#perl
23 October 2007
Total 38 pages. You are browsing page 3/38.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last
02:04 <****> ekim|irc-: that's the most recent stable version
02:04 <****> perl6 isn't out
02:04 <****> sorry
02:04 <****> is there any way in a foreach loop to know that you are on the last?
02:04 <****> the trunk is
02:04 <****> Somni: He didn't say it was, merely that he knew it was. ;)
02:04 <****> i just compiled 5.8.8 on solaris 9 not too long ago
02:04 <****> prakriti: iterate over indices
02:04 <****> prakriti: nothing canonical
02:04 <****> ekim|irc-: what trunk?
02:04 <****> fair enough
02:05 <****> tyvm
02:05 <****> DSterling: so like XML::LibXML?
02:05 <****> pravus: indeed
02:05 <****> pravus: you can also use XML::XPath, but it might be slower
02:05 <****> so, if i'm trying to install Date::Calc using the CPAN module, and it flips out with tons of "cast to pointer from integer of different size" warnings and eventually dies, is there something straightforward i can do?
02:05 <****> DSterling: it mentions something about gnome...
02:06 <****> pravus: yea, that's the libxml2 part
02:06 <****> pravus: It's developed by the gnome people for gnome, but not limited to gnome.
02:07 <****> gnomus sapien
02:07 <****> gnomus neanderthalis
02:07 <****> i guess i'll give it a shot. looks like i already have libxml2 installed.
02:08 <****> litheum: what system are you on?
02:08 <****> nickm__: ubuntu 7.10 x86-64
02:08 <****> whoa
02:08 <****> anno-: ubuntu 7.10 x86-64
02:09 <****> litheum: is "sudo aptitude install libdate-calc-perl" straightforward enough? :)
02:10 <****> hobbs: oooooh
02:10 <****> ah, someone who knows what he's talking about
02:10 <****> litheum: it even looks to be up to date
02:11 <****> hobbs: that's exciting
02:11 <****> i just decided it would be a good idea to try to upgrade the CPAN module so i'm in the middle of installing a dozen ridiculous dependencies
02:11 <****> yea debian/ubuntu has a surprising number of Perl packages, probably the largest besides CPAN itself
02:11 <****> litheum: ah, I see. That's actually not the best idea, but good luck chasing it all down :)
02:12 <****> it just said that almost everything failed to compile, and then said "OK!"
02:12 <****> i hate the CPAN module, damn
02:12 <****> hobbs++
02:12 <****> litheum: I usually like to install as much as possible via the packages, and then grab everything else with CPAN. that usually solves dependency issues pretty well
02:12 <****> looks like the package installed just fine
02:12 <****> DSterling: i totally agree
02:13 <****> i just had no idea so man individual perl modules were available through apt
02:13 <****> nor do i have any idea how to figure out which package to install to get a particular module, for that matter
02:14 <****> packages most match up with package names, with dashes instead of colons... you can also use apt-cache search or packages.ubuntu.com
02:15 <****> DSterling, The dash format is the distname, though most vendors distort that a little.
02:15 <****> DSterling: so... i should be able to guess that Date::Calc would be libdate-calc-perl?
02:15 <****> DSterling: Ever better yet, see dh-make-perl
02:15 <****> apeiron: ah.
02:15 <****> insert the module name between "lib" and "-perl"? that seems a bit far-fetched...
02:15 <****> litheum: the CPAN dist name, which is nearly the same thing
02:15 <****> bleh, apaprently i need Date::Format, too
02:15 <****> hobbs: hmm
02:16 <****> But yeah, have heard great things about dh-make-perl.
02:16 <****> apeiron: debian does almost no screwing with the package names. The only one I can think of that's "off" is libwww-perl
02:16 <****> hobbs: And perlmagick
02:16 <****> jagerman: cool
02:16 <****> which I expect is that way for historical reasons.
02:16 <****> hobbs, I mean the lib-$dist-perl parts.
02:17 <****> apeiron: right, that's what I'm talking about
02:17 * apeiron shrugs.
02:17 <****> apeiron: they follow that 99.999% -- but libwww-perl isn't liblwp-perl, and perlmagick isn't libimage-magick-perl :)
02:17 <****> hobbs: Though AIUI, installing perlmagick via CPAN runs into problems.
02:18 <****> jagerman, ImageMagick is very, very particular.
02:19 <****> Not to mention often buggy and crashy even when it is "working" :(
02:24 <****> imagetragick
02:24 <****> haha
02:25 <****> imagemagick depends on so many millions of libraries
02:25 <****> i'm amazed when it compiles
02:25 <****> you're not the only one
02:25 <****> ports can make anything compile.
02:25 <****> hobbs: thanks for pointing me towards the date::calc package in apt!
02:26 <****> litheum: no problem :)
02:26 <****> one happy customer
02:27 <****> I like that. Needy whiny jerks shatter my faith in IRC. (Hint: that's a joke)
02:28 <****> hobbs: are you happy?
02:32 <****> hello, I'm writing a perl script that reads in a file, grabs the start & stop locations for each line (using seek), grabs some text, and stores it in an array of hashes. this works fine, except i need to do this to work on very large files. I think tying the array to a file will fix the problem of running out of space for the array (freebsd barfs eventually) but when I run the program with a tied array, it apparently can't pull the hash from the array e
02:32 <****> err, tell, not seek ;)
02:33 <****> phurba: what are you tying it to? DBM stuff doesn't generally cope with references unless it says it does.
02:33 <****> i should add that i get the hash error when trying to sort the array
02:33 * iank GumbyGumby GumbyBRAIN dongs
02:33 <****> Gumbybrain, but i don't know, i wasn't talking about dongs?
02:33 <****> Gumbygumby is a tree native to australia where it is gumbybrain talking about dongs?
02:33 <****> GumbyBRAIN: Gumbybrain is not a tree.
02:33 <****> ... Don't i get the gumbybot... Is there a way to know.
02:33 <****> hobbs: the long and short of it is that i'm trying to sort a very large file by some data in each line
02:33 <****> :'D
02:33 <****> phurba: also, if you're that concerned over tightening things up, could you replace the hash with a string and extract the info when you need it?
02:34 <****> Can anyone confirm a segfault using readdir inside a chroot?
02:34 <****> hobbs: yeah i could do that, but i was hoping it would be a faster simple fix ;)
02:34 <****> sudo chroot /some/directory; perl < "open(DIR,
02:34 <****> phurba, you should do it in two parts as far as I am concerned. First: read the file line by line and insert the data into the database, Second select the data back sorted and use it
02:34 <****> phurba: that'd make it serialize easily, and drop your in-memory overhead
02:34 <****> sudo chroot /some/directory; perl < "open(DIR, '.'); print readdir(DIR);"
02:34 <****> Should segfault you.
02:35 <****> prakriti: the problem is the data is too large to keep in memory at once, which is why i'm tying it to the file
02:35 <****> instead of making it a hash i suppose i could just delmit the text and sort on that somehow
02:35 <****> phurba: you seriously want a RDBMS, if you're dealing with that much data
02:35 <****>{ts} } @unsorted;
02:35 <****> Ok...now it doesn't...
02:35 <****> dsterling: this is a one-time fix
02:36 <****> phurba: also, what kind of sort is it? If it's simple lexical or numeric, there's no shame in going out to sort(1)
Total 38 pages. You are browsing page 3/38.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last