compilation error of tensorflow-proto 0.1.0.0
Hello, Using ghc 8.2.2 (debian) , I got the following error while building the package with cabal 2.0: Building library for tensorflow-proto-0.1.0.0.. creating dist/build /media/travail/ghc/bin/ghc --make -fbuilding-cabal-package -O -static -dynamic-too -dynosuf dyn_o -dynhisuf dyn_hi -outputdir dist/build -odir dist/build -hidir dist/build -stubdir dist/build -i -idist/build -i. -idist/build/autogen -idist/build/global-autogen -Idist/build/autogen -Idist/build/global-autogen -Idist/build -I. -optP-include -optPdist/build/autogen/cabal_macros.h -this-unit-id tensorflow-proto-0.1.0.0-A707TW4oXcjUPzDlNSl3T -hide-all-packages -Wmissing-home-modules -package-db dist/package.conf.inplace -package-id proto-lens-0.2.2.0-4VJbI6HHabd74gTwpFhCiY -package-id proto-lens-protoc-0.2.2.3-KiMlOomTiwkJiIiDqOmBdc -package-id base-4.10.1.0 -XHaskell2010 Proto.Tensorflow.Core.Framework.AttrValue Proto.Tensorflow.Core.Framework.Graph Proto.Tensorflow.Core.Framework.NodeDef Proto.Tensorflow.Core.Framework.OpDef Proto.Tensorflow.Core.Framework.ResourceHandle Proto.Tensorflow.Core.Framework.Summary Proto.Tensorflow.Core.Framework.Tensor Proto.Tensorflow.Core.Framework.TensorShape Proto.Tensorflow.Core.Framework.Types Proto.Tensorflow.Core.Protobuf.Config Proto.Tensorflow.Core.Util.Event Proto.Tensorflow.Core.Framework.AllocationDescription Proto.Tensorflow.Core.Framework.CostGraph Proto.Tensorflow.Core.Framework.Function Proto.Tensorflow.Core.Framework.StepStats Proto.Tensorflow.Core.Framework.TensorDescription Proto.Tensorflow.Core.Framework.Versions Proto.Tensorflow.Core.Protobuf.Debug [ 1 of 18] Compiling Proto.Tensorflow.Core.Framework.AllocationDescription ( dist/build/autogen/Proto/Tensorflow/Core/Framework/AllocationDescription.hs, dist/build/Proto/Tensorflow/Core/Framework/AllocationDescription.o ) dist/build/autogen/Proto/Tensorflow/Core/Framework/AllocationDescription.hs:159:15: error: ⢠Couldn't match expected type âData.ProtoLens.MessageDescriptor AllocationDescriptionâ with actual type âData.Map.Map Prelude.String (Data.ProtoLens.FieldDescriptor AllocationDescription) -> Data.ProtoLens.MessageDescriptor AllocationDescriptionâ ⢠Probable cause: âData.ProtoLens.MessageDescriptorâ is applied to too few arguments In the expression: Data.ProtoLens.MessageDescriptor (Data.Map.fromList [(Data.ProtoLens.Tag 1, requestedBytes__field_descriptor), (Data.ProtoLens.Tag 2, allocatedBytes__field_descriptor), (Data.ProtoLens.Tag 3, allocatorName__field_descriptor), (Data.ProtoLens.Tag 4, allocationId__field_descriptor), ....]) (Data.Map.fromList [("requested_bytes", requestedBytes__field_descriptor), ("allocated_bytes", allocatedBytes__field_descriptor), ("allocator_name", allocatorName__field_descriptor), ("allocation_id", allocationId__field_descriptor), ....]) Does it mean that I have the wrong version of proto-lens ? * proto-lens Synopsis: A lens-based implementation of protocol buffers in Haskell. Default available version: 0.3.1.0 Installed versions: 0.2.2.0, 0.3.1.0 Homepage: https://github.com/google/proto-lens#readme License: BSD3 -- Fabien
On Tue, May 15, 2018 at 01:13:30PM +0200, Fabien R wrote:
Hello,
Using ghc 8.2.2 (debian) , I got the following error while building the package with cabal 2.0:
[...]
Does it mean that I have the wrong version of proto-lens ?
`cabal new-build --allow-newer` works (or at least, passes that specific compilation stage -- I get a `Can't run protoc on step [18 of 18]), so most likely yes! Can you check if it works on your machine?
On 15/05/18 14:55, Francesco Ariis wrote:
`cabal new-build --allow-newer` works (or at least, passes that specific compilation stage I'm using 'cabal install'. 'cabal new-build' seems to rebuild everything...
-- I get a `Can't run protoc on step [18 of 18]), Check that you have protobuf 3 (I installed it from the sources).
-- Fabien
`cabal new-build` should only rebuild things once, however. And it's generally better. On 05/16/2018 11:50 AM, Fabien R wrote:
On 15/05/18 14:55, Francesco Ariis wrote:
`cabal new-build --allow-newer` works (or at least, passes that specific compilation stage I'm using 'cabal install'. 'cabal new-build' seems to rebuild everything...
-- I get a `Can't run protoc on step [18 of 18]), Check that you have protobuf 3 (I installed it from the sources).
-- Fabien
_______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.
On 16/05/18 18:54, Vanessa McHale wrote:
`cabal new-build` should only rebuild things once, however. And it's generally better. Ok. I have another error now:
cabal -v new-build --allow-newer tensorflow-proto Building library for tensorflow-proto-0.1.0.0.. creating /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build /media/travail/ghc/bin/ghc --make -fbuilding-cabal-package -O -static -dynamic-too -dynosuf dyn_o -dynhisuf dyn_hi -outputdir /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build -odir /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build -hidir /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build -stubdir /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build -i -i/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build -i. -i/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/autogen -i/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/global-autogen -I/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/autogen -I/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/global-autogen -I/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build -I. -optP-include -optP/tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/autogen/cabal_macros.h -this-unit-id tensorflow-proto-0.1.0.0-inplace -hide-all-packages -Wmissing-home-modules -no-user-package-db -package-db /home/fabien/.cabal/store/ghc-8.2.2/package.db -package-db /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/packagedb/ghc-8.2.2 -package-db /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/package.conf.inplace -package-id proto-lens-0.3.1.0-606296feb88b5391f9c209775e99cb41e62920d11d1a70bc2dee6dc222293b57 -package-id proto-lens-protoc-0.3.1.0-701ad82b9718922791137f9ed26169a31fe583a6444a5c76dfa689acf9e7466e -package-id base-4.10.1.0 -XHaskell2010 Proto.Tensorflow.Core.Framework.AttrValue Proto.Tensorflow.Core.Framework.Graph Proto.Tensorflow.Core.Framework.NodeDef Proto.Tensorflow.Core.Framework.OpDef Proto.Tensorflow.Core.Framework.ResourceHandle Proto.Tensorflow.Core.Framework.Summary Proto.Tensorflow.Core.Framework.Tensor Proto.Tensorflow.Core.Framework.TensorShape Proto.Tensorflow.Core.Framework.Types Proto.Tensorflow.Core.Protobuf.Config Proto.Tensorflow.Core.Util.Event Proto.Tensorflow.Core.Framework.AllocationDescription Proto.Tensorflow.Core.Framework.CostGraph Proto.Tensorflow.Core.Framework.Function Proto.Tensorflow.Core.Framework.StepStats Proto.Tensorflow.Core.Framework.TensorDescription Proto.Tensorflow.Core.Framework.Versions Proto.Tensorflow.Core.Protobuf.Debug [ 1 of 18] Compiling Proto.Tensorflow.Core.Framework.AllocationDescription ( /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/global-autogen/Proto/Tensorflow/Core/Framework/AllocationDescription.hs, /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/Proto/Tensorflow/Core/Framework/AllocationDescription.o ) /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/global-autogen/Proto/Tensorflow/Core/Framework/AllocationDescription.hs:134:9: error: âdescriptorâ is not a (visible) method of class âData.ProtoLens.Messageâ | 134 | descriptor | ^^^^^^^^^^ /tmp/tensorflow-proto-0.1.0.0/dist-newstyle/build/x86_64-linux/ghc-8.2.2/tensorflow-proto-0.1.0.0/build/global-autogen/Proto/Tensorflow/Core/Framework/AllocationDescription.hs:173:15: error: Not in scope: data constructor âData.ProtoLens.MessageDescriptorâ Perhaps you meant one of these: âData.ProtoLens.AnyMessageDescriptorâ (imported from Data.ProtoLens.Reexport.Data.ProtoLens), âData.ProtoLens.FieldDescriptorâ (imported from Data.ProtoLens.Reexport.Data.ProtoLens) Module âData.ProtoLensâ does not export âMessageDescriptorâ. | 173 | Data.ProtoLens.MessageDescriptor Any hint ?
On Thu, May 17, 2018 at 12:17:29PM +0200, Fabien R wrote:
On 16/05/18 18:54, Vanessa McHale wrote:
`cabal new-build` should only rebuild things once, however. And it's generally better. Ok. I have another error now:
Try `cabal update` and run it again, if that fails file a bug in: https://github.com/tensorflow/haskell I see there are five pull-requests open, three of them older than a year; hopefully the repo has not been google'd.
This has been recently reported as
https://github.com/tensorflow/haskell/issues/180 -- I've done a bit of
analysis there, but it's not obvious what to do about it.
On Thu, May 17, 2018 at 3:36 AM Francesco Ariis
On Thu, May 17, 2018 at 12:17:29PM +0200, Fabien R wrote:
On 16/05/18 18:54, Vanessa McHale wrote:
`cabal new-build` should only rebuild things once, however. And it's generally better. Ok. I have another error now:
Try `cabal update` and run it again, if that fails file a bug in:
https://github.com/tensorflow/haskell
I see there are five pull-requests open, three of them older than a year; hopefully the repo has not been google'd.
_______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.
participants (4)
-
Andrew Pritchard -
Fabien R -
Francesco Ariis -
Vanessa McHale