You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After setting name to value contentequiv, I received another error:
Error while parsing file: value of attribute "name" is invalid; must be equal to "alternate-representation"
At this point I was able to resolve the issue and validate the epub.
My question is two-fold: Should this be flagged at this level of severity? Could the error report elaborate more on suggested values of attribute name?
And for my own edification, where is this requirement coming from?
The text was updated successfully, but these errors were encountered:
this has not come up in previous versions of epubcheck.
Right, this was introduced for v4.2.0 in 47d4926. We now need to tweak that to adapt to the latest changes mentioned above.
Should this be flagged at this level of severity?
The new rules now only say:
When Content MathML is included as per the previous condition, the given annotation-xml element's encoding attribute MUST be set to either of the functionally-equivalent values MathML-Content or application/mathml-content+xml, and its name attribute MUST be set to contentequiv.
So in your case (as far as I understand you have XHTML content in annotation-xml), there will be no restriction other than what is specified in HTML – i.e. the content model must be flow HTML content.
But if the annotation-xml contains Content MathML, the absence or incorrect value for the name attribute will still be reported as an ERROR, as this is a MUST in the spec.
Could the error report elaborate more on suggested values of attribute name?
Unfortunately this message is generated by the schema validator, so we cannot easily control the message details…
And for my own edification, where is this requirement coming from?
This was originally intended to restrict the element to fallback content, see the discussion in w3c/epub-specs#1212. But this point is moot now that this restriction has been removed in the spec :-)
Searching around, I was able to find what I thought was the solution to my problem (https://www.idpf.org/epub/30/spec/epub30-contentdocs.html#sec-xhtml-mathml-conf-content), but this has not come up in previous versions of epubcheck.
After setting name to value contentequiv, I received another error:
At this point I was able to resolve the issue and validate the epub.
My question is two-fold: Should this be flagged at this level of severity? Could the error report elaborate more on suggested values of attribute name?
And for my own edification, where is this requirement coming from?
The text was updated successfully, but these errors were encountered: