| 1 |
Hola. |
| 2 |
|
| 3 |
Continuing the tradition of being a lazy bugger and raiding from my |
| 4 |
previous work instead of doing something new, attached is a patch that |
| 5 |
forces unpack to go through a level of indirection. |
| 6 |
|
| 7 |
Essentially... rather then direct access to $DISTDIR, we create a |
| 8 |
tmpdir, symlink in the _known_ files that portage believes that ebuild |
| 9 |
uses, and the unpack function looks in that directory. |
| 10 |
|
| 11 |
Can be defeated by a unpack ${DISTDIR}/file call, but that's invalid |
| 12 |
anyways. |
| 13 |
|
| 14 |
Why do this? Well, mainly it's a catch for ebuilds that aren't |
| 15 |
listing files in their src_uri when fetch restricted. No file in uri, |
| 16 |
no file in accessed distdir essentially. |
| 17 |
|
| 18 |
It also has the nice side affect that if we have ever screw up src_uri |
| 19 |
parsing, the ebuild will puke rather then the internal portage error |
| 20 |
sliding by- that's not really the intention of the patch however, just |
| 21 |
a nice assert of sorts. |
| 22 |
|
| 23 |
Either way, it's attached. Comments? |
| 24 |
~harring |
Archives