On Jan 23, 2009, at 1:08 PM, Mathieu Bouchard wrote:
On Fri, 23 Jan 2009, Hans-Christoph Steiner wrote:
On Jan 23, 2009, at 10:37 AM, Mathieu Bouchard wrote:
So, do you really think that Tcl's GetText 'format' is being used at all outside of Tcl? It's not like it's the same format as used in GNU GetText.
Considering Tcl's msgcat is listed in the official GNU gettext manual as "fully portable", I think it is safe to say yes: http://www.gnu.org/software/automake/manual/gettext/Tcl.html
But Tcl's msgcat is using its own *.msg file format instead of *.po.
Here is one: http://www.gted.org/
Does it support Tcl's *.msg files, or only *.po ? I only see a mention of *.po.
"GNU gettext has support (since V 0.13 [1]) for Tcl msgcat and conversion tools from and to PO format. http://ftp.gnu.org/gnu/gettext/"
.hc
Do you need a well debugged and optimised library for accessing a hashtable when the language already has commands for accessing hashtables freely?
Do you really think that accessing a hashtable is all there is to localization? How about parsing the names of the locales (en_US, en_US.UTF-8, en_GB, etc.)? That is non-trivial and handled by msgcat, as are many other things we haven't thought of, but the people who have been working on gettext/msgcat since
Yeah, this is what most of the msgcat library does. If you remove the locale autodetection code from it, there's not much code left. I don't really mind setting the language myself though. Yet it'd be fine with me if the locale were autodetected.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal, Québec
----------------------------------------------------------------------------
The arc of history bends towards justice. - Dr. Martin Luther King, Jr.