Hello, Today I made some more progress with the implementation of `Typebable with kinds`. In the process, I had to update one line in the `deepseq` package, because it has code to `rnf` `TypeReps`, which now have an extra field. I made a separate branch in `deep-seq` where I was going to push the change, but when I try to do so I get the following error: git push -u origin typeable-with-kinds:typeable-with-kinds Counting objects: 30, done. Delta compression using up to 8 threads. Compressing objects: 100% (3/3), done. Writing objects: 100% (4/4), 377 bytes | 0 bytes/s, done. Total 4 (delta 2), reused 0 (delta 0) remote: W refs/heads/typeable-with-kinds packages/deepseq diatchki DENIED by refs/.* remote: error: hook declined to update refs/heads/typeable-with-kinds To ssh://git@git.haskell.org/packages/deepseq.git ! [remote rejected] typeable-with-kinds -> typeable-with-kinds (hook declined) error: failed to push some refs to 'ssh:// git@git.haskell.org/packages/deepseq.git' Any ideas what that's about? Perhaps, it is simply that I don't have permission to push to `deep-seq`? -Iavor
On 2015-02-10 at 02:20:39 +0100, Iavor Diatchki wrote: [...]
Any ideas what that's about? Perhaps, it is simply that I don't have permission to push to `deep-seq`?
You have to push to GitHub's upstream of deepseq: $ awk '/^libraries\/deepseq/ { print $4 }' packages ssh://git@github.com/haskell/deepseq.git you can do that ad-hoc by git push ssh://git@github.com/haskell/deepseq.git HEAD:master from inside libraries/deepseq it's also mentioned as a comment in the ./packages (as well as on the GHC Wiki, I think) file that some subrepos have to be pushed somewhere else Cheers, hvr
Aha, that makes sense, thanks!
-Iavor
On Tue, Feb 10, 2015 at 12:47 AM, Herbert Valerio Riedel wrote: On 2015-02-10 at 02:20:39 +0100, Iavor Diatchki wrote: [...] Any ideas what that's about? Perhaps, it is simply that I don't have
permission to push to `deep-seq`? You have to push to GitHub's upstream of deepseq: $ awk '/^libraries\/deepseq/ { print $4 }' packages
ssh://git@github.com/haskell/deepseq.git you can do that ad-hoc by git push ssh://git@github.com/haskell/deepseq.git HEAD:master from inside libraries/deepseq it's also mentioned as a comment in the ./packages (as well as on the
GHC Wiki, I think) file that some subrepos have to be pushed somewhere
else Cheers,
hvr
On further thought, would it be hard to update whatever hook is checking
for this, to give a bit more detail? If it is easy, we could update the
message to say something like ("You should not push here, please push to
URL instead").
It is nice to have documentation on the wiki, but it is even better if the
system is self-documenting, so one does not have to go searching on the
Internet.
On Tue, Feb 10, 2015 at 10:01 AM, Iavor Diatchki
Aha, that makes sense, thanks! -Iavor
On Tue, Feb 10, 2015 at 12:47 AM, Herbert Valerio Riedel < hvriedel@gmail.com> wrote:
On 2015-02-10 at 02:20:39 +0100, Iavor Diatchki wrote:
[...]
Any ideas what that's about? Perhaps, it is simply that I don't have permission to push to `deep-seq`?
You have to push to GitHub's upstream of deepseq:
$ awk '/^libraries\/deepseq/ { print $4 }' packages ssh://git@github.com/haskell/deepseq.git
you can do that ad-hoc by
git push ssh://git@github.com/haskell/deepseq.git HEAD:master
from inside libraries/deepseq
it's also mentioned as a comment in the ./packages (as well as on the GHC Wiki, I think) file that some subrepos have to be pushed somewhere else
Cheers, hvr
On 2015-02-10 at 19:14:12 +0100, Iavor Diatchki wrote:
On further thought, would it be hard to update whatever hook is checking for this, to give a bit more detail? If it is easy, we could update the message to say something like ("You should not push here, please push to URL instead").
Tbh, I don't know how easy it is; we use Gitolite (v2) for handling access control, and I'd have to investigate if there's a way to customize the access-denied message in useful way (as a function of git-repo and maybe ref-name)... I'd have to investigate, or have somebody with more experience tell me... :) Cheers, hvr
participants (2)
-
Herbert Valerio Riedel -
Iavor Diatchki