[Linux-ia64] Re: [BK PATCH] SN update

[Linux-ia64] Re: [BK PATCH] SN update

[David Mosberger]

[Forgot to cc the list.]

>>>>> On Thu, 15 May 2003 21:32:57 -0700, Jesse Barnes <jbarnes@sgi.com> said:

  Jesse> Ok, this is my first 'bksend' cset, so please let me know if
  Jesse> you need something different.

Please don't do that.  Either send regular patches or point me to a
tree I can pull from (actually, I still need to learn how to do this
"safely").

	--david

Re: [Linux-ia64] Re: [BK PATCH] SN update

[Martin Hicks]

[-- Attachment #1: Type: text/plain, Size: 568 bytes --]


On Thu, May 15, 2003 at 11:37:00PM -0700, David Mosberger wrote:
> [Forgot to cc the list.]
> 
>   Jesse> Ok, this is my first 'bksend' cset, so please let me know if
>   Jesse> you need something different.
> 
> Please don't do that.  Either send regular patches or point me to a
> tree I can pull from (actually, I still need to learn how to do this
> "safely").

I think it's "safe" to bk pull from anywhere.  There is always bk unpull
if you want to reverse the last pull.

mh

-- 
Wild Open Source Inc.                  mort@wildopensource.com

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

[Linux-ia64] Re: [BK PATCH] SN update

[Jesse Barnes]

On Thu, May 15, 2003 at 11:37:00PM -0700, David Mosberger wrote:
> [Forgot to cc the list.]
> 
> >>>>> On Thu, 15 May 2003 21:32:57 -0700, Jesse Barnes <jbarnes@sgi.com> said:
> 
>   Jesse> Ok, this is my first 'bksend' cset, so please let me know if
>   Jesse> you need something different.
> 
> Please don't do that.  Either send regular patches or point me to a
> tree I can pull from (actually, I still need to learn how to do this
> "safely").

Since I can't legally put a tree up, I'll send you a patch.

Thanks,
Jesse

Re: [Linux-ia64] Re: [BK PATCH] SN update

[David Mosberger]

>>>>> On Fri, 16 May 2003 08:09:01 -0400, Martin Hicks <mort@wildopensource.com> said:

  Martin> On Thu, May 15, 2003 at 11:37:00PM -0700, David Mosberger
  Martin> wrote:
  >> [Forgot to cc the list.]
  >> 
  Jesse> Ok, this is my first 'bksend' cset, so please let me know if
  Jesse> you need something different.
  >>  Please don't do that.  Either send regular patches or point me
  >> to a tree I can pull from (actually, I still need to learn how to
  >> do this "safely").

  Martin> I think it's "safe" to bk pull from anywhere.  There is
  Martin> always bk unpull if you want to reverse the last pull.

That's not safe in my mind.  What I'd like is an efficient way to
preview a pull (see a diff -u between the remote tree and where I'm
pulling into).  Note the "efficient" part.  Having to do a full clone
first doesn't qualify.

	--david

Re: [Linux-ia64] Re: [BK PATCH] SN update

[Jesse Barnes]

On Fri, May 16, 2003 at 10:49:58AM -0700, David Mosberger wrote:
> >>>>> On Fri, 16 May 2003 08:09:01 -0400, Martin Hicks <mort@wildopensource.com> said:
> 
>   Martin> On Thu, May 15, 2003 at 11:37:00PM -0700, David Mosberger
>   Martin> wrote:
>   >> [Forgot to cc the list.]
>   >> 
>   Jesse> Ok, this is my first 'bksend' cset, so please let me know if
>   Jesse> you need something different.
>   >>  Please don't do that.  Either send regular patches or point me
>   >> to a tree I can pull from (actually, I still need to learn how to
>   >> do this "safely").
> 
>   Martin> I think it's "safe" to bk pull from anywhere.  There is
>   Martin> always bk unpull if you want to reverse the last pull.
> 
> That's not safe in my mind.  What I'd like is an efficient way to
> preview a pull (see a diff -u between the remote tree and where I'm
> pulling into).  Note the "efficient" part.  Having to do a full clone
> first doesn't qualify.

I found myself wishing the same thing as I tried to get together a
patch for you.  I noticed that 'bk import' has a -C option which will
let you mess with the changeset before you commit it, but 'bk receive'
doesn't, AFAIK.  It would be nice if it did (or if it was easier to
modify an existing changeset).

Jesse

Re: [Linux-ia64] Re: [BK PATCH] SN update

[David Mosberger]

>>>>> On Fri, 16 May 2003 11:20:02 -0700, Jesse Barnes <jbarnes@sgi.com> said:

  >> That's not safe in my mind.  What I'd like is an efficient way to
  >> preview a pull (see a diff -u between the remote tree and where I'm
  >> pulling into).  Note the "efficient" part.  Having to do a full clone
  >> first doesn't qualify.

  Jesse> I found myself wishing the same thing as I tried to get together a
  Jesse> patch for you.

Pulling from upstream is a bit easier: since there is only one
upstream, what I do clone the upstream locally and then do an
unconditional pull.  Then I do a diff between the last version of the
tree and the current one.  It does require maintaining two trees,
but that's not too bad.

The scheme doesn't work well if you pull from many places, though (I
don't know about you, but somehow I'm _always_ short of diskspace, no
matter how big the drives are getting...).

	--david