[Python-Dev] (s)re crashing in regrtest (was SRE 0.9.8 benchmarks) (original) (raw)
Fredrik Lundh Fredrik Lundh" <effbot@telia.com
Thu, 3 Aug 2000 13:25:45 +0200
- Previous message: [Python-Dev] (s)re crashing in regrtest (was SRE 0.9.8 benchmarks)
- Next message: [Python-Dev] (s)re crashing in regrtest (was SRE 0.9.8 benchmarks)
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
mark favas wrote:
>> >>> p.match(500000*'x') >> Segmentation fault (core dumped) > >Andrew, would you mind adding a test case for that to the re test >suite? It's important that this doesn't come back! =20 In fact, on my machine with the default stacksize of 2048kb, = test_re.py already exercises this bug. (Goes away if I do an "unlimit", of = course.) So testing for this deterministically is always going to be dependent = on the platform. How large do you want to go (reasonably)? - although I guess core dumps should be avoided...
afaik, there was no test in the standard test suite that included run-away recursion...
what test is causing this error?
(adding a print statement to sre._compile should help you figure that out...)
- Previous message: [Python-Dev] (s)re crashing in regrtest (was SRE 0.9.8 benchmarks)
- Next message: [Python-Dev] (s)re crashing in regrtest (was SRE 0.9.8 benchmarks)
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]