[PATCH hglib] tests: handle the removal of `obsolete._enabled` in Mercurial

Matt Harbison mharbison72 at gmail.com
Tue May 21 21:53:59 EDT 2019


On Mon, 20 May 2019 11:42:49 -0400, Augie Fackler <raf at durin42.com> wrote:

> On Tue, May 07, 2019 at 10:21:28PM -0400, Matt Harbison wrote:
>> # HG changeset patch
>> # User Matt Harbison <matt_harbison at yahoo.com>
>> # Date 1557281819 14400
>> #      Tue May 07 22:16:59 2019 -0400
>> # Node ID 1febca34c800a64cc2c886d24920b936674cbd1c
>> # Parent  33b512aa8dba0cbe523188fbb62d30ae2125a236
>> tests: handle the removal of `obsolete._enabled` in Mercurial
>>
>> I'm not sure why we can't just set `experimental.evolution=all`, but it  
>> didn't
>> work.
>>
>> diff --git a/tests/test-hidden.py b/tests/test-hidden.py
>> --- a/tests/test-hidden.py
>> +++ b/tests/test-hidden.py
>> @@ -22,7 +22,7 @@
>>          super(test_obsolete_baselib, self).setUp()
>>          self.append('.hg/obs.py',
>>                      "import mercurial.obsolete\n"
>> -                    "mercurial.obsolete._enabled = True")
>> +                    "mercurial.obsolete.isenabled = lambda r, opt:  
>> True")
>>          self.append('.hg/hgrc','\n[extensions]\nobs=.hg/obs.py')
>
> We generally try to have hglib work across many hg versions - should we  
> do both instead of just one or the other?

I sent a V2 that does both.  V1 would support all the way back to 3.2 (Nov  
2014).  I'm not sure how far back we try to go.


More information about the Mercurial-devel mailing list