Re: [q-lang-users] This feels like a dumb question, but...

[Andrew B. <and...@ya...>]

=0AAlbert Graef wrote:=0A>=0A>> =0A>> -f F =3D true if (close (open F 0 0))=
 =3D ();=0A>>         =3D false otherwise;=0A>=0A>Note that -f is not a val=
id function name in Q. What you actually define=0A>here is a rule for unary=
 minus applied to the constructor term f F.=0A>=0AI had recognized it as an=
 invalid function name already, but I rather liked the look of "-f" rather =
than "_f" or (eww) "is_regular_file".  And besides, the interpreter let me =
get away with it!=0A=0A>> I understand that when the file does not exist, t=
he "(close (open F 0  =0A>> 0))" part does not reduce, but why does "=3D ()=
" part not look at  =0A>> "close (open "fetch.qaz" 0 0)" and say, "nope, no=
t equal"?  "eq" does  =0A>> the same thing.  Is there something that will f=
orce the equality  =0A>> check to happen?=0A>=0A>eq and (=3D) are different=
 things in Q. The former checks for syntactic=0A>equality (which is always =
defined), the latter is a comparison operator=0A>which only applies to cert=
ain data types (much like (<) and (<), etc.).=0A>See Section 7.2 of the man=
ual,=0A>http://q-lang.sourceforge.net/qdoc/qdoc_7.html#SEC27=0A>=0A>Of cour=
se, you can add your own rules for (=3D) to extend equality to any=0A>data =
types you need, but it is generally undesirable to have a very=0A>broad def=
inition of (=3D) as such rules might interfere with the specific=0A>cases.=
=0A=0AI (think that I) understand the =3D vs eq difference.  I was surprise=
d that when I tried to apply either one to an expression where the "open F =
0 0" had failed, all I got was a bigger expression.=0A=0A>So you'll have to=
 use eq instead. Or you could use a pattern match like so:=0A>=0A>-f F =3D =
true where () =3D close (open F 0 0);=0A>        =3D false otherwise;=0A=0A=
Aah, that's probably the version I was looking for.  So, if the evaluation =
fails in a "where" clause, the containing equation fails also.  Somehow I d=
id not manage to learn that from my reading of the manual.=0A=0A>But I thin=
k that the most convenient way to do these checks would be to=0A>use the st=
at/lstat functions from clib. For instance:=0A=0A... snip a bunch...=0A=0AI=
t's kind of amusing, I started out trying to write this way, but the only v=
ersion of stat that I could find in the manual seemed to take a FD, which r=
equired calling "open" anyhow, so since I only needed "-f" just then I sett=
led on a really dumb implementation.=0A=0AThanks for the advice,=0A=0A-andr=
ew=0A=0A=0A=0A=0A