10 Jun, 2014
38 commits
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
UI improvements 
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Remove thread variables Lets get rid of thread variables. It produces additional complexity and weird stack trace. Also part of !1133 - - - Also it fixes issue/merge_request close and reopen bug via API when no dashboard event and comment was created.
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Fix broken email threading The email threading support introduced in GitLab CE 6.9 is broken on several popular email clients (including Mail.app and Airmail on Mac OS X). This MR makes the following changes to improve email threading compatibility: * Subject of answers to an existing thread begins with `Re: ` (required by Mail.app) * The recipient of every email in a thread is stable (required by Mail.app ; otherwise it groups emails by sender) * Send a ‘In-Reply-To’ header along the ‘References’ header (for compatibility with the spec) In order to do this, these commits: * Change the `To:` field to `namespace/project` ; the actual receiver is now in the `Cc:` field. * Introduce the `mail_new_thread` and `mail_answer_thread` methods ; they format the message correctly for threading, and can generate the `Message-ID` automatically from a model instance. * Refactor the tests to shared behaviors for email threading. We've been using these patches at @capitainetrain for a few months now ; I just ported them to work nicely with the recent threading commits.
-
* send a ‘In-Reply-To’ header along the ‘References’ header * subject of answers to an existing thread begins with ‘Re: ’ This fixes threading with at least Mail.app and Airmail.
-
This fixes email threading in Mail.app, that doesn't like when a thread doesn't have stable recipients. For instance, here is a possible sender-recipient combinations before: From: A To: Me New issue From: B To: Me Reply on new issue From: A To: Me Another reply Mail.app doesn't see B as a participant to the original email thread, and decides to break the thread: it will group all messages from A together, and separately all messages from B. This commit makes the thread look like this: From: A To: gitlab/project Cc: Me New issue From: B To: gitlab/project Cc: Me Reply on new issue From: A To: gitlab/project Cc: Me Another reply Mail.app sees a common recipient, and group the thread correctly.
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Https docs
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Clarify that bbastov is the style of Hound CI 2.
-
Https in installation document
-
Project commit count for default branch Currently, the commit count displayed on a project's home page shows the total number of commits on the master branch. This can lead to situations in which a project may have no commits reported or very few if the project workflow does not involve the use of a branch named "master". Since number of commits can be an indicator of project health, showing 0 commits could make it harder to encourage adoption of a project. This MR alters the repository model's `commit_count` method to return the count for the default branch instead of master. See issue #299
-
Fix typo in CHANGELOG
-
Create a note when issue milestone was changed Part of #1207
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Gitlab shell version file
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Update installation doc with python-docutils Fixes #1293
09 Jun, 2014
2 commits
-
We need to change config in step 5
-
Fix for the broken routing specs