|
|
nycdiesel
|
nycdiesel
|
I'm working on a script which downloads the search results web page from a webserver. I know how I can do this but was just curious if there's a better way. I'm trying to workaround the /set: line too long error in mIRC when sockreading. So I've decided to write the file to disk and then parse the file though I'm still getting errors. The problem is all of the TAB chrs at the beginning of lines in the file. Any ideas?
|
|
|
|
Joined: Sep 2004
Posts: 237
Fjord artisan
|
Fjord artisan
Joined: Sep 2004
Posts: 237 |
Nice Code! Come up with that all on your own?  Perhaps its not too late to edit that and add the code.
|
|
|
|
nycdiesel
|
nycdiesel
|
Actually figured it out 
alias test2 {
var %idx = 0, %b, %chr
write -c txt.txt
fopen a page.txt
while (!feof) {
if ($ferr || $feof) { fclose a | return }
%b = $fread(a,4096,&var)
if (%b) {
while (%idx < %b) {
inc %idx
%chr = $bvar(&var,%idx)
if ($istok(10.13,%chr,46)) { bset &new %idx %chr }
elseif (!$istok(10.13.9,%chr,46)) { bset &new %idx %chr }
}
bwrite txt.txt -1 &new
%idx = 0
bunset &new
}
}
}
It's still a little slow and freezes mIRC for about 15-30 seconds, not too bad, I suppose.
Last edited by nycdiesel; 10/02/05 09:40 PM.
|
|
|
|
DaveC
|
DaveC
|
Im supprised this was all that was needed to reduce lines to managable sizes, but if it is it is. Use this its alot faster. alias test2 {
var %idx = 0, %b, %chr
write -c txt.txt
.fopen a page.txt
while (!feof) {
if ($ferr || $feof) { .fclose a | return }
%b = $fread(a,4096,&var)
if (%b) {
if (1) {
while ($bfind(&var, $v1, 9)) {
bset &var $v1 0
}
bwrite txt.txt -1 &var
}
}
}
}
The if (1) { is needed to preload $v1 for the while loop
|
|
|
|
nycdiesel
|
nycdiesel
|
Thanks Dave, much appreciated  I wasn't too familiar with those v1, v2 things but I did my homework, heh. Anyways I'm still encountering a problem when attempting to parse the file after calling test2 .. *shrugs* my second test alias echoed the contents of the file and it to the end of the file but it was only showing partial lines.
|
|
|
|
DaveC
|
DaveC
|
try this, alias test1 {
var %idx = 0, %b, %chr
write -c txt.txt
.fopen a page.txt
while (!feof) {
if ($ferr || $feof) { .fclose a | return }
%b = $fread(a,4096,&var)
if (%b) {
if (1) {
while ($bfind(&var, $v1, 9)) {
bset &var $v1 10
}
}
if (1) {
while ($bfind(&var, $v1, 60)) {
bset &var $v1 10
}
}
bwrite txt.txt -1 &var
}
}
}
I dumped the 0 byte as i remebered it terminates the line, following data well not read from that line. I now change TAB to LF and < to LF, this should be enough to break the lines down to managable sizes. Just remeber that you cant search for <TR> sine it well just be TR> at the start of a line, and also that there well be $null length lines in the file.
|
|
|
|
Iori
|
Iori
|
This writes it back to the same file, if you really want the extra file just change the blue page.txtalias testb {
bread page.txt 1 $file(page.txt) &a
breplace &a 9 10 60 10
bwrite [color:blue]page.txt[/color] 1 -1 &a
}
|
|
|
|
Iori
|
Iori
|
That last post of mine should have been directed to you. I noticed this in your code Should be they way you have it there "!feof" is just a string (which would always be true), same as Daves "if (1)" - //linesep | if !feof { echo -a $v1 } | if feof { echo -a $v1 }
|
|
|
|
DaveC
|
DaveC
|
This writes it back to the same file, if you really want the extra file just change the blue page.txtalias testb {
bread page.txt 1 $file(page.txt) &a
breplace &a 9 10 60 10
bwrite [color:blue]page.txt[/color] 1 -1 &a
} Oh that is way nicer method, I was just adjusting his code, since i saw the byte by byte checking, and thought Im sure u can just find the values u want in the var instead, should have stepped back and looked at what he wanted as a result, rather than how to improve his code. Goes the same for the while (!feof) { I didnt even look at it 
|
|
|
|
|
|