Re: [xml2rfc-dev] [Ext] RFC 7991 issue #37: Schema Issue, RFC 7991, In Section 2.12, <br>

[Julian Reschke <[e-mail]>]

On 10/4/2018 4:52 PM, Paul Hoffman wrote:
> On Oct 4, 2018, at 7:34 AM, Henrik Levkowetz <[e-mail]> wrote:
>>
>>
>> On 2018-10-04 16:25, Paul Hoffman wrote:
>>> On Oct 3, 2018, at 9:31 PM, Henrik Levkowetz <[e-mail]> wrote:
>>>>> Because putting a <br> in running text prevents you from being able
>>>>> to search for the whole phrase. This was one of the explicit reasons
>>>>> to prevent them everywhere, and titles and section names are
>>>>> particularly important examples. You might search for a title of "of
>>>>> Generating Datagrams" but would never think of searching for "of
>>>>> Generating<br>Datagrams" or even . "of Generating <br>Datagrams"
>>>>
>>>> I'm sorry, but I think this is incorrect.  In rendered context, whether
>>>> text/plain, text/html, or application/pdf, this would not be an issue --
>>>> the search tools would not see any <br> (in text and pdf because it
>>>> would not be part of the text; in html, because html search tools already
>>>> disregard html's <br>).
>>>>
>>>> In the context of an editor, working on the xml, a phrase search is just
>>>> as likely to fail because the text has been broken across a line, and
>>>> contains multi-space indentation, which again would prevent an in-editor
>>>> phrase search.  This is a reality I live with every day ,:-)
>>>
>>> Neither of those is what I was referring to: I was referring to searching the corpus of RFCs or Internet Drafts.
>>
>> Using which search tool?
> 
> grep for now, hopefully something more formal later when the corpus has more XML.

That doesn't make sense.

Using grep to search in markup files doesn't work reliably -- <br> 
wouldn't make things any more worse than they are already.

Best regards, Julian