1 |
On Tuesday 16 September 2008 21:04:59 Vaeth wrote: |
2 |
> Alan McKinnon wrote: |
3 |
> > "You asked me to do something. It didn't work |
4 |
> |
5 |
> But it is an annoyance if you leave your computer on during the three |
6 |
> days you are on the road to compile a load of new packages like e.g. |
7 |
> a new kde version, and when you return, compiling has not even started |
8 |
> because your first pacakge fortune-mod had a trivial file collision. |
9 |
|
10 |
Yes, it is indeed extremely annoying. As a KDE-4 and e17-svn user I'm all too |
11 |
aware.... But you have to consider the actual cost of a technique designed to |
12 |
increase convenience where the results are not determined. |
13 |
|
14 |
> > A failed emerge is by definition an error, and unpredictable. How can we |
15 |
> > expect software to dream up the best solution to an exception? |
16 |
> |
17 |
> Since the best solution to this exception is to finish that part of |
18 |
> the task which is not influenced by this error, I think the |
19 |
> expectation for this exception is clear. |
20 |
|
21 |
"Which is not influenced" - this is the crucial clause, the one that is |
22 |
fraught with error. Who is to say what "not influenced" actually means? A |
23 |
complete lack of any related dependencies is one workable way to scope it. It |
24 |
happens often enough that it's worth the effort to accommodate it. |
25 |
|
26 |
But it also represents an alternate unexpected code path and for that reason I |
27 |
believe it warrants a configuration option but not to be default. IOW, you |
28 |
have to ask for it to get it. |
29 |
|
30 |
If you haven't already guessed, I subscribe to the school of thought of "Don't |
31 |
Cause Unintended Side-Effects" |
32 |
|
33 |
-- |
34 |
alan dot mckinnon at gmail dot com |