summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* v1.6.1: correct missing version number updatesv1.6.1Edward Thomson2023-02-254-5/+5
|
* Merge pull request #6501 from libgit2/ethomson/v1.6Edward Thomson2023-02-255-5/+104
|\ | | | | v1.6
| * v1.6: update version numbersEdward Thomson2023-02-244-5/+5
| |
| * v1.6: update the changelogEdward Thomson2023-02-241-0/+99
|/
* Merge pull request #6503 from fxcoudert/hostandportEdward Thomson2023-02-244-15/+42
|\ | | | | Pass hostkey & port to host verify callback
| * tests: validate host and port for ssh tests when non-standardEdward Thomson2023-02-241-1/+11
| |
| * Pass hostkey & port to host verify callbackFrancois-Xavier Coudert2023-02-241-3/+11
| | | | | | | | Co-authored-by: Stefan Karpinski <stefan@karpinski.org>
| * net: parse urls or scp style paths in the same functionEdward Thomson2023-02-243-11/+20
|/
* Merge pull request #6500 from timrogers/timrogers/empty-docsEdward Thomson2023-02-231-1/+3
|\ | | | | Correct the definition of "empty" in the docs for `git_repository_is_empty`
| * Correct the definition of "empty" in the docs for `git_repository_is_empty`Tim Rogers2023-02-231-1/+3
|/ | | | | | | | | | | This improves the documentation for `git_repository_is_empty` which currently does not accurately describe libgit2's definition of "empty". It says that HEAD must point to the "unborn master branch", when in fact, this is not the case if the repo's `init.defaultBranch` configuration is set. If it is set, it will check that HEAD points there. Only if it is not set does it fall back to `master`.
* Merge pull request #6499 from libgit2/ethomson/sha256_fix_2Edward Thomson2023-02-232-0/+4
|\ | | | | odb: don't unconditionally add oid_type to stream
| * odb: don't unconditionally add oid_type to streamEdward Thomson2023-02-232-0/+4
| | | | | | | | | | | | `git_odb_stream` should have an `oid_type` to disambiguate; that's not necessary on non-experimental SHA256 builds. Avoid unnecessary ABI breakage for consumers and hide it behind an ifdef for now.
* | Merge pull request #6498 from libgit2/ethomson/httpproxy_env_testsEdward Thomson2023-02-231-21/+15
|\ \ | |/ |/| tests: always unset HTTP_PROXY before starting tests
| * tests: always unset HTTP_PROXY before starting testsEdward Thomson2023-02-231-21/+15
|/ | | | | | The remote::httpproxy::env test (correctly) reset the environment before running; however the other tests are also impacted by the environment and need to have it isolated.
* Merge pull request #6497 from fxcoudert/ssh-agent-nonfatalEdward Thomson2023-02-221-1/+3
|\ | | | | Make failure to connect to ssh-agent non-fatal
| * Make failure to connect to ssh-agent non-fatalFrancois-Xavier Coudert2023-02-221-1/+3
| | | | | | | | | | | | | | | | | | Fixes https://github.com/libgit2/libgit2/issues/3866 Has been applied in all Julia builds since 2017: https://github.com/JuliaLang/julia/pull/17459 Authored-by: Keno Fischer <kfischer@college.harvard.edu>
* | Merge pull request #6494 from libgit2/ethomson/clarify_readmeEdward Thomson2023-02-221-2/+3
|\ \ | |/ |/| README: clarify the linking exception
| * Update README.md Edward Thomson2023-02-221-5/+5
| | | | | | Update based on carlosmn suggestions.
| * README: clarify the linking exceptionEdward Thomson2023-02-201-4/+5
| | | | | | | | | | | | | | We say that you can link libgit2 "unmodified"... "without having to release its source code". Clarify that you can modify libgit2 - but you must release _its_ source code back - and you can link libgit2 without having to release _your software's_ source code.
* | Merge pull request #6493 from libgit2/ethomson/ownershipEdward Thomson2023-02-202-7/+43
|\ \ | |/ |/| Handle Win32 shares
| * repo: support safe.directory with %(prefix)/Edward Thomson2023-02-201-6/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Git for Windows does some truly bizarre things with paths that start with a forward slash; and expects you to escape that with `%(prefix)`. This syntax generally means to add the prefix that Git was installed to -- eg `/usr/local` -- unless it's an absolute path, in which case the leading `%(prefix)/` is just removed. And Git for Windows expects you to use this syntax for absolute Unix-style paths (in "Git Bash" or Windows Subsystem for Linux). Worse, the behavior used to be that a leading `/` was not absolute. It would indicate that Git for Windows should add the prefix. So `//` is required for absolute Unix-style paths. Yes, this is truly horrifying. Emulate that behavior, I guess, but only for absolute paths. We won't deal with the Git install prefix. Also, give WSL users an escape hatch where they don't have to think about this and can use the literal path that the filesystem APIs provide (`//wsl.localhost/...`).
| * repo: don't fail on strange win32 pathsEdward Thomson2023-02-182-1/+4
| | | | | | | | | | | | | | With some paths on Win32, we cannot identify the owner because it's on a file share (WSL2 or UNC). In that case, don't fail, but identify that the current user does not own the path. This matches Git for Windows behavior.
* | Merge pull request #6492 from cavaquinho/fix/bare-repo-oid-typeEdward Thomson2023-02-202-23/+51
|\ \ | |/ |/| #6491: Sets oid_type on repos open with git_repository_open_bare
| * Deleted unused variable.Miguel Arroz2023-02-171-1/+0
| |
| * #6491: Sets oid_type on repos open with git_repository_open_bareMiguel Arroz2023-02-172-22/+51
|/
* Merge pull request #6487 from libgit2/ethomson/strarray-deprecateEdward Thomson2023-02-161-0/+1
|\ | | | | stash: update strarray usage
| * stash: update strarray usageEdward Thomson2023-02-161-0/+1
|/
* Merge pull request #6330 from gitkraken-jacobw/partial-stashingEdward Thomson2023-02-164-32/+302
|\ | | | | stash: partial stash specific files
| * stash: fixes from code reviewEdward Thomson2023-02-162-57/+38
| |
| * stash: test save options initJacob Watson2022-07-141-0/+5
| |
| * stash: add `const` to argumentsJacob Watson2022-07-142-5/+7
| |
| * stash: better option validation for stash saveJacob Watson2022-07-141-6/+17
| |
| * stash: implement CI testingJacob Watson2022-07-141-0/+37
| |
| * stash: implement partial stashing by pathJacob Watson2022-07-142-26/+260
| |
* | Merge pull request #6486 from libgit2/ethomson/strarray-deprecateEdward Thomson2023-02-165-14/+28
|\ \
| * | strarray: remove deprecated declarationEdward Thomson2023-02-165-14/+28
|/ / | | | | | | | | | | `git_strarray_copy` is deprecated (and has been included in `deprecated.h` for some time). It should not have remained in the public `strarray.h`. Remove it.
* | Merge pull request #6480 from libgit2/ethomson/warnEdward Thomson2023-02-153-4/+5
|\ \ | | | | | | Fix some warnings in main
| * | tests: use static scope for test dataEdward Thomson2023-02-152-3/+3
| | |
| * | socket: mark flags as unusedEdward Thomson2023-02-151-1/+2
|/ /
* | Merge pull request #6479 from libgit2/ethomson/cloneEdward Thomson2023-02-151-1/+1
|\ \ | | | | | | clone: clean up options on failure
| * | clone: clean up options on failureEdward Thomson2023-02-141-1/+1
|/ /
* | Merge pull request #6456 from libgit2/ethomson/sha256_experimentalEdward Thomson2023-02-14196-530/+5143
|\ \ | | | | | | SHA256: more SHA256 support
| * | ci: build our own git on xenialEdward Thomson2023-02-131-2/+14
| | | | | | | | | | | | The git included with xenial is ancient, and lacks sha256 support.
| * | win32: adjust max path length for SHA256Edward Thomson2023-02-131-1/+1
| | | | | | | | | | | | | | | The longest path within a git repository is now a SHA256 format packfile. Adjust our max length checking to match.
| * | cmake: support older cmakesEdward Thomson2023-02-131-1/+1
| | |
| * | ci: actually build sha256 support on linuxEdward Thomson2023-02-121-1/+1
| | |
| * | clone: free connect optsEdward Thomson2023-02-122-0/+9
| | |
| * | repo: take an oid_type when initializingEdward Thomson2023-02-122-3/+24
| | |
| * | repo: dump backends on oid type changeEdward Thomson2023-02-121-1/+18
| | |
| * | transport: teach transports about oid types and SHA256Edward Thomson2023-02-1213-83/+298
| | |
View Lorry Controller Status