Re: Patch to implement progress in dd

[Sean Reifschneider]

On Tue, Apr 29, 2003 at 11:37:18PM -0600, Bob Proulx wrote:
>Hi Sean!

Hey there.  It's weird where you run into people, eh?

>It is an interesting item.  Please allow me to comment upon it.  I

Yeah, that's mostly why I submitted it.  I expected some discussion...

>don't think "progress=1" is of the right spirit for setting that type
>of option.  The 1 and 0 don't seem right here.

That was my first thought as well, but having the processing of long
arguments off in it's own code, possibly shared among other code, kind
of put me off.  In particular, that it was meant to work only for cases
where the --arg was the only argument...

I'd be tempted to convert to using getopt, but I don't really know how
that fits into fileutils, primarily from a portability standpoint...  I
could write specific handling for the "--progres=" option in main(), but
do I require that it be the first option?  Re-build argv/argc after
searching for that option?  Search for it and then ignore it later on?

Any thoughts on that front?

>Therefore the documentation changes need to go into the coreutils.texi
>file.

Ahh, thanks for the pointer.

>Jim has previously posted the following guidelines which I will
>reproduce here.  Perhaps I can beat him to the post.  :-)

Any thoughts on wether it's likely to require convincing?  I've been
wanting to make this patch for like a year, because it's just nice on
really big dds to know where you are in it...  The particular catalyst
for it right now is that Ant's hard drive is dieing and he picked up a
replacement from me today, and is trying dd the data around.  He really
has no good idea if it's making forward progress though, because it
*WILL* encounter read-errors at some point...

We pointed him at some better ways of getting the data off, but still
there are just times when it'd be nice to know how many blocks have been
read or written.

Thoughts?

Thanks,
Sean
-- 
 You must align your chakra to resist The Man.
Sean Reifschneider, Inimitably Superfluous <address@hidden>
tummy.com, ltd. - Linux Consulting since 1995.  Qmail, Python, SysAdmin
      Back off man. I'm a scientist.   http://HackingSociety.org/