Change code owners #2587
Change code owners #2587
Conversation
owen-mc-diffblue
requested review from
chrisr-diffblue,
kroening,
peterschrammel,
smowton and
tautschnig
as code owners
Leave code owners for test folders unspecified
owen-mc-diffblue
force-pushed
the
owen-jones-diffblue/update-codeowners-2
branch
from
a81147c to
52d71d2
Compare
CODEOWNERS
Outdated
| /.travis.yml @diffblue/devops @thk123 @forejtv @peterschrammel | ||
| /appveyor.yml @diffblue/devops @thk123 @forejtv @peterschrammel | ||
| # All remaining files are assumed to be low-risk | ||
| # Any two reviewers can be assigned |
There was a problem hiding this comment.
To be honest, I wouldn't really consider any of the CI configuration files "low risk." Changes to them may have impact on the bills paid. Indeed the CodeBuild configuration files should be added here. (I do agree that /scripts/ is low risk.)
|
@diffblue/devops and @forejtv Are you okay with being removed as code owners from |
There was a problem hiding this comment.
This PR failed Diffblue compatibility checks (cbmc commit: a81147c).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/79480656
Status will be re-evaluated on next push.
Please contact @peterschrammel, @thk123, or @allredj for support.
Common spurious failures:
- the cbmc commit has disappeared in the mean time (e.g. in a force-push)
- the author is not in the list of contributors (e.g. first-time contributors).
There was a problem hiding this comment.
Passed Diffblue compatibility checks (cbmc commit: 52d71d2).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/79482358
|
Personally I think we should leave the owner list as-is (i.e. the underlying #2580 should be merged, but drop this PR), but change the rules for how auto-review requests ought to be handled: When you receive an auto-review request, ASSIGN (cf. request-review-from) two people to do the actual reviewing (possibly yourself, possibly do it right now if it's trivial). The problem behaviour is that people get auto-review requests and ignore by default. If your response to an auto-request is assign reviewers if not already assigned then the problem goes away. |
|
As explained in #2580 this should not be merged in its current form. Removing the default code owners has unintended consequences. |
| /src/xmllang/ @kroening @tautschnig @peterschrammel | ||
| /src/nonstd/ @smowton @peterschrammel | ||
| /src/solvers/cvc @martin-cs @kroening | ||
| /src/solvers/flattening @martin-cs @kroening @tautschnig @peterschrammel |
| /src/goto-programs/ @kroening @tautschnig | ||
| /src/util/ @kroening @tautschnig | ||
| /src/solvers/refinement @romainbrenguier @peterschrammel | ||
| /jbmc/src/java_bytecode/ @peterschrammel @smowton |
There was a problem hiding this comment.
I think this isn't workable without someone from the Deeptest group as a code owner for java_bytecode as it needs to be changed frequently.
|
There doesn't seem to be enough support for reducing the number of code owners for each folder. Closing this PR. |
This is based on top of #2580. This PR only concerns the last two commits. See the old PR for more discussion.