Is this case now approved?
Olga
2010/3/5 ????? ???????????? <olga.kryzhanovska at gmail.com>:
> 2010/3/2 ????? ???????????? <olga.kryzhanovska at gmail.com>:
>> On Tue, Mar 2, 2010 at 10:44 PM, Glenn Skinner <glenn.skinner at covad.net>
>> wrote:
>>> On Mar 2, 2010, at 12:55 PM, Garrett D'Amore - sun microsystems wrote:
>>>
>>>> This project is an amendment to the Korn Shell 93 Integration project
>>>> (PSARC/2006/550 and PSARC/2007/035, PSARC/2008/094, PSARC/2008/344
>>>> and PSARC/2008/589) specifying the following additional
>>>> interfaces:
>>>> Addition of /usr/bin/xgrep
>>>>
>>>> Bug/RFE Number(s):
>>>>
>>>> 6929154 RFE: Add /usr/bin/xgrep (Augmented regular expressions
>>>> (conjunction, negation.))
>>>>
>>>>
>>>> Interface Stability Description
>>>> --------- --------- -----------
>>>> /usr/bin/xgrep Committed xgrep command
>>>> ksh93 'xgrep' built in Committed xgrep command
>>>
>>> Before I give this case my +1, I have some questions:
>>>
>>> The case's specification should define the syntax of the regular expressions
>>> the utility accepts. I suspect that the variants requested by the -E, -F,
>>> -G, and -P flags can be handled with references to the corresponding man
>>> pages. But what is the specification for -X mode REs (including the
>>> semantics of alternation and conjunction)?
>>
>> -X is like POSIX grep -E (egrep) but adds & as AND operator and ! as
>> NOT operator.
>>
>>>
>>> What is the difference between lenient and strict pattern interpretation?
>>
>> Strict will print syntax errors if the pattern does not match the
>> standard exactly. I leave the exact explanation to Glenn Fowler.
>
> Glenn's replied this:
> --------------------------------
> -O, --lenient sets the regcomp(3) REG_LENIENT flag
> in general if REG_LENIENT is on then certain constructs marked
> "unspecified" in the standard will be accepted, otherwise they
> may produce regcomp(3) errors
>
> e.g.
>
> invalid \char escape
>
> grep -S '\#' <<<'#'
>
> invalid [...] range endpoint
>
> grep -S '[a-q-z]' <<<'a'
>
> the intention (theory) is that RE's that make it through -S, --strict
> will make it through all conformant implementations
>
> as counterexamples roll in I tweak libast/reg*.c to make pratice match theory
> --------------------------------
>
> 'lenient' mode by default may be an option (i.e. look at Garrett
> D'Amore proposal for moving /usr/xpg4/bin to /usr/bin, AST grep would
> keep strong backwards compatibility sans the dreaded i18n bugs in the
> current /usr/bin/grep) if we replace /usr/bin/grep and friends with
> this grep implementation and use 'strict' mode for /usr/xpg4/bin/grep.
>
> Olga
> --
> , _ _ ,
> { \/`o;====- Olga Kryzhanovska -====;o`\/ }
> .----'-/`-/ olga.kryzhanovska at gmail.com \-`\-'----.
> `'-..-| / Solaris/BSD//C/C++ programmer \ |-..-'`
> /\/\ /\/\
> `--` `--`
>
--
, _ _ ,
{ \/`o;====- Olga Kryzhanovska -====;o`\/ }
.----'-/`-/ olga.kryzhanovska at gmail.com \-`\-'----.
`'-..-| / Solaris/BSD//C/C++ programmer \ |-..-'`
/\/\ /\/\
`--` `--`
