Hi, Michael

Thanks for review!

Michael Hunter wrote:
> On Wed, 09 Jan 2008 13:48:36 -0800
> Zhenghui.Xie at Sun.COM wrote:
>
>   
>> http://cr.opensolaris.org/~xzh/walkprop/
>>
>> This webrev includes
>> 1. the property-walk-through command
>> 2. a fix for the bug Jia pointed out
>> 3. a better/user-friendly display format (for instance, it will accept 
>> and report "true/false" for a boolean value parameter).
>>
>> Thanks for review!
>>
>> BTW, in case anybody curious, the up-to-date nwamcfg(1M) man page is on 
>> the way...
>>     
>
> nwamcfg.c:313,335 what is this suppose to be?  maybe 
> sizeof(ncu_prop_table)/sizeof(ncu_prop_table[0])
> way nit: personally I like SIZE or NELEMS for number of elements in a
> table and LEN for length of buffers because I think they carry more of
> the meaning then NUM
>
>   

I think what you mentioned is a valuable suggestion. The reason I used 
this format is to keep consistent with the format in libnwam_ncp.c 
line134 and libnwam_enm.c line381, where both of them have the similar 
table and used this format to calculate the table size.

> nwamcfg.c:826 boolean_t
>
> nwamcfg.c:894-6 nit: Above (896,8,...) you enclose single line blocks.
> Here in an else with a single line unbraced block and a braced else
> block.  I think you should make these consistent.  Personal choice is
> to always use blocks unless there is some if/else if/.../else block all
> with single line bodies but consistency is the more important goal.
>
> nwamcfg.c:910-915 way nit: All of the inner structure initializers have
> the extra ',' at the end.  That looks a little wierd.  The one after the
> last array element on 915 makes some sense and is specially allowed in
> the ANSI c specification due to automatic code generators FWIR.  In our
> less mechanical code writing duties it means you can add elements to
> the array without changing the last line thus limiting the size of your
> diff.  Yea, seems like a lot of thought for pretty modest returns.
>
> nwamcfg.c:1377 wonder if you should have a define for whatever an
> inband bad nwam_data_type_t is.
>
> nwamcfg.c:1404 const char *input_str?
>
> nwamcfg.c:1537,1573 nit like 894 above
>
> nwamcfg.c:2275,2314 The input is not guaranteed to be newline termianted.
>
>   

all accepted.

> libnwam_ncp.c:1122 you might assert that i doesn't get > prop_num
>   

I probably didn't get what you mean. Would you please elaborate?

Thanks!

-Jan


Reply via email to