-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 9 Sep 2007, at 13:24, Abe Timmerman wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, Sorry for the long output, but I hope it helps to state my confusion. I see no failing tests, yet it states "Failed 5/57 testprograms"
What are those test results for Abe? I see it's on MSWin32 but what are you testing?
I also do not agree with the no plan found diagnostics: C:\download\Tk-804.027-9793>perl595 -Mblib t\JP.t 1..0 # Skipped: locale's 'iso-8859-1' cannot represent Japanese. C:\download\Tk-804.027-9793>perl595 -Mblib t\Kr.t 1..0 # Skipped: locale's 'iso-8859-1' cannot represent Korean.
The TAP grammar we're working to has it as /^1..0\s+#\s*SKIP\b/i - so Skippped doesn't match.
It occurs to me that we could relax our parsing for any TAP stream that doesn't have an explicit version number - so we'd only require strict compliance for new TAP. We already have a mechanism for swapping in a different grammer per TAP version number.
But that would be at odds with one of our missions: to be more strict about malformed TAP. Personally I wouldn't mind making the parser a little more relaxed for old TAP - but I'm not sure what the history of the decision to be strict is so it's not really my call.
- -- Andy Armstrong, hexten.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iD8DBQFG5CEywoknRJZQnCERAjivAKDLV5yVaNR7cavbYt31ksFWN7xNNwCg01DT h4w1TdvZ23PoZk/jZiKXsrI= =Imwl -----END PGP SIGNATURE-----
