ps command

[Kamal R. Prasad]

Alright. It looks like you are taking care of it. Are there any other 
commands/utilities that need significant work for posix compliance?

thanks
-kamal

Garance A Drosihn wrote:

> At 9:54 PM +0530 10/14/04, Kamal R. Prasad wrote:
>
>>
>> I see that ps command needs a few additions to be compliant with
>> IEEE 1003.1-2001 POSIX std. Please confirm if I can provide a
>> patch that contains the following stuff:-
>>
>> -A, -e  : write information about all processes
>> -d : write information about all processes except session leaders
>> -f : shows command line options , change this to show the fields
>>      UID PID PPID C STIME TTY TIME CMD
>> -l : add the fields F, S
>> -g : grouplist : Write information for processes whose session
>>      leaders are given in grouplist. The application shall ensure
>>      that the grouplist is a single argument in the form of a
>>      <blank> or comma-separated list.
>> -G : grouplist : Write information for processes whose real group
>>      ID numbers are given in grouplist. The application shall
>>      ensure that the grouplist is a single argument in the form
>>      of a <blank> or comma-separated list.
>>
>> I'd appreciate info on who to send the patch to.
>
>
> There is some extensive work done by Cyrille Lefevre, and which
> was sent in as a PR.  I have incorporated some of that work, and
> plan to incorporate more of it.  (although probably not in the
> exact manner that Cyrille wrote it up).  I also have a few updates
> of my own for `ps', that I need to clean up and commit.  I was
> doing work in May and June.  In July I got bogged down with some
> other matters, and I have not gotten back to `ps' since then, but
> I do intend to get back to it once 5.3-release is official.
>
> At least some of the changes are incompatible with our present `ps'
> command, so we can't just throw them in and break the behavior that
> FreeBSD users are used to.  For instance, `-G' is already in our
> version of `ps' which is in the 5.x-branch, due to work done earlier
> this year.  `-A' is also supported, but it is (deliberately) not
> documented.  The *code* for `-g' is also in our source, but it is
> commented out because it conflicts with the historical definition
> of `-g' (which has always been supported in FreeBSD, so many people
> are used to typing it).
>