got patch: allow to strip path components Move some bits from the libexec helper to the main process so we know if the patch was generated by git or not and finally document the automatic stripping of a/ and b/ prefixes added by git-diff(1). ok stsp@

use test(1) -eq and -ne to compare integers, and reduce quoting This brings the rest of the regression test scripts in line with patch.sh.

remove trailing whitespaces

link libexec helpers against _p libraries when building with 'make PROFILE=1'

plug a memory leak in list_refs(), ok stsp@

plug a memory leak in got_fetch_pack(); patch by Mikhail

bump version number

CHANGES for 0.68

fgetc() returns int, not char; fixes -Werror build on armv7

sync dist file list

fix free() on uninitialized variable upon error in gw_output_file_blame() ok op@

tweak wording to make it clearer where the mentioned -u option applies

got patch: handle "\ No newline at end of file"

reuse apply_hunk in patch_file when creating a file

mention that 'got patch' is compatible with 'cvs diff -u' output

mention that 'got patch' can read diffs produced by Git, to avoid any doubt

fix pack progress object counter for loose objects Move pack progres object accounting to a single place. This makes it easier to account for the case were only loose objects are packed. A wrong amount of objects was reported before when packing loose ones.

in pack progress output, remove excluded objects from 'found' objects counter

fix a bug where 'gotadmin pack' packed too many objects unless -a was used

subordinate clauses preceding the main clause are offset by a comma

fix potential NULL deref in error path of got_object_idset_remove()

got patch: prefer new name if not /dev/null and not a git-style diff This fixes a common issue when for e.g. generating patches with $ diff -u foo.orig foo where 'got patch' failed because 'foo.orig' has an 'unexpected status'. prodded by naddy, ok stsp

const-ify string table

move got_patch file status checking in worktree.c check_file_status used got_worktree_status to check if the file was in an allowed state, but it's wrong since the callback is not invoked on unchanged files. While here also fix a relate bug: unlink(newpath) is in the wrong spot and ends up removing files even when it shouldn't, so move it early in the got_worktree_schedule_* error handling. Finally, update the appropriate test case. It was passing before because got_worktree_schedule_add returned GOT_ERR_FILE_STATUS, not because check_file_status failed. ok stsp@

explicitly include <unistd.h> for close(2)