1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
Hokay... for those users who are bored, have some time, and feel like |
5 |
hunting through portage's source I've opened bug #61881 ( |
6 |
http://bugs.gentoo.org/show_bug.cgi?id=61881 ). |
7 |
|
8 |
Basically, |
9 |
def funcx(a,b='',c='',d=''), |
10 |
does not mean you should call it as |
11 |
funcx(1,2,3,4) |
12 |
|
13 |
Specifying b, c, and d as optional args is just that- python doesn't |
14 |
require those args, and passing them in as positional is just |
15 |
- --asking-- for a bug that will cause random segments of code to implode |
16 |
as their args are switched around. It's *bad* practice, don't do it. |
17 |
|
18 |
Any users looking to chip in, please split patches against either |
19 |
portage-2.0.51_pre20 (currently package.masked), or the most current |
20 |
cvs snapshot in http://dev.gentoo.org/~ferringb/portage-cvs/ . Patches |
21 |
should be attached to bug #61881. Any questions you have regarding the |
22 |
source (fex: what does func xyz of portage.py do?) should be directed |
23 |
here. |
24 |
|
25 |
So... in summary, if you're looking for an excuse to familiarize |
26 |
yourself with the portage source, now would be the time, and would be |
27 |
appreciated. The sooner this is taken care of, the better. |
28 |
As is calling a function and specifying an optional arguement via |
29 |
position makes the caller, and the function prototype *very* fragile- |
30 |
prefix a new optional arg, or change the order of optional args in the |
31 |
prototype and the caller no longer is specifying args correctly, |
32 |
leading to bugs down the line. |
33 |
~brian |
34 |
-----BEGIN PGP SIGNATURE----- |
35 |
Version: GnuPG v1.2.4 (Darwin) |
36 |
|
37 |
iD8DBQFBLqyXvdBxRoA3VU0RAkbgAJ4gVX+Dvf3E9SLA1Kxe0PFsSkYYkgCdFLHm |
38 |
Wq9PUq35DHp+PeaS1hvngYI= |
39 |
=4+Gv |
40 |
-----END PGP SIGNATURE----- |
41 |
|
42 |
|
43 |
-- |
44 |
gentoo-portage-dev@g.o mailing list |