Without referencing anything specific in your code (since I haven't read
it since about first week you started) I'll just note that
internationalisation of any kind should be as close to the user
interface as you can possibly make it, without the assumption that the
user interfaces you have now will be the only ones you ever have.
That is all.
Adam K
Ovid wrote:
Hi all,
This will likely *not* be in Test::Harness 3.0 unless it turns out to
be easier/more important than I thought, but I think that
internationalization may be important. Right now, all text that it
output is hard-coded in TAP::Parser/TAP::Harness/etc. This has some
negative consequences.
For example, right now if you have more than 5 parse errors in your
TAP, the harness merely displays a message warning you about parse
errors and telling you to pass the -p switch to see all of them. (This
is important because if you get 1000 tests out of sequence because you
started your TAP with "ok 2", you get 1000 lines of TAP parse errors).
However, the -p switch only makes sense in the context of runtests, so
it's not appropriate to hardcode this knowledge into the harness.
My thought is to create a YAML phrasebook module (ripped off from
Barbie's, perhaps) that allows different phrasebooks to be embedded and
inherited. Thus, you could specify a 'parse error' message and
runtests could override just that one message for it's needs. However,
this also gives us internationalization because a full phrasebook might
be overridden in Cantonese so that the entire runtests output could be
easier to follow.
I have three questions:
1. How important is this feature?
2. Is a YAML phrasebook a sensible approach?
3. Where's my beer?
Cheers,
Ovid
--
Buy the book -- http://www.oreilly.com/catalog/perlhks/
Perl and CGI -- http://users.easystreet.com/ovid/cgi_course/
