On Sat, Feb 20, 2016 at 8:19 PM, Jim Meyering <[email protected]> wrote:
> On Sun, Feb 14, 2016 at 12:02 PM, Ulya Fokanova <[email protected]> wrote:
>> I've explored the following case:
>>
>> $ printf '12\n34\0' | LC_ALL=en_US.utf-8 grep -z '^[1-4]*$' | wc -c
>> 6
...
>> The bug also present with PCRE engine:
>>
>> $ printf '12\n34\0' | LC_ALL=en_US.utf-8 grep -z -P '^[1234]*$' | wc -c
>> 6
>> $ printf '12\n34\0' | LC_ALL=en_US.utf-8 grep -z -P '^[1-4]*$' | wc -c
>> 6
>
> Thank you for the analysis and the report.
> I have fixed the regex-oriented problem with the attached
> patch, but not yet the case using -P -z (PCRE + --null-data):
The -Pz/PCRE problem is more fundamental, and strikes
even with LC_ALL=C. This shows that with -Pz, anchors
still wrongly match at newlines, rather than at \0 bytes:
$ printf '\0a\nb\0' | LC_ALL=C src/grep -Plz '^a'
[Exit 1]
$ printf '\0a\nb\0' | LC_ALL=C src/grep -Plz '^b'
(standard input)
Fixing this is on PCRE's maint/README wish list with this item:
. Line endings:
* Option to use NUL as a line terminator in subject strings. This could now
be done relatively easily since the extension to support LF, CR, and CRLF.
