Brian got the ball rolling with his object-id patches to fixup the code base, but we also need to discuss how we approach our test infrastructure eventually.
C.f.: * jb/t8008-cleanup (2017-07-26) 1 commit - t8008: rely on rev-parse'd HEAD instead of sha1 value * sb/t4005-modernize (Remove hard coded sha1 values.) This presents an alternative approach in the second patch, which just marks tests as "SHA1" required instead of making the test hash-agnostic. The first patch is just an aid to show which tests need addressing, although there are some false positives (i.e. the test fails for other reasons than the hash dependency, such as corruption of data.) Stefan Stefan Beller (2): alter hash function: expose test dependencies on sha1 t6500: mark tests as SHA1 reliant sha1dc/sha1.c | 2 +- t/t6500-gc.sh | 6 +++--- t/test-lib.sh | 4 ++++ 3 files changed, 8 insertions(+), 4 deletions(-) -- 2.14.0.rc0.3.g6c2e499285
