# SUSv3 compliant sed tests.
# Copyright 2005 by Rob Landley <rob@landley.net>
-# Licensed under GPL v2, see file LICENSE for details.
+# Licensed under GPLv2, see file LICENSE in this source tree.
-. testing.sh
+. ./testing.sh
-# testing "description" "arguments" "result" "infile" "stdin"
+# testing "description" "commands" "result" "infile" "stdin"
# Corner cases
testing "sed no files (stdin)" 'sed ""' "hello\n" "" "hello\n"
testing "sed s//p" "sed -e s/foo/bar/p -e s/bar/baz/p" "bar\nbaz\nbaz\n" \
"" "foo\n"
testing "sed -n s//p" "sed -ne s/abc/def/p" "def\n" "" "abc\n"
+test x"$SKIP_KNOWN_BUGS" = x"" && {
testing "sed s//g (exhaustive)" "sed -e 's/[[:space:]]*/,/g'" ",1,2,3,4,5,\n" \
"" "12345\n"
+}
testing "sed s arbitrary delimiter" "sed -e 's woo boing '" "boing\n" "" "woo\n"
testing "sed s chains" "sed -e s/foo/bar/ -e s/bar/baz/" "baz\n" "" "foo\n"
testing "sed s chains2" "sed -e s/foo/bar/ -e s/baz/nee/" "bar\n" "" "foo\n"
testing "sed s [delimiter]" "sed -e 's@[@]@@'" "onetwo" "" "one@two"
+testing "sed s with \\t (GNU ext)" "sed 's/\t/ /'" "one two" "" "one\ttwo"
# branch
testing "sed b (branch)" "sed -e 'b one;p;: one'" "foo\n" "" "foo\n"
testing "sed T (!test/branch)" "sed -e 's/a/1/;T notone;p;: notone;p'" \
"1\n1\n1\nb\nb\nc\nc\n" "" "a\nb\nc\n"
+test x"$SKIP_KNOWN_BUGS" = x"" && {
# Normal sed end-of-script doesn't print "c" because n flushed the pattern
# space. If n hits EOF, pattern space is empty when script ends.
# Query: how does this interact with no newline at EOF?
testing "sed n (flushes pattern space, terminates early)" "sed -e 'n;p'" \
"a\nb\nb\nc\n" "" "a\nb\nc\n"
-# N does _not_ flush pattern space, therefore c is still in there @ script end.
-testing "sed N (doesn't flush pattern space when terminating)" "sed -e 'N;p'" \
+}
+# non-GNU sed: N does _not_ flush pattern space, therefore c is eaten @ script end
+# GNU sed: N flushes pattern space, therefore c is printed too @ script end
+testing "sed N (flushes pattern space (GNU behavior))" "sed -e 'N;p'" \
"a\nb\na\nb\nc\n" "" "a\nb\nc\n"
+
+testing "sed N test2" "sed ':a;N;s/\n/ /;ta'" \
+ "a b c\n" "" "a\nb\nc\n"
+
+testing "sed N test3" "sed 'N;s/\n/ /'" \
+ "a b\nc\n" "" "a\nb\nc\n"
+
testing "sed address match newline" 'sed "/b/N;/b\\nc/i woo"' \
"a\nwoo\nb\nc\nd\n" "" "a\nb\nc\nd\n"
"sed -e '/ook/d;a\' -e 'bang'" "woot\nbang\n" "" "ook\nwoot\n"
# Multiple files, with varying newlines and NUL bytes
+test x"$SKIP_KNOWN_BUGS" = x"" && {
testing "sed embedded NUL" "sed -e 's/woo/bang/'" "\0bang\0woo\0" "" \
"\0woo\0woo\0"
+}
testing "sed embedded NUL g" "sed -e 's/woo/bang/g'" "bang\0bang\0" "" \
"woo\0woo\0"
-echo -e "/woo/a he\0llo" > sed.commands
+test x"$SKIP_KNOWN_BUGS" = x"" && {
+$ECHO -e "/woo/a he\0llo" > sed.commands
testing "sed NUL in command" "sed -f sed.commands" "woo\nhe\0llo\n" "" "woo"
rm sed.commands
+}
# sed has funky behavior with newlines at the end of file. Test lots of
# corner cases with the optional newline appending behavior.
"" "one\ntwo"
testing "sed cat plus empty file" "sed -e 's/nohit//' input -" "one\ntwo" \
"one\ntwo" ""
+test x"$SKIP_KNOWN_BUGS" = x"" && {
testing "sed append autoinserts newline" "sed -e '/woot/a woo' -" \
"woot\nwoo\n" "" "woot"
+}
testing "sed insert doesn't autoinsert newline" "sed -e '/woot/i woo' -" \
"woo\nwoot" "" "woot"
testing "sed print autoinsert newlines" "sed -e 'p' -" "one\none" "" "one"
testing "sed selective matches noinsert newline" \
"sed -ne 's/woo/bang/p' input -" "a bang\nb bang" "a woo\nb woo" \
"c no\nd no"
+test x"$SKIP_KNOWN_BUGS" = x"" && {
testing "sed clusternewline" \
"sed -e '/one/a 111' -e '/two/i 222' -e p input -" \
"one\none\n111\n222\ntwo\ntwo" "one" "two"
+}
+testing "sed subst+write" \
+ "sed -e 's/i/z/' -e 'woutputw' input -; $ECHO -n X; cat outputw" \
+ "thzngy\nagaznXthzngy\nagazn" "thingy" "again"
+rm outputw
+testing "sed trailing NUL" \
+ "sed 's/i/z/' input -" \
+ "a\0b\0\nc" "a\0b\0" "c"
+testing "sed escaped newline in command" \
+ "sed 's/a/z\\
+z/' input" \
+ "z\nz" "a" ""
# Test end-of-file matching behavior
"hello\nthere"
testing "sed match EOF two files" "sed -e '"'$p'"' input -" \
"one\ntwo\nthree\nfour\nfour" "one\ntwo" "three\nfour"
-echo -ne "three\nfour" > input2
+# sed match EOF inline: gnu sed 4.1.5 outputs this:
+#00000000 6f 6e 65 0a 6f 6f 6b 0a 6f 6f 6b 0a 74 77 6f 0a |one.ook.ook.two.|
+#00000010 0a 74 68 72 65 65 0a 6f 6f 6b 0a 6f 6f 6b 0a 66 |.three.ook.ook.f|
+#00000020 6f 75 72 |our|
+# which looks buggy to me.
+$ECHO -ne "three\nfour" > input2
testing "sed match EOF inline" \
"sed -e '"'$i ook'"' -i input input2 && cat input input2" \
"one\nook\ntwothree\nook\nfour" "one\ntwo" ""
"GNU sed version \n" "" ""
# Jump to nonexistent label
-testing "sed nonexistent label" "sed -e 'b walrus' 2> /dev/null || echo yes" \
+test x"$SKIP_KNOWN_BUGS" = x"" && {
+# Incompatibility: illegal jump is not detected if input is ""
+# (that is, no lines at all). GNU sed 4.1.5 complains even in this case
+testing "sed nonexistent label" "sed -e 'b walrus' 2>/dev/null || echo yes" \
"yes\n" "" ""
+}
testing "sed backref from empty s uses range regex" \
"sed -e '/woot/s//eep \0 eep/'" "eep woot eep" "" "woot"
#testing "sed -g (exhaustive)" "sed -e 's/[[:space:]]*/,/g'" ",1,2,3,4,5," \
# "" "12345"
+# testing "description" "commands" "result" "infile" "stdin"
+
+testing "sed n command must reset 'substituted' bit" \
+ "sed 's/1/x/;T;n;: next;s/3/y/;t quit;n;b next;: quit;q'" \
+ "0\nx\n2\ny\n" "" "0\n1\n2\n3\n"
+
+testing "sed d does not break n,m matching" \
+ "sed -n '1d;1,3p'" \
+ "second\nthird\n" "" "first\nsecond\nthird\nfourth\n"
+
+testing "sed d does not break n,regex matching" \
+ "sed -n '1d;1,/hir/p'" \
+ "second\nthird\n" "" "first\nsecond\nthird\nfourth\n"
+
+testing "sed d does not break n,regex matching #2" \
+ "sed -n '1,5d;1,/hir/p'" \
+ "second2\nthird2\n" "" \
+ "first\nsecond\nthird\nfourth\n""first2\nsecond2\nthird2\nfourth2\n"
+
+testing "sed 2d;2,1p (gnu compat)" \
+ "sed -n '2d;2,1p'" \
+ "third\n" "" \
+ "first\nsecond\nthird\nfourth\n"
+
+# Regex means: "match / at BOL or nothing, then one or more not-slashes".
+# The bug was that second slash in /usr/lib was treated as "at BOL" too.
+testing "sed beginning (^) matches only once" \
+ "sed 's,\(^/\|\)[^/][^/]*,>\0<,g'" \
+ ">/usr</>lib<\n" "" \
+ "/usr/lib\n"
+
+testing "sed c" \
+ "sed 'crepl'" \
+ "repl\nrepl\n" "" \
+ "first\nsecond\n"
+
+testing "sed nested {}s" \
+ "sed '/asd/ { p; /s/ { s/s/c/ }; p; q }'" \
+ "qwe\nasd\nacd\nacd\n" "" \
+ "qwe\nasd\nzxc\n"
+
+testing "sed a cmd ended by double backslash" \
+ "sed -e '/| one /a \\
+ | three \\\\' -e '/| one-/a \\
+ | three-* \\\\'" \
+' | one \\
+ | three \\
+ | two \\
+' '' \
+' | one \\
+ | two \\
+'
+
+# first three lines are deleted; 4th line is matched and printed by "2,3" and by "4" ranges
+testing "sed with N skipping lines past ranges on next cmds" \
+ "sed -n '1{N;N;d};1p;2,3p;3p;4p'" \
+ "4\n4\n" "" "1\n2\n3\n4\n"
+
+testing "sed -i with address modifies all files, not only first" \
+ "cp input input2; sed -i -e '1s/foo/bar/' input input2 && cat input input2; rm input2" \
+ "bar\nbar\n" "foo\n" ""
+
+testing "sed understands \r" \
+ "sed 's/r/\r/'" \
+ "\rrr\n" "" "rrr\n"
+
+testing "sed -i finishes ranges correctly" \
+ "sed '1,2d' -i input; echo \$?; cat input" \
+ "0\n3\n4\n" "1\n2\n3\n4\n" ""
+
+# testing "description" "commands" "result" "infile" "stdin"
+
exit $FAILCOUNT