| 1 |
On 16/08/2016 18:40, Rich Freeman wrote: |
| 2 |
> On Tue, Aug 16, 2016 at 11:43 AM, Peter Humphrey <peter@××××××××××××.uk> wrote: |
| 3 |
>> On Tuesday 16 Aug 2016 16:58:20 hw wrote: |
| 4 |
>> |
| 5 |
>>> The messages are cryptic, and you have to guess what they are trying to |
| 6 |
>>> tell you. |
| 7 |
>> |
| 8 |
>> *Guess?* Good God, man. You don't guess - you use your experience and your |
| 9 |
>> logical reasoning to deduce likely cause and effect. |
| 10 |
>> |
| 11 |
> |
| 12 |
> Well, on this one I have to admit I sometimes feel like I'm breaking |
| 13 |
> on the divining rods when I get errors from portage. I don't think |
| 14 |
> anybody is really satisfied with them. The devs have been doing what |
| 15 |
> they can to improve this but I think the issue is that portage often |
| 16 |
> realizes that something is wrong, but it can't really tell what. |
| 17 |
> Often this is the result of poor developer practices. Just as often |
| 18 |
> adding --backtrack to the options fixes it. |
| 19 |
> |
| 20 |
|
| 21 |
I've trained my brain to recognize familiar output and mentally map what |
| 22 |
occurred the last time to the present time :-) |
| 23 |
|
| 24 |
I have also noticed that very often the bit that looks like an error is |
| 25 |
actually just info (think --verbose) and the real cause is often |
| 26 |
obscured inside a longer sentence. |
| 27 |
|
| 28 |
Practice I guess. |
| 29 |
|
| 30 |
-- |
| 31 |
Alan McKinnon |
| 32 |
alan.mckinnon@×××××.com |
Archives