| 1 |
On 12/22/2015 12:26 AM, Alexander Berntsen wrote: |
| 2 |
> On 21/12/15 20:57, Zac Medico wrote: |
| 3 |
>> Since the Manifest "stable mtime" behavior could have undiscovered |
| 4 |
>> bugs |
| 5 |
> This isn't exactly very motivating. "Since Foo could have bugs" |
| 6 |
> applies to pretty much any Python code >10 lines. |
| 7 |
|
| 8 |
We could include micro-optimization as a motivation, since this option |
| 9 |
will eliminate some stat, listdir, and utime calls that should be pretty |
| 10 |
negligible in comparison to the overall cost of generating the manifest. |
| 11 |
|
| 12 |
I'm confident enough in the reliability of the stable mtime code, so |
| 13 |
from my perspective, micro-optimization is the most valid motivation. |
| 14 |
However, I'm not so sure that it's a good idea to disable stable mtime |
| 15 |
by default, since it seems like a reasonable behavior with negligible cost. |
| 16 |
|
| 17 |
> The patch itself looks OK though. |
| 18 |
|
| 19 |
Thank you! |
| 20 |
-- |
| 21 |
Thanks, |
| 22 |
Zac |
Archives