Queries regarding transaction site

For technical support for all editions of CSS HTML Validator. Includes bug reports.
Post Reply
Gibbowolfie
Rank 0 - Newcomer
Rank 0 - Newcomer
Posts: 3
Joined: Wed Mar 23, 2005 6:01 am

Queries regarding transaction site

Post by Gibbowolfie » Wed Mar 23, 2005 6:13 am

I am researching validation applications for my company and I have a question regarding transaction websites.

We mainly implement transaction sites that use an engine that generates XHTML which is then combined with some static XML/XHTML to produce the final page sent to the browser. The page function might be for data entry and would probably feature some form of validation on entries.

My queries are around how such a site might be validated using CSE HTML Validator. I can see problems using the "follow links " options for batch operation, as following the submit link would cause the validation to fail. Also dependant upon data entered the link might go to a dynamically defined URL. Without data entered it would not be possible to get to some areas of the application.

I'm sure this would be an issue with a lot of commercial sites so perhaps there is a methodology that can be followed in order to validate.

Of course it should be down to developers to ensure that each page "fragment" is XHTML valid before the site is completed but because of the dynamic nature I consider it a good idea to have a gateway test.

I'd be interested in any professional/commercial users opnions on this subject as well as the official CSE view.

Warm regards,

Steve Gibbings.

Gibbowolfie
Rank 0 - Newcomer
Rank 0 - Newcomer
Posts: 3
Joined: Wed Mar 23, 2005 6:01 am

Post by Gibbowolfie » Wed Mar 23, 2005 6:30 am

I reread the FAQ and saw the ASP entry which I completely missed before. Is there no way to automate the final site search? I realise the issues of data entry are the biggest problem there but with a big site (400 pages) then doing it by hand would be a nightmare. Perhaps there is a way of disabling certain followed links so that at least the majority of the site would be checked. i.e. to the data entry page but not following the link on that page. That would greatly reduce the manual effort.

Steve.

User avatar
Albert Wiersch
Site Admin
Site Admin
Posts: 3451
Joined: Sat Dec 11, 2004 9:23 am
Location: Near Dallas, TX
Contact:

Post by Albert Wiersch » Wed Mar 23, 2005 9:34 am

Hello Steve,

Since you mentioned the ASP entry, I assume that you read about using the intgrated web browser to browse the site and check it as you browse. That is one way but of course it is manual.

With the Batch Wizard, you can exclude parts of a site using an option in the Target List Options tab.

You may also be able to attach data to URLs (like the GET method does) and then check the URL in the Batch Wizard. Example: http://www.yourdomain.com/yourform.asp? ... ar3=value3

If the above would work for your site, then you could then find a way to create a list of URLs to check your pages and then manually create a Batch Wizard target list with the list of URLs:
http://www.htmlvalidator.com/htmlval/v7 ... ormat.html

You should be able to also validate those XHTML fragments directly in CSE, as CSE is designed to recognize fragments in many cases.
Image
Albert Wiersch

Gibbowolfie
Rank 0 - Newcomer
Rank 0 - Newcomer
Posts: 3
Joined: Wed Mar 23, 2005 6:01 am

Post by Gibbowolfie » Thu Mar 24, 2005 3:51 am

Albert,

Thanks for you reply, very much appreciated.

Yes I read about using the internal browser and that would work for small areas when a new features are added. In fact I can see a real benefit as any usability issues might be detected that wouldn't if it was automatic.

However I was more looking at automating a check as this could be run over the whole site as a gateway check before deployment. With various elements of the final page coming from different groups of people (In different countries) and different technology sources it's something I'd like to be able to do efficiently.

What you suggest might work but I'd have to check with the developers of the transactional engine. Unfortunately it isn't plain old ASP.

I shall post back my findings as they might be useful to others.

Thanks again and regards,

Steve.

Post Reply