Re: [PATCH 0/5] New signing interface API with pluggable drivers
- From
brian m. carlson <sandals@crustytoothpaste.net>
- Date
- Aug 26, 2019, 23:15 UTC
- Message-ID
- <20190826231543.GD11334@genre.crustytoothpaste.net>
- In-Reply-To
- <tz1YiiRtWStGrH2sc42DyD-8bDtH1A52rOCCapct59Qos6jEikqscvusUs7QeOBRNmCF7L_AL1ezKurbTp6qEZCmEk7L9B28wH-TVQGBSdY=@pm.me>
On 2019-08-26 at 19:57:45, Ibrahim El wrote:
Show 12 quoted lines
> Following previous introduction mail [1], this first series of 5 patches is a re-write of the signing interface API in an effort to support easily the addition of new tools with minimal effort and also keeping backwards compatibility with current tools and configuration. > > All existing tests currently pass with backward compatibility. > > [1]: https://public-inbox.org/git/CACi-FhDeAZecXSM36zroty6kpf2BCWLS=0R+dUwuB96LqFKuTA@mail.gmail.com/T/#r43cbf31b86642ab5118e6e7b3d4098bade5f5a0a > > The patches are ordered as follow: > [1/5] - Adding Documentation files explaining the different changes using a design document and updates to the configuration part > [2/5] - Adding new files that define the signing interface API and also drivers for the existing GPG and GPGSM X.509 tools > [3/5] - Migrating the code to using the new signing interface API. Old GPG Interface code is commented and ommited > [4/5] - Removing the old GPG interface and updating the code to remove all gpg mentions from it to make it transparent to the signing tool that is being used > [5/5] - Duplicating existing signature related tests and updating them to using the new configuration aliases
I haven't done an in-depth review of this series, but I did point out a few things that stood out to me. I think the consensus on the list in the past was that for adding future tools, we'd like the drivers to be configuration-based so that Git need not learn about every signing tool. I think such a change would be welcome if done right.
-- brian m. carlson: Houston, Texas, US OpenPGP: https://keybase.io/bk2204