Re: select patch

Hi, Johannes.

Thanks, this is very nice.  Some more comments and suggestions.

Why do you need MATCH_PREFIX_IGNORECASE instead of just using MATCH_PREFIX
| MATCH_IGNORECASE ?

Is MAX_WINDOW_NAME_LENGTH == 42 reasonable?  The goal isn't to enforce
short window names, only to protect against strings not being
null-terminated.  Why not 1024 or even 4096?  Or maybe something already
enforces 42?

I'm a fan of descriptive variable names.  So compare_length is more
descriptive to me than cmplen.  This is motivated by seeing abbreviations
that were obvious after being explained but that required asking.

Many a bug can be prevented by using braces even on one line blocks where
they are not needed.  The mechanism is that someone adds a statement later
on and misses the missing braces.

window_name returns a char* that is not a copy (i.e., you don't own it).
 So modifying it isn't such a good idea.

Can you use a function pointer for your compare function so that you don't
need to have the pre-processing block?


Jeff Abrahamson
+33 6 24 40 01 57   <-- brièvement indisponible le 4 juillet
+44 7920 594 255    <-- will change 18 July

http://jeff.purple.com/
http://blog.purple.com/jeff/


On 19 June 2014 18:16, Johannes Altmanninger <[email protected]>
wrote:

>
> On 06/19/2014 07:07 PM, Johannes Altmanninger wrote:
>
>>
>> #define MATCH_PREFIX_IGNORECASE 0x0004
>>
> my bad this should be
> #define MATCH_PREFIX_IGNORECASE 0x0003
>

_______________________________________________
Ratpoison-devel mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/ratpoison-devel