2c3632d14f
Lots of code refactoring, simplification and cleanup. Lots of new unit-tests providing much higher code coverage. All courtesy of rillig at netbsd. Other significant changes: o new read-only variable .SHELL which provides the path of the shell used to run scripts (as defined by the .SHELL target). o variable parsing detects more errors. o new debug option -dl: LINT mode, does the equivalent of := for all variable assignments so that file and line number are reported for variable parse errors.
40 lines
1.3 KiB
Makefile
40 lines
1.3 KiB
Makefile
# $NetBSD: varmod-order-shuffle.mk,v 1.3 2020/08/16 20:43:01 rillig Exp $
|
|
#
|
|
# Tests for the :Ox variable modifier, which returns the words of the
|
|
# variable, shuffled.
|
|
#
|
|
# As of 2020-08-16, make uses random(3) seeded by the current time in seconds.
|
|
# This makes the random numbers completely predictable since there is no other
|
|
# part of make that uses random numbers.
|
|
|
|
NUMBERS= one two three four five six seven eight nine ten
|
|
|
|
# Note that 1 in every 10! trials two independently generated
|
|
# randomized orderings will be the same. The test framework doesn't
|
|
# support checking probabilistic output, so we accept that each of the
|
|
# 3 :Ox tests will incorrectly fail with probability 2.756E-7, which
|
|
# lets the whole test fail once in 1.209.600 runs, on average.
|
|
|
|
# Create two shuffles using the := assignment operator.
|
|
shuffled1:= ${NUMBERS:Ox}
|
|
shuffled2:= ${NUMBERS:Ox}
|
|
.if ${shuffled1} == ${shuffled2}
|
|
.error ${shuffled1} == ${shuffled2}
|
|
.endif
|
|
|
|
# Sorting the list before shuffling it has no effect.
|
|
shuffled1:= ${NUMBERS:O:Ox}
|
|
shuffled2:= ${NUMBERS:O:Ox}
|
|
.if ${shuffled1} == ${shuffled2}
|
|
.error ${shuffled1} == ${shuffled2}
|
|
.endif
|
|
|
|
# Sorting after shuffling must produce the original numbers.
|
|
sorted:= ${NUMBERS:Ox:O}
|
|
.if ${sorted} != ${NUMBERS:O}
|
|
.error ${sorted} != ${NUMBERS:O}
|
|
.endif
|
|
|
|
all:
|
|
@:;
|