Correcting Bad HTML Code on DB driven templates

Guidelines and suggestions on how to manage bad code

[Whitey]

We've run into a bit of trouble with some bad html code , validated with [validator.w3.org...] on our network of sites.

We have a custom built system, and the problem is that the html code is mixed in with our template scripting. Our developer is saying, that running an automated editing tool over these templates, over a large network of sites, will wreak havoc with the programming code.

Seperately my SEO has a nice CMS [ which will cost a fortune to convert to ] where the code is fixed automatically. That would be nice.

Does anyone have any suggestions to get us out of this fix efficiently?

[edited by: Whitey at 4:48 am (utc) on Oct. 18, 2006]

[tedster]

You may not achieve 100% valid code in this kind of situation -- but major errors absolutely can and should be fixed at the template level. Things like incorrect nesting, unclosed tags and so on are not inherent in a system like you describe.

Sometimes developers are not clear that browser error recovery routines may be dealing just fine with their code soup, but spiders are not creating a visual rendering and so their error recovery may not be what you hope. I look at it this way -- we need search engine listings more that a search engine needs any one of our sites. It's a very smart idea to make the signal as clean and clear as we can.