Mailing-List: contact reviews-help@aurora.incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: reviews@aurora.incubator.apache.org
Content-Type: multipart/alternative;
 boundary="===============0317713075365978206=="
MIME-Version: 1.0
Subject: Re: Review Request 26688: Fix errors in help rendering:
From: "Mark Chu-Carroll" <mchucarroll@twopensource.com>
To: "Zameer Manji" <zmanji@twopensource.com>,
 "Joshua Cohen" <jcohen@twopensource.com>
Cc: "Mark Chu-Carroll" <mchucarroll@twopensource.com>,
 "Aurora" <reviews@aurora.incubator.apache.org>
Date: Wed, 15 Oct 2014 16:46:43 -0000
Message-ID: <20141015164643.24413.50046@reviews.apache.org>
Auto-Submitted: auto-generated
Sender: "Mark Chu-Carroll" <noreply@reviews.apache.org>
References: <20141014164654.24426.19762@reviews.apache.org>
In-Reply-To: <20141014164654.24426.19762@reviews.apache.org>
Reply-To: "Mark Chu-Carroll" <mchucarroll@twopensource.com>

--===============0317713075365978206==
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


> On Oct. 14, 2014, 12:46 p.m., Joshua Cohen wrote:
> > src/test/python/apache/aurora/client/cli/test_help.py, line 75
> > <https://reviews.apache.org/r/26688/diff/1/?file=720844#file720844line75>
> >
> >     Are option names guaranteed to be unique? If not this test could potentially pass if any help output contains a plugin option name, not necessarily the help output for the command to which the plugin was registered.
> >     
> >     It's also possible for a plugin option name to appear in the help for another option, and not on its own, which would cause this test to succeed even if the plugin options themselves are not properly displayed?
> >     
> >     I guess what I'm getting at is would it be better to test for more than just the appearance of a string at any point in the output?
> >     
> >     (This may be based on incomplete understanding of how the client registers commands/options).

I'm trying to keep the test from being overly sensitive to changes. The problem with a lot of output testing is that it's incredibly brittle - even tiny changes to the output require the test to get rewritten.  So when possible, I'd prefer to have tightly focused tests, which look for specific problems.

We know that we had a rendering bug that was causing plugin options to get omitted from usage strings - so this test tries to specifically check that that's not happening anymore. If it fails, we know exactly what the problem is.


- Mark


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26688/#review56541
-----------------------------------------------------------


On Oct. 15, 2014, 12:41 p.m., Mark Chu-Carroll wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/26688/
> -----------------------------------------------------------
> 
> (Updated Oct. 15, 2014, 12:41 p.m.)
> 
> 
> Review request for Aurora, Joshua Cohen and Zameer Manji.
> 
> 
> Bugs: aurora-831
>     https://issues.apache.org/jira/browse/aurora-831
> 
> 
> Repository: aurora
> 
> 
> Description
> -------
> 
> - Put plugin-generated options into the correct order.
> - Include the option-name in the detailed help list.
> - Add missing metavars.
> 
> 
> Diffs
> -----
> 
>   src/main/python/apache/aurora/client/cli/__init__.py da9d5b6ba4d22ba1f444341b97bbcfaf7889a4a8 
>   src/main/python/apache/aurora/client/cli/options.py dc76c25b90acb9610e40b939e65c3cabf032649f 
>   src/main/python/apache/aurora/client/cli/standalone_client.py 20f4d7ef43ba336a2b6d02cbf5656c97bdfa2ea1 
>   src/test/python/apache/aurora/client/cli/test_help.py f73c8a3778b7d118ea2865f213b442a607fb4a7d 
> 
> Diff: https://reviews.apache.org/r/26688/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Mark Chu-Carroll
> 
>


--===============0317713075365978206==--