1 |
On Mon, May 7, 2018 at 11:24 PM, Brian Dolbec <dolsen@g.o> wrote: |
2 |
> On Mon, 7 May 2018 13:38:47 -0700 |
3 |
> Matt Turner <mattst88@g.o> wrote: |
4 |
> |
5 |
>> |
6 |
>> If there's a way to have repoman alert developers to deprecated |
7 |
>> dependencies in the same way we handle deprecated eclasses, I'd like |
8 |
>> to know about it. |
9 |
>> |
10 |
>> |
11 |
> |
12 |
> Currently there is not. |
13 |
> |
14 |
> Thinking out loud... It would mean parsing package.mask to generate |
15 |
> the list filtering out those with "masked for removal", from other |
16 |
> general mask reasons, but even that is not consistent. |
17 |
|
18 |
Thanks for the reply. |
19 |
|
20 |
One clarification: the x11-proto/* packages aren't package.masked (for |
21 |
removal or otherwise) -- just deprecated. They still exist just to |
22 |
provide a simpler transition to x11-base/xorg-proto. After all reverse |
23 |
dependencies are updated, they will indeed be tree cleaned. |
24 |
|
25 |
But yeah. I kind of like your idea. Perhaps some method of marking a |
26 |
package deprecated would be a good addition to PMS. I can imagine it |
27 |
being useful in a case like mine, and it would provide a good path |
28 |
towards tree cleaning. I've always thought it was kind of silly to |
29 |
start the 30-day removal timer only after the last dependency is gone |
30 |
from the tree, so maybe a way of marking a package deprecated could |
31 |
speed that up. |
32 |
|
33 |
The workflow I envisage would be |
34 |
|
35 |
(1) Package is marked as deprecated by some method (call it package.deprecated) |
36 |
|
37 |
repoman would now emit warnings when it finds a dependency on the |
38 |
deprecated package, informing developers of the necessary changes and |
39 |
speeding up the transition. |
40 |
|
41 |
(2) When the transition is complete, the deprecated package could be |
42 |
masked for removal under some <30 day mask or perhaps removed directly |
43 |
if it's been deprecated for >=30 days. |
44 |
|
45 |
|
46 |
I'll figure out the process for suggesting a future EAPI feature... :) |