Thanks Alex, A follow up question - please excuse me if I am misreading here. As per the doc, the third bit indicates sym - independent of the architecture. So isSym should just have 4 in its logic - No?
Regards, Kashyap On Sat, Feb 23, 2019 at 1:40 AM Alexander Burger <a...@software-lab.de> wrote: > Hi Kashyap, > > > I am having some trouble understanding the predicates and would really > > appreciate some explanation. > > > > /* Predicates */ > > #define isNil(x) ((x)==Nil) > > #define isTxt(x) (num(x)&1) > > #define isNum(x) (num(x)&2) > > #define isSym(x) (num(x)&WORD) > > #define isSymb(x) ((num(x)&(WORD+2))==WORD) > > #define isCell(x) (!(num(x)&(2*WORD-1))) > > This is miniPicoLisp, right? Only there we have isTxt(). > > > > I can understand isTxt and isNum - bit 1 set means it's text and bit 2 > set > > means its number. > > Correct. As you know, this is depicted in miniPicoLisp/doc/structures > > Primary data types: > num xxxxxx10 > sym xxxxx100 > pair xxxxx000 > > Raw data: > bin xxxxxxxx > txt xxxxxxx1 > > > > However, why is WORD used for isSym? Could it not have > > been just 4 and why is does it have to be sizeof(long) - which is also 4 > on > > 32bit > > The reason is that miniPicoLisp compiles both on 32- and 64-bit > architectures, > so sizeof(long) can be either 4 or 8. In that aspect it differs from the > full > PicoLisp versions, which are always specialized for one architecture. > > ☺/ A!ex > > -- > UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe >
