• Davidl (6/6) Apr 19 2007 if(AA) <-- invalid
• Dan (4/12) Apr 19 2007 For an AA, I believe this works out to length - the length of the char[]...
• Davidl (3/17) Apr 19 2007 no if(k) doesn't indicate an empty AA, it just tests whether it has ever...

Davidl <Davidl 126.com> writes:

if(AA) <-- invalid

char[][char[]] k;
if (k) // this code should be invalid, cause it could strongly be a typo  
or some porting mistake from some
C source

I hope if(identifier) only work for Bool & Pointer types

Dan <murpsoft hotmail.com> writes:

Davidl Wrote:

 if(AA) <-- invalid
 
 char[][char[]] k;
 if (k) // this code should be invalid, cause it could strongly be a typo  
 or some porting mistake from some
 C source
 
 I hope if(identifier) only work for Bool & Pointer types

For an AA, I believe this works out to length - the length of the char[][] keys;

if(k) I believe checks k.keys.length by virtue of being at *(&k), which is 0
until you assign it a buffer.  This seems like the right behavior to me for an
AA.

if(k) asks if k is a non-empty Associative Array.

Davidl <Davidl 126.com> writes:

 if(AA) <-- invalid

 char[][char[]] k;
 if (k) // this code should be invalid, cause it could strongly be a typo
 or some porting mistake from some
 C source

 I hope if(identifier) only work for Bool & Pointer types

 For an AA, I believe this works out to length - the length of the  
 char[][] keys;

 if(k) I believe checks k.keys.length by virtue of being at *(&k), which  
 is 0 until you assign it a buffer.  This seems like the right behavior  
 to me for an AA.

 if(k) asks if k is a non-empty Associative Array.

no if(k) doesn't indicate an empty AA, it just tests whether it has ever  
been used.
if k has been used and then cleared, still if(k) test would pass