hugodaniel / timerep

Reading and parsing time in Haskell

timerep

This started out as a simple project to parse and display time according to RFC 3339, but with time gained ability to convert to and from other RFCs as well.

You will find a change log in CHANGES.md.

Todos

• Get rid of the formatTime function, and use a more decent pretty printing lib.
• Do some proper parsing (it parses nicely, but this was just a quick hack that my mind produced while being stressed to finish this).

License

This project is licensed under a 3-clause BSD license as specified in the LICENSE file.

timerep's Issues

Bulding on GHC 7.2

I was wondering if you could address the build error on 7.2. It is preventing a project I'm working on from building on hackage. It looks like you may just need to add a FlexibleInstances pragma:

http://hackage.haskell.org/packages/archive/rfc3339/1.0.4/logs/failure/ghc-7.2

`formatTimeRFC2822` does not parse comments

The parser is not capable of parsing the example7 from the docs:

λ> parseTimeRFC2822 @String "Thu,          13\n     Feb\n  1969\n        23:32\n     -0330 (Newfoundland Time)"
Nothing

A simpler example with less whitespace:

λ> parseTimeRFC2822 @String "Fri, 03 Dec 2010 16:02:30 -0600 (CST)"
Nothing

Move all tests into the test suite

The testParse function and other test-related functions in the primary source of the package could go to the test suite. Maybe that also justifies to base the test suite on something like tasty.

Mark rfc3999 deprecated on Hackage

Currently the rfc3999 package links to git://github.com/HugoDaniel/RFC3339.git, which redirects to this repo. Until the user tries to resolve that git repo though it appears to them that there are two active rfc3999 libraries on Hackage, rfc3999 and timerep. Perhaps rfc3999 could be marked as deprecated, and its description updated with a link to timerep?

timerep-2.0.0 test suite crashes GHC 7.10.1

Citing from http://hydra.cryp.to/build/795204/nixlog/6/raw:

In-place registering Tests-2.0.0...
[1 of 1] Compiling Main             ( dist/build/TestsStub/TestsStub-tmp/TestsStub.hs, dist/build/TestsStub/TestsStub-tmp/Main.dyn_o )
Linking dist/build/TestsStub/TestsStub ...
ghc: panic! (the 'impossible' happened)
  (GHC version 7.10.1 for x86_64-unknown-linux):
        Don't understand library name Tests

This is a known bug in GHC. Maybe it's possible to work around it by simply renaming the test suite?

Please consider registering this package to Stackage

Stackage a set of Haskell packages that is known to compile together. I'd like to use timerep as a dependency on htoml, and register it on Stackage; but I cannot do both when your package is not already registered.

Thanks for considering!

timerep fails to build with time-1.11

$ stack --resolver nightly build timerep
timerep> stack-83dd043523002f44/timerep-2.0.1.0/Data/Time/RFC2822.hs:74:24: error:
timerep>     • Variable not in scope:
timerep>         parseTime :: TimeLocale -> String -> String -> Maybe ZonedTime
timerep>     • Perhaps you meant ‘parseTimeM’ (imported from Data.Time.Format)
timerep>    |
timerep> 74 |         parse format = parseTime defaultTimeLocale (toString' format) t'
timerep>    |                        ^^^^^^^^^

Cabal Install error with GHC 7.4.2 on Mac 10.7

Hi Hugo

I got the error message when installing,

Resolving dependencies...
Configuring rfc3339-1.0.4...
Preprocessing library rfc3339-1.0.4...
Building rfc3339-1.0.4...
[1 of 1] Compiling Data.Time.RFC3339 ( Data/Time/RFC3339.hs, dist/build/Data/Time/RFC3339.o )

Data/Time/RFC3339.hs:60:10:
Illegal instance declaration for RFC3339 String' (All instance types must be of the form (T a1 ... an) where a1 ... an are *distinct type variables*, and each type variable appears at most once in the instance head. Use -XFlexibleInstances if you want to disable this.) In the instance declaration forRFC3339 String'
cabal: Error: some packages failed to install:
rfc3339-1.0.4 failed during the building phase. The exception was:
ExitFailure 1

I have confirm " {-# LANGUAGE TypeSynonymInstances, FlexibleInstances #-}" in the RFC3339.hs, it's no wired...

How can i solve it?

Consider dropping GHC language pragma's from the cabal file

As they are explicitly included from the tops of the files they are needed in.

Failure on GHC 7.10

http://hackage.haskell.org/package/timerep points to this repo, so I'm reporting here.

[1 of 2] Compiling Data.Time.RFC2822 ( Data/Time/RFC2822.hs, dist/dist-sandbox-c7fca7f4/build/Data/Time/RFC2822.o )

Data/Time/RFC2822.hs:75:16:
    Ambiguous occurrence ‘defaultTimeLocale’
    It could refer to either ‘Data.Time.Format.defaultTimeLocale’,
                             imported from ‘Data.Time.Format’ at Data/Time/RFC2822.hs:39:1-23
                             (and originally defined in ‘time-1.5.0.1:Data.Time.Format.Locale’)
                          or ‘System.Locale.defaultTimeLocale’,
                             imported from ‘System.Locale’ at Data/Time/RFC2822.hs:43:1-20

Data/Time/RFC2822.hs:96:25:
    Ambiguous occurrence ‘defaultTimeLocale’
    It could refer to either ‘Data.Time.Format.defaultTimeLocale’,
                             imported from ‘Data.Time.Format’ at Data/Time/RFC2822.hs:39:1-23
                             (and originally defined in ‘time-1.5.0.1:Data.Time.Format.Locale’)
                          or ‘System.Locale.defaultTimeLocale’,
                             imported from ‘System.Locale’ at Data/Time/RFC2822.hs:43:1-20

Change the repo name to match the package name

Don't forget to update the cabal file. :)