Hi Darren,

On Sep 30, 2011, at 2:05 AM, Darren Kenny wrote:

> Hi Seth,
> 
> In general it looks good - I have some comments, but mainly nits TBH and
> could be ignored to avoid churn:
> 
> legacygrub.py:
> 
>      404 +            if (argdict.get('bootfs', None) is None and
>      405 +                argdict.get('rpool', None) is None):
> 
> - I know it's consistent with the rest of the code that I can see, but I
>  wonder why you don't do the following to check if a specific key is in
>  the dict():
> 
>       if 'bootfs' not in argdict and 'rpool' not in argdict:

  Originally, it was possible for those keys to exist but have None values;  
that's not the case anymore, though, so I'll change it.

> 
> 
> menulst.py:
> 
>      132 +            if idx is -1:
>      133 +                idx = self._find_command_index('title')
>      134 +            if idx is not -1:
>      135 +                self._cmdlist.insert(idx + 1, mlcmd)
>      136 +                return
> 
> - I'm not sure that I like the use of 'is' and 'is not' with integers,
>  while it does appear to work, it's certainly not in keeping with normal
>  Python coding practice, and even conflicts with similar code on the
>  lines directly preceding the above:
> 
>      127 +            if idx != -1:
>      128 +                self._cmdlist.insert(idx + 1, mlcmd)
>      129 +                return
> 

  That's true -- I've made them consistent.


> 
> bootconfig.py:
> 
>      516 +        if len(default_instances) == 0:
> 
> - While this works, it's common to use simply 'if not default_instances:'
>  to test this. This checks for both the value being None, and the list
>  (and strings BTW) having a length of 0.

  Yea, I know but all I really want to test is if default_instances has 0 items 
-- I know it's not None.  I'm constantly torn between Pythonisms and 
readability of code ;).

 Thanks,
 --S

> 
> Thanks,
> 
> Darren.
> 
> On 30/09/2011 05:55, Seth Goldberg wrote:
>> Hi,
>> 
>>   Can I please have a few reviewers for:
>> 
>>    https://cr.opensolaris.org/action/browse/caiman/sethg/7096624/webrev/
>> 
>>   This is a stopper and it's been fully tested (100% code coverage of the 
>> changes verified with the coverage tool) with a variety of menu.lst files.
>> 
>>  Thanks,
>>  --S
>> _______________________________________________
>> caiman-discuss mailing list
>> [email protected]
>> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss

_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss

Reply via email to