[Grok-dev] Why the grokproject specific addition to bootstrap.py of installing eggbasket?

[Maurits van Rees]

Roger Erens, on 2009-06-04:
> Thanks for the clearing this up to me.
>
> I have two remaining thoughts:
>
> I) When a user has already defined (and filled) an eggs-directory (e.g. 
> via the ~/.buildout/default.cfg file), the speed improvement will be 
> less pronounced. Thus the use case for this optimazitation is mainly 
> when a newly released tarball is to be processed by the eggbasket recipe?

Yes.

Well, the biggest reason for creating this recipe is to avoid failing
to install grok when one of the single points of failure is down
(servers that host eggs that grok depends on).  Better said: it lowers
the number of single points of failure.

> II) In the total workflow, the speed optimization may be even less 
> pronounced, as I still have a Bottle Neck Existing Between Keyboard And 
> Chair. Only after I come back from my coffee break that follows the step 
> 2) you described above, I find the prompt 'You may now run bin/buildout' 
> that is awaiting my command for some indetermined time. Now, I have to 
> go for a second cup of coffee to await the results of step 3) and 4).
>:-/
> This could be solved by adding another addition to bootstrap.py that 
> involves running just plain bin/buildout, right?

Yes, it could.  But then it would make sense to add this to the
original bootstrap.py instead of specifically in our version.  But it
is good that the user learns that he should run bin/buildout, as he
will usually be running that a lot more than the one-time bootstrap.

-- 
Maurits van Rees | http://maurits.vanrees.org/
            Work | http://zestsoftware.nl/
"This is your day, don't let them take it away." [Barlow Girl]