-null-

Grrrr...the value was being set again in another file that I never noticed Fixed now thanks.

In my php.ini I have error_reporting = E_ALL & ~E_DEPRECATED however I am still getting errors about deprecated functions (ereg_replace). I've checked the code (3rd party app) and they only seem to have set error_reporting like this error_reporting (E_ERROR | E_WARNING | E_PARSE); I assumed this would switch off the deprecated warning however it isn't. So what will?

I have the following xml document <links> <item> <name>[OoE]</name> <link>http://blahblahblah.com</link> <category>3</category> </item> <item> <name>-FaKe|</name> <link>http://moreblah.com</link> <category>3</category> </item> <item> <name>Cymru</name> <link>http://asdas.com</link> <category>3</category> </item> </links> and I use an xpath query to select nodes whose name begins with a certain character. This is all working well however when I access the nodes using nextSibling I am not getting what I expect. In my foreach foreach ($nodeList as $node) echo( $node->nodeValue ); echo( $node->nextSibling->nodeValue ); echo( $node->nextSibling->nextSibling->nodeValue ); } The first line gives me the value of the name node as I expect. However the second line echos nothing where I would expect the value of link and the third echos the value of link where I would expect category. Is this because the whitespace is being treated as a node and if so can I set it so whitespace is ignored.

I should hopefully get the full details of what hosting package my client has today so I can check it supports sqllite. Thanks.

Can anyone recommend any CMSs that do not require a database to store the data. I have a client that wants a website they are able to update themselves however their hosting doesn't provide a database.

Thanks That works for single character tags and I managed to work out how to match tags longer than one character as well.

I'd like to be able to allow only certain BBCodes in my forum. To do so I am trying to use regular expressions to parse all the bbcode and removing the unwanted codes. So far I have the expression that matches any simple bbcode #\[(.+?)\](.+?)\[/\\1\]#is However I'm not sure how to exclude the characters I want to keep (b, i, u a and s) from this expression. Is it possible to exclude characters with a dot match.

Thanks.  I'll have a look at netbeans :)

I've been using Dreamweaver for editing and have recently started using Eclipse for debugging. I find Eclipse really really slow however and now it's started throwing errors everytime I do anything.  Can anyone recommend any other free editors that allow debugging control.

Never mind, I solved it. The problem was with the contents of the file, not the regular expression.

I'm trying to extract some data from a hex file. The data I want is prefixed by a series of hex characters then a country name a couple more hex characters then a user name. I've attached a screen grab showing the data I'm extracting from. There can be multiple sections in the file like this, with different countries (from a preset list) and user names. And I'd like to extract all the countries and their associated user names. I have a regular expression that I believe should work and does so in Editpad Pro however it doesn't in PHP. This is my regex code preg_match_all('/\x00\x0E[\x07|\x06]\x00(a\x00u\x00s\x00t\x00r\x00i\x00a|b\x00r\x00i\x00t\x00a\x00i\x00n|f\x00r\x00a\x00n\x00c\x00e)\x00\x0E(.*?)\x00\x0E/Us',$data,$matches); [attachment deleted by admin]

Thanks, I thought that was the case.

This isn't so much my problem as the problem of a site I'm signed up to. The site tracks the URLs of the people you refer to it however one of my referrals is showing as untracked and another member who referred people from YouTube is showing as untracked as well. The programmer says the reason is because the people clicked a link in an email or typed the URL into the browser and I was wondering if that would be the only reason the URL would be untracked or whether there are other cases in which the referrer wouldn't be set. I'm pretty sure when I did ASP years ago there were cases where the referrer wouldn't be set and wondered if that was the same now for PHP.

I've had a look at Joomla but its multi site isn't quite what I'm looking for. The problem is that each of the sites will be in a different folder with the particular url pointing to that folder. Most of the multi-site functionality of the CMSs need the urls to point to the same place. Anyway it looks like I'm going to have to do some fiddling about to get exactly what I need.

I have recently made the switch from asp.net to php and am looking for a similar cms to the one I was previously working with in .net. I'm looking for a CMS that can administer multiple sites from the one admin. Something where the full CMS would be installed on a main domain but a basic "front end" section could be installed on separate sub-domain folders. The content on all the domains (main and sub) would be administered from the one admin system. It would also be ideal if the CMS allowed the design of the sites to be administered as well, something like wordpress but where the design could be copied over to all sites. There seem to be hundreds of CMSs in PHP, are there any that provide this kind of functionality. I think the answer is probably no but thought I'd ask just in case.