Commit Graph

31441 Commits

Author SHA1 Message Date
Vishnu Ks 9c1e409ea3 support: Update discount label to include nonprofit discount %. 2019-05-06 17:47:07 -07:00
Puneeth Chaganti 735b6cb761 digest: Remove code to gather new users and unread pms. 2019-05-06 17:43:53 -07:00
Puneeth Chaganti 77e0154800 test_users: Add assertion that guests can't see all realm users.
This commit adds coverage for the `UserProfile.can_access_all_realm_members`
method outside of tests in `test_digest`.
2019-05-06 17:43:53 -07:00
Puneeth Chaganti be762f9485 digest: Strip down the digest email removing a lot of fluff. 2019-05-06 17:43:52 -07:00
Puneeth Chaganti 7000834a2b digest: Display the text email also in the /digest view. 2019-05-06 17:43:40 -07:00
Yashashvi Dave 81bcdf2e94 streams: Fix minor false html element selection. 2019-05-06 17:30:33 -07:00
Yashashvi Dave bfabb86f39 stream hash: Update hash before updating settings UI elements. 2019-05-06 17:30:33 -07:00
Yashashvi Dave 6ae6da5d22 stream_ui_updates: Move function `update_add_subscriptions_elements`. 2019-05-06 17:30:33 -07:00
Yashashvi Dave 71e561de76 stream_ui_updates: Extract `update_subscribers_list` function. 2019-05-06 17:30:33 -07:00
Yashashvi Dave 7d7cf3d786 stream_ui_updates: Extract `update_subscribers_count` function. 2019-05-06 17:30:33 -07:00
Yashashvi Dave 4cd33f23bc stream_ui_updates: Refactor update-stream-privacy-type function. 2019-05-06 17:30:33 -07:00
Yashashvi Dave d86ee6b4f0 stream_ui_updates: Extract `update_stream_privacy` function. 2019-05-06 17:30:33 -07:00
David Wood f53a8f8bb6 settings: Rename `create_stream_permission` in templates.
This commit renames the `create_stream_permission` field in the
templates to `create_stream_policy`, matching the field used in the
database model. This matches what `invite_to_stream_policy` does and
will be clearer when the `waiting_period_threshold` is split into its
own field.
2019-05-06 16:30:01 -07:00
David Wood 34d810aac3 settings: Migrate to create_stream_policy structure.
This commit replaces the `create_stream_by_admins_only` setting with a
new `create_stream_policy` setting, which mirroring the structure of
the existing `invite_to_stream_policy`.

This is important preparation for migrating the waiting period feature
to be its own independent setting.

Fixes #12236.
2019-05-06 16:27:55 -07:00
Tim Abbott 3ecdabdc77 soft_deactivation: Add temporary nocoverage to fix CI. 2019-05-06 16:14:31 -07:00
Tim Abbott cf0fc7c221 test_link_embed: Fix unused variable.
This should have been in bc2ebd0f09.
2019-05-06 16:04:37 -07:00
Puneeth Chaganti 0884764421 url preview: Add python-magic dependency for content-type sniffing. 2019-05-06 12:37:32 -07:00
Puneeth Chaganti bc2ebd0f09 url preview: Refactor test code to create mock responses. 2019-05-06 12:37:32 -07:00
Puneeth Chaganti da33b72848 url preview: Use in-memory caching in dev environment. 2019-05-06 12:37:32 -07:00
Puneeth Chaganti 1f6306a5a7 url preview: Cleanup import ordering. 2019-05-06 12:37:32 -07:00
Puneeth Chaganti d56b16b275 url preview: Ignore open graph tags without a content attribute. 2019-05-06 12:37:32 -07:00
Puneeth Chaganti d02eb99831 url preview: Return generic parser <p> text as str (not bs4 string). 2019-05-06 12:37:32 -07:00
Wyatt Hoodes 2b3de816a8 test_classes.py: Add activation to default language.
Running the backend tests with a high number of processes can cause
unexpected errors with language changes.  When certain tests that change
the default language, (without explicitly overriding the teardown method
to reset the default language), interleave with other tests that are
expecting the language to be in English, discrepancies arise.

This fixes a common nondeterministic test failure with high levels of
parallelization.
2019-05-06 11:16:08 -07:00
Tim Abbott 1af4f8fe20 soft_deactivation: Add some explanatory comments.
This function still doesn't make sense in the way I'd like it to, but
this at least records what algorithm we're trying to implement.
2019-05-05 18:33:15 -07:00
Tim Abbott eb97e9fae0 soft_deactivation: Fix buggy handling of race condition.
If a soft deactivated user had a subscription double-toggled without
any new messages being sent in between, add_missing_messages might
incorrectly process those two subscription changes in the wrong order.

Fortunately, the failure mode was usually to throw this exception:

django.db.utils.IntegrityError: duplicate key value violates unique
constraint "zerver_usermessage_user_profile_id_message_id_4936d0df_uniq"
DETAIL:  Key (user_profile_id, message_id)=(4, 57) already exists.

Our unit tests actually had this precise setup some fraction of the
time, because a bit of the test setup code subscribed+unsubscribed the
target user without sending any messages in between, resulting in a
test failure something like 50% of the time.

The original exception was hard to reproduce reliably originally
(resulting in an extremely annoying nondetermnistic test failure), but
is easily reproducible by changing the "id" to "-id" in this change to
always mis-order the processing of those RealmAuditLog events.
2019-05-05 18:29:20 -07:00
Tim Abbott 8218bf101c soft_deactivation: Fix buggy handling of other streams.
Previously, our soft-deactivation logic incorrectly did not filter the
set of stream subscription changes to look at to only include the
target stream.

This could result in unspecified buggy behavior.
2019-05-05 18:29:20 -07:00
Tim Abbott 8e995deab5 soft_deactivation: Make the stream_messages mutation logic clearer.
This changes our code from something that's pretty nasty bad behavior
mutating objects during a loop to only somewhat bad behavior.
2019-05-05 18:29:20 -07:00
Tim Abbott a1d2b73790 soft_deactivation: Clarify loop logic around stream_messages.
Break will do the same thing as continue here, as each iteration will
have the same result, and it's also worth explaining why this isn't
one layer up in the loop setup.
2019-05-05 18:29:20 -07:00
Mayank Madan 99414e2d96 provision: Add support for Debian 10 Buster. 2019-05-05 17:57:19 -07:00
Mayank Madan aae01f5795 dependencies: Install line_profiler from our fork.
This is required for Python 3.7 support; line_profiler needs a release
with Cython rerun following the Python 3.7 release, and this achieves
that goal.
2019-05-05 17:53:00 -07:00
Mayank Madan a86eccb2b6 pgroonga: Upgrade from 2.1.6 to 2.1.8.
This is necessary to support postgres 11 in Debian Buster.
2019-05-05 17:52:32 -07:00
Puneeth Chaganti dc1571426e open graph: Use the complete URL for open graph URLs.
Closes #12199
2019-05-05 17:51:53 -07:00
Raymond Akornor e01d3be1ba tools: Remove check-urls linter.
This commit removes `tools/check-urls`. It was added as
a useful tool in preparation for the Django 1.10 migration.
Since we completed that migration, it is no longer needed.

Fixes #12180.
2019-05-05 17:47:41 -07:00
Anders Kaseorg bc9d7141b2 mypy: Enable strict_optional for test_outgoing_webhook_interfaces.
Signed-off-by: Anders Kaseorg <anders@zulipchat.com>
2019-05-04 23:15:58 -07:00
Vishnu Ks 176d37a513 tests: Make tests work in reverse order in mattermost_importer. 2019-05-04 08:49:34 -07:00
vinitS101 81b5a72252 admin_settings: Change maxfilesize to max_file_upload_size.
Renamed maxfilesize to max_file_upload_size for consistency.
Related to #12152.
2019-05-03 17:36:09 -07:00
vinitS101 3cb502974e compose: Change file upload error message to show correct file size.
The error message, that is displayed when a user tries to upload a file
larger than the allowed size, currently displays a hardcoded max file
size of 25MB.
Changed this to show the correct max_file_upload_size value that has
been set by the server admin.
2019-05-03 17:36:09 -07:00
Kouhei Sutou a046a0957b docs: Fix upgrade process for Ubuntu 16.04 -> 18.04.
We need to disable "CREATE EXTENSION pgroonga" in zulip-puppet-apply
by creating /usr/share/postgresql/10/pgroonga_setup.sql.applied.
Because PostgreSQL 10 isn't running in this case. If PostgreSQL 10
isn't running, we can't run "CREATE EXTENSION pgroonga".

We can't use pg_upgrade with PGroonga. PGroonga's install SQL
https://github.com/pgroonga/pgroonga/blob/master/data/pgroonga.sql has
conditions to support multiple PostgreSQL versions. So it's not safe
to use pg_upgrade. pg_upgrade copies metadata for PostgreSQL 9.5 to
PostgreSQL 10. We need to use pg_dump and pg_restore to upgrade
PGroonga correctly for PostgreSQL 10.
2019-05-03 17:35:32 -07:00
Kouhei Sutou 407ed45ec2 docs: Use systemctl instead of service.
Because memcached package on Ubuntu 18.04 supports systemd.
2019-05-03 17:33:29 -07:00
David Wood 16e9892ab2 webhooks/jira: Include issue title in messages. 2019-05-03 20:03:31 -02:30
Vishnu Ks 123bcea518 management: Don't use sys.exit(1).
Using sys.exit in a management command makes it impossible
to unit test the code in question.  The correct approach to do the same
thing in Django management commands is to raise CommandError.

Followup of b570c0dafa
2019-05-03 14:20:39 -07:00
Tim Abbott 687f1bcdaf test_classes: Fix missing flushing of per-request caches.
We should definitely be starting each test case with an empty copy of
the per-request caches, since their intended duration is even shorter
than a request.

This was masked by the fact that these caches are automatically
flushed when one makes an actual request to the Zulip API; so the
problems were only manifesting in tests like test_events, where we
call lower-level functions that access a per-request cache without
using the Zulip API.
2019-05-03 13:52:56 -07:00
Tim Abbott 5cddf07361 test_mattermost_importer: Fix buggy use of wrong paths.
This worked for me when testing locally because it was using a
constant path, and so was accessing the product under `var/` of the
previous test run.
2019-05-03 13:30:54 -07:00
Tim Abbott 0adb93db32 test_classes: Fix data import path reuse parallelism errors.
The make_import_output_dir helper function used a path determined
primarily by the filename of the fixture being used, and expected to
have complete control over that path for the duration of the test.

This resulted in nondeterministic errors if our two test classes that
ran Mattermost import code ran at the same time.
2019-05-03 13:23:09 -07:00
Shubham Padia 884c19c3bc settings: NAME_CHANGES_DISABLED disables the name change button.
Fixes #12251.

Previously when disabling name changes in server settings, instead
of realm settings, the name edit button did not get disabled.
Changing name resulted in a message stating `no changes made`.
2019-05-03 12:58:51 -07:00
Shubham Padia 8c01f396db settings: Add AVATAR_CHANGES_DISABLED server setting.
Fixes #12132.
Realm setting to disable avatar changes is already present.
The `AVATAR_CHANGES_DISABLED` setting now follows the same
2-setting model as `NAME_CHANGES_DISABLED`.
2019-05-03 12:58:51 -07:00
Shubham Padia 4994a6c2c9 realm: Add setting to disable avatar changes.
This is useful when syncing avatars from an integrated LDAP/active
directory.

The upload avatar and delete avatar buttons are hidden if avatar
changes are disabled and the user is a non-admin.
If the user has a gravatar set, then the user will not be able to
upload an image as their avatar if avatar changes are disabled.

Part of #12132.
2019-05-03 12:52:43 -07:00
Mateusz Mandera 63c170c390 circleci: Fix redis on Bionic.
Fixes #11079.
Because of "service redis-server start" not working properly, in "do
Bionic hack" another instance of the server was being started. However,
once in a while this would fail for unknown reasons causing flakes in
tests involving redis. The cause of the service not starting was the
error:
::1:6379: bind: Cannot assign requested address
(see https://github.com/antirez/redis/issues/3241)
This is fixed by editing /etc/redis/redis.conf to not attempt binding
on IPv6. We change "do Bionic hack" to a simple sed command,
changing the appropriate line in the file.
By default, the line is "bind 127.0.0.1 ::1".
We change it to "bind 0.0.0.0".
2019-05-03 11:33:39 -07:00
Tim Abbott b570c0dafa convert_mattermost_data: Don't sys.exit(1).
This behavior ends up being quite problematic for running unit tests;
the thing one is supposed to do in management commands is raise
CommandError.
2019-05-02 11:13:41 -07:00
Wyatt Hoodes 21c67ea1c2 test-backend: Change --parallel default to use `cpu_count`.
The number of processes to run the backend tests is currently a
hardcoded value, this commit transistions the default to be based on the
number of logical CPUs available.
2019-05-02 10:48:20 -07:00