[PATCH hglib] tests: add test for log() method errors

[Augie Fackler]

On Apr 27, 2013, at 5:50 AM, Alexander Plavin <me at aplavin.ru> wrote:

> # HG changeset patch
> # User Alexander Plavin <me at aplavin.ru>
> # Date 1367066976 -14400
> # Node ID 0584acb30b9d7d7831cf4d74c116539c7ff9768e
> # Parent  c635e6e7054f2600d36343eaa1770cb36549382a
> tests: add test for log() method errors
> 
> Add test to ensure that log() raises an error when unknow revision specified.
> As hg log behavior with empty repo changed recently,
> two variants are allowed in test.
> 
> diff -r c635e6e7054f -r 0584acb30b9d tests/test-log.py
> --- a/tests/test-log.py	Fri Apr 26 01:46:08 2013 +0400
> +++ b/tests/test-log.py	Sat Apr 27 16:49:36 2013 +0400
> @@ -1,3 +1,4 @@
> +from hglib.error import CommandError
> import common
> import hglib
> 
> @@ -17,8 +18,11 @@
>         self.assertEquals(revs[0], self.client.log('0')[0])
>         self.assertEquals(self.client.log(), self.client.log(files=['a']))
> 
> -    # def test_errors(self):
> -    #     self.assertRaisesRegexp(CommandError, 'abort: unknown revision', self.client.log, 'foo')
> -    #     self.append('a', 'a')
> -    #     self.client.commit('first', addremove=True)
> -    #     self.assertRaisesRegexp(CommandError, 'abort: unknown revision', self.client.log, 'bar')
> \ No newline at end of file
> +    def test_errors(self):
> +        try:
> +            self.assertRaisesRegexp(CommandError, 'abort: unknown revision', self.client.log, 'foo')
> +        except AssertionError:
> +            self.assertEqual(self.client.log('foo'), [])

Not sure I understand this - in the event of an assertion error, why are we asserting that log() returns nothing? Shouldn't this command either raise or return all the time in this circumstance?

> +        self.append('a', 'a')
> +        self.client.commit('first', addremove=True)
> +        self.assertRaisesRegexp(CommandError, 'abort: unknown revision', self.client.log, 'bar')


> _______________________________________________
> Mercurial-devel mailing list
> Mercurial-devel at selenic.com
> http://selenic.com/mailman/listinfo/mercurial-devel