RE: [cgmo-webcgm] QUESTION: ban 0x09, 0x0A, 0x0D from 'name' ?

[Robert Orosz <roboro@AUTO-TROL.com>]

OK, that raises the question, what is a format effector?  I assumed it was
talking about ANSI style escape sequences.  However, I don't see it defined
in sections 9.5.4.6 or 9.5.4.7 of CGM99.
 
What is a format effector?  Is it defined somewhere in CGM99 or one of the
normative references therein?
 
Rob

-----Original Message-----
From: Lofton Henderson [mailto:lofton@rockynet.com]
Sent: Tuesday, October 04, 2005 9:58 AM
To: Robert Orosz; 'Benoit Bezaire'; cgmo-webcgm@lists.oasis-open.org
Subject: RE: [cgmo-webcgm] QUESTION: ban 0x09, 0x0A, 0x0D from 'name' ?


At 09:32 AM 10/4/2005 -0600, Robert Orosz wrote:


Benoit and Lofton,

As I point out in my Chapter 3 review, T.14.5 prohibits the entire set of C0
control characters except for NUL.  So, the decision has already been made.
Section 3.1.1.3 needs some rewriting to clarify this.


T.14.5 says:
"Format effectors and ESC: Permitted Yes; Prohibited No;
Other C0 control codes (except NUL and ISO/IEC 2022 switching) are
prohibited."

Since these bad guys are format effectors, they are allowed by the PPF.

Here is another thought that has occurred to me during this dialog.  The 1.0
repertoire would allow, in the value of a 'name' ApsAttr, for example, the
character "<".  E.g., a valid value would be:
"my <special> group"

You can see where I'm going I guess:  XCF.  This would be problematic as an
XML attribute value, right?  (So can it be handled by CDATA section within
the XML attribute?  I'm not familiar enough to know the answer to that.)

Any case, I wonder if a warning note would be useful in 3.1.1.3, about
keeping in mind "all applicable XML rules", if the WebCGM content might ever
be the companion of an XCF.  Thoughts?

-Lofton.



-----Original Message-----
From: Benoit Bezaire [ mailto:benoit@itedo.com <mailto:benoit@itedo.com> ]
Sent: Tuesday, October 04, 2005 8:50 AM
To: cgmo-webcgm@lists.oasis-open.org
Subject: Re: [cgmo-webcgm] QUESTION: ban 0x09, 0x0A, 0x0D from 'name' ?


Did we get more feedback on this? was it discussed at the telecon?

This is essentially removing any whitespaces from 'name'. They would
become deprecated?

-- 
 Benoit   mailto:benoit@itedo.com <mailto:benoit@itedo.com> 

 
Monday, September 19, 2005, 6:30:09 PM, Lofton wrote:

LH> Ref: 
LH> http://lists.oasis-open.org/archives/cgmo-webcgm/200509/msg00093.html
<http://lists.oasis-open.org/archives/cgmo-webcgm/200509/msg00093.html> 
LH> Comments:  15
LH> ==========

LH> Probably not very controversial, but it is a technical change, strictly
LH> speaking...

LH> QUESTION:  should WebCGM fragment syntax ban 0x09, 0x0A, 0x0D from the
LH> 'name' production?

LH> DISCUSSION:  3.1.1.3, #3, 2nd bullet says, "shall not contain any
leading
LH> or trailing whitespace (#x09 | #x0a | #x0d | #x20)."  Dieter asks, "Does
LH> this mean, 0x09, 0x0A or 0x0D are fine in other places [within
objname]?"

LH> RECOMMENDATION:  as Dieter suggests, "They should be banned completely
from
LH> a name."

LH> Comments / objections?

LH> Regards,
LH> -Lofton.