HTML Field Value Search

stevelowe

Hi,

Well this is a unique request, just do not ask me why I need it!!
I have the source to an HTML page. I need to extrapolate each text box name and value to create a tab delimited or CSV.

Also there are dozens of these text boxes and each field name repeatds itself and is appended by 1 each time. So 'EPAYT' in the first field becomes EPAYT_1 in the second.

I have tried using split() and strtok () but they are not constant. What I need is a piece of code that will detect the field name EPAYT and then look for the value at X chars to the right.

I have exhausted my limited knowledge of PHP, can any one help please?

Running WinNT, Apache1.3 ,PHP 4.1.1.

Weedpacket

Well, I'm not entirely clear what on what you're describing, or what your motivation is or whether the approach you describe is the best (and it sounds like it's safer not to ask), but one strategy (and I'll be writing it as I think of it) might go:

(I will assume that EPAYT is never at the start of a string - if it is it would be a trivial matter to chage the string so that it's not...)

explode() on EPAYT. That way you get an array of strings, all (except the first) consisting of that bit of a string that follows an EPAYT.
Throw that first array element away - unset()ting it should be enough.
If you know X, then substr() will be sufficient to extract whichever bits take your fancy.
If on the other hand you need a regexp, all you need to keep in mind is that all of the strings in the array started with EPAYT before the explosion; so the regexp would generally start with '^' followed by a pattern matching what you're looking for.

Does that sound like it would do what you're after? On the sample string (again, I'm not entirel clear what the string is supposed to look like) fooEPAYT_1barEPAYT_2bazEPAYT_3wibble the array that results looks like

_1bar
_2baz
_3wibble

...and substr($array[n],1,1) would extract the digit from row n.

bobthedog

flexes his regular expression muscles

Ok - regular expression is gonna be your best bet...

something like this

preg_match_all("/\<input\s+type=text\s+name\s*=\s*[\"\']?(.+?)[\"\']?\s+value\s*=\s*[\"\'](.*?)[\"\'][^\>]*\>/i", "<input type=text name='fred_1' value='stuff'>", $arrMatches);

I know it looks messy - bit it aint that difficult...

it reads soemthing like this...

find anything starting "<input", followed by at least 1 whitespace character "\s+", followed by "type=text", followed by at least 1 whitespace character "\s+", followed by "name" and some white space, then an "=" and some white space, then an optional (one or 0) speech mark "[\"\']?", then the name with minimal matching ".+?" then an optional closing speechmark....

etc etc... the rest just follows this same pattern.

I am sure there is probably a nicer way of doing this, but hopefully it will help you in the right direction.

Oh, btw, the results are in $arrMatches, just take a look at the array using "print_r($arrMatches)" and then I'm sure you can work the rest out from there 🙂

Weedpacket

Note that the vBulletin software takes it upon itself to mangle most of the backslashes when syntax highlighting is used - perhaps change those to code pseudotags, bobthedog?

Heh, and I'm the one who does regexps for recreation....

bobthedog

hehehe - cool. cheers for the heads up weedpacket

stevelowe

Thanks gents,

I used the explode() in the end. It's got me past stage one anyhow. Managing Directors? Who needs 'em!!