cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Attributes as upper and lowercase

unknown1
1-Newbie

Attributes as upper and lowercase

Adepters

I am hoping someone in this group might have found a workaround for this.
Arbortext help support is currently looking at this issue as well.

THE PROBLEM:
I am typing in attributes in uppercase (i.e. "AB") and they are fine for that
session. When I save the document close it and reopen it the attributes are in
lowercase (i.e. "ab"). This is driving me NUTS! Is anyone else having this
problem?

Charles Angione
Harris GCSD
Work 407-727-6621
-
7 REPLIES 7

At 10:47 1999 03 24 -0500, Angione, Charles wrote:
>I am typing in attributes in uppercase (i.e. "AB") and they are fine for that
>session. When I save the document close it and reopen it the attributes
are in
>lowercase (i.e. "ab"). This is driving me NUTS! Is anyone else having this
>problem?

With NAMECASE GENERAL YES in the (explicit or implicit) SGML declaration,
the case of almost all "names" except entity names--e.g., element names,
attribute names, and other name tokens (including attribute values of
declared value NAME, NMTOKEN, NUTOKEN, ID, IDREF, and their plurals
and any enumerated list and probably others except for CDATA)--is
insignificant. Adept normalizes all such names to lowercase when
the document is parsed by the full parser. There is no way to change
this behavior. Note that the case is not normalized by the mini-parser.

(By way of brief summary, the full parser is what you get when reading
a document that does not have the ArborText comment in it or when you
"edit -cc"; the mini-parser is what you get when you read a document
that has the ArborText comment or when you "edit -nocc". More has been
written on this elsewhere and is available from our Tech Support.)

Note that the default SGML declaration used by Adept has NAMECASE GENERAL
YES. In fact, Adept does not yet fully support NAMECASE GENERAL NO, though
we are working on that for an upcoming release, since XML does operate
with NAMECASE GENERAL NO.

About the only way to avoid having your attribute values normalized to
lowercase is to change those attribute's declared values to CDATA. This
will allow a greater range of values to be entered (which may be
undesireable in some situations), but it also indicates that case is
significant therein so that Adept will not change the case.

If you only use the mini-parser, you can avoid such case normalization,
but (1) if you cause the full parser to be used to read the document at
any point or (2) if you use another parser (in another tool) to read the
document--which may well do its own case normalization which might even be
to force it all to uppercase--you will lose your original case. The only
compliant way to ensure your case is maintained (with NAMECASE GENERAL YES)
is to declare the attributes to be CDATA.

paul

"Angione, Charles" wrote:

> Adepters
>
> I am hoping someone in this group might have found a workaround for this.
> Arbortext help support is currently looking at this issue as well.
>
> THE PROBLEM:
> I am typing in attributes in uppercase (i.e. "AB") and they are fine for that
> session. When I save the document close it and reopen it the attributes are in
> lowercase (i.e. "ab"). This is driving me NUTS! Is anyone else having this
> problem?
>
> Charles Angione
> Harris GCSD
> Work 407-727-6621
> -

Charles,

>From my experience, Adept Editor v7.0.1 doesn't support the following sgml
declaration part :

NAMECASE GENERAL NO
ENTITY NO

This part would have assured the distinction between lower and uppercase
characters.
Every General Identifiers - Tags, attributes, values - are implicitly converted in
lowercase.
An attribute value "a" and "A" is not anymore distinguishable.

If your sgml declaration part is :

NAMECASE GENERAL YES
ENTITY YES

my SGML reference book require no distinctinction between lower and uppercase
characters.
An lowercase canonization is so equivalent to an uppercase canonization. In this
case, if you prefer
a uppercase canonization, you have to convert each tag in uppercase.

Good Luck

Pierre Colot

Yes, we are having that problem also, and it has caused us some headaches
and aggravation as well.... we would *love* to see a solution to this.

Thanks,

Kevin

Kevin Mitchell x43743
Programmer Analyst
Document Information and
Management Systems

If you using the attribute value as the source in a subsequent usetext (via
a fillval) and want the output of the usetext to consistently be uppercase,
add a highlight subchars to your usetext and set allcap="1". This also
relieves the user of remembering to enter the value as uppercase.

Cheers,

David

David S. Taylor
Project Manager, SGML Publishing
Institute for Research in Construction
National Research Council Canada

We have the same problem and normalize the completed instance with OMNIMARk
when complete.
cleccese
5-Regular Member
(To:unknown1)

We were having this issue as well, in 2016. NAMECASE GENERAL NO had no effect. I didn't read Paul's comment carefully, but he is correct. Adding

<!--Arbortext, Inc., 1988-2015, v.4002-->

To the top of our SGML file solved the problem.

Never noticed a problem, and we have numerous U/LC attributes. But all of our document have

<!--Arbortext, Inc., 1988-2016, v.4002-->

at the top and it all works well. The only thing is a generated ID, if its SGML it lowercases it, if you are XML it keeps it as typed

even if its a generated CamelCase id.

Ray

Announcements