Thanks everyone who has helped test so far! We uncovered a few last issues and things are looking great now!
koeien update the buildbot to 6.12.1 and we are now automatically building happstack against 6.8.3, 6.10.2, and 6.12.1. I fixed a bug in a test case and disabled the bogus XML test cases, and we are seeing green lights across the board!
On Fri, Dec 18, 2009 at 8:30 PM, stepcut <jer...@n-heptane.com> wrote: > Hello,
> Thanks everyone who has helped test so far! We uncovered a few last > issues and things are looking great now!
> koeien update the buildbot to 6.12.1 and we are now automatically > building happstack against 6.8.3, 6.10.2, and 6.12.1. I fixed a bug in > a test case and disabled the bogus XML test cases, and we are seeing > green lights across the board!
> It would be great to know if the latest sendfile and happstack from > darcs build correctly on Windows, and pass the test suites.
> As soon as we know FreeBSD sendfile is fixed, and that Windows is > working, we can release 0.4 and move on to 0.5!
> Thanks everyone for your help so far! > - jeremy
Thanks for keeping Happstack moving forward!
This is what I get on my virtual machine -- it breaks on happstack-server. It looks like "Handler.hs" imports System.Posix.IO, which is not available on Windows. Or am I missing some option?
==================== [...] src\Happstack\Server\HTTP\Handler.hs:39:7: Could not find module `System.Posix.IO': Use -v to see a list of the files searched for. cabal.exe: Error: some packages failed to install: happstack-server-0.4 failed during the building phase. The exception was: exit: ExitFailure 1 ====================
> On Fri, Dec 18, 2009 at 8:30 PM, stepcut <jer...@n-heptane.com> wrote: > > Hello,
> > Thanks everyone who has helped test so far! We uncovered a few last > > issues and things are looking great now!
> > koeien update the buildbot to 6.12.1 and we are now automatically > > building happstack against 6.8.3, 6.10.2, and 6.12.1. I fixed a bug in > > a test case and disabled the bogus XML test cases, and we are seeing > > green lights across the board!
> > It would be great to know if the latest sendfile and happstack from > > darcs build correctly on Windows, and pass the test suites.
> > As soon as we know FreeBSD sendfile is fixed, and that Windows is > > working, we can release 0.4 and move on to 0.5!
> > Thanks everyone for your help so far! > > - jeremy
> Thanks for keeping Happstack moving forward!
> This is what I get on my virtual machine -- it breaks on > happstack-server. It looks like "Handler.hs" imports System.Posix.IO, > which is not available on Windows. Or am I missing some option?
> ==================== > [...] > src\Happstack\Server\HTTP\Handler.hs:39:7: > Could not find module `System.Posix.IO': > Use -v to see a list of the files searched for. > cabal.exe: Error: some packages failed to install: > happstack-server-0.4 failed during the building phase. The exception was: > exit: ExitFailure 1 > ====================
On Fri, Dec 18, 2009 at 9:46 PM, stepcut <jer...@n-heptane.com> wrote: > Oops! Thanks for finding that! It is a bogus import that never should > have been committed.
> Fixed in head. Let me know if it works now.
> thanks again, > - jeremy
Yep, that did it! I tested a simple "hello world" program and it works. Is there a more serious way of testing it?
On Sat, Dec 19, 2009 at 1:12 AM, Paulo Tanimoto <ptanim...@gmail.com> wrote: > On Fri, Dec 18, 2009 at 9:46 PM, stepcut <jer...@n-heptane.com> wrote: >> Oops! Thanks for finding that! It is a bogus import that never should >> have been committed.
>> Fixed in head. Let me know if it works now.
>> thanks again, >> - jeremy
> Yep, that did it! I tested a simple "hello world" program and it > works. Is there a more serious way of testing it?
> Paulo
By the way, this is on GHC 6.10.4 (Haskell Platform) if you want to update the Wiki. I can give GHC 6.12.1 a try this weekend.
> Thanks everyone who has helped test so far! We uncovered a few last > issues and things are looking great now!
> koeien update the buildbot to 6.12.1 and we are now automatically > building happstack against 6.8.3, 6.10.2, and 6.12.1. I fixed a bug in > a test case and disabled the bogus XML test cases, and we are seeing > green lights across the board!
> It would be great to know if the latest sendfile and happstack from > darcs build correctly on Windows, and pass the test suites.
> As soon as we know FreeBSD sendfile is fixed, and that Windows is > working, we can release 0.4 and move on to 0.5!
> Thanks everyone for your help so far! > - jeremy
> --
> You received this message because you are subscribed to the Google > Groups "HAppS" group. > To post to this group, send email to happs@googlegroups.com. > To unsubscribe from this group, send email to happs+unsubscribe@googlegroups.com > . > For more options, visit this group at http://groups.google.com/group/happs?hl=en > .
On Sat, Dec 19, 2009 at 12:48 PM, Paulo Tanimoto <ptanim...@gmail.com> wrote: > Hi Jeremy,
> On Sat, Dec 19, 2009 at 12:26 PM, Jeremy Shaw <jer...@n-heptane.com> wrote: >> Hello,
>> Someone reported a build failure for happstack 0.3 on a 64-bit OS. Has >> anyone tried building Happstack 0.4 on a 64-bit OS?
>> thanks! >> - jeremy
> It works for me. I'm on Ubuntu 9.10 64-bit with GHC 6.10.4.
> Paulo
Still on Ubuntu 64-bit, now with GHC 6.12.1 (RC) after installing syb-with-class 0.6.1, I get the following error. It's not related to happstack, so it surprises me because you've got it to work with GHC 6.12.1 before, right?
src/HJScript/Lang.hs:30:4: parse error on input `rec' cabal: Error: some packages failed to install: HJScript-0.4.4 failed during the building phase. The exception was: ExitFailure 1 happstack-0.4 depends on HJScript-0.4.4 which failed to install. hsp-0.4.5 depends on HJScript-0.4.4 which failed to install. =======================
> Someone reported a build failure for happstack 0.3 on a 64-bit OS. Has > anyone tried building Happstack 0.4 on a 64-bit OS?
> thanks! > - jeremy
> On Dec 18, 2009, at 8:30 PM, stepcut wrote:
>> Hello,
>> Thanks everyone who has helped test so far! We uncovered a few last >> issues and things are looking great now!
>> koeien update the buildbot to 6.12.1 and we are now automatically >> building happstack against 6.8.3, 6.10.2, and 6.12.1. I fixed a bug in >> a test case and disabled the bogus XML test cases, and we are seeing >> green lights across the board!
>> It would be great to know if the latest sendfile and happstack from >> darcs build correctly on Windows, and pass the test suites.
>> As soon as we know FreeBSD sendfile is fixed, and that Windows is >> working, we can release 0.4 and move on to 0.5!
>> Thanks everyone for your help so far! >> - jeremy
>> --
>> You received this message because you are subscribed to the Google >> Groups "HAppS" group. >> To post to this group, send email to happs@googlegroups.com. >> To unsubscribe from this group, send email to happs+unsubscribe@googlegroups.com >> . >> For more options, visit this group at http://groups.google.com/group/happs?hl=en >> .
> --
> You received this message because you are subscribed to the Google Groups "HAppS" group. > To post to this group, send email to happs@googlegroups.com. > To unsubscribe from this group, send email to happs+unsubscribe@googlegroups.com. > For more options, visit this group at http://groups.google.com/group/happs?hl=en.
On Sat, Dec 19, 2009 at 1:12 AM, Paulo Tanimoto <ptanim...@gmail.com> wrote: > On Fri, Dec 18, 2009 at 9:46 PM, stepcut <jer...@n-heptane.com> wrote: >> Oops! Thanks for finding that! It is a bogus import that never should >> have been committed.
>> Fixed in head. Let me know if it works now.
>> thanks again, >> - jeremy
> Yep, that did it! I tested a simple "hello world" program and it > works. Is there a more serious way of testing it?
> Paulo
Well, now that I see there's a suite of tests, I ran:
On Dec 19, 2009, at 1:08 PM, Paulo Tanimoto wrote:
> Still on Ubuntu 64-bit, now with GHC 6.12.1 (RC) after installing > syb-with-class 0.6.1, I get the following error. It's not related to > happstack, so it surprises me because you've got it to work with GHC > 6.12.1 before, right?
koeien got it working in the buildbot. But he had to patch HJScript. He sent a patch upstream. No idea how long it will take for that to make it onto hackage..
Perhaps we should have koeien put the patch somewhere publicly accessible in the meantime?
> Hello. I don't think there are really any errors. I think it should > read:
>> Cases: 75 Tried: 75 Errors: 0 Failures: 0
> the 7099 stuff is just issues with runTestTT not refreshing the > display properly.
> If there were actual errors it would print out a line with a > descriptive identifier for the error. Something like:
> happstack:happstack-data:xml001:testname
Ah, I see what you mean. : )
It looks like we're good, especially with 6.10. If I can get koeien's patch, I can try it with GHC 6.12 on Windows. I say "try" because I don't have it installed there yet.
On Sat, Dec 19, 2009 at 01:30:01PM -0600, Jeremy Shaw wrote: > koeien got it working in the buildbot. But he had to patch HJScript. > He sent a patch upstream. No idea how long it will take for that to > make it onto hackage..
> Perhaps we should have koeien put the patch somewhere publicly > accessible in the meantime?
Perhaps Hackage should have an NMU policy? (Especially for fixes targeted to working with new systems while keeping compatibility with old ones.)
C:\>happstack-state-tests ### Failure in: happstack-state:3:saverProperties:2:File:1:prop_seqReadWrite Test failed with exception: DeleteFile: permission denied (The process cannot ac cess the file because it is being used by another process.) ### Failure in: happstack-state:3:saverProperties:2:File:2:prop_cutDrop Test failed with exception: DeleteFile: permission denied (The process cannot ac cess the file because it is being used by another process.) ### Failure in: happstack-state:3:saverProperties:3:Queue File:1:prop_seqReadWri te Test failed with exception: <<loop>> Cases: 25 Tried: 25 Errors: 0 Failures: 3 99 C:\>happstack-ixset-tests Cases: 1 Tried: 1 Errors: 0 Failures: 0
C:\>happstack-tests ### Failure in: happstack:3:happstack-state:3:saverProperties:2:File:1:prop_seqR eadWrite Test failed with exception: DeleteFile: permission denied (The process cannot ac cess the file because it is being used by another process.) ### Failure in: happstack:3:happstack-state:3:saverProperties:3:Queue File:1:pro p_seqReadWrite Test failed with exception: <<loop>> Cases: 75 Tried: 75 Errors: 0 Failures: 2 99 C:\>
On Dec 19, 2009, at 12:54 PM, Michael Steele wrote:
>> It would be great to know if the latest sendfile and happstack from >> darcs build correctly on Windows, and pass the test suites.
> I tried installing it on Windows XP 32bit with Haskell Platform > 2009.2.0.2 and MinGW 5.1.6.
> Happstack built okay, but it looks like some tests failed:
Very weird. Looks like some odd failure in the unit tests themselves. I think the DeleteFile error is coming from the removeDirectoryRecursive call. If you can figure out what went wrong, we can see if we need to fix something. I don't think this is an actual bug in happstack, so I am going to make the release anyway. If it turns out I am wrong, I'll just have to release 0.4.2 ;)
> Very weird. Looks like some odd failure in the unit tests themselves. > I think the DeleteFile error is coming from the > removeDirectoryRecursive call. If you can figure out what went wrong, > we can see if we need to fix something. I don't think this is an > actual bug in happstack, so I am going to make the release anyway. If > it turns out I am wrong, I'll just have to release 0.4.2 ;)
> - jeremy
After installing from hackage I ran the tests again. This time everything passed.
>> Very weird. Looks like some odd failure in the unit tests themselves. >> I think the DeleteFile error is coming from the >> removeDirectoryRecursive call. If you can figure out what went wrong, >> we can see if we need to fix something. I don't think this is an >> actual bug in happstack, so I am going to make the release anyway. If >> it turns out I am wrong, I'll just have to release 0.4.2 ;)
>> - jeremy
> After installing from hackage I ran the tests again. This time > everything passed.
