river-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Firmstone <j...@zeus.net.au>
Subject Re: New version of TaskManager
Date Mon, 13 Sep 2010 09:19:21 GMT
Patricia Shanahan wrote:
> Peter Firmstone wrote:
>> Thanks for the link, do you have a patch with some tests?  I can run 
>> them if you like.
>
> My next step is going to be to make the changes to the rest of the 
> code for the new interface, and start testing. I don't think there is 
> any need to commit until I've done at least some tests.

Good, that should buy me some time to fix trunk, in the mean time, I'm 
happy to run some tests for you, if you want, while your sorting the 
logging problem, or if you want the tests run on another architecture.

I'll be setting up a directory with a trunk replacement on the weekend, 
which can be reviewed and voted on.  This will be based on pepe with 
DynamicConcurrentPolicyProvider, RevokeableDynamicPolicy and 
StreamServiceRegistrar all removed, which I'll continue developing in 
pepe, when I merge back eventually, I don't want any Ghosts getting in 
the way, or failed experiments remaining in the trunk for that matter.

This should give me some time this week to review the commits over the 
previous months to make sure I haven't forgotten anything important.

Hopefully this will be approved quickly, so we can hone our merging skills.

Experimenting in trunk was efficient when I was the only committer, 
unfortunately it's proven quite counter productive of late.

>
>> Did you try Jonathan's recommendations for logging using a 
>> build.properties file and a logging.properties file in your home 
>> directory?
>
> I tried a build.properties and a logging file both in the trunk 
> directory, with the build.properties pointing to the logging file. I 
> checked the parameters being passed to a typical test, and it had 
> picked up my file, so the problem has to be in the file contents. I'll 
> do some more checking.
>
>> Jonathan's our build Guru, we'd be stuffed without his contributions.
>
> Indeed. I'm less knowledgeable than I really like in these areas, 
> because I did my original programming on punch cards. All my work on 
> large projects after that was done as part of a team that included a 
> build/tools guru who set everything up.

Well I'll bet you understand the intricacies of computing much better 
than many other programmers, having learned when everything ran much 
slower, with more time available to see what was happening,  computing's 
too great a field for anyone to know everything, which is why it's so 
important we all work together.

The hardest part is, knowing what we don't know, when we don't know 
about it.  - Humor, I can't plead complete ignorance, just ignorance & fear.

Cheers,

Peter.


Mime
View raw message