extracting contents from a file

safra

Hi,

I need to update a lot of pages on a remote server.

How can I extract specific parts of the page using php

for example:

has to become:

as there are several things to do I have to extract and validate the part "something.php" so I need to match the contents between <a href=" and "> and return it into a string.

How is this done in php?

Thanks,

feldon23

Ok, I'm confused.

You are using PHP to edit other PHP files remotely?

If so, the security concerns here are mind-boggling.

You would just search for every occurance of <a href="something.php with strpos(); and use substr() to insert things.

Or are you trying to extract the filename before the ? and the values after the ? from within the PHP script receiving the ? values?

If so,
the filename is $SERVER['PHP_SELF']
everything after the ? is stored in $SERVER['QUERY_STRING]
and you can also access the individual values after the ? with $_GET['']

music.php?city=nashville&type=blues&artist=Elvis

$SERVER['PHP_SELF'] = music.php
$SERVER['QUERY_STRING] = city=nashville&type=blues&artist=Elvis
$GET['city']=nashville
$GET['type']=blues
$_GET['artist']=Elvis

safra

I am trying to reduce work. The new files are first saved locally and afterwards uploaded to the server.

the problem is that I have several links (something.php is just an example) that's why I have to match the part between <a href=" and "> so I can validate the link and continue from there.

I think I have to use preg_match for this but those regular expressions are always so confusing to me.

Thanks,

planetsim

Well For regular expressions theres a very good article on PHPBuilder. I learnt from it.

feldon23

I am trying to reduce work. The new files are first saved locally and afterwards uploaded to the server.

the problem is that I have several links (something.php is just an example) that's why I have to match the part between <a href=" and "> so I can validate the link and continue from there.

I think I have to use preg_match for this but those regular expressions are always so confusing to me.

Again I am trying to figure out why you need to sift through and change PHP files on another server remotely instead of downloading those PHP scripts, changing the links, and re-uploading or storing the dynamic values that those PHP scripts use in a database somewhere. You don't have to answer, I'm just curious.

Again, in order to have a PHP script on one server edit the contents of a PHP script on another server will mean that your PHP script on server #2 will have to be editable/deletable by the public. This is a severe security issue.

safra

There are about 400 dynamically created pages all over the website. Then I first have to download them, change all of them manually and upload them.

Putting a little script together that scans the website for these particular pages, Check for the links (the script will have to look for 4 different links)/ update the link and print the page on the screen so I can copy/paste it and upload it back to the server, seems faster to me.

Once again I am not talking about changing scripts directly on the remote server and I am also not talking about a php script on one server changing php scripts on a different server, they are on the same server! I feel like you are sincerely doubting my intentions?

Thanks,

safra

planetsim,

I just searched through the website, I can't find the tutorial only the manual about several regular expressions functions in php.

I looked on php.net and went through user submitted contents about regukar expressions but so far I only found examples of how to extract text from a file which seeme to be an url and then make a link from it.

Weedpacket

Originally posted by safra
I just searched through the website, I can't find the tutorial only the manual about several regular expressions functions in php.

planetsim is referring to the article listed here. Its main drawback is that it discusses the POSIX-style regexps that ereg uses, which aren't as powerful as the Perl-compatible ones preg_*() use. But for tutorials on those in particular, I think Googling on Perl "regular expressions" tutorial ought to throw up some useful hits.

There is also a thread on this forum (I'm too lazy to look for it myself right now) that contains code that uses such regexps to identify and collect the href= attributes of all the <a> tags in a page (seach for, say, "match href attributes").

safra

thanks weedpacket, I found that thread. That is exactly what I need!