Author: Philip Hazel Date: To: Michael McCarty CC: pcre-dev Subject: Re: [pcre-dev] PCRE test differences
On Tue, 31 Mar 2009, Michael McCarty wrote:
> I've compiled PCRE 7.8 on Linux and ran pcretest (compiled with NOPOSIX
> defined) on the test data. It passes most of the tests without problem, but
> it apparently fails on testinput7, 8, and 9 (see attached output files).
>
> % pcretest -C
> PCRE version 7.8 2008-09-05
> Compiled with
> UTF-8 support
> Unicode properties support
> Newline sequence is ANY
> \R matches all Unicode newlines
> Internal link size = 2
> POSIX malloc threshold = 10
> Default match limit = 10000000
> Default recursion depth limit = 10000000
> Match recursion uses stack
>
> Are these differences anything I should be concerned about? Any advice is
> welcome.
I've taken a look at these, and they suggest to me that you ran the
tests in a non-standard way. Did you use the RunTest script? It looks as
if you ran tests 7,8, and 9 without the -dfa option because the output
shows that it used the normal matching function rather than the DFA
matcher. This explains why it did not work as expected.
