1 |
>>>>> "ZM" == Zac Medico <zmedico@g.o> writes: |
2 |
|
3 |
ZM> On 04/06/2010 07:22 AM, James Cloos wrote: |
4 |
>>>>>>> "ZM" == Zac Medico <zmedico@g.o> writes: |
5 |
>> |
6 |
ZM> You can configure eclass override behavior via eclass-overrides in |
7 |
ZM> /etc/portage/repos.conf, as documented in `man portage`. |
8 |
>> |
9 |
>> ,----< From that manpage > |
10 |
>> | When using eclass-overrides, due to bug #276264, you must ensure that |
11 |
>> | your portage tree does not contain a metadata/cache/ directory. |
12 |
>> `---- |
13 |
>> |
14 |
>> Which translates into "eclass-orderrides are completely and entirely |
15 |
>> useless, so don't bother. |
16 |
|
17 |
ZM> Well, it's roughly equivalent to the old default behavior (which you |
18 |
ZM> apparently preferred). However, the issue is now complicated by the |
19 |
ZM> fact that FEATURES=metadata-transfer is disabled by default, so when |
20 |
ZM> portage goes to pull cache directly from metadata/cache/, it won't |
21 |
ZM> be able to validate eclass changes since there are no eclass |
22 |
ZM> timestamps saved inside metadata/cache/. |
23 |
|
24 |
Portage does not need to validate eclass changes. |
25 |
|
26 |
-JimC |
27 |
-- |
28 |
James Cloos <cloos@×××××××.com> OpenPGP: 1024D/ED7DAEA6 |