Phobos.testing

Michel Fortin michel.fortin at michelf.com
Sun Oct 11 05:13:22 PDT 2009


On 2009-10-11 03:56:55 -0400, "Denis Koroskin" <2korden at gmail.com> said:

> I submitted a few Phobos bugs to bugzilla. They are still not 
> addressed.  Having 2-3 people with write access to Phobos is clearly 
> not enough -  there is not enough human power. That's bugzilla entries 
> are left without  answers, bugs are not fixed.
> 
> I don't submit them anymore. It just doesn't work. I see a lot of 
> quirks  in Phobos, huge performance problems (it allocates every time, 
> often  without any reason) and just typos.
> Given a direct svn access, I could easily fix some of them, but I'm too 
>  lazy to waste my time on creating one line long patches, making 
> bugzilla  reports, etc. And what then? Waiting like 3 years until they 
> are  addressed? No, thanks.

Somehow I wonder if a distributed versioning system wouldn't be better 
to encourage public participation and make it easy for maintainers to 
accept patches. It'd be easy for me and others to maintain their own 
fork of Phobos with their own fixes while we test them, and for Phobos 
maintainers to review, select and merge back in the mainline any 
addition (whole branches or single commits) made in those forks. It'd 
be a much more automated process than applying patches from bugzilla, 
and that way you don't have to give access to the mainline to a lot of 
people. It'd require people to know the tool though.

-- 
Michel Fortin
michel.fortin at michelf.com
http://michelf.com/




More information about the Digitalmars-d mailing list