#!/bin/sh

# Copyright 2007 by Denys Vlasenko <vda.linux@googlemail.com>
# Licensed under GPLv2, see file LICENSE in this source tree.

. ./testing.sh

# testing "test name" "options" "expected result" "file input" "stdin"
#   file input will be file called "input"
#   test can create a file "actual" instead of writing to stdout

testing "cut '-' (stdin) and multi file handling" \
	"cut -d' ' -f2 - input" \
	"over\n""quick\n" \
	"the quick brown fox\n" \
	"jumps over the lazy dog\n" \

abc="\
one:two:three:four:five:six:seven
alpha:beta:gamma:delta:epsilon:zeta:eta:theta:iota:kappa:lambda:mu
the quick brown fox jumps over the lazy dog
"

testing "cut -b a,a,a" "cut -b 3,3,3 input" "e\np\ne\n" "$abc" ""

testing "cut -b overlaps" \
	"cut -b 1-3,2-5,7-9,9-10 input" \
	"\
one:to:th
alphabeta
the qick \n" \
	"$abc" ""
testing "-b encapsulated" \
	"cut -b 3-8,4-6 input" \
	"\
e:two:
pha:be
e quic\n" \
	"$abc" ""
optional LONG_OPTS
testing "cut -b --output-delimiter overlaps" \
	"cut --output-delimiter='^' -b 1-3,2-5,7-9,9-10 input" \
	"\
one:t^o:th
alpha^beta
the q^ick \n" \
	"$abc" ""
SKIP=

testing "cut high-low error" "cut -b 8-3 input 2>/dev/null || echo err" "err\n" \
  "$abc" ""

testing "cut -b 2-1 error" "cut -b 2-1 input 2>/dev/null || echo err" "err\n" \
  "$abc" ""

testing "cut -c a-b" "cut -c 4-10 input" ":two:th\nha:beta\n quick \n" "$abc" ""
testing "cut -c a-" "cut -c 41- input" "\ntheta:iota:kappa:lambda:mu\ndog\n" "$abc" ""
testing "cut -c -b" "cut -c -39 input" \
  "one:two:three:four:five:six:seven\nalpha:beta:gamma:delta:epsilon:zeta:eta\nthe quick brown fox jumps over the lazy\n" \
  "$abc" ""
testing "cut -c a" "cut -c 40 input" "\n:\n \n" "$abc" ""
testing "cut -c a,b-c,d" "cut -c 3,5-7,10 input" "etwoh\npa:ba\nequi \n" "$abc" ""

testing "cut -f a-" "cut -d ':' -f 5- input" "five:six:seven\nepsilon:zeta:eta:theta:iota:kappa:lambda:mu\nthe quick brown fox jumps over the lazy dog\n" "$abc" ""

testing "cut show whole line with no delim" "cut -d ' ' -f 3 input" \
	"one:two:three:four:five:six:seven\nalpha:beta:gamma:delta:epsilon:zeta:eta:theta:iota:kappa:lambda:mu\nbrown\n" "$abc" ""

testing "cut with echo, -c (a-b)" "echo 'ref_categorie=test' | cut -c 1-15 " "ref_categorie=t\n" "" ""
testing "cut with echo, -c (a)" "echo 'ref_categorie=test' | cut -c 14" "=\n" "" ""

testing "cut with -c (a,b,c)" "cut -c 4,5,20 input" "det\n" "abcdefghijklmnopqrstuvwxyz" ""

testing "cut with -b (a,b,c)" "cut -b 4,5,20 input" "det\n" "abcdefghijklmnopqrstuvwxyz" ""

input="\
406378:Sales:Itorre:Jan
031762:Marketing:Nasium:Jim
636496:Research:Ancholie:Mel
396082:Sales:Jucacion:Ed
"
testing "cut with -d -f(:) -s" "cut -d: -f3 -s input" "Itorre\nNasium\nAncholie\nJucacion\n" "$input" ""
testing "cut with -d -f( ) -s" "cut -d' ' -f3 -s input && echo yes" "yes\n" "$input" ""
testing "cut with -d -f(a) -s" "cut -da -f3 -s input" "n\nsium:Jim\n\ncion:Ed\n" "$input" ""
testing "cut with -d -f(a) -s -n" "cut -da -f3 -s -n input" "n\nsium:Jim\n\ncion:Ed\n" "$input" ""

input="\

foo bar baz

bing bong boop

"
testing "cut with -d -s omits blank lines" "cut -d' ' -f2 -s input" "bar\nbong\n" "$input" ""

# substitute for awk
optional FEATURE_CUT_REGEX
testing "cut -DF unordered" "cut -DF 2,7,5" \
	"\
said and your
are
is demand. supply
forecast :
you you better,

Em: Took hate
" "" \
"Bother, said Pooh. It's your husband, and he has a gun.
Cheerios are donut seeds.
Talk is cheap because supply exceeds demand.
Weather forecast for tonight : dark.
Apple: you can buy better, but you can't pay more.
Subcalifragilisticexpialidocious.
Auntie Em: Hate you, hate Kansas. Took the dog. Dorothy."

# No delimiter found: print entire line regardless of -F RANGES
testing "cut -F1" "cut -d: -F1" \
	"the_only_field\n" "" \
	"the_only_field\n"
testing "cut -F2" "cut -d: -F2" \
	"the_only_field\n" "" \
	"the_only_field\n"
# No delimiter found and -s: skip entire line
testing "cut -sF1" "cut -d: -sF1" \
	"" "" \
	"the_only_field\n"
#^^^ the above is probably mishandled by toybox, it prints the line
testing "cut -sF2" "cut -d: -sF2" \
	"" "" \
	"the_only_field\n"
# -D disables special handling of lines with no delimiters, the line is treated as the 1st field
testing "cut -DF1" "cut -d: -DF1" \
	"the_only_field\n" "" \
	"the_only_field\n"
testing "cut -DF2" "cut -d: -DF2" \
	"\n" "" \
	"the_only_field\n"

optional FEATURE_CUT_REGEX LONG_OPTS
testing "cut -F preserves intermediate delimiters" \
	"cut --output-delimiter=: -F2,4-6,7" \
	"2:4		5  6:7\n" \
	"" "1 2  3  4\t\t5  6  7 8\n"
SKIP=

optional LONG_OPTS
testing "cut -f does not preserve intermediate delimiters" \
	"cut --output-delimiter=: -d' ' -f2,4-6,7" \
	"2:4:5:6:7\n" \
	"" "1 2 3 4 5 6 7 8\n"
SKIP=

testing "cut empty field" "cut -d ':' -f 1-3" \
	"a::b\n" \
	"" "a::b\n"
testing "cut empty field 2" "cut -d ':' -f 3-5" \
	"b::c\n" \
	"" "a::b::c:d\n"
testing "cut non-existing field" "cut -d ':' -f1,3" \
	"1\n" \
	"" "1:\n"

# cut -d$'\n' has a special meaning: "select input lines".
# I didn't find any documentation for this feature.
testing "cut -dNEWLINE" \
	"cut -d'
' -f4,2,6-8" \
	"2\n4\n6\n7\n" \
	"" "1\n2\n3\n4\n5\n6\n7"

optional LONG_OPTS
testing "cut -dNEWLINE --output-delimiter" \
	"cut -d'
' --output-delimiter=@@ -f4,2,6-8" \
	"2@@4@@6@@7\n" \
	"" "1\n2\n3\n4\n5\n6\n7"

testing "cut -dNEWLINE --output-delimiter 2" \
	"cut -d'
' --output-delimiter=@@ -f4,2,6-8" \
	"2@@4@@6@@7\n" \
	"" "1\n2\n3\n4\n5\n6\n7\n"

testing "cut -dNEWLINE --output-delimiter EMPTY_INPUT" \
	"cut -d'
' --output-delimiter=@@ -f4,2,6-8" \
	"" \
	"" ""
SKIP=

# This seems to work as if delimiter is never found.
# We test here that -d '' does *not* operate as if there was no -d
# and delimiter has defaulted to TAB:
testing "cut -d EMPTY" \
	"cut -d '' -f2-" \
	"1 2\t3 4 5\n" \
	"" "1 2\t3 4 5\n"
testing "cut -d EMPTY -s" \
	"cut -d '' -f2- -s" \
	"" \
	"" "1 2\t3 4 5\n"

exit $FAILCOUNT