From:
Donal K. Fellows <dona...@manchester.ac.uk>
Sat, 19 Jul 2014 22:45:00 +0100
On 17/07/2014 19:04, Joe English wrote:
> Trevor Davel wrote:>> This is a minor change to what is proposed in TIP 412 - mostly the>> introduction of msgcat::mcl - but I believe it would significantly>> improve the usability of [msgcat] for per-request dynamic locale.>> Agreed, esp. part (b). This is clearly the right way to go.
I also concur; it is entirely reasonable to have a mechanism that code
can use when it knows exactly what locale to use. Point (c) — the
refactoring once such an API exists — seems plain sensible; no point in
writing the code twice.
Here's some back-up reasoning. :-) We also need to figure out what to do
about locales and TclOO (and itcl, and so on) as the existing mechanism
— binding current locale to namespace — is unworkable (each object has
its own namespace, with no inheritance between them). Having a mechanism
that takes an explicit locale makes fixing things much simpler as it
means that the msgcat package doesn't need to understand TclOO's
inheritance mechanism itself; it can delegate to some class that holds
that piece of the smarts.
Donal (not claiming to be a locale specialist).
begin:vcard
fn:Donal K. Fellows
n:Fellows;Donal
org:The University of Manchester;School of Computer Science
adr:;;Oxford Road;Manchester;;M13 9PL;United Kingdom
email;internet:dona...@manchester.ac.uk
version:2.1
end:vcard
------------------------------------------------------------------------------
Want fast and easy access to all the code in your enterprise? Index and
search up to 200,000 lines of code with a free copy of Black Duck
Code Sight - the same software that powers the world's largest code
search on Ohloh, the Black Duck Open Hub! Try it now.
http://p.sf.net/sfu/bds
_______________________________________________
Tcl-Core mailing list
Tcl-...@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tcl-core