[PATCH v2 00/33] Yet more preparation for reference backends

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Thanks to David, Junio, and Peff for their comments on v1 of this
patch series [1]. I think I have addressed all of the points that were
brought up. Plus I fixed a pre-existing bug that I noticed myself
while adding some more tests; see the first bullet point below for
more information.

Changes between v1 and v2:

* Prefixed the patch series with three new patches that demonstrate
  and fix some bugs in reference resolution that I noticed while
  inspecting this series:

  * "t1404: demonstrate a bug resolving references" -- this
    demonstrates a bug that has existed since at least 2.5.0.
  * "read_raw_ref(): don't get confused by an empty directory"
  * "commit_ref(): if there is an empty dir in the way, delete it"

* Added a patch "read_raw_ref(): move docstring to header file".

* "ref_transaction_create(): disallow recursive pruning": Add a
  comment that `REF_ISPRUNING` must only be used with `REF_NODEREF`.

* "refs: don't dereference on rename": explain why we can't pass an
  `old_sha1` argument to `delete_ref()`. Inline `resolve_flags`
  constant to make the code more transparent.

* "add_update(): initialize the whole ref_update": move some more
  checks from `ref_transaction_update()` to `add_update()`.

* "refs: resolve symbolic refs first":
  * Remove unused `deleting` parameter from `lock_raw_ref()`
  * Fix a comment, add another.

* "lock_ref_sha1_basic(): only handle REF_NODEREF mode": initialize
  `ref_name` only once, as its value can be reused.

This patch series is also available from my GitHub repo [2] as branch
"split-under-lock".

[1] http://thread.gmane.org/gmane.comp.version-control.git/292772
[2] https://github.com/mhagger/git

David Turner (2):
  refs: allow log-only updates
  refs: don't dereference on rename

Michael Haggerty (31):
  t1404: demonstrate a bug resolving references
  commit_ref(): if there is an empty dir in the way, delete it
  read_raw_ref(): don't get confused by an empty directory
  safe_create_leading_directories(): improve docstring
  remove_dir_recursively(): add docstring
  refname_is_safe(): use skip_prefix()
  refname_is_safe(): don't allow the empty string
  refname_is_safe(): insist that the refname already be normalized
  commit_ref_update(): write error message to *err, not stderr
  rename_ref(): remove unneeded local variable
  ref_transaction_commit(): remove local variable n
  read_raw_ref(): rename flags argument to type
  read_raw_ref(): clear *type at start of function
  read_raw_ref(): rename symref argument to referent
  read_raw_ref(): improve docstring
  read_raw_ref(): move docstring to header file
  lock_ref_sha1_basic(): remove unneeded local variable
  refs: make error messages more consistent
  ref_transaction_create(): disallow recursive pruning
  ref_transaction_commit(): correctly report close_ref() failure
  delete_branches(): use resolve_refdup()
  verify_refname_available(): adjust constness in declaration
  add_update(): initialize the whole ref_update
  lock_ref_for_update(): new function
  unlock_ref(): move definition higher in the file
  ref_transaction_update(): check refname_is_safe() at a minimum
  refs: resolve symbolic refs first
  lock_ref_for_update(): don't re-read non-symbolic references
  lock_ref_for_update(): don't resolve symrefs
  commit_ref_update(): remove the flags parameter
  lock_ref_sha1_basic(): only handle REF_NODEREF mode

 builtin/branch.c                   |  19 +-
 cache.h                            |   5 +
 dir.h                              |  23 +
 refs.c                             |  96 ++--
 refs/files-backend.c               | 909 ++++++++++++++++++++++++++++---------
 refs/refs-internal.h               |  95 +++-
 t/t1400-update-ref.sh              |  41 +-
 t/t1404-update-ref-df-conflicts.sh |  76 +++-
 t/t1430-bad-ref-name.sh            |   2 +-
 t/t3200-branch.sh                  |   9 +
 10 files changed, 1013 insertions(+), 262 deletions(-)

-- 
2.8.1

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]