On Tue, 2011-01-11 at 11:13 -0800, Richard S. Crawford wrote:
> I'm retrieving CLOB data from an Oracle database, and cleaning up the HTML
> in it. I'm using the following commands:
>
> $content =
> strip_tags($description->fields['CONTENT'],'<p><ol><ul><li>');
> $content = preg_replace("/<p.*>/","<p>",$content);
>
> The second line is necessary because the <p> tag frequently comes with class
> or style descriptions that must be eliminated.
>
> This works on the whole except where the <p> tag with the style definition
> is broken up over two or more lines. In other words, something like:
>
> <p class = "bullettext" style = "line-height: normal
> border: 3;">
>
> In this case, the second line of my code does not strip the class or style
> definitions from the paragraph tag. I've tried:
>
> $content = nl2br($content)
>
> and
>
> $content = str_replace(chr(13),$content)
>
> and
>
> $content = preg_replace("/[".chr(10)."|".chr(13)."]/","",$content)
> (I've read that Oracle uses chr(10) or chr(13) to represent line breaks
> internally, so I decided to give those a try as well.)
>
> and
>
> $content = str_replace(array('\n','\r','\r\n'),$content)
>
> all to no avail; these all leave the line break intact, which means my
> preg_replace('/<p.*>/','<p>',$content) line still breaks.
>
> Anyone have any ideas?
>
If you don't have too many problems with the HTML code (like broken
tags, etc) then maybe you could use strip_tags() which will perform
better than a regex in this instance, and should work with tags over
multiple lines as well (although I can't say I've specifically tried
that)
Thanks,
Ash
http://www.ashleysheridan.co.uk
