[Aptitude-devel] Experimental package and package_pool implementation

[Piotr Galiszewski]

2010/7/15 Daniel Burrows <dburrows at google.com>:
> On Thu, Jul 15, 2010 at 11:57 AM, Daniel Burrows <dburrows at google.com> wrote:
>> On Thu, Jul 15, 2010 at 10:05 AM, Piotr Galiszewski
>> <piotr at galiszewski.pl> wrote:
>>> Today I have rebased my later patches to current code. Unfortunately
>>> there is one big problem with package_pool. Creating package list
>>> takes nearly 17 seconds on my machine. The problem is with
>>> boost::multi_index_container. I have changed this to std::vector and
>>> time is 0.3s now.
>>
>>  Interesting, I'd like to know why that is so we can avoid hitting it again.
>> Glad it's fixed, anyway.
>
>  Rather than spending your time on it, would you mind posting the SHA of
> the version with lousy performance so I can play with it and see what
> happened?  I have a suspicion about what caused it that I'd like to
> investigate.  (in any event, this doesn't *need* a multi index as far as I can
> see)
>
>  Is it cfac0549f1ae899cc1f18b07084eef1c9aa0b857?
>

yes that is this commit. Later I have rebased it, and overwritten, so
probably it is not available on gitirious

>  BTW: do you know about pkg->ID?  Packages are numbered from 0 to
> PackageCount()-1, and you can access that number.  In fact, it should be
> exactly the same as the number you use as the index into the pool.  No
> need for a side table to look packages up.
>

I have not known about this. But lazy loading is also helpful, so
probably this work is not a waste of time :)

>  Daniel
>
-- 
Regards
Piotr Galiszewski