[Python-Dev] regression test failure in test_tokenize? (original) (raw)

Fredrik Lundh Fredrik Lundh" <effbot@telia.com
Tue, 1 Aug 2000 23:16:15 +0200

andrew wrote:

On Wed, Aug 02, 2000 at 04:30:42AM +0800, Mark Favas wrote: >Current CVS (Wed Aug 2 04:22:16 WST 2000) fails on Tru64 Unix: >Is this just me, or an un-checked checkin? (I noticed some new sre = bits >in my current CVS version.) =20 testtokenize works fine using the current CVS on Linux; perhaps this is a 64-bit problem in sre manifesting itself?

I've confirmed (and fixed) the bug reported by Mark. It was a nasty little off-by-one error in the "branch predictor" code...

But I think I know why you didn't see anything: Guido just checked in the following change to re.py:

*** 21,26 **** # =20 ! engine =3D "sre" ! # engine =3D "pre" =20 if engine =3D=3D "sre": --- 21,26 ---- # =20 ! # engine =3D "sre" ! engine =3D "pre" =20 if engine =3D=3D "sre":