[SOLVED] EREG HELP!

[Rottingham]

Ok, This is my first attempt at using regex.. I've been coding for years and can't believe I survived this long, but the day is come that I must accept it... So I'm having a problem... I am trying to interpret 6000 lines of the following style

 

"792036": { d:"Holmes Heavy Duty Slide Bolt", p:"9.11", q:"1" },

 

My goal is to extract the first 6 numbers, everything withing the quotes after d: and everthing in the quotes after the p:

 

In this line, I'm hoping to get an array like such

 

array[0] = 792036

array[1] = Holmes Heavy Duty Slide Bolt

array[2] = 9.11

 

My regex statement of:

ereg("\"[1-9]{6}\"", $line, $regs);

 

gives me some encouraging results but frustating also...

 

Warning: Invalid argument supplied for foreach() in /home/macactio/public_html/topsoftweb/test.php on line 48
"101303": { d:"4015 NS 13 Key Blank", p:"0.00", q:"0" }, 

Warning: Invalid argument supplied for foreach() in /home/macactio/public_html/topsoftweb/test.php on line 48
"270625": { d:"Dor-O-Matic SCREW.1022 Dog Screws (25 Pack)", p:"64.00", q:"0" }, 
"271766" "271766": { d:"BRIG 691259 Key Blank, High Security", p:"45.00", q:"0" }, 

 

You will notice that my first line gets a warning, the second line gets a warning, but then the third line actually gets the number, albeit the " signs too. I need to remove the \"...\" from my regex expression. I can't figure out a) why it works on some of the lines and not others, and b) how will I get the rest of the parts I need? I think I have an idea but I'm going to show my ever simple function to see if someone can tell me why it works some times, and not on every line...

 

foreach($file_lines as $line)
{
unset($regs);
// Interpret Line
// "792036": { d:"Holmes Heavy Duty Slide Bolt", p:"9.11", q:"1" },

ereg("\"[1-9]{6}\"", $line, $regs); 

foreach($regs as $reg)
	echo $reg." ";

echo $line; echo "<br>"; 
}       

[dsaba]

Here's a preg (PCRE) solution, I hear its faster than ereg (POSIX)

you can use preg_match_all() to grab all the matches in 1 parse

 

~"([0-9]{6})": { d:"([^"]+)", p:"([0-9]\.[0-9]{2})", q:"([0-9])" }~

 

tested:

http://nancywalshee03.freehostia.com/regextester/regex_tester.php?seeSaved=yyvefddg

 

 

I also noticed in your error report above one of you lines of data does not follow the format you specified:

"271766" "271766": { d:"BRIG 691259 Key Blank, High Security", p:"45.00", q:"0" }

[Rottingham]

Thanks man, looks like a little more success... I changed my code to the following

 

foreach($file_lines as $line)
{
// Interpret Line
// "792036": { d:"Holmes Heavy Duty Slide Bolt", p:"9.11", q:"1" },

// Places three space separated words into $regs[1], $regs[2] and $regs[3].
//ereg("[0-9]{6}", $line, $regs); 
preg_match_all('~"([0-9]{6})": { d:"([^"]+)", p:"([0-9]\.[0-9]{2})", q:"([0-9])" }~', $line, $regs, PREG_SET_ORDER);

echo $regs[0][1].' ';

echo $line; echo "<br>"; 
}    

 

You can see the results here:

http://macaction.org/topsoftweb/test.php

 

Unfortunately, it still works on some lines but not the others.

[dsaba]

you need to show your parse/input data you're working with, the regex I supplied worked fine with what you showed, I cannot see what your problem if I don't see the input data

 

instead of going though each line of the input data:

foreach($file_lines as $line)

 

read it:

<?php
$data = file_get_contents('whatever.txt');
$pat = '~"([0-9]{6})": { d:"([^"]+)", p:"([0-9]\.[0-9]{2})", q:"([0-9])" }~';
preg_match_all($pat, $data, $out);
foreach ($out[0] as $k => $fullMatch) {
$num = $out[1][$k];
$d = $out[2][$k];
$p = $out[3][$k];
$q = $out[4][$k];
echo "$num<br>$d<br>$p<br>$q<br><br>";
}
?>

 

The matches array in my website is verbatim the same array that you will see spit out in the $out array from preg_match_all() with no special flags set.

[Rottingham]

Hmm...

 

You can view my results again at

http://macaction.org/topsoftweb/test.php

 

You can view the source file at

http://macaction.org/topsoftweb/parts_prices.txt

 

There are 6997 lines in that file of those parts, and I'm only getting 1400 results. I'm not sure what the deal is.

[dsaba]

this works:

~"([0-9]*)": { d:"([^"]*)", p:"([0-9]{1,}\.[0-9]{2})", q:"([0-9]{1,})" }~

 

it was because you had varying formats in your data

 

change + to * because the d:.. could be blank

 

changed p to accept 1 or more digits {1,}

 

changed the others accordingly..

 

<a href="http://www.regular-expressions.info/reference.html">Here some a reference to simple regex symbols/terms</a>

[Rottingham]

Thanks a ton!

 

That did the trick. I really appreciate your help.

[dsaba]

*edited last post

glad to help..

return the favor on the forums..