Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ghc-prim dependency mismatch #2

Closed
fauno opened this issue Apr 16, 2015 · 10 comments
Closed

ghc-prim dependency mismatch #2

fauno opened this issue Apr 16, 2015 · 10 comments

Comments

@fauno
Copy link

fauno commented Apr 16, 2015

the package description says it requires ghc-prim < 0.5 but the source says ghc-prim 0.4

@hvr
Copy link
Collaborator

hvr commented Apr 16, 2015

What do you mean by "the source says ghc-prim 0.4"?

PS: Do you mean http://hackage.haskell.org/package/deepseq-generics-0.1.1.2/revisions/ ? If so, how is this causing problems?

@fauno
Copy link
Author

fauno commented Apr 16, 2015

sorry i meant the package description from source says ghc-prim < 0.4

i ran into this issue while building pandoc 1.13.2.1 with ghc 7.10, since ghc provides base > 0.4 the build fails, but cabal didn't notice during source download because its database says ghc-prim requires base < 0.5

hope this is clearer

@hvr
Copy link
Collaborator

hvr commented Apr 16, 2015

So this happens if you cabal install deepseq-generics too? Which cabal-install version are you using?

@fauno
Copy link
Author

fauno commented Apr 16, 2015

  • cabal-install 1.22.2.0
  • ghc 7.10.1 which provides ghc-prim 0.4.0.0

cabal install deepseq-generics works correctly (because deepseq-generics.cabal from hackage requires ghc-prim < 0.5), but cabal fetch deepseq-generics and cabal build on the extracted sources fails because deepseq-generics.cabal says ghc-prim < 0.4

@hvr
Copy link
Collaborator

hvr commented Apr 17, 2015

Now I see; you should use cabal get deepseq-generics as that one copies the amended .cabal revision into the extracted source folder (and thus overwriting the 0th .cabal revision contained in the tarball).

I'm still wondering; why do you operate manually on the source tarball, rather than using the revision-aware cabal-facilities?

@fauno
Copy link
Author

fauno commented Apr 17, 2015

Herbert Valerio Riedel notifications@github.com writes:

Now I see; you should use cabal get deepseq-generics as that ones
overlay the amended cabal revision into the extracted source folder.

cool, thanks!

I'm still wondering; why do you operate manually on the source
tarball, rather than using the revision-aware cabal-facilities?

because in order to distribute a pandoc binary package for parabola0 i
need to provide a source tarball, so i use cabal fetch first, make a
source tarball and then build them using the distro pkgbuild1

http://hackcoop.com.ar

@dcoutts
Copy link

dcoutts commented Apr 17, 2015

@fauno in that case I suggest that you make your tools patch the tarball with the latest .cabal file from the index. For example you could cabal unpack (which by default will update the .cabal file), and then call tar to make an updated tarball.

@fauno
Copy link
Author

fauno commented Apr 17, 2015

Duncan Coutts notifications@github.com writes:

@fauno in that case I suggest that you make your tools patch the
tarball with the latest .cabal file from the index. For example you
could cabal unpack (which by default will update the .cabal file), and
then call tar to make an updated tarball.

great! :)

http://hackcoop.com.ar

@fauno
Copy link
Author

fauno commented Apr 18, 2015

i solved it with cabal get, thanks!

@fauno fauno closed this as completed Apr 18, 2015
@mimi1vx
Copy link

mimi1vx commented May 12, 2015

Isn't good problem solv ... not system ,

for example in rpm distributions is by default used cabal file from tar.gz archive with bad constrains ...

@hvr please release new minor.minor version with updated cabal file .. 0.1.1.3 ..

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants