Re: Using Hat with HUnit
Malcolm Wallace <[email protected]> Wed, 11 Oct 2006 16:00:34 +0100
| Newsgroups | gmane.comp.lang.haskell.hat |
|---|---|
| Organization | Dept of Computer Science, University of York |
| Message-ID | <[email protected]> |
"Katerina Barone-Adesi" <[email protected]> wrote: > hat-trans -i.. -iHUnit-1.0 HUnit-1.0/HUnitLang.lhs > Wrote Hat/HUnit-1.0/HUnitLang.hs > /usr/bin/ghc ... > HUnit-1.0/Hat/HUnitLang.hs As I think you can see, hat-trans and hmake are disagreeing about where the Hat-translated versions of the source-code should live. hat-trans thinks Hat/HUnit-1.0/HUnitLang.hs whilst hmake thinks HUnit-1.0/Hat/HUnitLang.hs (the inital portion is swapped around). The main reason for this is a differing interpretation of the -i flag (giving a directory where extra source files live). One workaround is to move all the HUnit code to the current directory, rather than a separate directory. Another workaround is to change the name of the HUnit directory, so it no longer begins with upper-case, e.g. hunit-1.0. I know this is somewhat fragile, but the only way some tools know how to distinguish between the filename of a simple module, and the pathname of a hierarchical module, is on the case of the first letter of the directory part. > I'd like to be able to trace my code; I'm indifferent to whether or > not I trace the code within HUnit. Any suggestions on how I can do > this? I hope one of these suggestions is useful. If not, please post again! Regards, Malcolm