[xml2rfc] List related bug in r1.31.

[fenner at research.att.com (Bill Fenner)]

>The nested list item has style="empty" (explicitly) in the xml but
>appears to be inheriting the outer list style.

Your example list doesn't have any style attribute in the
file that you posted.  (One possibility is that your XML editor
is using a DTD that says that the default is "empty" so removes the
attribute when writing if it's set to the default - the 1.31
DTD uses #IMPLIED instead of having a default, so this would be
fixed by updating the DTD; if you're using xxe I have an update
that I should release...)

Are the other places that you see this behavior

<t>"Historically the same machinery is used for both. While

and

<t>Editor's Note: Technology is running ahead of imagination

?

Those are the other two lists selected by //list[@style]//list[not(@style)]
(i.e., the list style inheritance would occur).

  Bill
>From elwynd at dial.pipex.com  Tue Sep  5 08:34:43 2006
From: elwynd at dial.pipex.com (Elwyn Davies)
Date: Mon Sep  4 23:34:52 2006
Subject: [xml2rfc] List related bug in r1.31.
In-Reply-To: <[e-mail]>
References: <[e-mail]>
Message-ID: <[e-mail]>



Bill Fenner wrote:
>> The nested list item has style="empty" (explicitly) in the xml but
>> appears to be inheriting the outer list style.
>>     
>
> Your example list doesn't have any style attribute in the
> file that you posted.  (One possibility is that your XML editor
> is using a DTD that says that the default is "empty" so removes the
> attribute when writing if it's set to the default - the 1.31
> DTD uses #IMPLIED instead of having a default, so this would be
> fixed by updating the DTD; if you're using xxe I have an update
> that I should release...)
>   
Ah!  I must admit I was relying on what xxe told me...I didn't check the 
raw XML for that instance.



> Are the other places that you see this behavior
>
> <t>"Historically the same machinery is used for both. While
>
> and
>
> <t>Editor's Note: Technology is running ahead of imagination
>
> ?
>
> Those are the other two lists selected by //list[@style]//list[not(@style)]
> (i.e., the list style inheritance would occur).
>   
I believe they were. I suspect that those two were inserted (by my 
co-editor) using xxe without explicitly setting the style now you 
mention it.

/Elwyn
>   Bill
>   
>From fenner at research.att.com  Tue Sep  5 06:15:40 2006
From: fenner at research.att.com (Bill Fenner)
Date: Tue Sep  5 05:15:47 2006
Subject: [xml2rfc] List related bug in r1.31.
References: <[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>


>Ah!  I must admit I was relying on what xxe told me...I didn't check the 
>raw XML for that instance.

The attribute editor will tell you, although subtly.  "style empty" will be
gray in the "attribute/value" table if it's defaulting, black if it's
actually present in the XML.

  Bill
>From elwynd at dial.pipex.com  Tue Sep  5 14:55:27 2006
From: elwynd at dial.pipex.com (Elwyn Davies)
Date: Tue Sep  5 05:55:29 2006
Subject: [xml2rfc] List related bug in r1.31.
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>



Bill Fenner wrote:
>> Ah!  I must admit I was relying on what xxe told me...I didn't check the 
>> raw XML for that instance.
>>     
>
> The attribute editor will tell you, although subtly.  "style empty" will be
> gray in the "attribute/value" table if it's defaulting, black if it's
> actually present in the XML.
>
>   Bill
>   
OK.. My awareness is now primed.  Presumably xxe displays "style empty" 
in the main window for the default because it isn't the empty string?  
Is that what your proposed fix was about - 'cos suppressing "style 
empty" would avoid this sort of thing?

It is a little confusing for nested lists since there is a difference 
between default and empty for nested lists.
Suggestion: Add style="inherited", make it the default and behave like 
the existing default.

/Elwyn

/Elwyn
>From fenner at research.att.com  Tue Sep  5 08:52:39 2006
From: fenner at research.att.com (Bill Fenner)
Date: Tue Sep  5 07:52:47 2006
Subject: [xml2rfc] List related bug in r1.31.
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>


>OK.. My awareness is now primed.  Presumably xxe displays "style empty" 
>in the main window for the default because it isn't the empty string?  

Up until xml2rfc 1.31, the DTD was mismatched with the behavior.
The DTD said "if there is no attribute, then it means 'empty'", so
xxe was believing the DTD.  However, xml2rfc has the inheritance
behavior that's documented.

The ideal thing would be for xxe to display the actual style used;
that's a bit difficult (I don't think it's feasible to express
this style attribute inheritance in CSS, so I'd have to write some
java to do it, and that just bumps it way down the priority list).
With the updated DTD, I can at least make the "[list style ___]"
correct, but it's harder to display the tags properly.

>It is a little confusing for nested lists since there is a difference 
>between default and empty for nested lists.
>Suggestion: Add style="inherited", make it the default and behave like 
>the existing default.

Yeah, there's been some discussion of something like this, and I actually
thought that the inherited list styling was going away in 1.31, but I
must have remembered the discussion wrong.

  Bill
>From fenner at gmail.com  Tue Sep  5 12:06:54 2006
From: fenner at gmail.com (Bill Fenner)
Date: Tue Sep  5 08:06:58 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <[e-mail]>
References: <[e-mail]>
Message-ID: <[e-mail]>

Fred, Nico,

  Have you had a chance to try the patch that I sent for table
formatting?  I'd like to see if it got you the formatting you wanted,
if you turn off <?rfc compact?> around the table and use style="all".
(If you don't want to patch your local xml2rfc, the experimental web
form is running the patched code [I have an update that makes the html
and txt forms consistent and doesn't require mucking with <?rfc
compact?>, but wanted to get some feedback before doing further
updates)

Thanks,
  Bill
>From elwynd at dial.pipex.com  Tue Sep  5 19:25:16 2006
From: elwynd at dial.pipex.com (Elwyn Davies)
Date: Tue Sep  5 10:25:20 2006
Subject: [xml2rfc] List related bug in r1.31.
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>



Bill Fenner wrote:
>> It is a little confusing for nested lists since there is a difference 
>> between default and empty for nested lists.
>> Suggestion: Add style="inherited", make it the default and behave like 
>> the existing default.
>>     
>
> Yeah, there's been some discussion of something like this, and I actually
> thought that the inherited list styling was going away in 1.31, but I
> must have remembered the discussion wrong.
>
>   Bill
>   
The thing that was agreed as I understand it was:
- style, hangIndent and format string would be inherited
- counter would not be inherited and defaults to the 'internal' list 
item counter if not specified (the predefined 'fmt' counter has been 
abolished).

This stops inherited 'format' lists screwing up cos they don't know what 
counter to use (a bug in 1.30)

/Elwyn
>From Nicolas.Williams at sun.com  Tue Sep  5 14:20:16 2006
From: Nicolas.Williams at sun.com (Nicolas Williams)
Date: Tue Sep  5 11:21:19 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>

On Tue, Sep 05, 2006 at 11:06:54AM -0400, Bill Fenner wrote:
> Fred, Nico,
> 
>  Have you had a chance to try the patch that I sent for table
> formatting?  I'd like to see if it got you the formatting you wanted,
> if you turn off <?rfc compact?> around the table and use style="all".
> (If you don't want to patch your local xml2rfc, the experimental web
> form is running the patched code [I have an update that makes the html
> and txt forms consistent and doesn't require mucking with <?rfc
> compact?>, but wanted to get some feedback before doing further
> updates)

Was the patch against 1.30 or 1.31?  Ah, 1.31.

Well, running 1.31 I get

xml2rfc: error: xml2rfc: error: missing Normative/Informative References
around input line 32

Context (format:  "file_basename:line_in_file:#elem_num:<elem ...>"):
    kitten-gssapi-extensions-iana-02.xml:31:#1:<rfc ipr="full3978"
    docName="draft-ietf-kitten-gssapi-extensions-iana-01.txt">

Er, why?

I added rfc3978 to the references section, and still no dice.
>From fenner at gmail.com  Tue Sep  5 15:57:18 2006
From: fenner at gmail.com (Bill Fenner)
Date: Tue Sep  5 11:57:27 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	 <[e-mail]>
	 <[e-mail]>
Message-ID: <[e-mail]>

On 9/5/06, Nicolas Williams <[e-mail]> wrote:
> xml2rfc: error: xml2rfc: error: missing Normative/Informative References
> around input line 32
>
> Context (format:  "file_basename:line_in_file:#elem_num:<elem ...>"):
>     kitten-gssapi-extensions-iana-02.xml:31:#1:<rfc ipr="full3978"
>     docName="draft-ietf-kitten-gssapi-extensions-iana-01.txt">
>
> Er, why?

With <?rfc strict="yes"?>, xml2rfc tries to enforce some of the rules
such as splitting references into normative and informative.  If you
don't have at least one references section with the
[case-insignificant] title "Normative References", "Normative
Reference", "Informative References", "Informative Reference", then
you will get this error.  (It would be nice if the error was reported
on the references element, not the rfc element).

If you just want to test the patch, turn off <?rfc strict?>.

If you do have a references section of the right kind, (i.e., the
<?rfc strict?> check is buggy) please send me the xml to test with.

  Bill
>From Nicolas.Williams at sun.com  Tue Sep  5 15:46:49 2006
From: Nicolas.Williams at sun.com (Nicolas Williams)
Date: Tue Sep  5 12:47:28 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>

On Tue, Sep 05, 2006 at 02:57:18PM -0400, Bill Fenner wrote:
> On 9/5/06, Nicolas Williams <[e-mail]> wrote:
> >xml2rfc: error: xml2rfc: error: missing Normative/Informative References
> >around input line 32
> >
> >Context (format:  "file_basename:line_in_file:#elem_num:<elem ...>"):
> >    kitten-gssapi-extensions-iana-02.xml:31:#1:<rfc ipr="full3978"
> >    docName="draft-ietf-kitten-gssapi-extensions-iana-01.txt">
> >
> >Er, why?
> 
> With <?rfc strict="yes"?>, xml2rfc tries to enforce some of the rules
> such as splitting references into normative and informative.  If you
> don't have at least one references section with the
> [case-insignificant] title "Normative References", "Normative
> Reference", "Informative References", "Informative Reference", then
> you will get this error.  (It would be nice if the error was reported
> on the references element, not the rfc element).

Ah, I had misnamed my references section.

Anyways, no, your patch did not cause a separator to be printed between
table rows...

Nico
-- 
>From fenner at gmail.com  Tue Sep  5 20:22:49 2006
From: fenner at gmail.com (Bill Fenner)
Date: Tue Sep  5 16:22:57 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	 <[e-mail]>
	 <[e-mail]>
	 <[e-mail]>
	 <[e-mail]>
Message-ID: <[e-mail]>

On 9/5/06, Nicolas Williams <[e-mail]> wrote:
> Anyways, no, your patch did not cause a separator to be printed between
> table rows...

Did you use table style="all" and <?rfc compact="no"?> ?

If so, would you please send me your xml file?

Thanks,
  Bill
>From Nicolas.Williams at sun.com  Tue Sep  5 19:29:02 2006
From: Nicolas.Williams at sun.com (Nicolas Williams)
Date: Tue Sep  5 16:29:41 2006
Subject: [xml2rfc] Question about tables
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>

On Tue, Sep 05, 2006 at 07:22:49PM -0400, Bill Fenner wrote:
> On 9/5/06, Nicolas Williams <[e-mail]> wrote:
> >Anyways, no, your patch did not cause a separator to be printed between
> >table rows...
> 
> Did you use table style="all" and <?rfc compact="no"?> ?
                    ^^^^^^^^^^^

Ah, that does it.  Sorry I missed that.

Nico
-- 
>From bortzmeyer at nic.fr  Wed Sep  6 13:32:02 2006
From: bortzmeyer at nic.fr (Stephane Bortzmeyer)
Date: Wed Sep  6 03:32:07 2006
Subject: [xml2rfc] can't read "rv(anchor)": no such element in array
Message-ID: <[e-mail]>

I've never seen that error and Google does no find it either:

% DISPLAY= xml2rfc sample-draft.xml sample-draft.txt
xml2rfc: error: can't read "rv(anchor)": no such element in array around input line 2

The sample-draft.xml starts with:

<?xml version="1.0"?>
<rfc ipr="full3978" docName="draft-bortzmeyer-foo-bar-00">
<front>
<title>foo</title>

And rnv and xmllint finds it valid.
>From bortzmeyer at nic.fr  Wed Sep  6 13:50:10 2006
From: bortzmeyer at nic.fr (Stephane Bortzmeyer)
Date: Wed Sep  6 03:50:15 2006
Subject: [xml2rfc] Re: can't read "rv(anchor)": no such element in array
In-Reply-To: <[e-mail]>
References: <[e-mail]>
Message-ID: <[e-mail]>

On Wed, Sep 06, 2006 at 12:32:02PM +0200,
 Stephane Bortzmeyer <[e-mail]> wrote 
 a message of 17 lines which said:

> xml2rfc: error: can't read "rv(anchor)": no such element in array around input line 2

Sorry, it was just a dangling <reference> (and the error message is
misleading). It works now.
>From fenner at gmail.com  Wed Sep  6 20:37:38 2006
From: fenner at gmail.com (Bill Fenner)
Date: Wed Sep  6 16:37:46 2006
Subject: [xml2rfc] Re: can't read "rv(anchor)": no such element in array
In-Reply-To: <[e-mail]>
References: <[e-mail]>
	 <[e-mail]>
Message-ID: <[e-mail]>

On 9/6/06, Stephane Bortzmeyer <[e-mail]> wrote:
> On Wed, Sep 06, 2006 at 12:32:02PM +0200,
>  Stephane Bortzmeyer <[e-mail]> wrote
>  a message of 17 lines which said:
>
> > xml2rfc: error: can't read "rv(anchor)": no such element in array around input line 2
>
> Sorry, it was just a dangling <reference> (and the error message is
> misleading). It works now.

Out of curiosity, does http://rtg.ietf.org/~fenner/ietf/xml2rfc-valid/
catch the problem?

Thanks,
  Bill
>From nobody at xyzzy.claranet.de  Tue Sep 19 08:00:18 2006
From: nobody at xyzzy.claranet.de (Frank Ellermann)
Date: Mon Sep 18 22:01:36 2006
Subject: [xml2rfc] 
	Re: Protocol Action: 'RFC 3978 Update to recognize the IETF 
	Trust' to BCP
References: <[e-mail]>
Message-ID: <[e-mail]>

The IESG wrote:

> - 'RFC 3978 Update to recognize the IETF Trust '
>    <draft-ietf-ipr-ietf-trust-update-03.txt> as a BCP
[...]
> http://www.ietf.org/internet-drafts/draft-ietf-ipr-ietf-trust-update-03.txt

> Implementation requires the community to be notified of minor
> changes to the RFC boilerplate.

Apparently the following two details are relevant for xml2rfc:

| Replace "Internet Society" with "IETF Trust" in the copyright
| statement in Section 5.4.

That gives (first line changed, quotes removed):

: Copyright (C) The IETF Trust (year).
:
: This document is subject to the rights, licenses and restrictions
: contained in BCP 78, and except as set forth therein, the authors
: retain all their rights.

Second detail:

| In Section 5.5 insert ", THE IETF TRUST" after "INTERNET SOCIETY".

That yields new screaming legalese (quotes removed):

: This document and the information contained herein are provided
: on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE
: REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY,
: THE IETF TRUST AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM
: ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
: ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT
: INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY
: OR FITNESS FOR A PARTICULAR PURPOSE.

Frank