Closed ysangkok closed 6 months ago
With the upper bound of attoparsec changed in the downloaded revision, the test suite does pass on GHC 8.10:
janus@gorm ~/flipstone/ftp-client-0.5.1.4
% cabal test --constraint='attoparsec>=0.14'
Resolving dependencies...
Build profile: -w ghc-8.10.7 -O1
In order, the following will be built (use -v for more details):
- ftp-client-0.5.1.4 (test:ftp-client-test) (first run)
Preprocessing test suite 'ftp-client-test' for ftp-client-0.5.1.4..
Building test suite 'ftp-client-test' for ftp-client-0.5.1.4..
Running 1 test suites...
Test suite ftp-client-test: RUNNING...
Test suite ftp-client-test: PASS
Test suite logged to:
/home/janus/flipstone/ftp-client-0.5.1.4/dist-newstyle/build/x86_64-linux/ghc-8.10.7/ftp-client-0.5.1.4/t/ftp-client-test/test/ftp-client-0.5.1.4-ftp-client-test.log
1 of 1 test suites (1 of 1 test cases) passed.
As a Hackage trustee, I have published a revision that bumps the upper bound on attoparsec
as requested in haskell-infra/hackage-trustees#325:
https://hackage.haskell.org/package/ftp-client-0.5.1.4/revisions/
Closing this since I took over the package.
Even after the revision bump allowing bytestring v0.11, the project doesn't compile on GHC 9.2 because of an additional constraint on attoparsec, even with the cabal.project for Cryptonite, Memory and Foundation.
By changing the upper bound of attoparsec to allow 0.14 (e.g. have it be
<0.15
, it compiles:(
testing is a separate issue which I will file once tasty-hspec allows GHC 9.2EDIT: As of Feb 20 2022, I have tested that the test suite passes with #24 andcabal test -w ghc-9.2.1 --constraint='attoparsec>=0.14'
)