Validation errors in the Guidelines HTML output
Hi all, I just discovered the VNU HTML validator, and I'm using it on another project; just for fun I ran it on a clean Guidelines build, and there seem to be about 150 errors, mostly minor and easily fixable. But it caused me to wonder why our build process didn't catch them, and I'm going to look into that. Validating modern HTML is notoriously difficult, but we should do the best we can, I think. Cheers, Martin
Just out of curiosity, are we using any of the HTML5 features that would prevent us from validating in standards-mode as XHTML1? pfs On Fri, Aug 14, 2015, at 08:46, Martin Holmes wrote:
Hi all,
I just discovered the VNU HTML validator, and I'm using it on another project; just for fun I ran it on a clean Guidelines build, and there seem to be about 150 errors, mostly minor and easily fixable. But it caused me to wonder why our build process didn't catch them, and I'm going to look into that. Validating modern HTML is notoriously difficult, but we should do the best we can, I think.
Cheers, Martin -- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council
PLEASE NOTE: postings to this list are publicly archived -- Paul Schaffner Digital Library Production Service PFSchaffner@umich.edu | http://www.umich.edu/~pfs/
On 15-08-14 07:48 AM, Paul Schaffner wrote:
Just out of curiosity, are we using any of the HTML5 features that would prevent us from validating in standards-mode as XHTML1?
Good question. I don't believe so, except that we're using the doctype for HTML5: <!DOCTYPE html SYSTEM "about:legacy-compat"> The interesting thing about the vnu checker is that it's a generic HTML checker, if I understand it correctly. Its the engine behind the W3C service now, and it's sensitive to HTML vs XHTML and provides some additional helpful output. One thing it's thrown up in a number of my projects (and in the Guidelines) is files with text runs that are not in Normalization Form C; that's a good thing to know about and and easy fix. It's also very fast and easy to use, so I'm thinking that whatever we're currently using in the build process, we could switch to vnu. Actually, if you look at validatehtml.xml: <project name="example" default="validate" basedir="."> <taskdef name="jing" classname="com.thaiopensource.relaxng.util.JingTask"/> <target name="validate"> <jing compactsyntax="true" rngfile=" ../../xhtml.rnc"> <fileset dir="Guidelines-web/en/html" includes="**/z*.html"/> </jing> </target> </project> it rather looks like the "z" there is disabling validation entirely (I don't think we have any files starting with "z"), but I haven't looked closely enough yet. It's possible that in the transition to HTML5, validation got disabled and was never re-enabled. Cheers, Martin
pfs
On Fri, Aug 14, 2015, at 08:46, Martin Holmes wrote:
Hi all,
I just discovered the VNU HTML validator, and I'm using it on another project; just for fun I ran it on a clean Guidelines build, and there seem to be about 150 errors, mostly minor and easily fixable. But it caused me to wonder why our build process didn't catch them, and I'm going to look into that. Validating modern HTML is notoriously difficult, but we should do the best we can, I think.
Cheers, Martin -- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council
PLEASE NOTE: postings to this list are publicly archived
participants (2)
-
Martin Holmes -
Paul Schaffner