Jason van Zyl <[EMAIL PROTECTED]> writes:
> Hi,
>
> I was just discussing with John how the module loading works now and I would
> like to start a developer's guide outlining how this works so we can decide
> on the final behaviour because John found a glitch in the ordering. I need
> to document this better anyway.
>
> So I wanted to nuke all the docs in HEAD because I want to start from
> scratch and pull it bits from the 2.1 tree as needed. We should keep
> publishing the docs from the 2.1 branch.
>
> So I would like to
>
> 1. Nuke all the docs in HEAD and start with a developer's guide
There are a few docs in the HEAD that may have had changes made to
them which aren't reflected by the branch. There has also been
another whole doc added.
I would much prefer that the develop guide be written BEFORE any docs
are removed from the HEAD.
> 2. I want to stop checking in generated docs, it's a PITA. I think for
> for now we can use a simple rsync script to copy the docs onto the apache
> machine efficiently. I'm using jon's approach of making a simple procedure
> that is not ideal but we should move toward not storing generated anything
> of anykind in CVS. This will only be for the HEAD docs anyway for now so it
> won't really affect anyone who isn't writing 3.0 docs.
+1, definitely.
> At any rate I'll make a tiny section of the developer's guide outlining the
> module searching/loading order so that we can clarify this.
Cool.
> We can probably reimport the tree soon too as it is almost entirely cleaned
> out now.
Can you clarify this? I don't understand.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
