office-formula message
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]
Subject: Re: [office-formula] New draft of openformula specification
- From: robert_weir@us.ibm.com
- To: office-formula@lists.oasis-open.org
- Date: Sun, 16 Jul 2006 20:09:21 -0400
"David A. Wheeler" <dwheeler@dwheeler.com>
wrote on 07/14/2006 06:12:51 PM:
>
> I'm going to back off the comparison description slightly, and insert
a TODO
> asking for help. Can anyone give me good text on how comparison
should
> work in spreadsheets in a modern international world when it's NOT
> case-sensitive?
> I've put in SOME text that makes sense, but help wanted.
> I'll release a new version of the spec in just a moment.
>
For what it is worth, Excel seems to process terminal
sigma correctly in their implementation of lower(). It also seems
to use collation order based string comparisons. Since spreadsheets
are more end-user business application than programming languages I think
users would expect this level of native language support.
http://www.unicode.org/reports/tr10/ is the standard
that I came up with. Since it is also the reference cited by James
Clark for the XSLT specification, I'll give it the presumption of correctness.
Note to implementors -- the easiest path getting this
right is to either to use Java (since collation support is included in
the standard libraries) or use ICU (International Components for Unicode
http://icu.sourceforge.net/) an open source project that gives similar
support in portable C code.
-Rob
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]