Re: [PATCH v2 0/4] pre-merge hook

2019-07-29 Thread Josh Steadmon
On 2019.07.29 22:29, Martin Ågren wrote: > On Fri, 19 Jul 2019 at 01:56, Josh Steadmon wrote: > > This series revives an old suggestion [...] to make merge honor > > pre-commit hook or a separate pre-merge hook. Since pre-commit does not > > receive any arguments (which the hook could use tell bet

Re: [PATCH v2 0/4] pre-merge hook

2019-07-29 Thread Josh Steadmon
On 2019.07.29 22:09, Martin Ågren wrote: > On Fri, 19 Jul 2019 at 01:56, Josh Steadmon wrote: > > * Martin's objection on 1/4 that the sample hook would always exit > > successfully is (I believe) incorrect. To test this, I ran > > "/bin/true && exec test 0 == 1" in a /bin/sh subshell, and it

Re: [PATCH v2 0/4] pre-merge hook

2019-07-29 Thread Martin Ågren
On Fri, 19 Jul 2019 at 01:56, Josh Steadmon wrote: > This series revives an old suggestion [...] to make merge honor > pre-commit hook or a separate pre-merge hook. Since pre-commit does not > receive any arguments (which the hook could use tell between commit and > merge) and in order to keep exi

Re: [PATCH v2 0/4] pre-merge hook

2019-07-29 Thread Martin Ågren
On Fri, 19 Jul 2019 at 01:56, Josh Steadmon wrote: > * Martin's objection on 1/4 that the sample hook would always exit > successfully is (I believe) incorrect. To test this, I ran > "/bin/true && exec test 0 == 1" in a /bin/sh subshell, and it > correctly had a non-zero exit status. I retr

[PATCH v2 0/4] pre-merge hook

2019-07-18 Thread Josh Steadmon
(I'm resending this cover letter with fixed Message-Id so that threading @ public-inbox works properly. Sorry for the noise.) I would like to revive discussion on this series; I have addressed most of the review comments on the original series sent by Michael, with the following exceptions: * Mar

[PATCH v2 0/4] pre-merge hook

2019-07-18 Thread Josh Steadmon
I would like to revive discussion on this series; I have addressed most of the review comments on the original series sent by Michael, with the following exceptions: * Martin's objection on 1/4 that the sample hook would always exit successfully is (I believe) incorrect. To test this, I ran "/