| |
| |
| |
|
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> then unmap it later
<1> Damn you merlyn, now I have to go listen to said mp3s.
<2> simcop2387: dont
<0> heh
<3> cfedde: what would you suggest then?
<1> Which also means I have to sing along. Which means my roommate will shortly be asking if I'm well.
<4> merlyn: I was trying something like that, but it wasn't turning out very nice
<0> or just strip them, counting the words
<2> simcop2387: depends. http is pretty flexable.
<0> then replace them back in offset by same count
<5> merlyn: that only works until you want to wrap documents using Tengwar from the PUA!
<0> What's the proper response to that?
<3> cfedde: http can't be used for realtime MMORPGs
<0> arrrghhgahhh13h43hsj? :)(
<2> simcop2387: why not?
<1> simcop2387: Sure it can! Just not well.
<3> cfedde: it'd be very cumbersome and slow
<2> simcop2387: ok.. if you say so.
<0> seriously, I'd do the word counting thing
<3> cfedde: the http headers themselves would make it very ineffecient
<0> as long as the wrapping can't turn one word into two
<6> ah, i recall that i have a good prog exercise
<3> cfedde: lots of short packets
<4> merlyn: Count the number of <color> tags?
<0> no. for aaa bbb <b> ccc ddd eee </b> fff ggg
<2> simcop2387: see now you've already got some implementation scheme in mind. What about persistant http 1.1 with keepalive and all that other goodness?
<0> you strip it, noting <b> after 2, and </b> after 5
<0> then do the wrap
<0> then walk the result, inserting <b> and </b> in the right places
<4> Hrmm. That could work.
<3> cfedde: that still sends headers for each "page"
<6> this is what i actually need to do in php, but is a pain in the *** in perl too
<4> There actually isn't a </endtag>, but that doesn't matter. How should I count the number of words a tag is after? Split by tags and keep a counter?
<0> I'd do a nibble
<3> cfedde: when the data i need to send is 10 bytes, and the headers end up totaling 30, its kind of a bad design for dialup users
<2> simcop2387: I guess my main point is that by the time you get beyond a simple udp or tcp protocol you might as well go with http.
<7> jjjj: what are you talking about?
<6> the prob is this: suppose you are given a list, where each ele is some string, and this list is sorted
<4> This is going to be so ugly. I'll give it a go.
<3> cfedde: who says i need something more than a simple protocol?
<5> "I mean, what does a hot blonde leaning over a foosball table exposing part of her breasts have to do with software development?"
<5> dabreegster: why would you be counting words?
<3> hobbs: ah you saw that ad too?
<7> hobbs: EVERYTHING!
<2> simcop2387: no one. I guess I just ***umed since you were asking you'd alredy ruled out the simple case.
<3> cfedde: nope, i've just never designed one before really
<6> you task is to: (1) partition this list, so that any element starting with the same alphabet will be grouped together.
<4> hobbs: merlyn recommended it and it seems like a better idea than trying to reinvert wrap(). I'm bad at that sort of task.
<1> hobbs: I have it on very good authority everything. At least, it did that time I was at the Rialto showing a friend how AUTOLOAD worked, and a hot blond leaned over a pool table.
<5> dabreegster: oh, I see. I'd still just rewrite wrapping with your modifications in place. It's not exactly a difficult task. Text::Wrap just makes it hard
<2> simcop2387: if you can serialize your data into 1500 bytes and don't need acknolegement then udp is the way to go.
<7> jjjj: we're not going to do your homework for you
<0> while (length) { if(s/^\s*(<.*>)//) {push @tokens, [scalar @words, $1]; redo} if (s/^\s+(\S+)//) { push @words, $1; redo } }
<6> e.g. result ((aaa, aab, aac...) (bb, b4...) ...)
<5> dabreegster: because it's got a bunch of options :)
<0> so when you push <...>, you note the length and the <>
<0> otherwise, you push a word
<0> then you fill "@words"
<0> and then walk through it word by word, replacing at the right time
<4> merlyn: OK. I'll try. Thanks.
<8> now, suppose listA is one of your sublist containing ele that starts with "a".
<8> (2) Now, for each listA, listB etc., you need to partition further so that
<1> I don't think he heard you when you said 'we're not doing your homework for you'.
<9> xahlee: Is there something you're actually trying to do this for?
<3> cfedde: yea i was figuring udp would be better, what i don't know is if i will need ack's
<5> dabreegster: or you just go through, do the same tokenization, fill lines with words until they're full, add newlines where needed, and insert those "tags" where they come up, without accounting them towards the length of a line :)
<4> hobbs: That actually seems easier. Hrmm, thanks. Dunno why simple stuff like that doesn't occur to me.
<5> dabreegster: and that way you don't end up with mysteriously rainbow-colored documents ;)
<8> each sublist has no more element than 40.
<4> hobbs: Aww, but those are fun.
<9> xahlee: Are you just going to keep listing the requirements without responding to any of the points we're raising?
<8> (3) partition that result so that each sublist has no more ele than 20.
<8> revdiablo: yes to the prev
<2> simcop2387: you need to walk thorugh some use case scenarios and get a feel for what your system needs. I know of some systems that use multiple communication channels for different things.
<8> so, generally speaking, the prob is like this:
<8> you are given a sql result of rows
<8> all ordered already,
<8> you want to display them in a web page
<5> no I don't.
<8> lke some index page of a book so that
<8> you have: A ... B ... C ... etc.
<8> then,
<8> under A
<0> I want someone else to display them. :)
<8> you have tables
<0> I'd use a computer!
<9> xahlee: Are you expecting us to do the whole thing for you?
<8> each table has 2 columns
<8> each column has no more than 20 items.
<0> each cat has seven kits
<8> now, as i was saying,
<9> Jesus, he's not going to stop, is he?
<2> I've heard about this thing called.. binary. maybe it would be useful.
<0> each sack has seven cats
<8> i need to do this in php, as a actual job i need to do
<0> each wife has seven sacks
<8> and it has been a pain in the ***
<8> this thing is very easy in func lang
<1> merlyn: I have added 4 internet clouds and randomly attached them to cl***es. I have also spent a good 10 minutes estimating how long this has taken me, and included a cost-per-documented method and arrow.
<8> and as i was saying,
<0> how many were going to st. ives!?
<8> this thing is also a pain in the *** in perl i'd imagine
<8> although prob better than php
<9> Are you trying to TRICK us into doing it for you?
<1> xahlee: SHUT THE **** UP
<8> because perl supports some func prog
<0> php supports dysfunctional programming
<9> At least he gets points for being clever
<8> merlyn: does php has map?
<9> But not too clever, really
<1> merlyn: Now, now. I'm sure php has a solve_xahlees_problem().
<5> xahlee: yes, it's just really broken because PHP _doesn't_ have proper anonymous functions
<0> Yes - because Rasmus and xahlee have identical problems!
<5> instead you call a function that does a string eval for you, puts the compiled code into a function, installs that at a randomly generated name in the symbol table, and hands you the name
<8> i mean, you guys should think about this
<5> and then all you have to do is provide array_map with the name of a random function to call! special!
<8> i have just dsecribed a prob,
<0> why should I think about it?
<8> meanhwile, under a lot of **** attitude
<3> GumbyBRAIN: i've never even heard of number 9. its doing number 13 while she's 11ing your 5
<10> even if its more than a number.
<7> xahlee - we really don't care. most people in the channel could solve your problem quite simply..
<8> and now i finished described the prob, as well as answer some of the bad accusations in question form,
<8> still getting ****.
<10> That should go for hours.
<0> exactly!
<5> GumbyBRAIN: And I say... does anybody really know what time it is? Does anybody really care?
<10> Is anybody able to successfully use that to do it.
<8> i have also explained, where this prob came from, and how it actually turned up.
<8> AND, i actually gave a positive perl point over PHP
<7> xahlee - yes, but you HAVENT ASKED A QUESTION
<8> but jesus, one bag of perl morons keeps at their no unstanding
<11> hi xahlee, do you have a perl question?
<0> xahlee - when you come in, act in a way that gets you hated, and then you continue to ask a question, even when people said "shut up", you now expect an ANSWER? This is what we call a "troll"
<7> You haven't even asked us to do the whole thing for you.
<9> xahlee: But you're still asking us to do work for us?
<0> truly
<12> hey
<9> Err
<9> For you
<5> beth: ooh ooh me pick me pick me!
<0> and I say this even having a lot of sympathy for you
<8> merlyn: don't try to start to get me angry. Did you really think that the guy who told me to shut up should be taken seriously?
<0> perhaps more than you deserve around here.
<13> Perl and Python walk into a bar ....
<8> are we, now, serious??
<0> xahlee - see how it feels to be trolled! ha ha!
Return to
#perl
or
Go to some related
logs:
loading jpeg sdl
#centos
FC5 ssh on
#perl
gentoo 810 driver gma 950
#web
yum missing dependency kernel-i686 fglrx
xine wmv the source seems encrypted
#openzaurus
libxine1-main + ubuntu
|
|