Hmm... this is an interesting problem.
But I am a bit confused: what is your criteria for sorting international
names? For example, for the Janpanese names, are the actual names stored
(with Japanese characters), or just the alphabetized version of the them? If
the actual Japanese characters are
Right, first you have to establish your sort algorithm.
Do it manually on paper first with a sampling of data. Most romanic languages
would be relatively easy...however it's not all trivial. What comes first? é
è ê ë ? What about special characters limited in even romanic languages? þ Ð
Right, first you have to establish your sort algorithm.
Do it manually on paper first with a sampling of data. Most romanic languages
would be relatively easy...however it's not all trivial. What comes first? é
è ê ë ? What about special characters limited in even romanic languages? þ Ð
Actually Qing Xia, how are chinese phonebooks done? Is it by pinyin?
I grew up here in the US, dunno how alot of commonplace things are in China
~|
Adobe® ColdFusion® 8 software 8 is the most important and dramatic release to
Good question, actually, upon further clarification, the Japanese names
appear to be using the English alphabet, or at least the Westernized
alphabet. However, I'll give an example.
The following shows an alphabetical list of last names:
Littrell
Liu
Liénard
Åquist
Shouldn't
I know that you could probably sort a field by first assigning a
collation such as the following
Really not trivial.
For some languages, a collation is not enough.
I may be wrong, but I think I've learned that in Spanish for instance,
the double l or ch are not considered as a two letters but
Littrell
Liu
Liénard
This is a good example.
Liénard comes at the end, because apparently the sort uses the ASCII
collating sequence in which é comes after all other non accented characters.
A solution would be to have two columns for names, like
Name key
-
Liénard
Yep, you are right, in China, the phonebooks and dictionaries are sorted
ascending using pinyin. So, in this case, if the Chinese names are stored
in the DB using pinyin, then it is actually easy to sort because there will
be no symbols outside of the 26 English alphabet letters.
On Thu, Jul 10,
I think Lienard (how do you bring up the symbol on a QWENTY keyboard) is
after Liu because that e comes after the 26 English alphabet letters. Same
reason why Aquist (with that A) is at the end of the list.
On Thu, Jul 10, 2008 at 12:16 PM, Vince Collins [EMAIL PROTECTED] wrote:
Good question,
I think Lienard (how do you bring up the symbol on a QWENTY keyboard) is
after Liu because that e comes after the 26 English alphabet letters.
It come at the end in the ISO character set, but alphabetic sort should
not use the ASCII code
Thus the problem.
--
Aha. Interesting.
BTW, if you wouldn't mind sharing, I'd love to learn from your function
which converts iso-8859-1 strings to upper case keys.
I have not had a chance to work with international characters in the DB, so
this sounds really interesting.
On Thu, Jul 10, 2008 at 12:52 PM, Claude
I have not had a chance to work with international characters in the
DB, so
this sounds really interesting.
I'm sending it by private eMail, I'm affraid that posting it here will
make the message too large.
It works works for strings in ISO, and may be used for databases in ISO,
not UTF-8
--
: Sorting International Names Challenge!
I have not had a chance to work with international characters in the
DB, so
this sounds really interesting.
I'm sending it by private eMail, I'm affraid that posting it here will
make the message too large.
It works works for strings in ISO, and may
13 matches
Mail list logo
