RE: [cgmo-webcgm] a question about font-name normalization

["Cruikshank, David W" <david.w.cruikshank@boeing.com>]

I recollection is both fonts will be normalized.

 

Dave

 

Technical Fellow - Graphics/Digital Data Interchange
Boeing Commercial Airplane
206.544.3560, fax 206.662.3734
david.w.cruikshank@boeing.com

 

---

From: Lofton Henderson [mailto:lofton@rockynet.com]
Sent: Monday, June 16, 2008 6:13 PM
To: cgmo-webcgm@lists.oasis-open.org
Subject: [cgmo-webcgm] a question about font-name normalization

All --

From the Needham minutes:

[[[
Font mapping syntax
o The question 
of case sensitivity for font names was discussed.
.. Resolution: Full normalization of font names will be 
implemented by the
vendors (names are converted to lower case and dashes, 
underscores,
and whitespace are compressed out) before mapping is 
attempted
o The question of the usefulness of the font scaling 
attributes was discussed.
.. Resolution: 
Remove the scale factors
o 
The two use cases of “conditional” and 
“unconditional” font mapping were
discussed at length. Conditional font 
mapping means that if a font is available on
a system no font mapping will 
take place. Unconditional font mapping forces the
mapping to take place 
whether a font is available on a system or not.
Unconditional font mapping is 
really a styling function. All of the vendors
currently to unconditional font 
mapping.
.. Resolution: Add an attribute 
to control whether mapping is forced or not
o A proposal 
from Airbus was submitted by Ulrich dealing with a list of fonts 
for
substitution. After much discussion, the group decided to allow a list of 
fonts in
the font mapping target using a CSS style font list. Display might 
contain
something like “Arial, ‘Times New Roman’, monospace”.
.. 
Resolution: Redesign ACI syntax to support CSS 
style font list
o Proposed new syntax snipped for font 
mapping
<!ELEMENT fontMap 
( defaultFont?, maplist* ) >
<!ELEMENT defaultFont EMPTY 
>
<!ATTLIST defaultFont
useFont CDATA #REQUIRED >
<!ELEMENT 
maplist EMPTY >
<!ATTLIST maplist
forceSubstitution ( yes | no ) 
“yes”
cgmFont CDATA #REQUIRED
substitutionList CDATA #REQUIRED 
>
]]]

Okay, here is a snapshot:

1.) the cgmFont gets 
the above full-normalization treatment -- whitespace, upper-lower case, HYPHEN, 
UNDERSCORE (some of this being ISOLatin1 dependent, as we observed);

2.) 
substitutionList uses CSS syntax and gets the CSS normalization 
treatment.

QUESTION.  When the processor is attempting a match of 
cgmFont against the font names in the metafile, does it:
1a.) do the full 
normalization on the font names in the metafile?
1b.) or do the full 
normalization on the "cgmFont" XML attribute value?
1c.) or 
both?

-Lofton.