# regression tests for the POSIX sed utility
function DATA
{
for f
do test -f $f && continue
case $f in
[0-7][0-7][0-7])
echo $f > $f
chmod $f $f
;;
nul.in) print '\0AAAfooZZZ' > $f
;;
nul.out)print '\0AAAbarZZZ' > $f
;;
esac
done
}
TEST 01 'empty script and passive command combinations'
EXEC -f script -n
NOTE '=, -n, blank lines'
INPUT script $'\n=\n'
INPUT - $'a\nb\nc'
OUTPUT - $'1\n2\n3'
EXEC -f script
NOTE 'comments, including #n'
INPUT script $'#n\n=\n#comment'
EXEC -f script
NOTE 'empty script'
INPUT script
OUTPUT - $'a\nb\nc'
EXEC -f script
NOTE 'do-nothing script'
INPUT script $'# nothing\n'
EXEC -f script
NOTE 'substitution; regexp dot; &; substitution flag'
INPUT script $'s/./&&/\n s/./&x/\n s/./&y/2\n s/./&z/5'
OUTPUT - $'axya\nbxyb\ncxyc'
TEST 02 'line counting, ranges, overlaps'
EXEC -f script
NOTE 'd overlap'
INPUT script $'1,2d\n 2,6d\n 4,5d\n 7,10d'
INPUT - $'1\n2\n3\n4\n5\n6\n7\n8'
OUTPUT - $'3\n6'
EXEC -f script -e 1,2d -e 2,6d
NOTE 'd disjoint range'
INPUT script $'4,5d\n 7,10d'
EXEC -f script -n
NOTE 'd bypassing end of range'
INPUT script $'2,5d\n 1,3p\n 1,6p'
OUTPUT - $'1\n1\n6'
EXEC -f script
NOTE 'negation, address $, print'
INPUT script $'3,$!d\n $d\n p\n 4,5d'
OUTPUT - $'3\n3\n4\n5\n6\n6\n7\n7'
TEST 03 'regexp addresses, append, insert, change'
EXEC -f script
INPUT script $'/a/a\\\nA\\\nA\n/a/a\\\nB\n/d/i\\\nD\n/c/c\\\nC\\\nC\n$i\\\nE'
INPUT - $'a\nb\nc\nd'
OUTPUT - $'a\nA\nA\nB\nb\nC\nC\nD\nE\nd'
TEST 04 'braces'
EXEC -f script
INPUT script $'3,7{\n/a/s/a/&&/\n/b/,/./s/./&&&/\n}'
INPUT - $'a\nb\na\nb\na\na\nb\na\nb'
OUTPUT - $'a\nb\naa\nbbb\naaaa\naa\nbbb\na\nb'
TEST 05 'hold, get combinations'
EXEC -f script
NOTE 'hold, get'
INPUT script $'$!H\n$!d\n$G'
INPUT - $'1\n2\n3\n4'
OUTPUT - $'4\n\n1\n2\n3'
EXEC -f script
NOTE 'hold, exchange, get, brace without newline'
INPUT script $'1{h\nd\n}\n3x\n$g'
OUTPUT - $'2\n1\n3'
EXEC -f script
NOTE 'quit'
INPUT script $'a\\\nx\nq\n='
OUTPUT - $'1\nx'
TEST 06 'next, regexp $'
EXEC -f script
INPUT script $'s/$/x/\nn\ns/./&y/\nN\ns/.../&z/\n='
INPUT - $'a\nb\nc'
OUTPUT - $'ax\n3\nby\nzc'
TEST 07 'substitute flag combinations'
EXEC -f script -n
NOTE 's newline p'
INPUT script $'/../s/./&\\\n/\np\nP\ns/\\n//p'
INPUT - $'a\nbc\nd'
OUTPUT - $'a\na\nb\nc\nb\nbc\nd\nd'
EXEC -f script -n
NOTE 'write, flag w'
INPUT script $'/../s/./&\\\n/\nw result\ns/\\n//w result'
OUTPUT result $'a\nb\nc\nbc\nd'
OUTPUT -
EXEC -f script -n
NOTE 's w file'
INPUT script $'s/b/--/gw result'
INPUT - $'abcabc'
OUTPUT result $'a--ca--c'
EXEC -f script -n
NOTE 's pw file'
INPUT script $'s/b/--/gpw result'
OUTPUT result $'a--ca--c'
OUTPUT - $'a--ca--c'
EXEC -f script
NOTE 's pw file'
INPUT script $'s/b/--/gpw result'
OUTPUT result $'a--ca--c'
OUTPUT - $'a--ca--c\na--ca--c'
TEST 08 'character classes, flag g'
EXEC -f script
INPUT script $'1s/[a-z][a-z]*//\n2s/[[:digit:]][[:digit:]]*//\n3s/[^a-z]/X/g\n'
INPUT - $'AZ`abyz{\n/0189:\na1b2c3d'
OUTPUT - $'AZ`{\n/:\naXbXcXd'
EXEC $'s/[\\]/X/g'
INPUT - $'[]./:,\\'
OUTPUT - $'[]./:,X'
EXEC $'s/[^./]/X/g'
OUTPUT - $'XX./XXX'
EXEC $'s/[][]/X/g'
OUTPUT - $'XX./:,\\'
EXEC $'s.[.]\\..X.g'
OUTPUT - $'[]X:,\\'
TEST 09 'null matches'
EXEC -f script
INPUT script $'1,2s/a*/x/g'
INPUT - $'123\naaa'
OUTPUT - $'x1x2x3x\nx'
TEST 10 'longest match, unmatched subexpressions'
EXEC -f script
INPUT script $'s/\\(...\\)*\\(..\\)*/:\\1:\\2:/'
INPUT - $'abc\nabcd\nabcde\nabcdef\nabcdefg'
OUTPUT - $':abc::\n::cd:\n:abc:de:\n:def::\n:abc:fg:'
TEST 11 'metacharacters in substition'
EXEC -f script
INPUT script $'1s/$/\\&/\n2s/$/\\b/\n3s/$/\\\\/\n4s/$/\\//\n5s&$&\\&&'
INPUT - $'1\n2\n3\n4\n5'
OUTPUT - $'1&\n2\b\n3\\\n4/\n5'
TEST 12 'branch combinations'
EXEC -f script
NOTE 'branches'
INPUT script $':x\n/a/{\ns/a/x/\nbx\n}'
INPUT - $'aaa\nb\nabca'
OUTPUT - $'xxx\nb\nxbcx'
EXEC -f script
NOTE 'long labels may be truncated'
INPUT script $':longlabel\n/a/s/a/x/\ntlonglabel'
EXEC -f script
NOTE 'jump to end of script'
INPUT script $'3b\n/a/s/a/x/g'
OUTPUT - $'xxx\nb\nabca'
EXEC -f script -n
NOTE 'end of bracket range'
INPUT script $'/c/d\n/a/,/d/{\n\t/b/,/c/{\n\t\t=\n\t}\n}'
INPUT - $'a\nb\nc\nd\na'
OUTPUT - $'2\n4\n5'
EXEC -f script
NOTE 'end of change range'
INPUT script $'/a/,/b/{\n\t/b/,/c/c\\\nx\n}'
OUTPUT - $'a\nc\nd\nx'
EXEC -f script
NOTE 'end of change range'
INPUT - $'a\nb\nc\na\nc\nb\nd'
OUTPUT - $'a\nc\nx\nd'
EXEC -f script
NOTE 'end of change range'
INPUT script $'/a/,/b/c\\\nc'
INPUT - $'a\nb\na'
OUTPUT - $'c\nc'
TEST 15 'weird delimiters, remembered expression'
EXEC -f script
INPUT script $'1s1\\(.\\)\\11\\11\n2s.\\(\\.\\)\\..\\.\\1.\n3s*\\(.\\)\\**\\*\\1*\n4s&\\(.\\)\\&&\\1\\&&\n\\1\\(\\1\\)\\11s//\\1/\n\\&\\(\\&\\)\\&&s//\\&b&/'
INPUT - $'a1b\nabc\nabc\na&b\n11b\na&&'
OUTPUT - $'1b\n.ac\n*c\naa&b\n1b\na&b&&'
TEST 16 '7680-char line, backreference'
EXEC -f script
INPUT script $'s/.*/&&&&&&&&/\ns//&&&&&&&&/\ns//&&&&&&&&/\nh\ns/[^8]//g\ns/^\\(.*\\)\\1\\1\\1$/\\1/\ns//\\1/\ns//\\1/\ns//\\1/p\ng\ns/^\\(.*\\)\\1\\1\\1$/\\1/\ns//\\1/\ns//\\1/\ns//\\1/\ns/\\(.*\\)\\1/\\1/'
INPUT - $'123456787654321'
OUTPUT - $'88\n123456787654321'
TEST 17 'r from w file, nonexistent r'
EXEC -f script -n
INPUT script $'r not_a_file\nr OUTPUT\nw OUTPUT'
INPUT - $'1\n2\n3'
OUTPUT - $'1\n1\n2\n1\n2\n3'
TEST 18 'eof combinations'
EXEC -f script
NOTE 'eof in n'
INPUT script $'a\\\n1\nn'
INPUT - $'a'
OUTPUT - $'a\n1'
EXEC -f script
NOTE 'eof in N'
INPUT script $'a\\\n1\nN'
OUTPUT - $'1'
TEST 19 'transliterate'
EXEC -f script
INPUT script $'y/abc/ABC/\ny:/\\\\\\:.:\\/.\\::'
INPUT - $'abcABCabcdef\n1/:2.'
OUTPUT - $'ABCABCABCdef\n1\\.2:'
EXEC -f script
INPUT script $'y/(/\\n/;y/)/\\n/'
INPUT - $'(a)(b)(c)'
OUTPUT - $'\na\n\nb\n\nc\n'
TEST 20 'N, D, G combinations'
EXEC -f script
NOTE 'N, D'
INPUT script $'=\nN\np\nD\ns/.*/x/'
INPUT - $'a\nb\nc'
OUTPUT - $'1\na\nb\n2\nb\nc\n3'
EXEC -f script
NOTE 'D, G initial states'
INPUT script $'1D\nG\nh'
OUTPUT - $'b\n\nc\nb\n'
TEST 21 'a,c,r interaction'
EXEC -f script
INPUT script $'$!a\\\nA\n$!r INPUT\n$!a\\\nB\n1,2c\\\nC'
INPUT - $'a\nb\nc'
OUTPUT - $'A\na\nb\nc\nB\nC\nA\na\nb\nc\nB\nc'
TEST 22 'substitution null string combinations'
EXEC -f script
NOTE 'global substitution for null string'
INPUT script $'s/a*/b/g'
INPUT - $'aaa\nccc'
OUTPUT - $'b\nbcbcbcb'
NOTE 'count 1 substitution for null string'
EXEC -f script
INPUT script $'s/a*/b/1'
OUTPUT - $'b\nbccc'
NOTE 'count 2 substitution for null string'
EXEC -f script
INPUT script $'s/a*/b/2'
OUTPUT - $'aaa\ncbcc'
EXEC -f script
NOTE 'count 3 substitution for null string'
INPUT script $'s/a*/b/3'
OUTPUT - $'aaa\nccbc'
TEST 23 'perverse semicolons'
EXEC $'s;\\;;x;'
INPUT - $'a;b'
OUTPUT - $'axb'
TEST 24 'multiple files'
EXEC -n '$=' INPUT INPUT INPUT
INPUT - $'a;\nb;\nc'
OUTPUT - $'9'
TEST 25 'weird chars, line folding'
EXEC -n $'s/.*/&&&&&&&&&&/\nH\nH\nH\nG\nl'
INPUT - $'\ta\a\\\n'
OUTPUT - $'\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\\n\\n\\n\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\\n\\a\\\\\\n\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\t\\\na\\a\\\\\\n\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\\n\\ta\\a\\\\$\n\\n\\n\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\\n\\a\\\\\\n\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\t\\\na\\a\\\\\\n\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\ta\\a\\\\\\\n\\ta\\a\\\\\\n\\n\\n$'
EXEC -n $'s/.*/&&&&&&&&&&/\nH\nH\nH\nG\nl'
INPUT - $'\ta\002\\\n'
OUTPUT - $'\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\n\\\\\\ta\\002\\\\\\ta\\002\\\\\\n\\n\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\t\\\na\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\n\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\n\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\n\\ta\\\n\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\t\\\na\\002\\\\\\ta\\002\\\\$\n\\n\\n\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\\n\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\n\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\\n\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\n\\ta\\002\\\\\\ta\\002\\\\\\ta\\\n\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\ta\\002\\\\\\n\\\n\\n\\n$'
TEST 26 'multiple w files, multiple inputs, OUTPUT1 empty'
EXEC -f script -n
INPUT - $'1\n2\n3\n4\n5\n6\n7\n8\n9'
INPUT script $'9,$d\nw out9\n8,$d\nw out8\n7,$d\nw out7\n6,$d\nw out6\n5,$d\nw out5\n4,$d\nw out4\n3,$d\nw out3\n2,$d\nw out2\n1,$d\nw out1'
OUTPUT -n out1
OUTPUT out2 $'1'
OUTPUT out3 $'1\n2'
OUTPUT out4 $'1\n2\n3'
OUTPUT out5 $'1\n2\n3\n4'
OUTPUT out6 $'1\n2\n3\n4\n5'
OUTPUT out7 $'1\n2\n3\n4\n5\n6'
OUTPUT out8 $'1\n2\n3\n4\n5\n6\n7'
OUTPUT out9 $'1\n2\n3\n4\n5\n6\n7\n8'
TEST 27 'assorted warnings'
DIAGNOSTICS
EXEC -n /o/p
NOTE 'trailing newline omitted'
INPUT -n - $'hello\ngoodbye'
OUTPUT - $'hello\ngoodbye'
TEST 28 'assorted diagnostics'
DIAGNOSTICS
EXEC = not_a_file
EXEC -e :x -e :x
EXEC r/dev/null
EXEC -f not_a_file
EXEC
EXEC 0p
EXEC //p
EXEC bx
EXEC /
EXEC /a
EXEC $'/\\'
EXEC /a/
EXEC 1,c
EXEC 1,/
EXEC 1,2,3p
EXEC $'/\\(/p'
EXEC $'/\\1/p'
EXEC s/a/b
EXEC s/a/b/q
EXEC s/a/b/3g3
EXEC s/a/b/gg
EXEC s/a/b/pp
EXEC s/a/b/wnot_a_file
EXEC y/a
EXEC y/a/
EXEC y/a/b
EXEC y/aa/bb/
EXEC y/aa/ab/
EXEC y/a/bb/
EXEC y/aa/b/
EXEC 1,2=
EXEC 1,2q
EXEC :
EXEC $'\\'
EXEC a
EXEC c
EXEC e
EXEC f
EXEC i
EXEC j
EXEC k
EXEC m
EXEC o
EXEC r
EXEC s
EXEC u
EXEC v
EXEC w
EXEC y
EXEC z
EXEC $'{'
EXEC $'}'
EXEC 1
EXEC pq
EXEC dq
EXEC aq
EXEC $'!!p'
EXEC $'1#'
EXEC $'a\\'
EXEC $'w .'
EXEC $'s /a/b/'
EXEC $'s/a/b/w .'
TEST 29 'customary extensions'
BODY {
INFO "$a" 'semicolon usable as newline'
case $a in
INFO "$a" 'semicolon terminates a label'
INFO "$a" 'space after semicolon'
INFO "$a" 'space before semicolon'
;;
esac
INFO "$a" 'previous regular expression can be abbreviated as //'
case $a in
*y*) a=$(print ab | COMMAND '
/b/=
:x
s///
/a/s/.*/static/
/b/s/.*/dynamic/
')
INFO "$a" 'label scope'
;;
esac
INFO "$a" 'space optional in r and w commands'
INFO "$a" '\ may precede a non-special character in regular expression'
INFO "$a" '\ may precede a non-special character in substitution text'
INFO "$a" 'spaces allowed between addresses'
}
TEST 30 'hold buffer line join'
EXEC -One '
/^Package:/{
s/^Package:[[:space:]]*\<\([[:alnum:].+-]*$1[[:alnum:].+-]*\).*/\1/
h
}
/^Description:/{
s/^Description:[[:space:]]*\(.*\)/\1/
H
g
s/\
/ - /
p
}
'
INPUT - 'Package: grep
Essential: yes
Priority: required
Section: base
Installed-Size: 488
Debian-Maintainer: Wichert Akkerman <wakkerma@debian.org>
Maintainer: Carl Worth <cworth@handhelds.org>
Architecture: arm
Version: 2.4.2-1
Provides: rgrep
Pre-Depends: libc6 (>= 2.1.2)
Conflicts: rgrep
Filename: ./grep_2.4.2-1_arm.ipk
Size: 119438
MD5Sum: 67fa4cb756f951fda7b7a5d4da2ab523
Description: GNU grep, egrep and fgrep.
The GNU family of grep utilities may be the "fastest grep in the west".
GNU grep is based on a fast lazy-state deterministic matcher (about
twice as fast as stock Unix egrep) hybridized with a Boyer-Moore-Gosper
search for a fixed string that eliminates impossible text from being
considered by the full regexp matcher without necessarily having to
look at every character. The result is typically many times faster
than Unix grep or egrep. (Regular expressions containing backreferencing
will run more slowly, however.)
Package: sed
Priority: required
Section: base
Installed-Size: 180
Debian-Maintainer: Wichert Akkerman <wakkerma@debian.org>
Maintainer: Carl Worth <cworth@handhelds.org>
Architecture: arm
Version: 3.02-6
Pre-Depends: libc6 (>= 2.1.2)
Filename: ./sed_3.02-6_arm.ipk
Size: 12338
MD5Sum: c893daf6fef70813b566db8ed8c06950
Description: The GNU sed stream editor.
sed reads the specified files or the standard input if no
files are specified, makes editing changes according to a
list of commands, and writes the results to the standard
output.'
OUTPUT - 'Package: grep - GNU grep, egrep and fgrep.
Package: sed - The GNU sed stream editor.'
TEST 31 'file access'
DATA 000 644
DIAGNOSTICS
EXEC 's/./x/' 000 644
OUTPUT - $'x44'
EXEC 's/./x/' 644 000
TEST 32 'embedded \0'
SAME OUTPUT nul.out
TEST 33 'multibyte'
EXPORT LC_CTYPE=C.UTF-8
EXEC $'y/\303\240\303\242\303\251\303\250\303\252\303\253\303\256\303\257\303\264\303\266\303\271\303\273\303\247/aaeeeeiioouuc/'
INPUT - $'chaine de caract\303\250res d\303\251finie en utf-8'
OUTPUT - $'chaine de caracteres definie en utf-8'
EXEC $'y\342\202\254\303\240\303\242\303\251\303\250\303\252\303\253\303\256\303\257\303\264\303\266\303\271\303\273\303\247\342\202\254aaeeeeiioouuc\342\202\254'