great! changed and tested:
On 08/30/13 12:52 PM, Xiaobo Shen wrote:
On 08/30/13 11:14 AM, Erik Trauschke wrote:
Cool. I updated the test by checking the actual output. "grep -v"
On 08/30/13 11:03 AM, Xiaobo Shen wrote:
On 08/29/13 05:21 PM, Xiaobo Shen wrote:
On 08/29/13 04:51 PM, Erik Trauschke wrote:Here is a enhanced version, which verifies human version is not
On 08/29/13 04:37 PM, Xiaobo Shen wrote:Ok. I fixed it and re-ran the test. the previous tests would never
please see the following link: thanks.
Have you tried to run the new test case without your fix and checked
that it actually fails?
Because looking at your test case it doesn't seem like it would find
fail because it did not capture the bug.
displayed if identical with version number and the version is still
Instead of using exit=1 you can just use "grep -v".
I would also rename the package to something else, 'humaniden' sounds
like some new kind of humanoid race they found at the bottom of the
maybe human2 or human-ident.
Otherwise this looks good to me.
probably does not capture the bug, because it still produce some output.
Please see the following link:
You can do that but you'd have to make sure the test is run in the correct locale, otherwise the dates pkg info prints might be in a different format.
Maybe the easiest way is grepping for "Version: 0.9.8.18$" (note the $)
This way it will fail if anything is printed after the 0.9.8.18.
[pkg-discuss] Re: Review request 17355511 pkg info displays human-version even when identical to actual version