Darren J Moffat wrote:
> Garrett D'Amore wrote:
>>> All the other non filename fields in the pfiles(1) output
>>> are on a different line to the filename. So I'd highly recommend
>>> that offset go on the same line as mode: or on its own line.
>>> If that is done then an offset of 0 should be printed for
>>> things that are seekable.
>>
>> If you put it on the same line as the mode:, then you wind up having
>> to line wrap. This makes the content less usable for *human* consumers.
>
> But putting offset: on the filename line will make that wrap quicker
> too, especially for pfiles run on an editor running against an deep
> file tree eg:
>
> 5: S_IFREG mode:0644 dev:182,65547 ino:377482 uid:101 gid:10 size:16384
> O_RDWR|O_CREAT|O_EXCL
> /builds/onnv-bugfixes/usr/src/cmd/cmd-crypto/elfsign/.elfsign.c.swp
>
> Thats a vim process editing a file not that deep in ON.
Good point.
>
>> It seems like folks writing *software* should probably not be
>> consuming pfiles output, but possibly using procfs directly.
>
> Indeed but that should imply we deliberately make the output worse
> than it is just now, so putting offset on a new line by its own would
> be my preference.
>
Okay, well, I'll let the project team (Dan Mick) comment on this.
Personally I don't have that strong a feeling; Dan I chatted for all of
about 5 minutes debating where the best way to expose this was.
Possibly the best solution is to introduce a parseable format, but I
didn't feel comfortable volunteering Dan for the additional work.
- Garrett
