* org/members: display 2FA state
* fix comment typo
* lay down UserList bases
* add basic test for previous methods
* add comment for UserList type
* add valid two-fa account
* test new UserList methods
* optimize MembersIsPublic by side loading info on GetMembers + fix integrations tests
* respect fmt rules
* use map for data
* Optimize GetTwoFaStatus
* rewrite by using existing sub func
* Optimize IsUserOrgOwner
* remove un-used code
* tests: cover empty org + fix import order
* tests: add ErrTeamNotExist path
* tests: fix wrong expected result
Previously, when formatting commit messages, anything
that looked like SHA1 hashes was turned into a link
using regex. This meant that certain phrases or numbers
such as `777777` or `deadbeef` could be recognized as a commit
even if the repository has no commit with those hashes.
This change will make it so that anything that looks
like a SHA1 hash using regex will then also be checked
to ensure that there is a commit in the repository
with that hash before making a link.
Signed-off-by: Gary Kim <gary@garykim.dev>
when you use gitea as OAuth2 provider, the /api/v1/user should return
user primary email as identifier, which is unique in OAuth2 clients.
this patch use convert.ToUser replace all u.APIFormat in api requests,
return primary email when caller is yourself or admin.
* Strip newlines from SSH keys before adding them
Fixes: https://github.com/go-gitea/gitea/issues/7500
* add test for CheckPublicKeyString
* add one more test
* simplify test
* further simplify
* make fmt
* Include thread related headers in issue/coment mail
Make it so mail programs will group comments from an issue into the same
thread by setting Message-ID on initial issue and then using In-Reply-To
and References headers to reference that later on.
* Add tests
* more tests
* fix typo
The update call on the user call races if there is more than one
repository creation concurrently, leading to incorrect count of
repos. Split things in two, so that we call the update for last
visibility (which isn't problematic if it races, since it can only
ever be best-effort anyway). This way we can atomically increment
the count of repos.
* only return head: null if source branch was deleted
* add URL into GetPullRequest
* TestPullRequest_APIFormat
* log error if it is not Err(Branch)NotExist
* Store original author info for migrated issues and comments
Keep original author name for displaying in Gitea interface and also
store original author user ID for potential future use in linking
accounts from old location.
* Add original_url for repo
Store the original URL for a migrated repo
Clean up migrations/tests
* fix migration
* fix golangci-lint
* make 'make revive' happy also
* Modify templates to use OriginalAuthor if set
Use the original author name in templates if it is set rather than the
user who migrated/currently owns the issues
* formatting fixes
* make generate-swagger
* Use default avatar for imported comments
* Remove no longer used IgnoreIssueAuthor option
* Add OriginalAuthorID to swagger also
* Fix#6946 by checking PullRequest ID on pushing
* Ensure we have the owner name, the pr attributes and the the issue
* Fix TestSearchRepo by waiting till indexing is done
* Update integrations/repo_search_test.go
* changes as per @mrsdizzie
* missing comma
* Spelling mistake
* Fix full pushing environment
* Show Pull Request button or status of latest PR in branch list
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* Do not show pull request button on deleted branches
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* Do not show commit divergence on deleted branches
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* Use XORMs Get instead of limit
* Links pull request ID and use smaller labels for displaying the pull request status
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* Handle error when getting latest pull request
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* Indent template
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* Check error when loading issue
Signed-off-by: Mario Lubenka <mario.lubenka@googlemail.com>
* fix extra newlines when copying from diff
See https://bugzilla.mozilla.org/show_bug.cgi?id=1273836
Basically, the <pre><code> seems to add a forced newline that is not
possible to get rid of via CSS, so I replaced it with just a <code>.
Secondly, .lines-type-marker also forced a newline in the copied text,
but that was possible to get rid of via user-select.
Safari still has a extraneous newline in the copied text of unknown
origin, but this should not block stop this PR.
* simplify .line-type-marker
* fix selector
* remove erronous ^^^
* Fix empty split diff
* Fix arc-theme-green
* fix add comment
* ensure line-num is copied too
* Update templates/repo/diff/box.tmpl
Co-Authored-By: zeripath <art27@cantab.net>
* attempt to fix safari via removing <code>
* remove useless whitespace at the end of 'class'
* remove inter-tag whitespace for code <td>s
* more inter-tag removal
* final inter-tag removal
* attempt to fix empty line copy
* move and comment getLineContent
* fix golint
* make background grey for missing added code
* Make diff line-marker non-selectable
* Move to use data-* as per @mrsdizzie
* fix missing line nums
* Add a minimum-width to force right-align of the line num
* Move line-type-marker into separate column
On merge we walk the merge history and ensure that all lfs objects pointed to in
the history are added to the base repository. This switches from relying on having git-lfs installed on the server, (and in fact .gitattributes being correctly installed.)
* Use go method to calculate key fingerprint
* add gitea copyright
* use native go method only for built-in server
* refactor and add tests
* add gitea copyright
