"Sander Striker" <striker@apache.org> writes:
> I'm sure Alex wants to contribute in that department.
>
> I think it is wise to 'educate' the developers in the mean time, so
> the same mistakes won't be introduced while others are being fixed.
What I'm getting at is, it'll be better to figure out some code
changes first and *then* write the new material for HACKING, based on
the tangible experiences in the code.
That patch to HACKING would have had no effect on people's coding.
The generic practices that it recommends most of us here already know.
Those who don't know quickly find out anyway, because of all the peer
review. I don't think there's any place in Subversion itself where we
read arbitrary length data into a fixed-length array; if there is,
it's because it slipped by a bunch of people who are well aware it's a
no-no.
The stuff about security boundaries is helpful, but it wasn't specific
enough to actually change anyone's behavior (I didn't know what I was
supposed to do differently, for instance). It needs concrete examples
from Subversion to be effective, that's all I'm saying.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Apr 23 18:03:09 2002