Commit Graph

57 Commits

Author SHA1 Message Date
Edward Thomson
ee5da720e5 reference_create: validate loose names
Validate loose reference names on Win32.
2014-12-16 10:08:56 -06:00
Carlos Martín Nieto
753e17b0f5 peel: reject bad queries with EINVALIDSPEC
There are some combination of objects and target types which we know
cannot be fulfilled. Return EINVALIDSPEC for those to signify that there
is a mismatch in the user-provided data and what the object model is
capable of satisfying.

If we start at a tag and in the course of peeling find out that we
cannot reach a particular type, we return EPEEL.
2014-11-22 18:55:22 +01:00
Carlos Martín Nieto
5915d7001b branch: consider an empty upstream remote config as not found 2014-11-17 14:28:22 +01:00
Carlos Martín Nieto
82374d9825 branch: add getter for the upstream remote name
This gets the value from branch.<foo>.remote.
2014-11-08 20:00:17 +01:00
Carlos Martín Nieto
c327d5db8b transaction: rename lock() to lock_ref()
This leaves space for future expansion to locking other resources
without having to change the API for references.
2014-10-09 16:29:30 +02:00
Carlos Martín Nieto
ab8d9242f5 Introduce reference transactions
A transaction allows you to lock multiple references and set up changes
for them before applying the changes all at once (or as close as the
backend supports).

This can be used for replication purposes, or for making sure some
operations run when the reference is locked and thus cannot be changed.
2014-09-30 15:44:32 +02:00
Jacques Germishuys
c7dd0a56bf Use p_snprintf also in tests 2014-08-05 20:51:59 +02:00
Edward Thomson
0cee70ebb7 Introduce cl_assert_equal_oid 2014-07-01 14:40:16 -04:00
Carlos Martín Nieto
4ee2543c5a refs: failing test for concurrent ref access
If we remove a reference while we're iterating through the packed refs,
the position in the iterator will be off.
2014-06-02 16:46:47 +02:00
Arthur Schreiber
9d6c3d2853 Refs: Extend unicode test for branch creation.
This adds another assertion to ensure that the reference name inside
the git_reference struct returned by `git_branch_create` is returned as
precomposed if `core.precomposeunicode` is enabled.
2014-05-30 15:15:54 +02:00
Arthur Schreiber
1a90b1e3f1 Refs: Add a unicode test for git_branch_move.
This tests that decomposed branch names are correctly precomposed when
passed to `git_branch_move` and `core.precomposeunicode` is enabled.
2014-05-30 14:53:28 +02:00
Russell Belfer
8a2ef218b9 Don't always test composed-insensitive lookups
Only on a filesystem that is composed/decomposed insensitive,
should be testing that a branch can be looked up by the opposite
form and still work correctly.
2014-05-08 14:48:27 -07:00
Russell Belfer
43a0413524 Pass unconverted data when iconv doesn't like it
When using Iconv to convert unicode data and iconv doesn't like
the source data (because it thinks that it's not actual UTF-8),
instead of stopping the operation, just use the unconverted data.
This will generally do the right thing on the filesystem, since
that is the source of the non-UTF-8 path data anyhow.

This adds some tests for creating and looking up branches with
messy Unicode names.  Also, this takes the helper function that
was previously internal to `git_repository_init` and makes it
into `git_path_does_fs_decompose_unicode` which is a useful in
tests to understand what the expected results should be.
2014-05-08 13:52:46 -07:00
Russell Belfer
99dfa47039 Some further sandboxing cleanups to tests
Trying to find other issues where tests may not clean up quite
properly when they are through...
2014-05-02 09:21:33 -07:00
Sven Strickroth
fa13ee2d7c Add GIT_BRANCH_ALL to git_branch_t enum
git_branch_t is an enum so requesting GIT_BRANCH_LOCAL | GIT_BRANCH_REMOTE is not possible as it is not a member of the enum (at least VS2013 C++ complains about it).

This fixes a regression introduced in commit a667ca8298 (PR #1946).

Signed-off-by: Sven Strickroth <email@cs-ware.de>
2014-04-16 18:51:16 +02:00
Carlos Martín Nieto
8350437112 reflog: follow core.logallrefupdates
On bare by default, or when core.logallrefupdates is false, we must not
write the reflog.
2014-03-20 19:18:49 +01:00
Carlos Martín Nieto
1afe140043 refdb: don't update when there's no need
If the caller wants to update a ref to point to the same target as it
currently has, we should return early and avoid writing to the reflog.
2014-03-19 15:54:33 +01:00
Carlos Martín Nieto
4b7e1b9e92 refs: append to the HEAD reflog when updating the current branch
When we update the current branch, we must also append to HEAD's reflog
to keep them in sync.

This is a bit of a hack, but as git.git says, it covers 100% of
default cases.
2014-03-17 17:47:47 +01:00
Ben Straub
494be429ad Merge pull request #2144 from linquize/branch-f-current
Do not allow git_branch_create() to force update branch
2014-03-02 09:00:00 -06:00
Linquize
1d08b72e4f Add unit test to show git_branch_create() fails if attempt to force create current branch 2014-03-02 19:14:20 +08:00
Carlos Martín Nieto
15284a2c5a refs: move current_id before the reflog parameters
Keep the reflog parameters as the last two, as they're the optional
parameters.
2014-02-10 14:52:28 +01:00
Russell Belfer
2d9291943c Merge pull request #2099 from libgit2/bs/more-reflog-stuff
More reflogness
2014-02-07 16:14:17 -08:00
Carlos Martín Nieto
5367ec4b84 refs: add an unconditional delete
Add it under the git_reference_remove() name, letting the user pass the
repo and name, analogous to unconditional setting/creation.
2014-02-05 12:07:57 +01:00
Carlos Martín Nieto
b7ae71ecf2 refs: catch cases where the ref type has changed
If the type of the on-disk reference has changed, the old value
comparison should fail.
2014-02-05 12:07:57 +01:00
Carlos Martín Nieto
f44fd59ed7 refs: check the ref's old value when deleting
Recognize when the reference has changed since we loaded it.
2014-02-05 12:07:57 +01:00
Carlos Martín Nieto
878fb66f57 refs: bring conditional symbolic updates to the frontend
Bring the race detection goodness to symbolic references as well.
2014-02-05 12:07:57 +01:00
Carlos Martín Nieto
d6236cf662 refs: add tests for conditional updates 2014-02-05 12:07:56 +01:00
Ben Straub
010cec3ac2 Add reflog params to git_repository_detach_head 2014-02-04 20:50:40 -08:00
Ben Straub
0d847a3159 Reset helpers: use revparse instead 2014-02-03 14:08:40 -08:00
Arthur Schreiber
50ad7cc208 Add git_reference_is_note. 2014-02-02 18:20:38 +01:00
Ben Straub
db092c1955 Allow tests to run without user config 2014-01-30 16:10:18 -08:00
Ben Straub
a1710a28f6 Enhance testing of signature parameters 2014-01-30 15:53:52 -08:00
Ben Straub
59bb1126e0 Provide good default reflog messages in branch api 2014-01-30 15:53:52 -08:00
Ben Straub
e871d89b28 Ensure moving a branch updates the reflog 2014-01-30 15:52:14 -08:00
Ben Straub
ccf6ce5c89 Ensure renaming a reference updates the reflog 2014-01-30 15:52:13 -08:00
Ben Straub
540c1809f4 Add reflog parameters to git_branch_move 2014-01-30 15:52:13 -08:00
Ben Straub
48110f67e4 Deleting a branch deletes its reflog 2014-01-30 15:52:13 -08:00
Ben Straub
b31ebfbc66 Add reflog params to git_branch_create 2014-01-30 15:52:13 -08:00
Ben Straub
67c4716f74 Add passing reflog tests 2014-01-30 15:51:00 -08:00
Carlos Martín Nieto
b25d87c9cd branch: move to git_buf when outputting newly-allocated strings
Internally we already did everything with git_bufs, so this is just
exposing those functions with public names.
2014-01-27 04:44:05 +01:00
Carlos Martín Nieto
0b28217bda refs: remove the _with_log differentiation
Any well-behaved program should write a descriptive message to the
reflog whenever it updates a reference. Let's make this more prominent
by removing the version without the reflog parameters.
2014-01-15 13:32:43 +01:00
Vicent Marti
4e1f517c61 Merge pull request #1920 from libgit2/cmn/ref-with-log
Reference operations with log
2013-12-18 09:33:45 -08:00
Russell Belfer
11bd7a034b More tests of canceling from callbacks
This covers diff print, push, and ref foreach.  This also has a
fix for a small memory leak in the push tests.
2013-12-12 11:14:51 -08:00
Russell Belfer
25e0b1576d Remove converting user error to GIT_EUSER
This changes the behavior of callbacks so that the callback error
code is not converted into GIT_EUSER and instead we propagate the
return value through to the caller.  Instead of using the
giterr_capture and giterr_restore functions, we now rely on all
functions to pass back the return value from a callback.

To avoid having a return value with no error message, the user
can call the public giterr_set_str or some such function to set
an error message.  There is a new helper 'giterr_set_callback'
that functions can invoke after making a callback which ensures
that some error message was set in case the callback did not set
one.

In places where the sign of the callback return value is
meaningful (e.g. positive to skip, negative to abort), only the
negative values are returned back to the caller, obviously, since
the other values allow for continuing the loop.

The hardest parts of this were in the checkout code where positive
return values were overloaded as meaningful values for checkout.
I fixed this by adding an output parameter to many of the internal
checkout functions and removing the overload.  This added some
code, but it is probably a better implementation.

There is some funkiness in the network code where user provided
callbacks could be returning a positive or a negative value and
we want to rely on that to cancel the loop.  There are still a
couple places where an user error might get turned into GIT_EUSER
there, I think, though none exercised by the tests.
2013-12-11 10:57:49 -08:00
Russell Belfer
96869a4edb Improve GIT_EUSER handling
This adds giterr_user_cancel to return GIT_EUSER and clear any
error message that is sitting around.  As a result of using that
in places, we need to be more thorough with capturing errors that
happen inside a callback when used internally.  To help with that,
this also adds giterr_capture and giterr_restore so that when we
internally use a foreach-type function that clears errors and
converts them to GIT_EUSER, it is easier to restore not just the
return value, but the actual error message text.
2013-12-11 10:57:49 -08:00
Edward Thomson
5588f07360 Clean up warnings 2013-12-09 11:40:44 -05:00
Carlos Martín Nieto
f21051297c refs: expose has_log() on the backend
The frontend used to look at the file directly, but that's obviously not
the right thing to do. Expose it on the backend and use that function
instead.
2013-12-09 15:55:11 +01:00
Carlos Martín Nieto
8d5ec9106a refs: expose a way to ensure a ref has a log
Sometimes (e.g. stash) we want to make sure that a log will be written,
even if it's not in one of the standard locations. Let's make that
easier.
2013-12-09 15:55:11 +01:00
Carlos Martín Nieto
6f13a30565 reflog: write to the reflog following git's rules
git-core only writes to the reflogs of HEAD, refs/heads/ and,
refs/notes/ or if there is already a reflog in place. Adjust our code to
follow these semantics.
2013-12-09 15:55:11 +01:00
Carlos Martín Nieto
13c9e44af9 reflog: remove git_reflog_append_to()
This was a convenience method for the refs front-end to do the reflog
writing. This is now done in the backend and it has no more reason for
being.
2013-11-23 14:55:02 +01:00