Re: [xml2rfc-dev] [Ext] Re: [Rfc-markdown] [xml2rfc] <br> is back, was: New xml2rfc release: v2.32.0

[Julian Reschke <[e-mail]>]

On 05.11.2019 20:59, Heather Flanagan wrote:
>
>
>> On Nov 5, 2019, at 2:47 PM, Julian Reschke <[e-mail]> wrote:
>>
>> On 05.11.2019 20:37, Heather Flanagan wrote:
>>>
>>>
>>>> On Nov 5, 2019, at 2:30 PM, Julian Reschke <[e-mail]> wrote:
>>>>
>>>> On 05.11.2019 19:55, Heather Flanagan wrote:
>>>>> ...
>>>>>
>>>>> Matching what you see aside, do you have any concerns with my requirements?
>>>>>
>>>>> Heather
>>>>
>>>> Yes, of course.
>>>
>>> To be clear, for my own sanity:
>>>
>>> Yes, of course you have concerns with the functional requirement I described, or
>>
>> You said that "<br/>" should be ignored when the document/section title
>> is referenced. I believe that's wrong; it should be treated as whitespace.
>
> OK. I’m assuming it’s an implementation detail to make sure we don’t end up with double spaces, then?

That's actually a good question. In general, in text content, we already
collapse multiple whitespace characters to a single one (exceptions are
only <artwork> and <sourcecode>).

>>> Yes, of course you have concerns with how the requirement is technically implemented?
>>
>> Well, xml2rfc currently does not implement <br>, but is does implement
>> the unicode line break character
>> (<https://www.fileformat.info/info/unicode/char/2028/index.htm>). *That*
>> implementation carries over the line break into the ToC, and I believe
>> that's wrong, and it also does not seem to be what you want. (I did not
>> check what happens with <xref> or <reference>, but I can check that if
>> that's of interest).
>
> It’s something fixable. I expect Henrik is working on it.

Yes, sure. But that requires somebody to say what the expected output
is. That's what I've been asking for the whole time.

>>> ...
>>> How <br> is implemented in the different outputs is going to be contingent on the requirements of that format and what is considered valid. So, of course we want valid HTML. Whether that means a transformation from what the XML says into what the HTML requires is an implementation I don’t have an opinion on. My job is to offer the functional requirements, not tell the developer how to do the job.
>>> ...
>>
>> I agree with that, but I note that RFC 7992 tried to actually do that :-).
>
> We’re trying to do a lot of things.

Let me rephrase this. I always thought that RFC 7992 is way over the top
in making requirements on the HTML output. I think when doing 7992bis,
we'll need to take out lots of these requirements, and just let the
developer find the best HTML output.

Best regards, Julian