# $NetBSD: cond-cmp-numeric-eq.mk,v 1.6 2023/06/01 20:56:35 rillig Exp $ # # Tests for numeric comparisons with the == operator in .if conditions. # This comparison yields the same result, whether numeric or character-based. .if 1 == 1 .else . error .endif # This comparison yields the same result, whether numeric or character-based. .if 1 == 2 . error .endif .if 2 == 1 . error .endif # Scientific notation is supported, as per strtod. .if 2e7 == 2000e4 .else . error .endif .if 2000e4 == 2e7 .else . error .endif # Trailing zeroes after the decimal point are irrelevant for the numeric # value. .if 3.30000 == 3.3 .else . error .endif .if 3.3 == 3.30000 .else . error .endif # As of 2020-08-23, numeric comparison is implemented as parsing both sides # as double, and then performing a normal comparison. The range of double is # typically 16 or 17 significant digits, therefore these two numbers seem to # be equal. .if 1.000000000000000001 == 1.000000000000000002 .else . error .endif # Because an IEEE 754 double can only hold integers with a mantissa of 53 # bits, these two numbers are considered the same. The 993 is rounded down # to the 992. .if 9007199254740993 == 9007199254740992 .else . error .endif # The 995 is rounded up, the 997 is rounded down. .if 9007199254740995 == 9007199254740997 .else . error Probably a misconfiguration in the floating point environment, \ or maybe a machine without IEEE 754 floating point support. .endif # There is no = operator for numbers. # expect+1: Malformed conditional (!(12345 = 12345)) .if !(12345 = 12345) . error .else . error .endif # There is no === operator for numbers either. # expect+1: Malformed conditional (!(12345 === 12345)) .if !(12345 === 12345) . error .else . error .endif all: @:;