock the conversion.
However, babel-plugin-rewire-ts is slow, buggy, nonstandard,
confusing, and unmaintained. It’s incompatible with running our ES
modules as native ES modules, and prevents us from taking advantage of
modern tools for ES modules. So we want to excise all use of
__Rewire__ (and the disallow_rewire, override_rewire helper functions
that rely on it) from the tests and remove babel-plugin-rewire-ts.
Commits 64abdc199e and
e17ba5260a (#20730) prepared for this by
letting us see where __Rewire__ is being used. Now we go through and
remove most of the uses that are easy to remove without modifying the
production code at all.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This ensures that anything that looks like a full release will
identify itself as such. It requires that the version have a tag at
the current commit (though it doesn't need to be pushed yet), as well
as the commits exist in `main` or a release branch in the remote.
dmypy is innately incompatible with `local_partial_types` being
`False` as it uses fined-grained incremental mode for caching.
We need to enable `local_partial_types` for mypy as well so that the two
will behave the same way.
This requires us to add additional type annotation in certain siuations
but that's fine.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
mypy daemon performs significantly better than running the regular
mypy cli tool when we type check the entire codebase multiple
times locally.
This adds running mypy daemon as an option for both
`tools/run-mypy` and `tools/lint`.
To ensure daemon messages like "Daemon started", "Daemon stopped"
won't get printed we filter any output that starts with "Daemon".
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Support for this header was removed in Chrome 78, Safari 15.4, and
Edge 17. It was never supported in Firefox.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Tornado finishes the request automatically. Avoids this error, hidden
until commit 81f7192ca3 (#22301), when
the browser tab is closed:
Traceback (most recent call last):
File "/srv/zulip-py3-venv/lib/python3.8/site-packages/tornado/web.py", line 1683, in _execute
result = await result
File "tools/run-dev.py", line 280, in prepare
await self.finish()
tornado.iostream.StreamClosedError: Stream is closed
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Since python-debian is GPL licensed, our script that imports it should
arguably be GPL licensed as well.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
`postgresql-14.4` is a notable upgrade in the PostgreSQL series, as it
fixes potential database corruption from `CREATE INDEX CONCURRENTLY`
statements which are run while rows are modified[1]. However, it also
requires an upgrade from `libllvm9` to `libllvm10`, which means it is
not installed by a mere `apt-get upgrade`.
Add the `--with-new-pkgs` flag to all of the potentially relevant
`apt-get upgrade` calls, so that this (and similar) packages are
upgraded successfully.
[1]: https://www.postgresql.org/docs/release/14.4/
Add none-checks, rename variables (to avoid redefinition of
the same variable with different types error), add necessary
type annotations.
This is a part of #18777.
Signed-off-by: Zixuan James Li <359101898@qq.com>
The production CI image starts `rabbitmq-server` but does not stop it,
which leaves a stale `/var/run/rabbitmq/pid` file in the image.
`rabbitmqctl wait --timeout 600 /var/run/rabbitmq/pid`, which is run
after starting the rabbitmq node, reads the PID file and waits for the
PID to be running, and for rabbitmq's port to be responding to pings.
If it reads an old PID file before the new PID is written, it
aborts (all but the first and last lines are output from `rabbitmqctl
wait` that is hidden by `/etc/init.d/rabbitmq-server`):
```
* Starting RabbitMQ Messaging Server rabbitmq-server
Waiting for pid file '/var/run/rabbitmq/pid' to appear
pid is 341
Waiting for erlang distribution on node 'rabbit@fc8f64d6acdb' while OS process '341' is running
Error:
process_not_running
* FAILED - check /var/log/rabbitmq/startup_\{log, _err\}
```
If it failed, the `production-upgrade` script tried to start
`rabbitmq` again -- despite it already still starting in the
background. These two attempts conflicted, and often one or both
failed.
Stop `rabbitmq-server` when building the image, which removes the
stale PID file.
Initializing the Zulip client opens a long-lived TCP connection due to
connection pooling in urllib3. In Github Actions, the network kills
such requests after ~270s, making the later `send_message` call fail.
Use a singular call to `zulip.Client()` early on to verify the
credentials, and do not cache the resulting client object. Instead,
re-create it during the final step when it is needed, so we do not run
afoul of bad TCP connection state.
This would ideally be fixed via connection keepalive or retry at the
level of the Zulip module.
The error message a user gets from the linter when using the
data-tippy-allowHTML attribute now conveys the fact that the
<template> tag is supposed to hold the tooltip content. This
might make understanding the correct workflow easier for
someone who encounters this error.
We previously forked tornado.autoreload to work around a problem where
it would crash if you introduce a syntax error and not recover if you
fix it (https://github.com/tornadoweb/tornado/issues/2398).
A much more maintainable workaround for that issue, at least in
current Tornado, is to use tornado.autoreload as the main module.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The previous commit changed our response for the root domain on a
self-hosted Zulip server containing no realms to be a 404.
This made the HTTP headers verification done here less robust than
would be ideal, in that one could imagine a bug that makes every URL
404 passing that test.
So we instead change the check to verify the /server_settings API,
which doesn't have that problem.
To provide a smoother experience of accessing a web public stream,
we don't ask user to login unless user directly requests a
`/login` URL.
Fixes#21690.
As a consequence:
• Bump minimum supported Python version to 3.8.
• Move Vagrant environment to Ubuntu 20.04, which has Python 3.8.
• Move CI frontend tests to Ubuntu 20.04.
• Move production build test to Ubuntu 20.04.
• Move 3.4 upgrade test to Ubuntu 20.04.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
We had skipped these in #14693 so we could keep generating a friendly
error on Python 3.5, but we gave that up in #19801.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This will allow Zulip release announcements to credit contributions
made to Zulip projects beyond the server in our release announcements.
Fixes#19044.
Specifically, this desupports:
android 4.4.3-4.4.4
baidu 7.12
ie 11
kaios 2.5
op_mini all
although we’ve already been blocking IE 11 since 3.0 (#14662).
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The most notable change here is that when you are adding
subscribers to a stream as part of creating the stream,
you can now use the same essential pill-based UI for
adding users as we do when you edit subscribers for an
existing stream.
We don't try to exactly mimic the edit-stream UI or
implementation, since when you are adding subscribers
during create-stream, we are just updating a list in
memory, whereas in the edit-stream UI, we immediately
send info to the server.
Fixes#20499
Removes `LEGACY_PREV_TOPIC` which is no longer needed due to the
message edit history migration.
Also remove additions to the linter exclude list that were added
earlier in this commit series.
We fix the mutation of caller and other bad patterns, as well as
adding explicit typing to make the code readable.
We also update the OpenAPI documentation for previously
undocumented `prev_strem` field in the `/get-message-history`
endpoint for API validation testing.
Co-authored-by: Lauryn Menard <lauryn.menard@gmail.com>
These types will help make iteration on this code easier.
Note that `user_id` can be null due to the fact that
edit history entries before March 2017 did not log
the user that made the edit, which was years after
supporting topic edits (discovered in test deployment
of migration on chat.zulip.org).
Co-authored-by: Lauryn Menard <lauryn.menard@gmail.com>
Changes in a529dc8 to raise exception for invalid file name
has removed support for passing full file paths.
This commit fixes it.
Thanks to Steve Howell (showell) for reporting this.
We are going to move to this code organization for
managing streams:
stream_create.js
stream_create_subscribers.js
stream_edit.js
stream_edit_subscribers.js
The modules stream_create.js and stream_edit.js historically
manage the entire process of creating and editing stream
data (respectively).
Going forward both will delegate most of the subscriber-specific
pieces to either stream_create_subscribers or stream_edit_subscribers.
The stream_*_subscribers modules will be somewhat similar in
nature, but the way that we manage subscribers at creation time
is a bit different than how we manage subscribers at edit time.
This is mostly a pure code move. A few small tweaks:
* The create() function is new.
* The new module doesn't assume a `pill_widget`
global.
This module represents the truly re-usable code
that can be shared during these two user actions:
* edit-stream subscribers (now)
* create-stream subscribers (future)
In both situations the input pill has (or will have)
essentially the same behavior, and the next commit
will tighten up the abstraction.
(The two processes will both also use fairly similar
ListWidgets, but the mechanics of managing the list
are going to be different, so we do not intend
to keep around stream_subscribers_ui in its current
name. More on that later.)
This simplifies some of our dependencies.
As an example, we really don't want compose.js
to depend on stream_subscribers_ui.js, since
the former doesn't use any actual UI code from
the latter.
We also rename the two functions here:
invite_user_to_stream -> add_user_ids_to_stream
remove_user_from_stream -> remove_user_id_from_stream
(The notion of "inviting" somebody to a stream is
somewhat misleading, since there is really no invitation
mechanism; you just add them.)
Apart from naming changes this is a verbatim code move.
Finally, we eliminate a little bit of test cruft--the
`override` helper already ensures that a function gets
called at least once during a test.
These tests have been historically difficult to maintain.
We have pretty good direct test coverage on the
components used by stream_edit.
The code tested here was mostly glue code and jQuery
code, which the node tests are particularly poorly
suited for testing.
Note that we lose 100% line coverage on
stream_settings_containers.js, but that module
is literally a single-line function to describe
a jQuery container, and the node tests for that
would be more convoluted than helpful.
We save the preferred theme in localstorage so that user doesn't
have to re-select the theme on every reload. Users on slow
computers might see flash of a theme change, if it happens.
For aliases that will no longer be listed, see the third column of
grep '^L ' zulip-py3-venv/lib/python3.*/site-packages/pytz/zoneinfo/tzdata.zi
Time zones previously set to an alias will be canonicalized on demand.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
A recent Postgres upstream release appears to have broken PGroonga.
While we wait for https://github.com/pgroonga/pgroonga/issues/203 to
be resolved, disable PGroonga in our automated tests so that Zulip
CI passes.
As seen in
https://chat.zulip.org/#narrow/stream/9-issues/topic/edit.20history.20bug/near/1320430,
clicking such a link takes you to the user's default view if the click
handler throws an exception before doing preventDefault().
There hrefs also have the negative effect of having your browser claim
that clicking the link will navigate you to the default view, which it
won't.
Comes with a linter rule to prevent future instances, since it seems
there are some recently added ones, though they are likely the result
of copy/paste.
As a preparatory step to refactoring json_success to accept
request as a parameter, removes custom lint python check for
calling json_success without a parameter.
This PR changes how the Pan & Zoom feature of images displayed in the
attachment lightbox are handled.
The existing method of using a canvas element is replaced by the Panzoom
library (timmywil/panzoom). This library is lightweight and has 0
transitive dependencies.
This fixes#20759 where the issue is that the viewport of a zoomed image
was not expanding to fill the available space on the page. Switching to
this new library also solves several other UX issues:
* Images are no longer blurred when in Pan & Zoom mode.
* The zoom behavior itself uses focal point zooming: zooming occurs
where the cursor is on the image instead of at the center of the
image, reducing the need for extra panning.
* CSS transitions are used for a more visually pleasing experience
when switching images, toggling zoom off, etc.
* The library has the potential to open other file types which
leaves that option open for us in the future.
Wordle has recently become a thing and it uses green, yellow and white (or
black in dark mode) large square unicode characters to let people share their
gameplay. Zulip converts the white and black large square unicode characters to
emojis, but not the green and yellow ones. This causes the Wordle grid to be
misaligned when shared on Zulip.
This commit adds green and yellow large square emojis to our emoji list to fix
the problem.
We move the stream subscribed/unsubscribed bookend info from
js files to bookend handlebar.
Tweaked by tabbott to override the check-templates indentation logic.
As a consequence:
• Bump minimum supported Python version to 3.7.
• Move Vagrant environment to Debian 10, which has Python 3.7.
• Move CI frontend tests to Debian 10.
• Move production build test to Debian 10.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Navigation key presses like `Up` and `PageUp` with an empty recipient
boxes will now close the compose and propagate the keypress to the message
list or recent topics, depending upon the active view.
This extends behavior we've had for a long time with focus in the
compose box itself.
The development environment installs PostgreSQL from the OS, not PGDG,
so we should install the non-PGDG PGroonga package to match. This is
required on Debian 10 where postgresql-12-pgdg-pgroonga does not exist.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
When you use nyc, its code instrumentation transforms
the code so that line numbers and columns no longer
make sense, and the long stack trace is likely to cause
more confusion than convenience.
We want to encourage a workflow where you debug your
node tests using the normal (and much quicker mode)
before running `--coverage`.
do_delete_users had two bugs:
1. Creating the replacement dummy users
with active=True
2. Creating the replacement dummy users with email domain set to
realm.uri, which may not be a valid email domain.
Prior commits fixed the bugs, and this migration fixes the pre-existing
objects.
The existing callsites of this are via `source` or being inline'd into
the startup of a new host; in both of these cases, the surrounding
script is already `set -eu`. However, if run as a standalone tool, it
should also configure itself to catch checksum failures and other
problems.
This is a fairly straightforward extraction.
It's good to test this with Iago, and then go into
Manage Streams and add/remove subscribers for a stream
like devel.
I copy/pasted two small functions that will soon
diverge from stream_edit. The get_stream_id function
will either use a module variable (since we're
generally only editing subscribers for one stream, and
we already have the singleton assumption with
`input_pill`) or a more strict CSS selector. And then
get_sub_for_target depends on get_stream_id. We may not
always need full subs, anyway, and when we adapt some
of this code for creating streams, things are likely to
change.
I stopped exporting a couple functions that have no
callers outside of this module.
The main entry point for the module is
enable_subscriber_management.
We continue to export invite_user_to_stream and
remove_user_from_stream, which should possibly be just
pulled into their own module to lessen some
dependencies, but they don't have too much baggage,
since they just wrap channel calls.
Appending to bytes in a loop leads to a quadratic slowdown since
Python doesn’t optimize this for bytes like it does for str.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Given that these values are uuids, it's better to use UUIDField which is
meant for exactly that, rather than an arbitrary CharField.
This requires modifying some tests to use valid uuids.
It should not use the configured zulip username, but should instead
pull from the login user (likely `nagios`), or an explicit alternate
provided PostgreSQL username. Failure to do so results in Nagios
failures because the `nagios` login does not have permissions to
authenticated the `zulip` PostgreSQL user.
This requires CI changes, as the install tests install as the `zulip`
login username, which allowed Nagios tests to pass previously; with
the custom database and username, however, they must be passed to
process_fts_updates explicitly when validating the install.
I have looked at maybe ~100 errors in the last week as part
of fixing the tooling, and it's quite common to want to just
see what the improved file would look like. Now I show the
desired output with line numbers.
I also try to encourage devs to scroll up, since newbies
often don't do that for some reason when confronted with
error output.
Finally, I add some color. I try to repeat myself without
color for certain things in case colors on certain
backgrounds are hard to read.
A fast way to test this is to just break up a long tag
into two lines.
`Press Enter to send` used to hide `Send` button, we remove that
behaviour.
We show the current state of `Enter` hotkey action via text below
`Send` button which can toggle behaviour on click.
get_object_from_key should be used when trying to fetch a Confirmation
object. There are some places that need to make
Confirmation.objects.filter(...) queries, so we can't completely ban the
pattern, but we can ban .get(...) and
.filter(..., confirmation_key=..., ...).
Now we only tokenize the file once, and we pass
**validated** tokens to the pretty printer.
There are a few reasons for this:
* It obviously saves a lot of extra computation
just in terms of tokenization.
* It allows our validator to add fields
to the Token objects that help the pretty
printer.
I also removed/tweaked a lot of legacy tests for
pretty_print.py that were exercising bizarrely
formatted HTML that we now simply ban during the
validation phase.
This accomplishes a few things:
* lighten the load for the main validation loop
* defer indentation checks until we are sure the author
even knows how to match up tags
* add some info to the Token objects that we may soon
consume in our pretty-printer
We now complain about programmers who don't use
4-space indents in template files, rather than
letting the pretty printer fix them.
This is partly just to simplify the pretty printer
code (in future commits), but it also makes the
symptom more obvious to newbie developers. They
are probably just as able to react to the direct
error messages as they are able to figure out how
to read diffs from the pretty printer and grok
the --fix syntax. And once they learn the convention
and configure their editor, it should then be a
one time problem.
We now create tokens for whitespace and text, such that you
could rebuild the template file with "".join(token.s for
token in tokens).
I also fixed a few bugs related to not parsing
whitespace-control tokens.
We no longer ignore template variables, although we could do
a lot better at validating them.
The most immediate use case for the more thorough parser is
to simplify the pretty printer, but it should also make it
less likely for us to skip over new template constructs
(i.e. the tool will fail hard rather than acting strange).
Note that this speeds up the tool by almost 3x, which may be
slightly surprising considering we are building more tokens.
The reason is that we are now munching efficiently through
big chunks of whitespace and text at a time, rather than
checking each individual character to see if it starts one
of the N other token types.
The changes to the pretty_print module here are a bit ugly,
but they should mostly be made irrelevant in subsequent
commits.
This tool has been unmaintained since our initial code
sweep to fix templates, and it has possibly bit-rotted
during unrelated code sweeps like introducing mypy, etc.
It's not documented anywhere.
The preferred method now is to run:
./tools/check-templates --fix
String 'Here are a few messages I understand:'(next commit) was failing
./tools/check-capitalization check because of the capital I. I added
'I understand' to the IGNORED_PHRASES list in tools/lib/capitalization.py.
Adding "I" was working as well but didn't seem to me as a very great fix.
Strangely enough, adding " I " to the list made the test fail again
(With a lot of failed strings this time) as mentioned in the following
CZO thread.
Relevent CZO chat -
https://chat.zulip.org/#narrow/stream/49-development-help/topic/capitalization.20confusion.2E
In https://github.com/jorisroovers/gitlint/pull/246 I split the
gitlint package into gitlint and gitlint-core, where the latter avoids
pinning exact versions of its requirements so we can use it again.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The main focus is on improving the instructions around claiming issues
to try to create less issue claiming spam.
Additionally, we haven't done a detailed update in a few years, and
some of the content is stale/irrelevant.
I rewrote most of tools/lib/pretty-printer.py, which
was fairly easy due to being able to crib some
important details from the previous implementation.
The main motivation for the rewrite was that we weren't
handling else/elif blocks correctly, and it was difficult
to modify the previous code. The else/elif shortcomings
were somewhat historical in nature--the original parser
didn't recognize them (since they weren't in any Zulip
templates at the time), and then the pretty printer was
mostly able to hack around that due to the "nudge"
strategy. Eventually the nudge strategy became too
brittle.
The "nudge" strategy was that we would mostly trust
the existing templates, and we would just nudge over
some lines in cases of obviously faulty indentation.
Now we are bit more opinionated and rigorous, and
we basically set the indentation explicitly for any
line that is not in a code/script block. This leads
to this diff touching several templates for mostly
minor fix-ups.
We aren't completely opinionated, as we respect the
author's line wrapping decisions in many cases, and
we also allow authors not to indent blocks within
the template language's block constructs.
In cases where an opening tag is so long that we stretch
it to 2+ lines of code, we should try to use block-style
formatting in the template code.
Unfortunately, we have lots of legacy code that violates
this concept, so this is a timid fix.
There are also legit use cases like textarea where we
probably need to keep the ugly template syntax for things
to render properly.
We disallow this HTML:
junk-text-before-open-tag<p>
This is a paragraph.
</p>
We rarely see the above mistake, but we want to eliminate
the possibility to be somewhat rigorous, and so that we
can eliminate a pretty-printer mis-feature.
Previously, running `./tools/run-dev.py` when provision was required
would lead to a warning along the lines of:
```
Before we run tests, we make sure your provisioning version
is correct by looking at var/provision_version, which is at
version 165.1, and we compare it to the version in source
control (version.py), which is 165.2.
It looks like you checked out a branch that has added
dependencies beyond what you last provisioned. Your command
is likely to fail until you add dependencies by provisioning.
Do this: `./tools/provision`
If you really know what you are doing, use --skip-provision-check to
run anyway.
```
The assumption that we're trying to run tests might cause some
confusion, especially if its the first time you're seeing the
provision warning. Hence, we reword the first paragraph to avoid
making that assumption.
The second paragraph has also been slightly altered, since (1) it's
possible that we didn't checkout a different branch, but eg just
rebased with upstream and (2) we might not be on a VM.
The warning you'd get after this commit would be along the lines of:
```
Provisioning state check failed! This check compares
`var/provision_version` (currently 165.2) to the version in
source control (`version.py`), which is 164.6, to see if you
likely need to provision before this command can run
properly.
The branch you are currently on expects an older version of
dependencies than the version you provisioned last. This may
be ok, but it's likely that you either want to rebase your
branch on top of upstream/main or re-provision your machine.
Do this: `./tools/provision`
If you really know what you are doing, use --skip-provision-check to
run anyway.
```
or along the lines of:
```
Provisioning state check failed! This check compares
`var/provision_version` (currently 165.2) to the version in
source control (`version.py`), which is 167.2, to see if you
likely need to provision before this command can run
properly.
The branch you are currently on has added dependencies beyond
what you last provisioned. Your command is likely to fail
until you add dependencies by provisioning.
Do this: `./tools/provision`
If you really know what you are doing, use --skip-provision-check to
run anyway.
```
The `current_queue_size` key in the queue monitoring stats file was
the local queue size, not the global queue size -- d5a6b0f99a
renamed the function, but did not adjust the queue monitoring JSON,
despite the last use of it having been removed in cd9b194d88.
The function is still used to mark "we emptied our queue," and it
remains a reasonable metric for that.
It's better for this to catch all exit(...) calls with non-zero exit
code, given the purpose is to catch all exits with failure, as opposed
to only exit(1).
Unhandled exceptions propagating to process_queue were not caught there,
causing improper logging - errors didn't land in errors.log as expected.
Exceptions should be caught and explicitly logged by the process_queue
logger. Exceptions occurring during consuming events are caught and
handled inside the worker's logic - however those that happen while
setting up the worker were not addressed at all, and that's the core bug
we mean to address here.
Furthermore, in multi-threaded mode we want the autoreload mechanism to
be working - which it doesn't without catching the exceptions. The
correct approach is to - again - catch the exception, log it and then
send SIGUSR1 signal to trigger exit and autoreload.
Unless this was working around a specific bug that has somehow
persisted for eight years, this was just pointlessly defeating
Python’s bytecode cache every time you started the dev server.
This reverts commits deffda072f,
59228f7458, and
ae45217671.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This tool helps catch common typos in code and documentation, which is
particularly useful for our many contributors who are not native
English speakers.
The config is based on the codespell that I ran in
https://github.com/zulip/zulip/pull/18535.
CircleCI expected 3434; GitHub Actions expects 1001. This is the
reason zulip-ci.yml currently needs to mess with the permissions in
/__w.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This will be useful to let users enable/disable
sharing read receipts once we add that feature.
Note: Added "I've" to IGNORED_PHRASES in
tools/lib/capitalization.py to avoid capitalization
errors for the label text of this setting.
We make zero invalid value for message_content_delete_limit_seconds and
for handling the case of "Allow to delete message any time", the API-level
value of message_content_delete_limit_seconds is "anytime" and "None"
as the DB-level value. We also use these values for message retention
setting, so it helps maintain consistency.
On a 2 GiB, 1 CPU system, webpack would hit the Node.js heap
limit (which is half of physical memory up to 4 GiB, on 64-bit
systems).
Signed-off-by: Anders Kaseorg <anders@zulip.com>
gitlint has a bunch of pinned requirements that hold back important
upgrades and conflict with other packages’ requirements. The gitlint
author has rejected proposals to unpin them because it might increase
the amount of maintenance he needs to do
(https://github.com/jorisroovers/gitlint/pull/133). That decision is
his to make, but _somebody_ needs to do the maintenance, so we
delegate it to Debian and Ubuntu. If that means using a significantly
older version of gitlint, that’s a tradeoff we need to make to keep
the rest of our requirements current.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The semgrep --dangerously-allow-arbitrary-code-execution-from-rules
flag is deprecated and no longer used.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This commit adds a new module settings_defaults.js which calls
the functions in settings_display passing appropriate container
element and settings object as parameters.
We also add one more parameter for_realm_settings to some of the
functions in settings_dislay to differentiate between the user
and realm-level settings.
When the upstream provides a chunked response, proxying this header
causes a protocol-level miscommunication.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This environment variable is not a thing and has never been a thing,
while the path it purportedly pointed to does not exist and has never
existed. It appears to have been inexplicably both cargo-culted and
renamed from test-js-with-casper.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Alex Vandiver
rht
Lauryn Menard
David Rosa
Aman Agrawal
Anders Kaseorg
Tim Abbott
jai2201
Barissa-Imran
madrix01
Ganesh Pawar
Sahil Batra
evykassirer
Riken Shah
Purushottam Tiwari
Adam Sah
Dinesh
Julia Bichler
Zixuan James Li
Matt Keller
Zixuan James Li
Raghav Luthra
patricklauer
Abhishek Reddypalle
Steve Howell
Lauryn Menard
Mateusz Mandera
Austin Riba
Puneeth Chaganti
Alya Abbott
Eeshan Garg
Palash
YashRE42
Vishnu KS
Kevin Scott
Kartik Soneji
sahil839
Priyank Patel
Philipp Seßner