| 1 |
On Fri, Aug 26, 2005 at 03:02:13PM -0700, Drake Wyrm wrote: |
| 2 |
> Brian Harring <ferringb@g.o> wrote: |
| 3 |
> > B) EAPI is pretty much bash env template switching |
| 4 |
> [snip] |
| 5 |
> |
| 6 |
> Perhaps the EAPI handling could be implemented using eclasses, rather |
| 7 |
> than something in the deep, dark, python-based internals. |
| 8 |
|
| 9 |
Effectively the implementation is essentially an eclass, but won't |
| 10 |
wind up in $PORTDIR/eclass due to the fact it's also slightly bound to |
| 11 |
python side. |
| 12 |
For example, the ebuild build operation class knows not to command the |
| 13 |
ebuild processor to execute the configure phase for eapi0, 'coz that |
| 14 |
hook doesn't exist. For >eap0, it commands it. That's about the |
| 15 |
extent of python side awareness at this point, beyond checking min/max |
| 16 |
eapi support. |
| 17 |
~harring |
Archives