summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2020-01-15t5604: make hash independentLibravatar brian m. carlson1-2/+3
To make our values hash independent, we turn the directory of the object into "Y" and the file name into "Z" after having sorted items by their name. However, when using SHA-256, one of our file names begins with an "a" character, which means it sorts into the wrong place in the list, causing the test to fail. Since we don't care about the order of these items, just sort them after stripping actual hash contents, which means they'll work with any hash algorithm. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5601: switch into repository to hash objectLibravatar brian m. carlson1-2/+2
This test performs a clone from outside any repository. Consequently, the hash algorithm used defaults to SHA-1. When the test is running with SHA-256, this results in an object ID that is not usable by the rest of the test. In order to ensure that we provide a usable value, switch into the source repository before hashing. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5562: use $ZERO_OIDLibravatar brian m. carlson1-1/+1
This test uses $_z40 to express an all-zeros object ID, which doesn't work for SHA-256. Use $ZERO_OID instead, which is the right size for all hash values. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5540: make hash size independentLibravatar brian m. carlson1-4/+2
Use regex values based on $OID_REGEX instead of hard-coding them based on expected object ID lengths. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5537: make hash size independentLibravatar brian m. carlson1-2/+6
This test modifies a pkt-line stream with sed to change a line with "shallow" to "unshallow". However, this modification is dependent on the size of the hash in use; with SHA-256, the pkt-line length is different, leading to the sed command having no effect. Use test_oid_cache to specify the correct values for each hash so that the test continues to work. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5530: compute results based on object lengthLibravatar brian m. carlson1-12/+14
Compute the various pkt-line values based on the length of the object IDs in use. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5512: abstract away SHA-1-specific constantsLibravatar brian m. carlson1-8/+9
Adjust the test so that it computes variables for object IDs instead of using hard-coded hashes. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5510: make hash size independentLibravatar brian m. carlson1-4/+5
Use $OID_REGEX instead of hard-coding 40-based regular expressions. Change invocations of cut with a hard-coded constant to split using a delimiter instead. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5504: make hash algorithm independentLibravatar brian m. carlson1-8/+9
Instead of hard-coding invalid object IDs in this test, use test_oid to look up ones of the appropriate length. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5324: make hash size independentLibravatar brian m. carlson1-3/+10
There are some offsets in the commit graph files used to corrupt data. Compute these offsets for both SHA-1 and SHA-256 so that the test works with either. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5319: make test work with SHA-256Libravatar brian m. carlson1-4/+18
This test corrupts various locations in a multi-pack index to test various error responses. However, these offsets differ between SHA-1 indexes and SHA-256 indexes due to differences in object length. Use test_oid to look up the correct offsets based on the algorithm. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5319: change invalid offset for SHA-256 compatibilityLibravatar brian m. carlson1-2/+2
When using SHA-1, the existing value of the byte we use is 0x13, so writing the byte 0x07 serves to corrupt the test and verify that we detect corruption. However, when we use SHA-256, the value at that offset is already 0x07, so our "corruption" doesn't work and the test fails to detect it. To provide a value that is truly out of range, let's use 0xff, which is not likely to be a valid value as the high byte of a two-byte offset in a multi-pack index this small. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t5318: update for SHA-256Libravatar brian m. carlson1-1/+1
When running with SHA-256 as the hash algorithm, the hash version octet is 2 instead of 1. Pick the right value depending on the hash algorithm and use it where we look for the existing value. To ensure the test checking for invalid data passes, use 3 as the test value for an invalid hash version. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4300: abstract away SHA-1-specific constantsLibravatar brian m. carlson1-94/+94
Adjust the test so that it computes values for object IDs instead of using hard-coded hashes. Move the heredocs later in the tests so we can take advantage of computed values. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4204: make hash size independentLibravatar brian m. carlson1-1/+1
Use $OID_REGEX instead of a hard-coded regular expression. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4202: abstract away SHA-1-specific constantsLibravatar brian m. carlson1-62/+65
Adjust the test so that it computes values for object IDs instead of using hard-coded hashes. Additionally, update the sanitize_output function to sanitize the index lines in diff output, since it's clear from the assertions in question that we are not interested in the specific object IDs. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4200: make hash size independentLibravatar brian m. carlson1-1/+2
Instead of hard-coding a fixed length example object ID in the test, look one up using the translation tables. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4134: compute appropriate length constantLibravatar brian m. carlson1-2/+3
Instead of using a specific invalid hard-coded object ID, generate one of the appropriate length by looking one up in the translation tables. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4066: compute index line in diffsLibravatar brian m. carlson1-5/+5
Since the object ID used in the index line will differ between different algorithms, compute these values instead of hard-coding them. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-01-15t4054: make hash-size independentLibravatar brian m. carlson1-1/+2
Instead of hard-coding the length of an object ID when creating a tree, generate it based on $ZERO_OID. Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-10Merge branch 'am/pathspec-from-file'Libravatar Junio C Hamano9-28/+434
A few commands learned to take the pathspec from the standard input or a named file, instead of taking it as the command line arguments. * am/pathspec-from-file: commit: support the --pathspec-from-file option doc: commit: synchronize <pathspec> description reset: support the `--pathspec-from-file` option doc: reset: synchronize <pathspec> description pathspec: add new function to parse file parse-options.h: add new options `--pathspec-from-file`, `--pathspec-file-nul`
2019-12-09Sync with Git 2.24.1Libravatar Junio C Hamano47-87/+991
2019-12-06The fourth batchLibravatar Junio C Hamano1-0/+18
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-06Merge branch 'dl/lore-is-the-archive'Libravatar Junio C Hamano14-17/+18
Publicize lore.kernel.org mailing list archive and use URLs pointing into it to refer to notable messages in the documentation. * dl/lore-is-the-archive: doc: replace LKML link with lore.kernel.org RelNotes: replace Gmane with real Message-IDs doc: replace MARC links with lore.kernel.org
2019-12-06Merge branch 'jk/lore-is-the-archive'Libravatar Junio C Hamano13-30/+30
Doc update for the mailing list archiving and nntp service. * jk/lore-is-the-archive: doc: replace public-inbox links with lore.kernel.org doc: recommend lore.kernel.org over public-inbox.org
2019-12-06Merge branch 'tg/perf-remove-stale-result'Libravatar Junio C Hamano2-11/+5
PerfTest fix to avoid stale result mixed up with the latest round of test results. * tg/perf-remove-stale-result: perf-lib: use a single filename for all measurement types
2019-12-06Merge branch 'jk/send-pack-check-negative-with-quick'Libravatar Junio C Hamano1-1/+3
Performance tweak on "git push" into a repository with many refs that point at objects we have never heard of. * jk/send-pack-check-negative-with-quick: send-pack: use OBJECT_INFO_QUICK to check negative objects
2019-12-06Merge branch 'hi/grep-do-not-return-void'Libravatar Junio C Hamano1-1/+1
Code cleanup. * hi/grep-do-not-return-void: grep: don't return an expression from pcre2_free()
2019-12-06Merge branch 'rs/use-skip-prefix-more'Libravatar Junio C Hamano5-22/+16
Code cleanup. * rs/use-skip-prefix-more: name-rev: use skip_prefix() instead of starts_with() push: use skip_prefix() instead of starts_with() shell: use skip_prefix() instead of starts_with() fmt-merge-msg: use skip_prefix() instead of starts_with() fetch: use skip_prefix() instead of starts_with()
2019-12-06Merge branch 'rs/simplify-prepare-cmd'Libravatar Junio C Hamano1-2/+1
Code cleanup. * rs/simplify-prepare-cmd: run-command: use prepare_git_cmd() in prepare_cmd()
2019-12-06Merge branch 'rs/test-cleanup'Libravatar Junio C Hamano10-23/+16
Test cleanup. * rs/test-cleanup: t7811: don't create unused file t9300: don't create unused file test: use test_must_be_empty F instead of test_cmp empty F test: use test_must_be_empty F instead of test -z $(cat F) t1400: use test_must_be_empty t1410: use test_line_count t1512: use test_line_count
2019-12-06Merge branch 'sg/assume-no-todo-update-in-cherry-pick'Libravatar Junio C Hamano2-1/+31
While running "revert" or "cherry-pick --edit" for multiple commits, a recent regression incorrectly detected "nothing to commit, working tree clean", instead of replaying the commits, which has been corrected. * sg/assume-no-todo-update-in-cherry-pick: sequencer: don't re-read todo for revert and cherry-pick
2019-12-06Merge branch 'sg/osx-force-gcc-9'Libravatar Junio C Hamano2-4/+5
TravisCI update. * sg/osx-force-gcc-9: ci: build Git with GCC 9 in the 'osx-gcc' build job
2019-12-06Git 2.24.1Libravatar Johannes Schindelin3-2/+10
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06Sync with 2.23.1Libravatar Johannes Schindelin46-87/+983
* maint-2.23: (44 commits) Git 2.23.1 Git 2.22.2 Git 2.21.1 mingw: sh arguments need quoting in more circumstances mingw: fix quoting of empty arguments for `sh` mingw: use MSYS2 quoting even when spawning shell scripts mingw: detect when MSYS2's sh is to be spawned more robustly t7415: drop v2.20.x-specific work-around Git 2.20.2 t7415: adjust test for dubiously-nested submodule gitdirs for v2.20.x Git 2.19.3 Git 2.18.2 Git 2.17.3 Git 2.16.6 test-drop-caches: use `has_dos_drive_prefix()` Git 2.15.4 Git 2.14.6 mingw: handle `subst`-ed "DOS drives" mingw: refuse to access paths with trailing spaces or periods mingw: refuse to access paths with illegal characters ...
2019-12-06Git 2.23.1Libravatar Johannes Schindelin3-2/+10
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06Sync with 2.22.2Libravatar Johannes Schindelin45-87/+975
* maint-2.22: (43 commits) Git 2.22.2 Git 2.21.1 mingw: sh arguments need quoting in more circumstances mingw: fix quoting of empty arguments for `sh` mingw: use MSYS2 quoting even when spawning shell scripts mingw: detect when MSYS2's sh is to be spawned more robustly t7415: drop v2.20.x-specific work-around Git 2.20.2 t7415: adjust test for dubiously-nested submodule gitdirs for v2.20.x Git 2.19.3 Git 2.18.2 Git 2.17.3 Git 2.16.6 test-drop-caches: use `has_dos_drive_prefix()` Git 2.15.4 Git 2.14.6 mingw: handle `subst`-ed "DOS drives" mingw: refuse to access paths with trailing spaces or periods mingw: refuse to access paths with illegal characters unpack-trees: let merged_entry() pass through do_add_entry()'s errors ...
2019-12-06Git 2.22.2Libravatar Johannes Schindelin3-2/+10
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06Sync with 2.21.1Libravatar Johannes Schindelin44-87/+967
* maint-2.21: (42 commits) Git 2.21.1 mingw: sh arguments need quoting in more circumstances mingw: fix quoting of empty arguments for `sh` mingw: use MSYS2 quoting even when spawning shell scripts mingw: detect when MSYS2's sh is to be spawned more robustly t7415: drop v2.20.x-specific work-around Git 2.20.2 t7415: adjust test for dubiously-nested submodule gitdirs for v2.20.x Git 2.19.3 Git 2.18.2 Git 2.17.3 Git 2.16.6 test-drop-caches: use `has_dos_drive_prefix()` Git 2.15.4 Git 2.14.6 mingw: handle `subst`-ed "DOS drives" mingw: refuse to access paths with trailing spaces or periods mingw: refuse to access paths with illegal characters unpack-trees: let merged_entry() pass through do_add_entry()'s errors quote-stress-test: offer to test quoting arguments for MSYS2 sh ...
2019-12-06Git 2.21.1Libravatar Johannes Schindelin3-2/+14
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06Merge branch 'fix-msys2-quoting-bugs'Libravatar Johannes Schindelin1-5/+20
These patches fix several bugs in quoting arguments when spawning shell scripts on Windows. Note: these bugs are Windows-only, as we have to construct a command line for the process-to-spawn, unlike Linux/macOS, where `execv()` accepts an already-split command line. Furthermore, these fixes were not included in the CVE-2019-1350 part of v2.14.6 because the Windows-specific quoting when spawning shell scripts was contributed from Git for Windows into Git only in the v2.21.x era. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06mingw: sh arguments need quoting in more circumstancesLibravatar Johannes Schindelin1-2/+3
Previously, we failed to quote characters such as '*', '(' and the likes. Let's fix this. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06t7415: drop v2.20.x-specific work-aroundLibravatar Johannes Schindelin1-1/+1
This reverts the work-around that was introduced just for the v2.20.x release train in "t7415: adjust test for dubiously-nested submodule gitdirs for v2.20.x"; It is not necessary for v2.21.x. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06mingw: fix quoting of empty arguments for `sh`Libravatar Johannes Schindelin1-1/+1
When constructing command-lines to spawn processes, it is an unfortunate but necessary decision to quote arguments differently: MSYS2 has different dequoting rules (inherited from Cygwin) than the rest of Windows. To accommodate that, Git's Windows compatibility layer has two separate quoting helpers, one for MSYS2 (which it uses exclusively when spawning `sh`) and the other for regular Windows executables. The MSYS2 one had an unfortunate bug where a `,` somehow slipped in, instead of the `;`. As a consequence, empty arguments would not be enclosed in a pair of double quotes, but the closing double quote was skipped. Let's fix this. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06mingw: use MSYS2 quoting even when spawning shell scriptsLibravatar Johannes Schindelin1-1/+2
At the point where `mingw_spawn_fd()` is called, we already have a full path to the script interpreter in that scenario, and we pass it in as the executable to run, while the `argv` reflect what the script should receive as command-line. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06mingw: detect when MSYS2's sh is to be spawned more robustlyLibravatar Johannes Schindelin1-1/+14
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06Sync with 2.20.2Libravatar Johannes Schindelin43-82/+935
* maint-2.20: (36 commits) Git 2.20.2 t7415: adjust test for dubiously-nested submodule gitdirs for v2.20.x Git 2.19.3 Git 2.18.2 Git 2.17.3 Git 2.16.6 test-drop-caches: use `has_dos_drive_prefix()` Git 2.15.4 Git 2.14.6 mingw: handle `subst`-ed "DOS drives" mingw: refuse to access paths with trailing spaces or periods mingw: refuse to access paths with illegal characters unpack-trees: let merged_entry() pass through do_add_entry()'s errors quote-stress-test: offer to test quoting arguments for MSYS2 sh t6130/t9350: prepare for stringent Win32 path validation quote-stress-test: allow skipping some trials quote-stress-test: accept arguments to test via the command-line tests: add a helper to stress test argument quoting mingw: fix quoting of arguments Disallow dubiously-nested submodule git directories ...
2019-12-06Git 2.20.2Libravatar Johannes Schindelin3-2/+20
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-12-06submodule: defend against submodule.update = !command in .gitmodulesLibravatar Jonathan Nieder1-0/+2
In v2.15.4, we started to reject `submodule.update` settings in `.gitmodules`. Let's raise a BUG if it somehow still made it through from anywhere but the Git config. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
2019-12-06t7415: adjust test for dubiously-nested submodule gitdirs for v2.20.xLibravatar Johannes Schindelin1-1/+1
In v2.20.x, Git clones submodules recursively by first creating the submodules' gitdirs and _then_ "updating" the submodules. This can lead to the situation where the clone path is taken because the directory (while it exists already) is not a git directory, but then the clone fails because that gitdir is unexpectedly already a directory. This _also_ works around the vulnerability that was fixed in "Disallow dubiously-nested submodule git directories", but it produces a different error message than the one expected by the test case, therefore we adjust the test case accordingly. Note: as the two submodules "race each other", there are actually two possible error messages, therefore we have to teach the test case to expect _two_ possible (and good) outcomes in addition to the one it expected before. Note: this workaround is only necessary for the v2.20.x release train; The behavior changed again in v2.21.x so that the original test case's expectations are met again. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-12 19:22:48 +0000