1 |
On Mon, Sep 11, 2006 at 04:02:53PM -0700, Chris White wrote: |
2 |
> On Monday 11 September 2006 15:22, Ciaran McCreesh wrote: |
3 |
> > * Otherwise, try again with ``._cfg0001_name``, then ``._cfg0002_name`` and |
4 |
> > so on (base ten is used for the number part) until a usable filename is |
5 |
> > found. |
6 |
> For what purpose are the older cfg[number]_name files kept around? I ask |
7 |
> because I would anticipate the default behavior for replacing configuration |
8 |
> files with their pending updates to be picking the newest update. That said, |
9 |
> the previous versions would not serve a purpose, or is there something I |
10 |
> don't see? |
11 |
Consider a case as follows: |
12 |
1. package foo provides /etc/bar - the contents of the file differ depending on the USE flags. |
13 |
2. an old version of foo is already installed, with USE='a' |
14 |
3. user does "USE='a b' emerge foo" - /etc/._cfg0000_bar is created. |
15 |
4. user then realizes he actually wanted USE=c as well, so does "USE='a b c' emerge foo" - this provides /etc/._cfg0001_bar. |
16 |
|
17 |
You now have 3 files, non-identical: |
18 |
/etc/bar |
19 |
/etc/._cfg0000_bar |
20 |
/etc/._cfg0001_bar |
21 |
|
22 |
The user now runs etc-update or dispatch-conf or whatever tool they use |
23 |
to manage their configurations. |
24 |
|
25 |
The 'try again rule to find a usable filename' is specifically intended |
26 |
for these cases where the configuration management is NOT used between |
27 |
more than two changes of a configuration file (Ideally it should be, but |
28 |
that's a different discussion). |
29 |
|
30 |
-- |
31 |
Robin Hugh Johnson |
32 |
E-Mail : robbat2@g.o |
33 |
GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 |