Ah yes plurals!
Certainly it is challenging to try and capture language semantics in
an easy modifiable way.
The best option I have seen so far with in the Chandler libraries is
the PyICU ChoiceFormat class.
I use it in the mail service to format numeric variations in English
strings.
For example:
DOWNLOAD_START_MESSAGES = ChoiceFormat(_(u"1#%(accountName)s:
Downloading %(numberOfMessages)s message...|1<%(accountName)s:
Downloading %(numberOfMessages)s messages..."))
The above statement says if the value passed to the ChoiceFormat is 1
then use:
%(accountName)s: Downloading %(numberOfMessages)s message...
Otherwise:
%(accountName)s: Downloading %(numberOfMessages)s messages...
Note that in this case 0 or any negative numbers are not going to be
passed to the ChoiceFormat.
Since the entire ChoiceFormat string appears in the localizers po
file, he or she is free
to add language specific numeric string variations to the translation.
-Brian
On Dec 2, 2007, at 10:14 PM, Philippe Bossut wrote:
Interesting. Note that they're wrong saying that semitic languages
have only 2 forms. Arabic at least has 3: singular, dual (for 2 of
something) and plural. Hebrew also but only some nouns have the
dual form. Apparently, the GNU still needs a couple more iterations...
Cheers,
- Philippe
Heikki Toivonen wrote:
http://ed.agadak.net/2007/12/one-potato-two-potato-three-potato-four
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "chandler-dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/chandler-dev
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "chandler-dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/chandler-dev
