Re: [office-formula] Not worth a JIRA issue - 5.1 Functionnamesignore case. ?

[Eike Rathke <erack@sun.com>]

Hi Dennis,

On Monday, 2010-02-01 06:33:44 -0800, Dennis E. Hamilton wrote:

> 1. Does the ignoring of case also apply to the use URI-mappable qualified
> names to identify the "namespace" of an extension-provided function and
> special ones like "LEGACY"?  (The model is equivalent to the scheme used for
> the Java language.)

Yes.

> 2. Is this one of those case-is-ignored but don't change the case that the
> producer used situations?

No. We had a recommendation that implementations should write upper case
function names. I'm not sure at the moment if and how that survived
Patrick's edits.

> 3. Or is it one of those situations where case might be honored but to be
> safe, never use names that differ only in case and expect them to identify
> separate functions?

As said, case is ignored.

> 4. If case is ignored for function names, does this impact other names
> (tables, arrays, sheets, defined functions, and cell coordinates)? 

Why should it? Why introduce unrelated topics?
We don't define tables, arrays don't have names, defined functions are
function names (or were you referring named expressions? We have them
defined as case-consistent, not sensitive), for cell coordinates we say
that the column name *shall* be written uppercase (nevertheless IMHO
a reading implementation should be lenient in what it accepts). The only
definition we seem to be lacking is the case insensitivity for
SheetLocator (sheet names), didn't find that quickly at least.

  Eike

-- 
Automatic string conversions considered dangerous. They are the GOTO statements
of spreadsheets.  --Robert Weir on the OpenDocument formula subcommittee's list.

PGP signature