Code review comment for lp:~lifeless/subunit/nopassthrough

Revision history for this message
Jonathan Lange (jml) wrote :

On Wed, Aug 5, 2009 at 1:18 AM, Robert Collins<email address hidden> wrote:
> On Tue, 2009-08-04 at 08:53 +0000, Jonathan Lange wrote:
>> Review: Approve
>> > This adds --no-passthrough to most filers, which is useful for ensuring there
>> > is no noise in a stream. This is important for some uses, like when outputting
>> > structured data other programs will be reading.
>>
>> Hey Rob,
>>
>> This seems like a good feature & I can't fault the code. From the description in the MP though, I was expecting the behaviour to be that --no-passthrough makes the script error out when there is noise.
>>
>> Perhaps the option should be 'discard-noise' or something similar?
>
> disard-passthrough perhaps?
>
> I hesitate to call the other content 'noise' - its typically things like
> 'make' output, or extra debugging that faulty tests aren't capturing.
>
> So I'd like a neutral name for it.
>

I agree. I'm stumped for the right noun though.

jml

« Back to merge proposal