| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
Comments:
<0> "Amce"?!
<1> maybe i could release a perl plugin for irssi that ignores buubot when mangling!
<2> Yaakov, you know there is a program which I have forogten the name of which outputs the context data for almost any english sentance, you could use it to improve mangle to only replce with world that match the correct noun/adjective/verb usage
<0> oooh, I know this one!!!! err....
<3> DoctorMO: That would only work if the database included that metdata, which it doesn't.
<3> a
<0> bugger, my bookmarks fail me again
<2> DoctorMO, it just has a hash thesaurus?
<3> We might be able to do it with wordnet, though.
<3> Yaakov, it is moby-thes
<3> Yaakov, at dict.org
<0> ah, the Brill tagger
<0> there's perl modules for POS tagging
<2> integral, that wasn't the program I've used, but if it does the same thing then cool
<0> it's yet another HMM thingie
<2> integral, HMM?
<3> integral: write a better version!
<0> hidden markov model
<3> In any case, I have sworn off it due to the HATE--so it is irrelevant to me.
<2> integral, too many mathematicians not enough architects?
<0> I think we need more zombies
<2> to eat the perl code?
<2> or to file bug reports?
<0> to eat the mathematicians
<4> architects are everywhere. they're almost as abundant as real estate agents.
<4> they're like all the other 'techs'
<0> .oO( mathetects )
<4> archi tech
<4> pharmacy tech
<4> lab tech
<4> mathematicians are like opticians
<4> they fit you with the equations you need and ***ist the mathmologists.
<5> anybody know how I find the owner of a specific file?
<5> apart from calling ls -la
<0> steve: perldoc -f stat
<6> steve: Type 'perldoc -f stat' in your shell or go to http://perldoc.perl.org/functions/stat.html
<0> gah, I wish I had strace on darwin
<7> eval: perldoc -f perldoc -q
<6> yango: Error: Can't find string terminator ";" anywhere before EOF at (eval 141) line 16.
<7> mangle: perldoc -f perldoc -q
<6> perldoc -f perldoc -q
<0> yay, collaborative, real-time module hating!
<4> a cpan rating hot of the press!
<8> hello, i want to regex the content in a file. the end result should be a file or a string. i have found two ways which work: http://www.rafb.net/paste/results/WmLnyo99.html but neither seems very optimal to me, is there a better way to do this?
<6> The paste WmLnyo99 has been copied to http://erxz.com/pb/2072
<4> why can't it report with a link too?
<0> To join in the general hatefest, and to downvote the module just because alias did, click here: >>*<<.
<9> zzctb: second one is the way to go if you want it in a string
<9> zzctb: first one is rather horrible but could be used with $^I, if you make some changes
<8> Botje: thanks, is there any way without need to open the file at all?
<10> integral: I'm all for that.
<11> well... that ****s
<11> ISP power outage for 5 hours
<0> merlyn: are you using pty-pipe much?
<11> it's being used in production.
<0> ah, so it works. hrm
<11> it was absolutely necessary for this project.
<4> I can't find it. Should I be looking at search.cpan.org or annocpan? I just wanted to see what the comment was if any...
<0> mjl69: ratings.cpan.org
<4> thanks!
<4> wow
<1> that *is* bad
<12> hey guys
<12> can you have alok at this its not outputting anythign at the moment
<12> http://rafb.net/paste/results/mnVUxk17.html
<6> The paste mnVUxk17 has been copied to http://erxz.com/pb/2073
<11> integral - are you having problems?
<0> merlyn: yep, I'm not getting any StderrEvents
<11> I'm seeing them just fine in my app
<11> however, your app may be buffering since stderr is a pipe
<0> I've definately got a "select STDERR; $| = 1"
<11> ok, weird.
<13> Or use IO::Handle; STDERR->autoflush( 1 );
<13> This has the advantage of not using the select() sideeffects
<0> sideeffects?
<0> I'm selecting STDOUT straight back afterwards.
<11> sideefects? :)
<13> See, that's what I mean.
<13> Suppose STDOUT wasn't the old selected filehandle.
<13> You'd have to my $oldselected = select STDERR; $| = 1; select $oldselected;
<0> no, I know what I'm doing. I'm in a brand new process
<0> selecting STDOUT is a *good* thing :)
<12> :(
<13> Easier just to do the filehandle trick and modify STDERR directly
<11> or select((select STDERR, $|=1)[0])
<11> then you don't need to load extra code. :)
<11> that list trick is mine
<10> Sickening.
<11> but stderr in a perl process is already unbuffered.
<13> Oooooh... evil. I like it... in a crazy obfuscated sort of way.
<13> merlyn: only if stderr is on a tty
<10> select(scalar(select STDERR,$|=1))
<11> oh
<11> $| = 1, select $_ for select STDERR
<10> Heh
<11> that's sickeningly better
<10> Yeah
<12> guys any ideas on the script ?
<13> Can we abuse do{} somehow?
<12> http://erxz.com/pb/2073
<10> s/(?{select STDERR})/$|=1;select($1)/e #I bet that doesn't work
<13> No.. that'd stringify the handle
<14> i got a perl regex question: perl -e '$var=a; $var =~ s/a*?/ # /g; print "$var\n"; ' gives
<14> # # #
<0> LeoNerd: it's ugly: select do { my $f = select STDERR; $|++; $f }
<14> can anyone tell me why
<13> integral: heh.. :/ Yeah...
<10> LeoNerd: Yeah, but it's a ***y idea
<0> $^R?
<10> m4n: Because nothing matches 3 times.
<13> I prefer the list one for cuteness...
<14> buu: i didnt get you
<13> Any string I match 3 times must be true.
<10> m4n: The regex, 'nothing', matches 3 times in a 1 character string.
<0> buu: s/(?{select STDERR})/$|=1;select($^R)/e ?
<10> String: "a"; you match "|a", "|a|", "a|"
<14> buu: then if I make the string a to aa, I get 5 #
<10> Hrm
<11> sub { $| = 1; select shift } -> (select STDERR);
<10> String: "a"; you match "||a", "||a|", "a||"
<10> integral: Think it would work?
<0> buu: I do
<10> I like it.
<10> m4n: If you have aa you can match 5 times like so: "||aa", "|a|a", "a||a", "a|a|", "aa||"
<14> buu: sorry, can you tell me what you mean by "| |aa" ?
<10> eval: print "matched at ", pos() for "a" =~ /a*/g;
<6> buu: matched at matched at
<10> bah
<10> m4n: The pipes surround what the regex is matching
<10> eval: $_="a"; print "matched at ", pos() for /a*/g;
<6> buu: matched at matched at
<10> USELESS
<10> eval: $_="a"; print "matched at ", pos() while /a*/g;
<6> buu: matched at 1 matched at 1
<14> ah i see
<10> eval: $_="a"; s/a*/1/g; $_
<6> buu: 11
<10> Guess not.
<14> buu: thanks I will play some more with that
<11> eval:"aa" =~ /(a?)(a?)(?{push @results, "$1|$2"; 0}}/g; \@results
<6> merlyn: Error: Sequence (?{...}) not terminated or not {}-balanced in regex; marked by <-- HERE in m/(a?)(a?)(?{ <-- HERE push @results, "$1|$2"; 0}}/ at (eval 141) line 15.
<11> eval:"aa" =~ /(a?)(a?)(?{push @results, "$1|$2"; 0})/g; \@results
<6> merlyn: ['a|a']
<10> Hrmph.
Return to
#perl
or
Go to some related
logs:
audiowrapper source
tcpwrapped : definition
netinstall fedora repomd.xml
ksh xargs too long
#css
debian xlibg6
#linux
vino xgl Ubuntu
basbryan
useradd with homedir ubuntu
|
|