The avatar logic has been moved from Project to Repository as this makes
caching easier. The logic itself in turn has been changed so that the
logo file names are cached in Redis. This cache is flushed upon pushing
a commit but _only_ if:
1. The commit was pushed to the default branch
2. The commit actually changes any of the logo files
If no branch or commit is given the cache is flushed anyway, this
ensures that calling Repository#expire_cache without any arguments still
flushes the avatar cache (e.g. this is used when removing a project).
Fixesgitlab-org/gitlab-ce#14363
Instead of flushing the behind/ahead counts for all branches upon every
push we now only flush the cache of branches that actually need to have
these statistics recalculated. There are now basically 2 scenarios and
their effects:
1. A user pushes a commit to the default branch, this results in the
cache being flushed for all branches.
2. A user pushes to a non default branch, this results in _only_ the
cache for that branch being flushed.
The existing code (Repository#expire_cache) remains backwards compatible
with the previous behaviour, the new behaviour is only applied when a
branch name is passed as an argument. This ensures that when for example
a project is deleted the cache for all branches is flushed.
This caches the output of the following methods:
* Repository#empty?
* Repository#has_visible_content?
* Repository#root_ref
The cache for Repository#has_visible_content? is flushed whenever a
commit is pushed to a new branch or an existing branch is removed.
The cache for Repository#root_ref is only flushed whenever a user
changes the default branch of a project. The cache for Repository#empty?
is never explicitly flushed as there's no need for it.
If a branch is deleted with an open merge request, amended offline, and then pushed again,
GitLab doesn't bother to update the merge request even though the last commit ID and/or
code may have changed. This MR ensures that each push will update any relevant merge
requests and adds a system note if this happens as well.
Closes#2926
I also revamped the tests on "closing reference commits" (= "Fixes #xxxx" for example).
Now, there are two different contexts:
- when the commits with "closing reference" are pushed to the default branch,
- when the commits with "closing reference" are pushed to a non-default branch.
Closesgitlab-org/gitlab-ce#2190.
In an empty repository, pushing a new branch not called "master" would
leave HEAD in an unknown state, causing ambiguity if another branch
were pushed. This could in turn cause a new protected branch to be
created and cause the default branch to change.
* Closes#1561
* Closes#1576
* Closes https://github.com/gitlabhq/gitlabhq/issues/8883
This commit disables process_commit_messages()
for the initial push to the default branch.
This fixes the mass SQL statements (~500000) that were executed during
the initial push of the linux kernel for example.
1) Adds a DB migration for all services to toggle on push, issue, and merge events.
2) Upon an issue or merge request event, fire service hooks.
3) Slack service supports custom messages for each of these events. Other services
not supported at the moment.
4) Label merge request hooks with their corresponding actions.
Drew Blessing
Kamil Trzcinski
Douglas Barbosa Alexandre
Yorick Peterse
Jacob Vosmaer
ashleys
tiagonbotelhoandYorick Peterse
James Lopez
Douwe Maan
Stan Hu
Robert Speicher
Maël Valais
Dmitriy Zaporozhets
Hannes Rosenögger
Vinnie Okada
Jeroen van Baarsen
Marco Wessel
Ciro Santilli
Marin Jankovski
Marin Jankovski
Hiroyuki Sato