[xml2rfc] <rfc obsoletes=...> handling in html output

fenner at gmail.com (Bill Fenner) Fri, 04 August 2006 06:05 UTC

From: fenner at gmail.com (Bill Fenner)
Date: Fri, 04 Aug 2006 06:05:01 +0000
Subject: [xml2rfc] <rfc obsoletes=...> handling in html output
In-Reply-To: <9C1CAB2B65E62D49A10E49DFCD68EF3E69523B@il-mail.actelis.net>
References: <9C1CAB2B65E62D49A10E49DFCD68EF3E69523B@il-mail.actelis.net>
Message-ID: <ed6d469d0608040604i4a063c87g460fc10817017b09@mail.gmail.com>
X-Date: Fri Aug 4 06:05:01 2006

Edward,

  Please try this patch.  It makes the space inside "(if approved)"
non-breaking and updates the parenthesis detection in the link
generator.

Thanks,
  Bill
-------------- next part --------------
A non-text attachment was scrubbed...
Name: if-approved.diff
Type: application/octet-stream
Size: 1335 bytes
Desc: not available
Url : http://drakken.dbc.mtview.ca.us/pipermail/xml2rfc/attachments/20060804/c76ebf03/if-approved.obj
>From fenner at gmail.com  Sat Aug  5 10:48:52 2006
From: fenner at gmail.com (Bill Fenner)
Date: Sat Aug  5 06:48:57 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <20060728191537.GQ11384@binky.Central.Sun.COM>
References: <CD9382A8-3CC3-4442-8A66-5EC6D7D2761A@cisco.com>
	 <20060728191537.GQ11384@binky.Central.Sun.COM>
Message-ID: <ed6d469d0608050648r1678ec00ke7efaa524276bd8f@mail.gmail.com>

Nico, Fred,

  I've implemented a texttable rendering mode that outputs dashed
lines instead of blank lines in tables.  ("What blank lines?", you
say?  Turn off compact.)

  Fred, with this patched xml2rfc, and the following patch to your source file:

--- RCP-Hop-by-hop.xml.orig     2006-08-05 09:43:13.000000000 -0400
+++ RCP-Hop-by-hop.xml  2006-08-05 09:44:06.000000000 -0400
@@ -182,7 +182,8 @@

         <?rfc needLines="30" ?>

-        <texttable anchor="table_example"
+        <?rfc compact="no" ?>
+        <texttable style="all" anchor="table_example"
                    title="Fields in RCP Hop-by-Hop Option">
           <preamble>The fields in the header are as follows:</preamble>

@@ -283,6 +284,7 @@

           <c></c>
         </texttable>
+        <?rfc compact="yes"?>

         <t>The premise is that any relevant number can be approximately
         represented as a "mantissa" (an integer) shifted left "exponent" bits

I get your edited output (except the header remains hyphens, not equals signs).

Note: I did not think *at*all* about the "style" name, or whether it
should be a different attribute for the table, or what - this is
solely for experimentation right now.

  Bill
-------------- next part --------------
A non-text attachment was scrubbed...
Name: table-style-all.diff
Type: application/octet-stream
Size: 1346 bytes
Desc: not available
Url : http://drakken.dbc.mtview.ca.us/pipermail/xml2rfc/attachments/20060805/23f6bbc0/table-style-all.obj
>From fenner at research.att.com  Sat Aug  5 08:32:48 2006
From: fenner at research.att.com (Bill Fenner)
Date: Sat Aug  5 07:32:53 2006
Subject: [xml2rfc] Reference indentation: per-section or across sections?
Message-ID: <200608051432.k75EWmJf028576@bright.research.att.com>


The RFC Editor has asked to have the same reference indentation across
multiple references sections.  When using <?rfc rfcedstyle="yes"?>,
the references are indented to the maximum length of the tag, and
the RFC Editor uses relatively short tags (i.e., never tags like
"I-D.weber-krb-wg-kerberos-initial-authentication").  When using <?rfc
rfcedstyle="no"?>, the references are indented to about 7 characters,
so a longer tag will appear on a different line.

If using rfcedstyle with long tags, you can end up with silly looking
cases like

5.1.  Normative References

   [I-D.weber-krb-wg-kerberos-initial-authentication]  Weber, J.,
                                                       "Kerberos Initial
                                                       Authentication
                                                       Methods", draft-
                                                       weber-krb-wg-
                                                       kerberos-initial-
                                                       authentication-00
                                                       (work in
                                                       progress), I-D
                                                       Status expired,
                                                       June 2003.

5.2.  Informative References

   [RFC2119]                                           Bradner, S., "Key
                                                       words for use in
                                                       RFCs to Indicate
                                                       Requirement
                                                       Levels",
                                                       RFC 2119, BCP 14,
                                                       Current
                                                       Status BEST
                                                       CURRENT PRACTICE,
                                                       March 1997.

However, without rfcedstyle, you get the existing output [no change]:

5.1.  Normative References

   [I-D.weber-krb-wg-kerberos-initial-authentication]
              Weber, J., "Kerberos Initial Authentication Methods",
              draft-weber-krb-wg-kerberos-initial-authentication-00
              (work in progress), I-D Status expired, June 2003.

5.2.  Informative References

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", RFC 2119, BCP 14, Current Status BEST
              CURRENT PRACTICE, March 1997.

I think the change to having the indentation the same across all
references sections will affect a very small corner case (those using
symrefs="yes" but very very short symref tags; shorter than 'RFCxxxx',
or those using symrefs="no" and the first section has fewer than 10
references but there are more than 10 combined), so I don't intend to
make the change conditional on rfcedstyle.  Does anyone have an objection
to that?

Thanks,
  Bill
>From EdwardB at actelis.com  Sat Aug  5 21:42:42 2006
From: EdwardB at actelis.com (Edward Beili)
Date: Sat Aug  5 10:42:50 2006
Subject: [xml2rfc] <rfc obsoletes=...> handling in html output
Message-ID: <9C1CAB2B65E62D49A10E49DFCD68EF3E69534D@il-mail.actelis.net>

Bill,
Your patch has solved the problem.
Thanks,
-Edward

-----Original Message-----
From: Bill Fenner [mailto:fenner@gmail.com] 
Sent: Friday, August 04, 2006 16:05
To: Edward Beili
Cc: xml2rfc@lists.xml.resource.org
Subject: Re: [xml2rfc] <rfc obsoletes=...> handling in html output

Edward,

  Please try this patch.  It makes the space inside "(if approved)"
non-breaking and updates the parenthesis detection in the link generator.

Thanks,
  Bill