Showing information from all \*only commands in XML output

[albert-github]

This fix reverts the changes from #381 (i.e issue #4214 "htmlonly content appears in generated XML output (Origin: bugzilla 646002)", as XML is seen as an output format that can be further processed and therefore should contain all possible information.

[mosra]

Thanks for doing this! 👍

One "stretch goal" -- the documentation is stating this:

When you want to insert a HTML fragment that has block scope like a table or list which should appear outside <p>..</p>, this can lead to invalid HTML. You can use \htmlonly[block] to make doxygen end the current paragraph and restart it after \endhtmlonly.

Ideally that would hold also for the XML output, so either <htmlonly> being inside a <para>, or <htmlonly> not being wrapped in a <para> (instead of current <htmlonly block="yes"> being inside a <para>). Basically so it doesn't need to be extracted out of the paragraph on the client side.

How hard would it be to implement it like this?

[mosra]

In other words, the output looking like this instead:

<para>Text. <htmlonly> 
HTML
</htmlonly></para>
<htmlonly>
HTML with block
</htmlonly>
<para><rtfonly> 
RTF
</rtfonly> <manonly> 
Man
</manonly> <latexonly> 
LaTeX
</latexonly>

[albert-github]

No idea at the moment how difficult this would be and whether this would be conceptual OK (as it is not really a "dump" anymore).
When really necessary I think best is to submit an issue for this as soon as this proposed pull request has been integrated in master on github.

[mosra]

Okay, no problem. Asked only in case it would be easy to do on your side -- I can do the paragraph extraction on my side as well ;)