regular expression help

[mohamed42]

Argh. I can't figure out this regular expression stuff. My friend wrote a regular expression for me, but it doesn't look at all like a PHP regular expression. Here it is:

^ +(.+) ([0-9]{2}) (.{2}) (.{1,15}) +([A-Za-z0-9 ]{1,17})? +(.{1,15}) +([A-Za-z0-9 ]{1,17})? +([A-Za-z]{1,15})? +([A-Z0-9]{17}) +([0-9]+/.[0-9]{2})? +([0-9]+)? +([0-9]+/.[0-9]{2})? +([0-9]+)?$

Can someone translate this into the sort of regex that PHP will use, please? I'd appreciate it muchly. If you want to take a look at the document that he looked at to put this together... it's at:
http://www.hyperion-data.net/vehicles/carlist.txt
the only lines we're interested in are the ones that look like this:
1P838 02 FO TBIRD TBIRD YELLOW 1FAHP60A42Y101420 0.00 38141.13 210
I think I could figure it out, but there are instances where some parts of the line aren't there in some lines and others aren't there in other lines. And that just makes my head hurt. some help please? thanks much

[Anon]

It should work with the preg_* functions:

http://www.phpbuilder.com/manual/ref.pcre.php

[Anon]

This could be done with a single regexp, though as your friend's
example shows, it can get really hairy really quick (though I have to
say that example looks unnecessarily hairy, and I can see cases where it
would fail). To make it simpler, I'm going to do some extra preperation.

This is done for each $line of the file.

Since every $line is the same length, and the start of each field are
aligned in columns, we can find out where a field starts without having
to do a regexp for it. This is a good thing, because if one of the
fields is missing, we'd have to try and figure out which one it was.
This is what makes a one-regexp approach hairy.

list($stockno, $yr, $mk, $carline, $model_description, $color,
$serialno, $list_price, $miles, $inv_amt, $days) = sscanf($line, '%8s %2s %2s %15s %20s %15s %17s %10s %7s %9s %4s');

Okay, but most of these will have extra padding spaces still in them.
Let's trim them all down.

$stockno = trim($stockno);
$yr = trim($yr);
$mk = trim($mk);
$carline = trim($carline);
$model_description = trim($model_description);
$color = trim($color);
$serialno = trim($serialno);
$list_price = trim($list_price);
$miles = trim($miles);
$inv_amt = trim($inv_amt);
$days = trim($days);

Now, some of these will contain silly results, from lines that don't
belong to the ones we want. We know the format of each field, so now we
can use regexps to recognise them.

$stockno_good = preg_match('/^{\w+$/',$stockno);} // The stock number looks good.

$yr_good = ($yr == '' || preg_match('/^{\d\d$/',$yr));} // The year might not be available, in which case $yr would be empty

$mk_good = ($mk == '' || preg_match('/^{\w\w$/',$mk));} // The make (optional)

$carline_good = ($carline == '' || preg_match('/⁽ |\w)+$/',$model_description)); // Carline (which may contain spaces)

$model_description_good = ($model_description == '' || preg_match('/⁽ |\w)+$/',$model_description));

$color_good = ($color == '' || preg_match('/⁽ |\w)+$/',$color));

$serialno_good = preg_match('/^{\w+$/',$serialno);} // Serial number (required)
//Incidentally, if serial numbers are always 17 characters long, /^\w{17}$/ would do)

$list_price_good = ($list_price == '' || preg_match('/^{\d+.\d{2}$/',$list_price));} // Which is some digits, a decimal point, and two more digits

$miles_good = ($miles == '' || preg_match('/^{\d+$/',$miles));} //Odometer reading

$inv_amt_good = ($inv_amt == '' || preg_match('/^{\d+.\d{2}$/',$inv_amt));}

$days_good = ($days == '' || preg_match('/^{\d+$/',$days));}

Now we have a valid line if all the fields are good:

if($stockno_good && $yr_good && $mk_good && $carline_good &&
$model_description_good && $color_good && $serialno_good &&
$list_price_good && $miles_good && $inv_amt_good && $days_good)
{
Do whatever is required; the information from the line is available in
$stockno, $yr, $mk, $carline, $model_description, $color, $serialno,
$list_price, $miles, $inv_amt, and $days.
}