Chris Shoemaker's (Jul) patch for account-*-balance refactoring and code

Chris Shoemaker c.shoemaker at cox.net
Sat Oct 8 01:08:09 EDT 2005


On Fri, Oct 07, 2005 at 06:40:43PM -0400, Joshua Sled wrote:
> Log Message:
> -----------
> Chris Shoemaker's (Jul) patch for account-*-balance refactoring and code format/comment massage.


:( 

Ok, I'm pretty keen on never again having to do what I just spent this
evening doing.  I know this was mostly my own fault (not Josh's), but
I want to choose the most convenient way of preventing it from
happening again, which depends on you other devs.

Here's what happened: At some time long after sending those patches, I
made more changes to Account.[ch].  Unless I go out of my way to make
a new patch containing just the new changes, the new changes just get
wrapped up into the same patch when the patch is refreshed.  Since the
patch I sent hadn't been ack'd, nack'd, or commented on, I figured it
would be ignored until I resubmitted, so I didn't make a new patch.

When Josh did apply the patch, it was, of course, the version from
July, the only version I ever mailed.  Unfortunately, this left me
with a large patch containing *mostly* stuff just applied.  Filtering
out that remaining new stuff turned out to be more painful than one
might think.

So how can I prevent this?  One solution would be for me to not ever
change patches after I mail them.  The disadvantage of this is that
sometimes multiple changes group naturally and/or logically into one
patch.  Also, it does raise the ambiguity of patch ordering: I can't
know in which order the patches will be applied, so I guess I must
pick an order and note the dependence.  It's also a little more work.

Another solution would be for anyone to check with me for a newer
version if they're applying an old patch.  That would probably save
work for everybody, since I have to keep my patch stack fresh anyway
in order to code.

Another solution would be to only apply patches "soon" after
submission.  IOW, we'd agree that the patches expired after some time,
and I would be free to change and resubmit after that time.

Any other suggestions or ideas?

-chris

p.s. All things considered, things could have been *much* worse.  If
I'd sent more than just the small sample of patches, I could've been
sorting for a week instead of just an evening.  But, I do want to send
a much larger batch of patches pretty soon, so let's settle on
something that won't incapacitate me.


More information about the gnucash-patches mailing list