| 1 |
>>>>> On Mon, 19 May 2014, Jeroen Roovers wrote: |
| 2 |
|
| 3 |
> Are you sure this is useful? |
| 4 |
|
| 5 |
>> elisp-site-regen() { |
| 6 |
> [...] |
| 7 |
>> - if [[ ! -d ${T} ]]; then |
| 8 |
>> - eerror "elisp-site-regen: Temporary directory ${T} does not exist" |
| 9 |
>> - return 1 |
| 10 |
>> - fi |
| 11 |
|
| 12 |
> "The package manager must define the following environment variables. |
| 13 |
|
| 14 |
> [...] |
| 15 |
|
| 16 |
> "T [...]The full path to a temporary directory for use by the |
| 17 |
> ebuild."[1] |
| 18 |
|
| 19 |
So is the theory. The test exists because this wasn't always the case |
| 20 |
in practice [2]. |
| 21 |
|
| 22 |
I'd rather leave it there, because if T is empty then files will be |
| 23 |
written to / instead. There is no sandbox in post{inst,rm}, so being |
| 24 |
extra careful won't harm. |
| 25 |
|
| 26 |
Ulrich |
| 27 |
|
| 28 |
> [1] http://dev.gentoo.org/~ulm/pms/5/pms.html#x1-11800011.1 |
| 29 |
[2] http://paludis.exherbo.org/trac/ticket/517 |
Archives