New 1.14 RC date?

[Jason Earl]

Ian Clatworthy <ian.clatworthy at internode.on.net> writes:

> Jason Earl wrote:
>
>> On the other hand, at least *I'll* be able to get back to being able
>> to test the conversion.  So that's something.  Who knows, it's even
>> possible that the brisbane-core format won't even work as well for
>> Emacs as the 1.9 format repositories.
>
> I haven't forgotten the fast-import bug affecting the Emacs
> conversion.  I've just been *extra* busy on EOL conversion & tuning
> brisbane-core for bulk data entry (fast-import) and log. FWIW, I fixed
> a bug this morning that was causing me grief converting the last 5K of
> the 105K revisions in Emacs. I'm rerunning a conversion right
> now. That particular bug wasn't brisbane-core related so it will help
> the conversion to 1.9 or 1.14 as well.

I need to just be quiet.

I was just trying to throw my weight around a bit, and re-reading it I
sound more petulant than anything.  I apologize.  I'll try and add more
smileys to my next several emails.

The real problem is that I am *dying* to play with the brisbane-core
stuff, but I just can't justify moving my machines to a non-release
version of bzr.  I really want brisbane-core included in the next
release, and in my quest to be persuasive I am afraid that I went a
little too far.

I'm confident that brisbane-core will be awesome, and I truly appreciate
the help and support that the bzr development team has given me.
Re-reading my comments it would be easy to assume that I am ungrateful
for the help and unimpressed by the rapid progress that bzr has made.
Let me assure you that nothing could be further from the truth.

I actually saw the new commit to the lp:bzr-fastimport branch, and I
have also noticed that you have made several changes over the last few
weeks that were related directly to Emacs (I reload launchpad.net about
once an hour to see what's changed).  I really appreciate your commits.

I will admit that in my private branch that I use to actually do my
Emacs conversions that I just made _sticky_blobs in cache_manager.py a
anydbm database.  It takes up a lot of disk space, but the conversion is
actually faster on my hardware (or at least it used to be, you've made
some changes that probably have changed that), and I don't have to make
two passes.  Now that I think about it I don't believe I have ever had a
successful import using fast-import-info.

I realize that you are ridiculously busy landing the EOL stuff and
working on brisbane-core.  What's more, I *want* you to work on
brisbane-core :).  That's likely to be the format that Emacs ends up
using for its repository anyway.  That's especially true if the core bzr
devs think that the brisbane-core format is likely to finally become the
one format to rule them all.

> In fact, I convert Emacs at *least* once or twice a week to the
> brisbane-core format. I benchmark as much as I can using Emacs as my
> sample data. Branching outside a shared-repo is a bit slower still.
> Otherwise, brisbane-core is looking extremely good IMO.

Did I mention that I am *dying* to play with this stuff?

>> I'll try not to be impatient.
>
> You're *very* close to the top of my list. But I feel that all the
> stuff I've been doing - log of DIRs, EOL, fast-import -> brisbane
> core, reviewing brisbane-core code - will get you there faster
> anyhow. I realise that doesn't reduce your frustration in trying to
> get a conversion to 1.9 done but please bear with me a little longer

Pay no attention to my frustration.  Besides, I am definitely not
frustrated with you.  I just was trying to weigh in in favor of getting
brisbane-core into 1.14.  I probably should have just said something
like:

+1 postponing 1.14 for brisbane-core

and been done with it.  Not that my +1 should count for anything.  I'm
just an end user with delusions of grandeur.

I have far more experience lurking on these lists than posting on them.
I need to go sit in the corner for a bit until I can learn to be a good
boy.

I am very grateful that I was curious enough about bzr to begin the
conversion process for the Emacs team.  As I said before, I was just
some random lurker on that list looking for a way that I could help out
a bit.  I have since learned a great deal about both Emacs and bzr
(mostly about bzr), and I have become convinced that helping out Free
Software projects is the best way imaginable to improve my own hacking
skills.  I only wish that I could be more helpful right now.  A huge
part of the reason that I keep refreshing launchpad.net to see what has
changed is that I have learned quite a bit just watching bzr-fastimport
progress.  That project is small enough that I don't feel overwhelmed
studying it.

As such, I don't believe that I am ever likely to be able to repay the
growing debt of gratitude that I owe you personally.

So if I have seemed even a little ungrateful please accept my humblest
of apologies.  I just want to start playing with the brisbane-core code.
In my excitement I got a bit carried away.

Jason