bug#33468: A bug with yes and --help

[Pádraig Brady]

On 15/02/19 10:32, Assaf Gordon wrote:
> Hello Eric and all,
> 
> 
> Thanks for the quick and detailed review.
> I've amended all the issues you mentioned.
> 
> On 2019-02-13 8:20 p.m., Eric Blake wrote:
>>>   15 files changed, 46 insertions(+), 141 deletions(-)
>>
>> Nice diffstat.
> 
> These are of course Bernhard's improvements,
> I just did the testing (and some minor things).
> 
>>> diff --git a/NEWS b/NEWS
>>
>> Is "argument" better than "option" here?  Or, maybe:
>>
>> now always process --help and --version options, regardless of any other
>> arguments present before any optinoal -- end-of-options marker.
> 
> I've used your phrasing, and also separated "nohup" from the rest
> of the programs, as it does not accept --help/--version anywhere,
> just as first arguments.
> 
> Attached updated patches, with tests.
> 
> comments welcomed,
>   - assaf
> 
>   P.S.
> There is at least one change in behavior, not sure if this is
> bad enough to be a regression or doesn't really matter:
> 
>    $ yes-OLD me -- --help | head -n1
>    me -- --help
> 
>    $ yes-NEW me -- --help | head -n1
>    me --help

I wouldn't worry about that.
If it could be controlled by POSIXLY_CORRECT all the better,
but still not worth worrying about.
It would be appropriate for a test, at least for documentation.

this is looks ready to land.

thanks!
Pádraig