[Ltru] Re: Test suite for language tags?

["Mark Davis" <[e-mail]>]

That's odd. Testing in Java on the file
http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagTest.txt(which
includes those cases) the only failures I get are the ones that
repeat extensions (like en-a-bbb-a-ccc), which can't be tested for with
regex. I don't think I'm using any regex syntax that differs between Java
and Perl (there are very few instances).

Note: I added a number of mechanically generated items to the test file.

Mark

On 9/17/06, Stephane Bortzmeyer <[e-mail]> wrote:
>
> On Sat, Sep 16, 2006 at 04:28:01PM -0700,
> Mark Davis <[e-mail]> wrote
> a message of 98 lines which said:
>
> > BTW, I had updated my regex to the final spec for 4646. Here is a
> > single Perl or Java regex that does most of the parse:
>
> Isn't it too lax? When testing it in a Perl script, I find it accepts
> all my well-formed tags (OK) but also accepts wrongly:
>
> fr-Latn-F is well-formed
> en-a-bbb-a-ccc is well-formed
> tlh-a-b-foo is well-formed
> abcdefghi-012345678 is well-formed
> ab-abc-abc-abc-abc is well-formed
> ab-abcd-abc is well-formed
> ab-ab-abc is well-formed
> ab-123-abc is well-formed
> ab-abcde-abc is well-formed
> ab-1abc-abc is well-formed
> ab-ab-abcd is well-formed
> ab-123-abcd is well-formed
> ab-abcde-abcd is well-formed
> ab-1abc-abcd is well-formed
> ab-a-b is well-formed
> ab-a-x is well-formed
> ab--ab is well-formed
> ab-abc- is well-formed
> ab-c-abc-r-toto-c-abc is well-formed
> abcd-efg is well-formed
> aabbccddE is well-formed
>

_______________________________________________
Ltru mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/ltru