On 29/10/2013 4:04 AM, Gord Tomlin wrote:
But I now see from what Mr. Gilmore has been saying that I am perhaps
wanting too much from a computer language. I will need to depend on the
programmer actually doing the proper things in all of his/her
programs. So
that when I use chorizo in English it will show up before cider, but if
I-the-programmer know that Spanish is the "locale", that it is after
ciudad. I don't know how this can be done generically. Especially if one
throws words in Latin characters into a list with words in Greek
characters.
"Intelligent" string collation of the sort you are suggesting is
probably something that would find a better home in a run time library
or in a utility product, as opposed to being built right into a
computer language. In fact, a Google search for "locale sensitive
string comparison" shows that such facilities are available for
several languages.
IBM did a very good job with ICU.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
