I've found a two argument arctan indispensable and have always had these
in my startup:
atan2=: 13 : '2p1|((0>1{"1 y)*1p1)+2p1+_3&o. %/"1 y' NB. arrays OK,
0<result<2p1 atan3=: 13 : '(_3&o. %/"1 y)+1p1*(0>{:"1 y)*(*{."1 y)'
NB. _1p1<result<1p1
For example:
(180%1p1)* atan2 1 1
45
(180%1p1)* atan2 _1 _1
225
(180%1p1)* atan3 1 1
45
(180%1p1)* atan3 _1 _1
_135
Patrick
On 3/20/23 8:15 PM, John Baker wrote:
While computing star rise, transit and set times I encountered a need for an
atn2 verb. This is basically a tweaked arctan verb that handles ratios like
(arctan a%b) when a and b might be 0. Atn2 is in FORTRAN libraries. I ended up
rolling my own j verb but surely something so standard must be floating around.
If anyone has any atn 2’s please post. I like to compare implementations for
stuff like this.
Jdb
analyzethedatanotthedrivel.org
----------------------------------------------------------------------
For information about J forums seehttp://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
