| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
We only want them shortened by the filter, which calls to_s
|
| | |
|
| |
|
|
| |
Except for Note, which still overrides it.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Snippet model was missing project association
|
| |
|
|
|
| |
Now there is a single source of information for which attribute a model
uses to be referenced, and its special character.
|
| | |
|
| |
|
|
|
|
|
|
| |
Prior, comparison would use the Ruby object's ID, which got out of sync
after a Spring fork and would result in erroneous test failures.
Now we just check that the compared object is a Commit and then compare
their underlying raw commit objects.
|
| |\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Change percent_complete rescue value from 100 to 0
The percent_complete method returns a value of 100 when a
ZeroDivisionError occurs. That seems like a very strange default for an
error case, and results in a bug when a milestone has no corresponding
issues (new, empty milestones show 100% completion). This commit changes
the rescue value to 0, and subsequently fixes #1656, which reported this
problem.
See merge request !714
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
The percent_complete method returns a value of 100 when a
ZeroDivisionError occurs. That seems like a very strange default for an
error case, and results in a bug when a milestone has no corresponding
issues (new, empty milestones show 100% completion). This commit changes
the rescue value to 0, and subsequently fixes #1656, which reported this
problem.
|
| |\ \ |
|
| | |\ \
| | | |
| | | | |
Validate wiki page creation
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
**What does this do?**
It adds validation to the creation of a wiki page, that way the user gets real
feedback instead of just a 404 page if the name of the wiki page was invalid
**Why is this needed?**
There are a lot of characters that are not allowed in the creation of a wiki
page, there is even a small text that is saying: Please don't use spaces.
Although we have that text there, we don't actually validate on this. This
commit adds validation on the title and gives the user actual feedback.
**What issues does this fix?**
Fixes http://github.com/gitlabhq/gitlabhq/issues/5357
Fixes https://github.com/gitlabhq/gitlabhq/issues/8565
Fixes https://github.com/gitlabhq/gitlabhq/issues/3913
Fixes https://github.com/gitlabhq/gitlabhq/issues/8166
Signed-off-by: Jeroen van Baarsen <jeroenvanbaarsen@gmail.com>
|
| |\ \ \ \
| |/ / / |
|
| | |\ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Fix Zen Mode not closing with ESC key
### What does this MR do?
This MR fixes the ESC key not leaving Zen Mode (fullscreen).
### Why was this MR needed?
This has been broken since GitLab v7.8. The `change` event for `zen-toggle-comment` was never fired. See:
http://stackoverflow.com/questions/19505011/checkbox-checked-with-prop-does-not-fire-events-attached-to-change?answertab=votes#tab-top
### What are the relevant issue numbers?
* Closes #1025
* Closes https://github.com/gitlabhq/gitlabhq/issues/9018
See merge request !710
|
| | | | | |
| | | | |
| | | | |
| | | | | |
Closes #1025
|
| |/ / / /
| | | |
| | | |
| | | | |
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
|
| |\ \ \ \
| |_|/ /
|/| | /
| | |/
| |/| |
|
| | |\ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Add link_to_label helper
The primary purpose of this change was to make the actual labels on `Labels#index` clickable.
See merge request !690
|
| | | | |
| | | |
| | | |
| | | |
| | | | |
Now the user can click the labels themselves or the "X open issues"
text.
|
| | | | | |
|
| | |\ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Update ffaker gem
Adds a version requirement which it didn't have before, at 2.0.0. This version has breaking API changes in that the namespace is now `FFaker` instead of `Faker`.
Prior, if a new developer checked out the repo, ran `bundle install` and then tried `rake dev:setup`, seeding would fail due to having the latest version of ffaker without the API changes in this MR.
Also updates `spring` version and the binstubs, just for kicks.
See merge request !686
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Adds a version requirement which it didn't have before, at 2.0.0. This
version has breaking API changes in that the namespace is now `FFaker`
instead of `Faker`.
|
| | | | | | |
|
| | |\ \ \ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Better handle label references that aren't actually references
Fixes #1690
See merge request !705
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | | |
Fixes #1690
|
| | |\ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Fix Markdown preview not working in Edit Milestone page
### What does this MR do?
This MR removes the automatic Zen Mode URL update when a hash is present and makes Markdown preview work again in the Edit Milestone page. I think the intent was to make it possible to link to a full-screen edit with a URL (e.g. http://foo/bar/issues/1/edit#fullscreen), but I don't this has ever worked. Perhaps a future MR can support this.
### Why was this MR needed?
A JavaScript error would be seen in the Milestone Edit page:
```javascript
Uncaught error, unrecognized expression: $('.zennable input[type=checkbox]##md-preview-holder')
```
In the Milestone Edit page, apparently the use of the hash `#md-preview-holder` causes the Zen Mode JavaScript to attempt to parse the hash for the `fullscreen_` prefix and look up the checkbox based on a unknown ID.
### Are there points in the code the reviewer needs to double check?
If we want to keep this hash update, an alternative fix is to add an `if` check in the beginning of `checkboxFromLocationHash`:
```coffeescript
if (window.location.hash.indexOf('#' + ZenMode.fullscreen_prefix) == -1)
return null
```
### What are the relevant issue numbers?
* Closes #1687
* Closes https://github.com/gitlabhq/gitlabhq/issues/9325
See merge request !711
|
| | | | |_|_|/
| | |/| | |
| | | | | |
| | | | | |
| | | | | | |
Closes #1687
Closes https://github.com/gitlabhq/gitlabhq/issues/9325
|
| |\ \ \ \ \ \
| |/ / / / /
|/| | | | | |
Use .md as extention for wiki pages
|
| |/ / / / /
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
**What does this do?**
It makes sure that when you create a wiki page via the web interface, the
extention is .md instead of .markdown
**Why is this needed?**
When you're using Gollum locally, it will create pages with the .md extention.
Also .md is the best known extention for markdown. This fix will make sure that
if you're using gollum or the webinterface, the extention will be the same.
**What issues does this fix?**
Fixes https://github.com/gitlabhq/gitlabhq/issues/5204
Signed-off-by: Jeroen van Baarsen <jeroenvanbaarsen@gmail.com>
|
| |\ \ \ \ \ |
|
| | |\ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Refactor permission checks with issues and merge requests project settings
Slight refinement for read/write permission checks for 04d44522 as discussed in !691. Currently it appears that there is [no role that only has read-only access to issues](https://gitlab.com/gitlab-org/gitlab-ce/blob/master/doc/permissions/permissions.md), but it could be possible.
/cc: @DouweM
See merge request !704
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
`merge_requests_enabled`
|
| | |\ \ \ \ \ \
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Remove Rack Attack monkey patches and bump to version 4.3.0
I finally got these monkey patches into Rack Attack v4.3.0, so GitLab no longer needs them. Hooray!
See: https://github.com/kickstarter/rack-attack/pull/128
See merge request !693
|
| | | | | | | | | |
|
