I can vouch that when built correctly 7.9 works fine on Windows. Wonder
why he didn't answer your question about how it was built? There were
changes in how to build PCRE after version 7.0. He can't just drop the
source into what worked for 6.7 and recompile. Tell him to read
Non-Unix-Use and configure using cMake. He should use a fresh source dir
and build in a separate dir. Verify using Runtest.bat. (some problems
are expected with RunTest if using stack for recursion on Windows).
Regards,
Sheri
Mart Goodall wrote:
> ------- You are receiving this mail because: -------
> You are on the CC list for the bug.
>
> http://bugs.exim.org/show_bug.cgi?id=867
>
>
>
>
> --- Comment #4 from Mart Goodall <mart.goodall@???> 2009-07-28 21:08:21 ---
>
>
> -----Original Message-----
> From: admin@??? [mailto:admin@bugs.exim.org] On Behalf Of Philip
> Hazel
> Sent: Tuesday, July 28, 2009 3:30 PM
> To: Mart Goodall
> Subject: [Bug 867] "\w" no longer functions
>
> ------- You are receiving this mail because: -------
> You are on the CC list for the bug.
> You reported the bug.
>
> http://bugs.exim.org/show_bug.cgi?id=867
>
>
>
>
> --- Comment #3 from Philip Hazel <ph10@???> 2009-07-28 20:29:30
> ---
> On Tue, 28 Jul 2009, Mart Goodall wrote:
>
>
>> I kept the compile as close to my production as possible. I have the same
>> failure - see screen output below:-
>>
>> PCRE version @PCRE_MAJOR@.@PCRE_MINOR@@PCRE_PRERELEASE@ @PCRE_DATE@
>>
>
> ...Something strange there ... why isn't it giving the real version
> number? I see from the original report that you are using Windows. How
> did you do the build?
>
>
>> re> "[^\w\s. |-]"
>> data> Maureen Hubbard
>> 0: M
>>
>>
>> PCRE version @PCRE_MAJOR@.@PCRE_MINOR@@PCRE_PRERELEASE@ @PCRE_DATE@
>> Compiled with
>> UTF-8 support
>> Unicode properties support
>> Newline sequence is LF
>> \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
>>
>>
>> Hope this helps
>>
>
> Not a lot, I'm afraid. On my Gentoo Linux box with exactly the same
> compile time options, it works fine. So it's something to do with your
> Windows version - I do not run Windows, so there is no way I can test
> for myself.
>
> I have three suggestions: (1) Try using autocallout to see what is going
> on. This is what I get when I run pcretest (note the "C" option after
> the pattern):
>
> PCRE version 7.9 2009-04-11
>
> re> "[^\w\s. |-]"C
> data> Maureen Hubbard
> --->Maureen Hubbard
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> +0 ^ [^\w\s. |-]
> No match
> data>
>
>
> same test gives
> PCRE version @PCRE_MAJOR@.@PCRE_MINOR@@PCRE_PRERELEASE@ @PCRE_DATE@
>
> re> "[^\w\s. |-]"C
> data> Maureen Hubbard
> --->Maureen\x20Hubbard
> +0 ^ [^\w\s. |-]
> +11 ^^
> 0: M
>
>
> This shows that it is testing every character, and they all fail.
> Another option you can use is "D" to show the compiled code:
>
> PCRE version 7.9 2009-04-11
>
> re> "[^\w\s. |-]"D
> ------------------------------------------------------------------
> 0 36 Bra
> 3 [\x00-\x08\x0b\x0e-\x1f!-,/:-@[-^`{}-\xff] (neg)
> 36 36 Ket
> 39 End
> ------------------------------------------------------------------
> Capturing subpattern count = 0
> No options
> No first char
> No need char
> data>
>
>
> PCRE version @PCRE_MAJOR@.@PCRE_MINOR@@PCRE_PRERELEASE@ @PCRE_DATE@
>
> re> "[^\w\s. |-]"D
> ------------------------------------------------------------------
> 0 36 Bra
> 3 [\x00-\x1f\x21-\x2c\x2f-rt-vx-{}-\xff] (neg)
> 36 36 Ket
> 39 End
> ------------------------------------------------------------------
> Capturing subpattern count = 0
> No options
> No first char
> No need char
>
> If your output is the same as mine, things are really weird...
>
> My final suggestion is that you edit pcre_internal.h, and change the "0"
> in line 50 into "1" and then re-compile. This will insert debugging
> statements into the code, and they will generate some output when you
> run the test. This may perhaps give some clue as to what is going on.
>
>
>
> PCRE version @PCRE_MAJOR@.@PCRE_MINOR@@PCRE_PRERELEASE@ @PCRE_DATE@
>
> re> "[^\w\s. |-]"
> ------------------------------------------------------------------
> [^\w\s. |-]
>
>>> start branch
>>>
> length=6 added 0 c=[
> length=39 added 33 c=
>
>>> end branch
>>>
> end pre-compile: length=40 workspace=36
> Length = 40 top_bracket = 0 top_backref = 0
> Options=00000000
> 0 36 Bra
> 3 [\x00-\x1f\x21-\x2c\x2f-rt-vx-{}-\xff] (neg)
> 36 36 Ket
> 39 End
> ------------------------------------------------------------------
> data> Maureen Hubbard
>
>>>>> Match against: Maureen Hubbard
>>>>>
> start non-capturing bracket
> bracket 0 tail recursion
> ims reset to 00
> match() returned 1 from line 926 >>>> returning 1
> 0: M
>
> Good luck
> mart
>
> Philip
>
>
>
