boolean - Turn off Warning: Extension: Conversion from LOGICAL(4) to INTEGER(4) at (1) for gfortran? -

i intentionally casting array of boolean values integers warning:

warning: extension: conversion logical(4) integer(4) @ (1) 

which don't want. can either

(1) turn off warning in makefile?

or (more favorably)

(2) explicitly make cast in code compiler doesn't need worry?

the code looking this:

a = (b.eq.0) 

where a , b both size (n,1) integer arrays. b filled integers ranging 0 3. need use type of command again later a = (b.eq.1) , need a integer array 1 if , if b requested integer, otherwise should 0. these should act boolean values (1 .true., 0 .false.), going using them in matrix operations , summations converted floating point values (when necessary) division, logical values not optimal in circumstance.

specifically, looking fastest, vectorized version of command. easy write wrapper testing elements, want vectorized operation efficiency.

i compiling gfortran, whatever methods used work in ifort compiling intel compilers down road.

update:

both merge , where work example in question. performance metrics on these , select best vectorization. interested in how work matrices, not arrays, not original question post new 1 unless wants expand answer how might adapted matrices.

i have not found compiler option solve (1).

however, type conversion pretty simple. the documentation gfortran specifies .true. mapped 1, , false 0.

note conversion not specified standard, , different values used other compilers. specifically, should not depend on exact values.

a simple merge trick scalars , arrays:

program test   integer :: int_sca, int_vec(3)   logical :: log_sca, log_vec(3)    log_sca = .true.   log_vec = [ .true., .false., .true. ]    int_sca = merge( 1, 0, log_sca )   int_vec = merge( 1, 0, log_vec )    print *, int_sca   print *, int_vec end program 

to address updated question, trivial merge:

a = merge(1, 0, b == 0) 

this can performed on scalars , arrays of arbitrary dimensions. latter, can vectorized compiler. should consult manual of compiler that, though.

the where statement in casey's answer can extended in same way.

---

since convert them floats later on, why not assign them floats right away? assuming a real, like:

a = merge(1., 0., b == 0)