> -----Original Message----- > From: Jochem Maas [mailto:jochem@xxxxxxxxxxxxx] > Sent: 18 June 2007 13:18 > To: tedd > Cc: [php] PHP General List > Subject: Re: generating an html intro text ... > > > tedd wrote: > > At 11:39 AM +0200 6/14/07, Jochem Maas wrote: > >> original string: > >> > > ... > > > > The problem as I see it is covering all the possibilities that may occur > > even if the text is well formed. Like what if someone introduces a span > > that sets a color for a paragraph, such as: > > > > <span color:"yellow"; >Dolore magna aliquam erat volutpat ut wisi enim > > ad minim veniam quis nostrud. Consectetuer adipiscing elit sed diam > > nonummy nibh euismod tincidunt ut laoreet exerci tation ullamcorper > > suscipit lobortis! <b>Decima eodem modo </b>typi qui nunc nobis videntur > > parum clari fiant sollemnes in.<span> > > > > And the </b> tag as well as the </span> tag is outside the 256 limit? > > > > You would have to search out and pull in all closing tags. > > > > So, I guess an algorithm could be: > > roughly speaking yes this is what is would do, except: > > > > > First, grab 256 characters -- The string. If The string is shorter, then > > quit. > > the algo should only be counting 'content characters', i.e. > anything that is > html markup should not go towards the string length count, > additionally html entities > such as '&' should be considered as a single character. > > > > > Second, determine what tags are not closed. > > > > Third, create closing tags and add them to the end of The string (in > > proper order). > > > > Fourth, then remove the same number of non-html characters from the end > > of The string. > > what the code should do (mmore or less) is quite clear - writing something > flexible & robust to actually do it (and do it fast) is quite > another matter. > > I have been looking at Edward Vermillon's code but I suspect that > what he sent > me is not quite what I'm looking for for a number of reasons: > > 1. it deals primarily with custom bbcode like markup > 2. I have a couple of doubts about the handling of html entities > 3. performance > > that said I still have to look at it in depth before making any real > conclusions as to it's viability (and or the possiblity to rework the > code to fit my needs). > > I'm also looking at an alternative where by I go through the > string and truncate it at the character (or characters that > represent an html entity) that reresents the Nth 'content character' > and then feeding the truncated string to the Tidy extension and let it > figure out the html cleaning part ... does anyone have experience > using tidy > to clean (make valid) html snippets using Tidy, that they would > like to share? > A few thoughts I've had on this problem: Assuming it is well formed HTML, you could use a stack. Parse the string putting all opening tags on the stack and then removing them when the close tag is found. This will leave you with all the un-closed tags and in the correct order. Remember that some tags can be self closing (<br /> etc). Could you use one of the XML extensions to build a DOM of the snippet? I've not used them much myself, but I'm guessing they could handle most of the tag stripping and matching... HTH, Edward -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
