Patchwork [7,of,8] tests: conditionalize test-phases.t output for chg

login
register
mail settings
Submitter Pulkit Goyal
Date March 26, 2020, 7:36 a.m.
Message ID <eedfec1725f230d9d074.1585208214@workspace>
Download mbox | patch
Permalink /patch/45902/
State Accepted
Headers show

Comments

Pulkit Goyal - March 26, 2020, 7:36 a.m.
# HG changeset patch
# User Pulkit Goyal <7895pulkit@gmail.com>
# Date 1585148595 -19800
#      Wed Mar 25 20:33:15 2020 +0530
# Node ID eedfec1725f230d9d074ffca2d543b119bc18954
# Parent  71f078cbf9ea839829229a8935a489200d1f8d36
tests: conditionalize test-phases.t output for chg

IIUC chg loads the ProgrammingError class which leads to it directly referencing
it in error output. This makes the test pass on chg.
Yuya Nishihara - March 26, 2020, 11:33 a.m.
On Thu, 26 Mar 2020 13:06:54 +0530, Pulkit Goyal wrote:
> # HG changeset patch
> # User Pulkit Goyal <7895pulkit@gmail.com>
> # Date 1585148595 -19800
> #      Wed Mar 25 20:33:15 2020 +0530
> # Node ID eedfec1725f230d9d074ffca2d543b119bc18954
> # Parent  71f078cbf9ea839829229a8935a489200d1f8d36
> tests: conditionalize test-phases.t output for chg
> 
> IIUC chg loads the ProgrammingError class which leads to it directly referencing
> it in error output. This makes the test pass on chg.

Perhaps the exception would be caught at different place and printed
differently.
Pulkit Goyal - March 30, 2020, 2:02 p.m.
On Thu, Mar 26, 2020 at 5:05 PM Yuya Nishihara <yuya@tcha.org> wrote:
>
> On Thu, 26 Mar 2020 13:06:54 +0530, Pulkit Goyal wrote:
> > # HG changeset patch
> > # User Pulkit Goyal <7895pulkit@gmail.com>
> > # Date 1585148595 -19800
> > #      Wed Mar 25 20:33:15 2020 +0530
> > # Node ID eedfec1725f230d9d074ffca2d543b119bc18954
> > # Parent  71f078cbf9ea839829229a8935a489200d1f8d36
> > tests: conditionalize test-phases.t output for chg
> >
> > IIUC chg loads the ProgrammingError class which leads to it directly referencing
> > it in error output. This makes the test pass on chg.
>
> Perhaps the exception would be caught at different place and printed
> differently.

I see. Fun fact: on py3 with chg, this change is not needed.

Patch

diff --git a/tests/test-phases.t b/tests/test-phases.t
--- a/tests/test-phases.t
+++ b/tests/test-phases.t
@@ -896,11 +896,13 @@  Check we deny its usage on older reposit
   $ hg --config "phases.new-commit=internal" commit -m "my test internal commit" 2>&1 | grep ProgrammingError
   ** ProgrammingError: this repository does not support the internal phase
       raise error.ProgrammingError(msg)
-  mercurial.error.ProgrammingError: this repository does not support the internal phase
+  mercurial.error.ProgrammingError: this repository does not support the internal phase (no-chg !)
+  ProgrammingError: this repository does not support the internal phase (chg !)
   $ hg --config "phases.new-commit=archived" commit -m "my test archived commit" 2>&1 | grep ProgrammingError
   ** ProgrammingError: this repository does not support the archived phase
       raise error.ProgrammingError(msg)
-  mercurial.error.ProgrammingError: this repository does not support the archived phase
+  mercurial.error.ProgrammingError: this repository does not support the archived phase (no-chg !)
+  ProgrammingError: this repository does not support the archived phase (chg !)
 
   $ cd ..