[PATCH stable?] import: respect diff.noprefix config (issue4639)

[Ryan McElroy]

On 1/22/2016 7:15 PM, Pierre-Yves David wrote:
>
>
> On 01/22/2016 10:43 AM, Siddharth Agarwal wrote:
>> On 1/22/16 06:51, Ryan McElroy wrote:
>>> # HG changeset patch
>>> # User Ryan McElroy <rmcelroy at fb.com>
>>> # Date 1453473839 28800
>>> #      Fri Jan 22 06:43:59 2016 -0800
>>> # Branch stable
>>> # Node ID 40555a31b98b027c91b1ccf61477151b3307deb2
>>> # Parent  158bdc8965720ca4061f8f8d806563cfc7cdb62e
>>> import: respect diff.noprefix config (issue4639)
>>>
>>> Previously export-import round-trips would not work if 
>>> diff.nopreofix was
>>> truthy, because import did not pay attention to that setting. Now we
>>> inspect
>>> that setting unless an explicit strip level is specified.
>>
>> I don't think this is the right solution here -- much of the time hg
>> import is used on a repository that's different from the one where hg
>> export was used. Could we figure out from the patch whether to use -p0
>> or -p1?
>
> I'm pretty unenthusiastic for this solution too. changing the behavior 
> of import based on an option meant for diff display seems sloppy to me.
>
> I all cases, this is too much of a behavior change for the freeze. 
> Let's discuss it in February.
>

Then we need something in bugzilla that makes it more clear if an issue 
is eligible for a fix during the freeze.

Bug-squash-a-thons like the one we held today are useless if we can't 
figure out the right bugs to work on.

Cheers,

~Ryan