svn commit: r191330 - head/usr.bin/ncal

[Robert Watson]

On Tue, 21 Apr 2009, David Schultz wrote:

> On Tue, Apr 21, 2009, Roman Divacky wrote:
>>> Also, before this change, ncal was already full of convoluted buffer 
>>> handling, arbitrary buffer sizes, and little to no bounds checking. This 
>>> commit adds more magic numbers and fragile buffer handling code, and 
>>> generally makes an already hairy program even less scrutable. This isn't 
>>> your fault, but it would be nice if we could make ncal better before it 
>>> gets much worse. For instance, you might use snprintf() or asprintf() 
>>> instead of an extra half dozen calls to memcpy() with various offsets.
>>
>> yes, thats true. do you want me to revert this? I am perfectly fine with 
>> having locally modified cal that supports this highlighting and not share 
>> this with world at all.
>
> I don't care (although some other people on this thread seem to); I'm just 
> encouraging you to clean things up a little before making the code even less 
> maintainable.

The usual moral seems to apply: people who make cosmetic changes should expect 
cosmetic criticisms.  If they aren't happy to receive the criticism, they had 
best refrain from the changes.  Likewise modifying style(9), making gratuitous 
style changes, re-spelling computer science non-words in comments, etc.

Robert N M Watson
Computer Laboratory
University of Cambridge