[Omake] test/object/Test14

Aleksey Nogin anogin at hrl.com
Tue Jul 31 17:30:25 PDT 2007


On 31.07.2007 15:57, Jason Hickey wrote:

> To me this is a bit frightening--trying to fix naming problems by 
> introducing some short-term policies in 0.9.8.x.  It isn't always easy 
> to undo the "fix" when it reaches the naming branch, because of conflicts.

My impression was that this issue would "survive" the naming changes in 
one form, or another. In particular, when you have nested method calls 
(A.foo calling B.bar), I expect that the exact same solution would still 
be required to separate the "this" members that came from A (and need to 
be exported) and "this" members that came from B (and need to be hoisted 
back into B).

In short, I attempted to create a permanent solution to the problem that 
I did not expect to go away. In case I am wrong and it does fully go 
away with naming changes, then the right thing to do would probably be 
reverting my changes and removing object hoisting from 0.9.8.5 (and 
whatever jumbo branches that do not have what it takes to make this work 
right).

In any case, I just do not want to release 0.9.8.5 in a form that will 
break too much. In particular, I do not want it to break scripts that I 
need for work (which were the source of most of the "foreach" test cases 
I've been adding).

Aleksey

P.S. Right now my only TODOs for 0.9.8.5 are Dynamic/Shared and today's 
hosting stuff, specifically:

- Make sure we agree on the Shared AKA Dynamic (so that we do not 
release something that will be renamed in the next release)

- Check with Jason whether he thinks my today's hosting changes are good 
enough for 0.9.8.5, or whether he prefers to revert them and go for some 
other solution (like "no object hoisting") for 0.9.8.5

-- 
Aleksey Nogin, Research Scientist
Advanced Technologies Department, Information & System Sciences Lab
HRL Laboratories, LLC, Malibu, CA


More information about the OMake-Devel mailing list