No, I think you've got the zen exactly right.
The incomplete coverage is historical. It's sort of grown organically
with the tags people wanted most coming first.
The downside of too many tags is that it gets more rigid. The closing
tags are expected and the penalty for their absence may be a bit harsh.
You're welcome to contribute the classes back, but there may be a bit of
push-back from users of the current set. Maybe not.
There are probably ways to add the new tags without breaking existing
applications.
Martin Hudson wrote:
> First, I am grateful for all the work that has been done to produce
> this project.
>
> Second, I noticed that the tag classes defined are not a complete
> representation of the HTML tags available. I may have missed a
> fundamental usage approach but noticed, for example, that there is no
> ‘H2’ specific tag. The filters appear to find the opening tag as “H2”
> but it is not associated with the end tag in any way, nor is the text
> that appears between it easily extractable without resorting to serial
> processing of the list. So, am I missing something in the approach or
> is the tag list incomplete?
>
> Third, due perhaps to my ignorance about the overall ‘zen’ of the
> parser, I created a clone of the Span.java class, edited it to create
> an H2Tag.java class, and then registered it as a compound tag. It
> works beautifully. This worked for many other tags. So, if I am not
> completely missing the point, should I contribute these back to
> increase the tag coverage?
>
> Martin N. Hudson
>
> devIS - Development InfoStructure
>
|
