[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Here's how to process XML documents written in German

• From: " ." <oscar.vives@gmail.com>
• To: "xml-dev@lists.xml.org" <xml-dev@lists.xml.org>
• Date: Fri, 1 Feb 2013 13:23:28 +0100

On 31 January 2013 23:23, Jim Melton <jim.melton@oracle.com> wrote:
> Roger,
>
> I'd like to observe that this is in no way an "XML problem", nor should an
> XML solution be pursued (e.g., identity transforms).  This is a file
> encoding problem in the form of a character normalization form.  If one
> deals with information sources that randomly encode characters using both
> precomposed and decomposed forms, or characters based on appearance on one
> medium or another, then one must decide how to resolve the problem based on
> those facts.
>

Maybe this is not a "XML problem", but how you use XML problem.  A
writter style problem.

If you are XML author,  choosing währung for the money is a bad idea,
like choosing  l0l0ll0Ol0lOl1OOll  is a bad identifier (for humans).
Maybe l0l0ll0Ol0lOl1OOll is a valid name, but will create confusion
(for humans).  So you avoid währung, and you choose any other word to
describe the same parameter.

As for XML technologies. I don' know. I am not a XML expert like most
people in this mail list.

In MySQL you have the concept of "collation", that include stuff like
how to alfabetically order texts, when to match two strings. Collation
decides if  searching for "Los Ángeles" you find "los angeles", "Los
ángeles" and other strings that are composed of a string of different
bytes.   I don't know if XPath have "collations" that could be used to
tell him to search währung and trigger wen it finds the different ways
to write währung.



-- 
--
ℱin del ℳensaje.