Re: [PATCH 3/3] fast-import: add 'strip-if-invalid' mode to --signed-commits=<mode>
- From
Junio C Hamano <gitster@pobox.com>
- Date
- Nov 8, 2025, 18:32 UTC
- Message-ID
- <xmqqjz00e5ns.fsf@gitster.g>
- In-Reply-To
- <20251105061918.3688870-4-christian.couder@gmail.com>
Christian Couder <christian.couder@gmail.com> writes:
> t/t9305-fast-import-signatures.sh | 118 ++++++++++++++++++++++++++++- > 6 files changed, 226 insertions(+), 28 deletions(-)
Unfortunately all these tests that assume that explicit-sha256 repository as a subdirectory exists would fail when the topic is merged to 'seen' and the tree is built without the optional Rust support. This is because brian's f6581e23 (repository: require Rust support for interoperability, 2025-10-27) changes a couple of tests to require RUST prerequisite. One of them is what creates the explicit-sha256 repository.
I do not think this topic to preserve or strip GPG signatures particularly cares about the dual hash interoperability, so can you rearrange the tests in this series to avoid crashing with the other topic?
Thanks.