| |
| |
| |
|
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
Comments:
<0> joobie: The limitation is that you're using a regex. :)
<1> sorry hobbs I was wrong: )
<2> m/<a href="([^"]+)">([^<]+)/
<0> joobie: to match something that regexes don't.
<3> hobbs, you are telling me it's not possible with regex
<3> it is.
<0> joobie: Do you know what a regex is?
<3> regex matches a string u give it
<3> it can be html
<3> it can be plain text
<4> joobie: you were told how to do it. take that for what it's worth. no one here WANTS to parse HTML with a regex, so we aren't experts at it and no one's going to tell you.
<3> regardless of the string you are matching, being htmkl or not
<3> the concept is still the same.
<0> joobie: Sorry, wrong.
<5> do you know ANYTHING about language processors?
<0> joobie: Regular expressions match regular languages.
<6> what kind of language parser
<2> jpeg, that's just silly, never parsing html.
<6> processor*
<2> umm
<4> bluebeard_: shaddup
<3> hobbs
<0> joobie
<3> you need to read more about regex
<1> joobie, why write something that someone else has already wrote? (and probably better than you'll ever can write)
<5> no, you do
<3> regex does not just match regular lanugages
<2> html is Content-type: text/*
<3> it matches text
<3> any text u give it
<2> you should be able to parse text/* with perl.
<3> nod
<0> joobie: Who the **** is this "u" person?
<3> thanks bluebeard
<3> u can
<0> bluebeard_: of course. And fortunately for us PERL CONSISTS OF MORE THAN REGEXES
<2> and?
<3> hobbs
<3> the point is
<0> bluebeard_: and that is all :)
<3> you're sitting there saying perl regex can't parse html
<3> that's wrong.
<2> how else would you do that?
<3> perhaps you should saying something like 'use a perl module to parse html rather than straight regex'
<1> joobie, What's the point about writing something that already exists, and you can use easily?
<3> that'd be more supporting your point
<2> umm
<4> no, he's saying there ARE BETTER TOOLS FOR THE JOB
<0> joobie: come back and collect your $10 when you have a working regex
<3> because you are wrong in saying perl regex cant parse html
<3> it can.
<2> what would the perlmodule be doing?
<0> joobie: until then, shut up :)
<3> dood
<2> scanning the text, just like a m// would.
<3> i havent used regex in a while
<0> bluebeard_: handing off to a LALR parser most likely
<3> but i did years ago
<3> and im 100% certain
<1> you can parse HTML using ***embly if you want, but I don't understand the point of it
<3> it can parse through html
<3> if you think it cant
<3> u need more skill in it
<3> exactly ofer
<0> joobie: WHO THE **** IS U?
<2> and what would the parser do?
<2> scan the text.
<2> why add a module when you can create a simple regex?
<3> exactly
<7> hmm... using "enter" as the only punctuation. using "u" to mean "you", and blatantly starting arguments...
<7> 1+2+3=troll?
<8> yay the internet.
<2> that's adding unecicarry bloat.
<0> bluebeard_: From a falsehood you can infer anything
<0> bluebeard_: you can't create a simple regex
<1> joobie, do what you want to do, but don't expect people to help you solve problems in a stupid way
<3> ofer
<3> i accept ur point
<1> good
<3> hobbs is saying it's impossible for perl regeex to parse html
<3> that's what im arguing
<5> the only thing you're succeeding in doing by arguing about regexs is wasting time
<1> start with HTML::LinkExtractor as hobbs told you 10 minutes ago
<2> why would you add a module when it's not neccicary?
<0> bluebeard_: Show me a "simple regex" that even tokenizes HTML.
<4> bluebeard_: why reinvent the wheel?
<0> bluebeard_: I know a 20-line one that does XML, but HTML is based on SGML, and more complicated.
<1> 20 line regex?
<1> that's crazy
<1> :)
<3> ur presuming the entire html needs to be parsed
<5> and html is full of garbage from the nonstandard browser dark ages
<2> you want the link url, and the link text?
<3> but if it's only 2% of the entire html
<0> bluebeard_: no
<3> why parse the entire thing with a module like that
<0> bluebeard_: I want tags, attribute, and text
<2> he didn't!
<3> exactly.
<0> joobie: because it's a context-sensitive language
<2> if you need the rest, add the module, if not, don't bloat your code.
<0> joobie: which means that if you pick up in the middle of the file and ignore context (like a regex) THEN YOU DON'T ****ING KNOW WHAT YOU'RE PARSING
<3> i agree bluebeard.
<0> SO
<0> ****ING
<0> STUPID
<0> SO
<0> STUPID
<3> hobbs
<8> perl is serious business.
<0> SO
<0> ****ING
<0> STUPID
<2> lol
<0> SO
<3> who said i was ignoring context?
<4> bluebeard_: why do you think modules are bloat?
<0> ****ING
<0> STUPID
<3> if context varies
<3> that's where the regex adopts to it
<0> joobie: you, when you said you were using a regex
<3> like i said, why go to the trouble of parsing the whole html doc
<3> when it's not even 1% of it that u wnat
<3> just use regex
<2> they take KB.
<2> that is bleat
<2> *bloat
<0> joobie: Because you don't know your context until you've parsed the whole doc up to that point
<3> really?
<3> i do
<2> if they take up uneccicary space or ram, they're bloat.
<3> because this page is fairly static
<3> you're presuming i dont know the page im parsing
<0> joobie: how do you know there wasn't a CDATA block starting on line 1?
<3> which is a presumtion you should not make.
<3> because no-one modifies the site
<4> bluebeard_: are you using a 486 or something? wtf is a few K?
<3> exactly
<0> bluebeard_: There's nothing unnecessary about using a solution that's tested and actually works
<8> didn't bill gates once say 64 KB would be enough for any personal computer?
<2> that's a bad stance, jpeg.
<3> just to get that 1% of html
<3> i have to load the module
<3> waste ram
<3> waste cpu cycles
<3> then grab that 1% i want
<4> bluebeard_: are you kidding?
<3> then keep that crap in ram
<3> why
<5> code in C, you won't have that pesky perl interpreter taking up ram :P
Return to
#perl
or
Go to some related
logs:
python subprocess.popen hide
shorten ubuntu shntool
perl list all loaded packages
all ebuilds that could satisfy nvidia-glx have been masked
#perl
#math
gnome commander conf file
#linux
#linux
#php
|
|