46 bytes, text/x-phabricator-request
|Details | Review|
This is necessary for no-single-locale-fennec work, (because I'm adding a new Build for the same platform that only has a different symbol but not platform
This changes a bunch of symbols, notably: * Group name BMR-L10n(*) ==> BMR(*) * Groups the BMR symbol itself for en-US Nightly [so BMR ==> BMR(N)] * Changes Source Signing fron symbol Bs to Srcs * Groups Source signing beetmoving as BM(Srcs) instead of symbol BM-S * Balrog complete update submission changes from c-Up(N) to c-Up(Ns) * Beetmover en-US changes from symbol BM-S to BM(Ns) [used in Fennec] * Beetmover Checksums (en-US) changes from BMcs(N) => BMcs(Ns) [fennec] * Checksums signing (en-US, fennec) changes from cs(N) => cs(Ns)
:Aryx and :RyanVM do either of you have any concerns/opinions as to if any of these group/symbol changes are bad from your seats?
(In reply to Justin Wood (:Callek) from comment #1) > * Changes Source Signing fron symbol Bs to Srcs > * Groups Source signing beetmoving as BM(Srcs) instead of symbol BM-S "Srcs" sounds ambiguous and people might think it's the abbreviation for "sources" and it has also has the potential to not remain unique in the long term. With the trailing "s" for "signing" and "src", how about "src-s" which is a different styles than Ns but afaik only build-ish jobs use capital letters so SRCs would also violate a style.
Agree that Srcs sounds like "sources" to me. "Ns" is also ambiguous to me since that's already used for Nightly signing jobs. Probably fine as long as it's properly grouped and is a legit signing job, but I just want to advocate for consistency with whatever we end up doing. Thanks for doing this cleanup!
I think most of the changes are great. However, I'm not sure that changes the that lead to `BM(Srcs)`, `BMcs(Ns)` etc. make sense. I wonder if it would be clearer to add a `s` group for all signing jobs instead?
Pushed by email@example.com: https://hg.mozilla.org/integration/autoland/rev/24dc3e24fd89 Define many treeherder symbols based on the parent task. r=aki
(In reply to Tom Prince [:tomprince] from comment #5) > I think most of the changes are great. However, I'm not sure that changes > the that lead to `BM(Srcs)`, `BMcs(Ns)` etc. make sense. I wonder if it > would be clearer to add a `s` group for all signing jobs instead? Possibly, I really want to make the Treeherder stuff more isolated as a whole, rather than how I did it here, but this was needed to support other work I've been doing...
Pushed by firstname.lastname@example.org: https://hg.mozilla.org/comm-central/rev/54215be7c803 Port bug 1485861 - Define many treeherder symbols based on the parent task. rs=bustage-fix
You need to log in before you can comment on or make changes to this bug.