October 2006 - Hat - Haskell.org

trying to get hat-make to swallow my code
by Tom Cooper 26 Oct '06

26 Oct '06

Hello, I've written a haskell program which is not behaving quite as I'd expect. Up till now, I've used Debug.Trace to understand it, but this has become a major pain, and I was hoping to use hat. When I run hat-make on the code, I get thousands of warnings followed by the following errors: ./Hat/EHM2Alt.hs:316:32: Not in scope: `Hat.Ix.grange' ./Hat/EHM2Alt.hs:331:29: Not in scope: `Hat.Ix.gindex' ./Hat/EHM2Alt.hs:342:29: Not in scope: `Hat.Ix.ginRange' Is this enough of a clue to track down what is happening? I'm not sure what other information would be relevant. I'm using GHC on Windows. One of my modules imports Ix, and several of them import Array. Any help would be greatly appreciated. Thanks. Tom.

2 2

2.05
by Thomas Davie 25 Oct '06

25 Oct '06

Malcolm, I'm trying to build hat with ghc 6.6 - I was under the impression that you'd updated it to be compatible, or am I missing something? At the moment I've modified the make files to get rid of the -package {base|lang|util} arguments, but I still get errors re PackedString (because it's been replaced with ByteString for non unicode, and String for unicode). Am I need of rolling back to 6.5, or is there a quick fix? Bob

4 4

hmake -hat looks for file in /usr/include, not PWD
by David Morse 24 Oct '06

24 Oct '06

"hmake -hat -package GLUT Main" fails with: "Fail: /usr/include/hat/DoublyLinked.hx: openFile: does not exist (No such file or directory)". DoublyLinked.hs is sitting right in the directory from which I'm compiling, I'm not sure why the -hat flag causes this digging in /usr/include. Deleting "-hat" causes successful compilation. The directory tree looks like this: lanes/ ByKey.hs DoublyLinked.hs IfPrime.hs Main.hs MaybeBool.hs Traffic.hs Traffic/ lanes/Traffic/ Data.hs Sim.hs Units.hs This is hat 2.02-12 and hmake-3.09-3 on Debian stable x86. This is the full text of the build: _______________________________________________________________________________ abu ~/sync/lanes hmake -hat -package GLUT Main hat-trans MaybeBool.hs Creating directories Hat Wrote Hat/MaybeBool.hs /usr/bin/haskell-compiler -package GLUT -c -package hat -o Hat/MaybeBool.o Hat/MaybeBool.hs hat-trans Traffic/Units.hs Creating directories Hat Hat/Traffic Wrote Hat/Traffic/Units.hs /usr/bin/haskell-compiler -package GLUT -c -package hat -o Hat/Traffic/Units.o Hat/Traffic/Units.hs hat-trans DoublyLinked.hs Wrote Hat/DoublyLinked.hs /usr/bin/haskell-compiler -package GLUT -c -package hat -o Hat/DoublyLinked.o Hat/DoublyLinked.hs hat-trans ByKey.hs Wrote Hat/ByKey.hs /usr/bin/haskell-compiler -package GLUT -c -package hat -o Hat/ByKey.o Hat/ByKey.hs hat-trans IfPrime.hs Wrote Hat/IfPrime.hs /usr/bin/haskell-compiler -package GLUT -c -package hat -o Hat/IfPrime.o Hat/IfPrime.hs hat-trans Traffic/Sim.hs Fail: /usr/include/hat/DoublyLinked.hx: openFile: does not exist (No such file or directory) _______________________________________________________________________________ in Traffic/Sim.hs the line including DoublyLinked looks like this: "import qualified DoublyLinked as DL"

3 3

Doom!
by Thomas Davie 24 Oct '06

24 Oct '06

Error while trying to get the new darcs version... LappyBob:~/Documents/Work tatd2$ darcs get darcs.haskell.org:/home/ darcs/hat Applying patch 332 of 1401... Unapplicable patch: Thu Jul 5 17:45:57 BST 2001 thorsten * - Hatobserve made primary "observation" tool. hat-observe.c now obsolete. - Heavily improved interactive mode for Hatobserve. - First integration of Hatobserve and hat-detect: Algorithmic Debugging Sessions can immediately be started on an observed equation. darcs failed: Error applying hunk to file ./src/hattools/Hatobserve.hs At a guess - this will be down to not having a case-sensitive file system. Bob

2 1

Interim release: Hat-2.05
by Malcolm Wallace 24 Oct '06

24 Oct '06

Several people have reported that Hat does not compile without errors on ghc-6.4.1, and now that ghc-6.6 is available, the Hat-2.04 release certainly does not compile there either. As an interim measure, I have just made available a 'snapshot' release of Hat-2.05, specifically to ensure that it compiles with ghc-6.6. This is very much a stop-gap. There have been huge changes in the standard base library package since the last full release of Hat, which are not reflected in 2.05. (For example, Parsec and Data.FiniteMap are no longer in base, whilst Data.Map, Data.Set and so on have been added. Hat still has the older API.) The Hat team is planning to rectify these problems shortly, but we are unlikely to be able to make a full release before the end of the year. I hope the 2.05 release is sufficient to meet at least some people's needs in the meantime. Regards, Malcolm

1 0

Committing changes
by Thomas Davie 23 Oct '06

23 Oct '06

I have a few changes to commit to the hat repo, but I'm unsure if I should or not, Malcolm, what state is the repo in, should I commit now, or wait until post-darcs? Bob

3 2

hat-trail: attempt to read beyond end of file
by Barzille 18 Oct '06

18 Oct '06

Hi there, I just started to use HAT. After Installation, which was hell ^^, I tried Tutorial Part 1. After compiling $ hmake -hat Insort the output is >hat-trans Insort.hs >Creating directories Hat >Wrote Hat/Insort.hs >ghc -c -package hat -o Hat/Insort.o Hat/Insort.hs > >Hat/Insort.hs:16:0: > Warning: Pattern match(es) are overlapped > In the definition of `hsort': hsort _ p = ... > >Hat/Insort.hs:30:0: > Warning: Pattern match(es) are overlapped > In the definition of `hinsert': hinsert _ _ p = ... >ghc -package hat -o Insort Hat/Insort.o After that I run: $ ./Insort >agmoprr Thats nice, I think. But when I start hat-trail $ hat-trail Insort the output is: >hat-trail: attempt to read beyond end of file >hat-trail: offset = 0x9012cc02, filesize = 0x2008 >hat-trail: errno = 0 (Success) I don't think that this is nice, is it? Has anybody some hints for me? Thanks in advance -- View this message in context: http://www.nabble.com/hat-trail%3A-attempt-to-read-beyond-end-of-file-tf243… Sent from the Haskell - Hat mailing list archive at Nabble.com.

4 14

Failed to load interface for `Hat.PreludeBasic'
by David Morse 17 Oct '06

17 Oct '06

This appears to be a Hat installation problem. Running the command "hmake -v -hat Test" yields the results: ------------------------------------------------------------------------------------ hat-trans DoublyLinked.hs Wrote Hat/DoublyLinked.hs /usr/bin/haskell-compiler -v -c -package hat -o Hat/DoublyLinked.o Hat/Doub lyLinked.hs Glasgow Haskell Compiler, Version 6.4.1, for Haskell 98, compiled by GHC version 6.4.1 Using package config file: /usr/lib/ghc-6.4.1/package.conf Hsc static flags: -static -funregisterised *** Checking old interface for Hat.DoublyLinked: *** Parser: *** Renamer/typechecker: Hat/DoublyLinked.hs:11:0: Failed to load interface for `Hat.PreludeBasic': Could not find module `Hat.PreludeBasic': locations searched: Hat/PreludeBasic.hi Hat/PreludeBasic.hi-boot *** Deleting temp files Deleting: /tmp/ghc7184.hc Warning: deleting non-existent /tmp/ghc7184.hc ------------------------------------------------------------------------------------- I have explicitly enabled hat via ghc-pkg -l lists "hat" not in parenthesis, so I think the package is enabled. ---------------------------------------------------------------- When apt-getting hat for ubuntu 6.06, note the troubling output of the package manager "ghc-pkg: cannot find package hat" - this was before the hat package was enabled -- could this be the problem? If so, should I more properly ask the package maintainers about this? Setting up hmake (3.10-1.1) ... Setting up hat (2.04-0ubuntu2) ... Setting up hat-ghc6 (2.04-0ubuntu2) ... ghc-pkg: cannot find package hat Reading package info from stdin ... done. Saving old package config file... done. Writing new package config file... done. ------------------------------------------------------------------------------------ Possibly related: "[Hat] hat installation problem" http://www.haskell.org/pipermail/hat/2006-February/000255.html ------------------------------------------------------------------------------------- Any idea what the problem could be? CC me in the reply please, I don't subscribe.

2 1

using Hat with unsafeCoerce
by roconnor＠theorem.ca 16 Oct '06

16 Oct '06

I have a program that is standalone expect for a dependency on unsafeCoerce. Is it possible to use Hat to trace through this program? -- Russell O'Connor <http://r6.ca/> ``All talk about `theft,''' the general counsel of the American Graphophone Company wrote, ``is the merest claptrap, for there exists no property in ideas musical, literary or artistic, except as defined by statute.''

2 1

Using Hat with HUnit
by Katerina Barone-Adesi 11 Oct '06

11 Oct '06

I'm trying to run hat on some unit tests. They use HUnit-1.0, which I have under the tests directory, in tests/HUnit-1.0/; I've been interpreting the tests using runhaskell. When I run hmake -package parsec -i.. -iHUnit-1.0 RunTests; ./RunTests, everything works exactly as it does when I'm interpreting. When I add a -hat to the options, I get the following error: hat-trans -i.. -iHUnit-1.0 HUnit-1.0/HUnitLang.lhs Wrote Hat/HUnit-1.0/HUnitLang.hs /usr/bin/ghc -package parsec -i.. -iHUnit-1.0 -c -package hat -o HUnit-1.0/Hat/HUnitLang.o HUnit-1.0/Hat/HUnitLang.hs ghc-6.4.2: error: directory portion of "HUnit-1.0/Hat/HUnitLang.o" does not exist (used with "-o" option.) If I manually create HUnit-1.0/Hat: hat-trans -i.. -iHUnit-1.0 HUnit-1.0/HUnitLang.lhs Wrote Hat/HUnit-1.0/HUnitLang.hs /usr/bin/ghc -package parsec -i.. -iHUnit-1.0 -c -package hat -o HUnit-1.0/Hat/HUnitLang.o HUnit-1.0/Hat/HUnitLang.hs ghc-6.4.2: file `HUnit-1.0/Hat/HUnitLang.hs' does not exist If I switch into the HUnit-1.0 directory and run "hmake -hat HUnitLang.lhs": hmake -hat HUnitLang.lhs hat-trans HUnitLang.lhs Wrote Hat/HUnitLang.hs /usr/bin/ghc -c -package hat -o Hat/HUnitLang.o Hat/HUnitLang.hs Hat/HUnitLang.hs:52:38: Could not find module `Hat.PreludeBuiltin': use -v to see a list of the files searched for In the fourth argument of `Hat.Hat.uapp1', namely `aioError' In the definition of `hassertFailure': hassertFailure fmsg p = Hat.Hat.uapp1 p59v23v59v60 p59v23v59v29 p aioError hioError (Hat.Hat.uapp1 p59v32v59v60 p59v32v59v40 p auserError huserError (Hat.Hat.uapp2 p59v43v59v60 p59v55v59v56 p (+++) (*++) (ghunitPrefix p59v43v59v53 p) fmsg)) 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? Katerina Barone-Adesi

2 1