January 2005 - Haskell-Cafe

what is a stack overflow?
by S. Alexander Jacobson 24 Jan '05

24 Jan '05

GHC assumes the user knows the difference between the heap and the stack. I don't. No matter how much heap I specify on the GHCi command line, I get a stack overflow exception. I have no idea what that means or how to remedy it. Hints? Note: My program is basically creating a few 100k item FiniteMaps. I don't think that should exceed the memory on my laptop.... -Alex- ______________________________________________________________ S. Alexander Jacobson tel:917-770-6565 http://alexjacobson.com

1 0

File path programme
by Mark Carroll 24 Jan '05

24 Jan '05

I tried writing a little command-line utility to find the relative path of one thing from another thing (with Unix-like systems in mind). For example, $ ./pathfromof /etc/init.d/ /etc/X11/XF86Config-4 ../X11/XF86Config-4 $ ./pathfromof /tmp/baz/ /tmp/foo/ . $ ls -l /tmp/baz lrwxr-xr-x 1 markc markc 8 2005-01-20 12:01 /tmp/baz -> /tmp/foo It turned out surprisingly complex, though, and doesn't feel very neat or tidy at all, nor is it very portable given that I couldn't find generic library functions for manipulating bits of filepaths. Anyhow, it's at http://www.chiark.greenend.org.uk/~markc/PathFromOf.hs and may yet have egregious bugs. It seems to me like it could certainly be improved in various ways. If anyone has any thoughts, as to how I could improve my style, make more use of standard libraries, etc., I'd certainly appreciate them. Thanks, Mark

7 7

HaXmL and the IO Monad
by John Goerzen 24 Jan '05

24 Jan '05

I've been playing with HaXmL lately. I've had a need to convert one XML document to another, and HaXmL was very nice for that. Along the way, I've discovered that I need to do some I/O as part of the conversion (specifically due to timezone-related calculations). Much of HaXML is based around the CFilter type: type CFilter = Content -> [Content] That is, most of the stuff one does in HaXML is a function with that type. Try as I might, I could not figure out a nice way to integrate I/O into this system. A simple IO CFilter wouldn't work, since there has to be input available when it runs. Content -> IO [Content] didn't work either, since the IO ripples up all the way to the top, messing up the nice code everywhere. I finally resorted to unsafePerformIO to get it done. But I'd like to know the proper solution. I do hope that this is not it :-) Thanks, John

4 3

Re: [Haskell-cafe] weird behavior using HUnit
by Keith Wansbrough 24 Jan '05

24 Jan '05

> > GHC's memory profiling? > > I just gave it a try: when compiled with -prof -auto-all, my program > is memory hungry in both cases, so I cannot really compare... [cc'd back to list in case anyone else finds this useful; hope that's OK] Ah. The cost-centre annotations get in the way of the optimiser, preventing certain optimisations from being performed. It's possible that the new entry point you introduce with testParseLog is similarly preventing something from being inlined. The way to tell (sadly!) is to look at the intermediate code being generated by GHC. In fact there are several intermediate codes to look at. Most of the optimisation work is done while the program is in a form called "Core"; if you look at the result of the optimiser (sometimes called the "simplifier", confusingly) just before this Core is translated to STG you will get the most readable and useful information. http://www.haskell.org/ghc/docs/latest/html/users_guide/options-debugging.h… tells you how, and also explains how to read the output. -ddump-simpl is probably what you want, but it's a while since I've had a look - I could be wrong. Compare what you get with and without that HUnit line that's causing trouble, and see how the critical function is affected. HTH. --KW 8-)

1 0

weird behavior using HUnit
by Frédéric Gobry 24 Jan '05

24 Jan '05

Hello, I just experienced a weird problem while using HUnit. My code (a simple apache log analyser) used to run in constant space (thanks to the good suggestions and examples I received from the list :-)), and I wanted to add some unit tests. To my great surprise, the code started to eat up lots of memory again. I've isolated the problem down. It seems that if I declare my test suite with the line: testParseLog = HUnit.runTestTT (HUnit.TestList [_testUpdate]) then my code does not run in constant space anymore. Note that the testParseLog is not actually used, but only exported from the module, so that I can run it with ghc -e testParseLog. The code is attached. The line in question is at the end of ParseLog.hs, and I compiled the whole with ghc-6.2.2 (debian testing) Any hint ? thanks in advance, Frédéric -- Frédéric Gobry Infoscience DIT-KIS / EPFL Tel: +41216932288 http://people.epfl.ch/frederic.gobry

1 1

Randomly balanced binary trees
by Dominic Fox 23 Jan '05

23 Jan '05

Some more code that might be of interest to someone. Based on a suggestion by Alistair Turnbull, I've written a Haskell implementation of a randomly balanced binary tree. This merges tree updates with a stream of random numbers, which are used to shuffle the order of nodes in the tree so that it remains balanced. The code is posted together with a short explanation here: http://codepoetics.com/poetix/index.php?p=72 I would be especially interested in any suggestions relating to optimisation. Dominic

1 0

can't 'make install-docs'
by Daniel Fischer 21 Jan '05

21 Jan '05

Hi, I've just installed ghc-6.2.2, but couldn't get the docs made. I got the following: Text/Regex.raw-hs Text/Regex/Posix.raw-hs Text/Show.raw-hs Text/Show/Functions.raw-hs \ --package=base \ --dump-interface=base.haddock \ --use-index=../doc-index.html --use-contents=../index.html \ /bin/sh: line 1: t: command not found gmake[2]: [html/haddock.css] Error 127 (ignored) /usr/bin/install -c -m 644 html/* /usr/local/share/ghc-6.2.2/html/libraries/base /usr/bin/install: cannot stat `html/*': No such file or directory gmake[2]: *** [install-docs] Error 1 gmake[1]: *** [install-docs] Error 1 gmake[1]: Leaving directory `/usr/lib/ghc-6.2.2/libraries' gmake: *** [install-docs] Error 1 What can I do about it? helpless, Daniel

1 0

RE: [Haskell-cafe] Can't do basic time operations with System.Time
by Simon Marlow 21 Jan '05

21 Jan '05

On 20 January 2005 22:12, John Goerzen wrote: > I have a simple desire. I have a string that I need to parse as a > date/time string in the local timezone, then convert it to standard > seconds-since-epoch format. This is trivial in C, Perl, Python, etc. > but seems impossible to do reliably in Haskell. I'm hoping someone > can tell me where I've gone wrong. > > My initial idea was to parse the string, generate a CalendarTime > object, use toClockTime to convert it to seconds-since-epoch, and > then grab the value out of that. There were some problems, though: > what to put in ctTZ and ctIsDST? I initially loaded them with 0 and > False, though that turned out to be clearly wrong. > > My next idea was to take the ctTZ result from toCalendarTime, then use > that for all the CalendarTime objects I'd generate. That was wrong, > too, during daylight savings time. I don't think I should have to > re-implement the algorithms for determining whether a given date is > under DST worldwide. > > What I really need is just a wrapper around mktime() that just calls > mktime and doesn't do anything else. All the post-processing that > System.Time is doing is what's causing me trouble. I'd just like it > to ignore ctTZ, ctTZName, and ctIsDST. That would yield behavior > identical to the C library. If you set ctTZ to the UTC offset outside DST, and set ctIsDST to False, then toClockTime will do the DST calculation for you. ctTZ is the actual UTC offset (ie. it is adjusted for DST). Hence, toClockTime ignores ctIsDST when performing the conversion: it isn't necessary. I can't tell what's wrong here, if anything (general brokenness in the System.Time API notwithstanding). Ah *lightbuilb* - I think I see: you want to take a calendar time without any TZ information, assume it is a time at the current geographical location, and produce a CalendarTime for that. Hmm, it looks like the current API doesn't support that - toClockTime assumes you know the timezone. That's a shame. There's one way that's almost correct: take your initial calendar time, set ctTZ to your non-DST UTC offset, then call toClockTime and toCalendarTime on the result, and you'll get an appropriate ctTZ for that time. Plug the new ctTZ into the original time, and toClockTime again (you can shortcut this by adjusting the ClockTime directly). I'll leave it up to you how to handle the boundaries properly :-) Cheers, Simon

4 4

Re: [Haskell-cafe] Re: Hugsvs GHC (again)was: Re: Somerandomnewbiequestions
by Udo Stenzel 21 Jan '05

21 Jan '05

Ben Rudiak-Gould <Benjamin.Rudiak-Gould(a)cl.cam.ac.uk> schrieb am 21.01.05 02:37:34: > If you're reading from a random-access file, there's no way [select] can tell > you when the file data is buffered, because it doesn't know which part > of the file you plan to read. The OS may try to guess for readahead > purposes, but select()'s behavior can't depend on that guess. The Glibc documentation says, "select determines if there is data available (more precisely, if a call to read(2) will not block)." I think, this is reasonably precise. The OS does know, where you are going to read (at the file pointer) and if you seek() or pread() instead, well, that is no call to read(2) and may change everything. Thus the question is, does select() reliably tell if read() would block or does it check for something else? Is the documentation wrong (on some platforms)? > This is another example of why the world would be better off with the > file/stream model. Have I convinced anyone? Uhm, well, me thinks, open() and read() on files quite accurately model a stream already... Udo. __________________________________________________________ Mit WEB.DE FreePhone mit hoechster Qualitaet ab 0 Ct./Min. weltweit telefonieren! http://freephone.web.de/?mc=021201

3 2

RE: [Haskell-cafe] Re: Hugsvs GHC (again)was:Re: Somerandomnewbiequestions
by Simon Marlow 21 Jan '05

21 Jan '05

On 21 January 2005 16:27, Keean Schupke wrote: > Having read around I have found that select does return readable for > all file IO on a block device... > > I wonder if ghc could use non-blocking mode (files opened with the > O_NONBLOCK) flag? GHC does use the O_NONBLOCK flag. Cheers, Simon

1 0