On Thu, Sep 16, 2010 at 2:51 PM, Anthony Molinaro <[email protected]> wrote: > Hi Christian, > > I would say however you would like to manage the backlog is fine. I'm
I meant the existing issues that have patches that have been pending for months: https://issues.apache.org/jira/browse/THRIFT-652 https://issues.apache.org/jira/browse/THRIFT-743 https://issues.apache.org/jira/browse/THRIFT-787 I have the time and incentive. Anything I can help with? > not really at all familiar with haskell, but if you can find at least > one other thrift person (probably by asking on the users list, or maybe > one of the other commits knows enough) to try out your patches I think > we'd probably be able to commit them. > > If you can manage to integrate the building and testing of haskell > in the make system we might even be able to elevate it to a language > which is automatically built and tested. Anything other than patching autotools and the usual suspects? > So please feel free to create new issues and split your changes up, > or just reuse current ticket. THRIFT-901 created for the bits not redundant. I'll just try the existing patches for where it was redundant and comment on the appropriate issues as "tried it, it works, please commit" as appropriate. > Thanks, > > -Anthony > > On Tue, Sep 14, 2010 at 11:33:30PM -0400, Christian Lavoie wrote: >> Ah, had never heard of JIRA. >> >> Looks like my patch is partially redundant with stuff that's on JIRA >> (THRIFT-743; with the first 2 items on my list) but does a bit more. >> Should I go ahead and post with the redundancy as is? With the >> redundancy removed? >> >> Anything I can do to help with the 'backlog' of patches in the haskell >> bindings (a whole 3 patches, all of which I have time to go run >> through and test, if needed)? >> >> On Tue, Sep 14, 2010 at 11:22 PM, Mark Slee <[email protected]> wrote: >> > Hi Christian, >> > >> > Don't think your patch actually came through as an attachment. Can you >> > post your patch to our JIRA page? That's the right place to document it >> > and it'll let others comment. >> > >> > https://issues.apache.org/jira/browse/THRIFT >> > >> > Cheers, >> > Mark >> > >> > -----Original Message----- >> > From: Christian Lavoie [mailto:[email protected]] >> > Sent: Tuesday, September 14, 2010 6:28 PM >> > To: [email protected] >> > Subject: [PATCH] Make haskell bindings work >> > >> > The following patch generally makes the haskell bindings compile with >> > ghc-6.12.3, without -fglasgow-exts, with -Wall -Werror, with -XArrows; >> > though I haven't actually tested this very thoroughly yet, it does >> > compile and work with a trivial IDL file. >> > >> > In particular, it: >> > * Adds 'length' to the list of imported symbols from the Prelude in >> > generated code (not sure how the generated bindings ever compiled >> > without that) >> > * Stops using 'rec' and 'proc' as identifiers, they're keywords when >> > using -XArrows >> > * Adds needed LANGUAGE pragmas (in order to not need -fglasgow-exts) >> > * Adds OPTIONS_GHC pragmas needed to bypass -Wall -Werror >> > * Cleans up a few style bugs / nitpicks in order to compile with >> > -Wall when it was trivial (and avoided more OPTIONS_GHC pragmas) >> > >> > Thoughts? >> > >> > -- >> > Have fun, Christian >> > http://linkedin.christianlavoie.net >> > >> > "I won't let you fall apart." >> > >> >> >> >> -- >> Have fun, Christian >> http://linkedin.christianlavoie.net >> >> "I won't let you fall apart." > > -- > ------------------------------------------------------------------------ > Anthony Molinaro <[email protected]> > -- Have fun, Christian http://linkedin.christianlavoie.net "I won't let you fall apart."
