[xml2rfc] ABNF parser broken?

[fenner at gmail.com (Bill Fenner)]

On Jan 27, 2008 10:47 AM, Pete Resnick <[e-mail]> wrote:
> Is it just my document (see
> <http://resnick1.qualcomm.com/2822upd.xml> for a sample), or is the
> ABNF parser on xml2rfc busted?

This is something that I started struggling with when I started trying
to get a consistent story about spacing.

You use:

                    <figure>
<artwork type="abnf">quoted-pair     =   ("\" (VCHAR / WSP)) / obs-qp</artwork>
                    </figure>

which is perfectly sensible, except that the validator wants vertical
whitespace - it ignores the first line (or assumes that it's blank, or
I don't quite know what) and it wants a trailing newline (after all,
lines end with newline).  So I think you'll find that

                    <figure>
<artwork type="abnf">
quoted-pair     =   ("\" (VCHAR / WSP)) / obs-qp
</artwork>
                    </figure>

validates (but may or may not give you the spacing properties you
want, see previous endless discussion).

  Bill
>From lars.eggert at nokia.com  Mon Jan 28 11:18:50 2008
From: lars.eggert at nokia.com (Lars Eggert)
Date: Mon Jan 28 01:20:25 2008
Subject: [xml2rfc] Mr I. Property is authoring a lot of drafts...
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>

On 2008-1-25, at 19:23, ext Marshall Rose wrote:
>> the format of 1id-index.txt seems to have changed. It now indicates  
>> whether an IPR disclosure has been filed for a draft, e.g.,
>>
>>  "Profile for Datagram Congestion Control Protocol (DCCP)  
>> Congestion ID 4:
>>  TCP-Friendly Rate Control for Small Packets (TFRC-SP)", Sally  
>> Floyd, Eddie
>>  Kohler, Intellectual Property, 18-Nov-07, <draft-ietf-dccp- 
>> ccid4-01.txt,.ps>
> ...
>> I'd be good to fix this before the IETF submission deadline comes.
>
> thanks for catching this. the scripts have been updated to ignore  
> this. it should take about 24 hours before the files themselves are  
> corrected.

Thanks for fixing this!

However, note that there are other issues that still exist with 1id- 
index.txt. For example:

   "TCP Friendly Rate Control (TFRC): Protocol Specification",  
University
   London, Sally Floyd, Jitendra Padhye, Joerg Widmer, Intellectual  
Property,
   25-Jan-08, <draft-ietf-dccp-rfc3448bis-04.txt,.pdf>

I don't know why it says "University London" as an author, instead of  
"Mark Handley" (or why the "College" is missing.)

Something's still fishy with the scripts.

Lars
>From julian.reschke at gmx.de  Mon Jan 28 10:20:30 2008
From: julian.reschke at gmx.de (Julian Reschke)
Date: Mon Jan 28 01:20:49 2008
Subject: [xml2rfc] cref  or discussion location
In-Reply-To: <p06250116c3c28e475cf5@[192.168.2.1]>
References: <p06250116c3c28e475cf5@[192.168.2.1]>
Message-ID: <[e-mail]>

Pete Resnick wrote:
...
> - Is there any reason that cref is defined as taking CTEXT as against 
> being identical to t? I really wanted to include an eref in a cref.

So do I (1st use case: use cref to hyperlink to a bug tracking system).

rfc2629.xslt allows that for HTML output (it includes an XSLT that 
strips extensions so that xml2rfc can be used for TXT generation).

> ...

BR, Julian
>From mrose at dbc.mtview.ca.us  Mon Jan 28 09:31:35 2008
From: mrose at dbc.mtview.ca.us (Marshall Rose)
Date: Mon Jan 28 09:31:47 2008
Subject: [xml2rfc] Mr I. Property is authoring a lot of drafts...
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>


> However, note that there are other issues that still exist with 1id- 
> index.txt. For example:
>
>   "TCP Friendly Rate Control (TFRC): Protocol Specification",  
> University
>   London, Sally Floyd, Jitendra Padhye, Joerg Widmer, Intellectual  
> Property,
>   25-Jan-08, <draft-ietf-dccp-rfc3448bis-04.txt,.pdf>
>
> I don't know why it says "University London" as an author, instead  
> of "Mark Handley" (or why the "College" is missing.)
>
> Something's still fishy with the scripts.

sorry, it's not the scripts, it's the 1id-abstracts.txt file that the  
secretariat is putting out. the scripts parse the file based on the  
same rules that have been in place for the last N years.

last week, i added a thing to deal with the "Intellectual Property".

if someone has a thing of lists that we should ignore like  
"University London", i guess i can add those.

however, perhaps the secretariat might want to review their  
procedures for producing this file, and understanding why it's gone  
sideways...

/mtr