Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
d18339fbcc
NetBSD/tests/bin/sh/t_here.sh
kre 8d9df45b43 Add sub tests to the quoting test case, to demonstrate a parsing 
bug (and its fix (depending upon whether the test is run against
an unfixed, or fixed, shell).

An obvious indication of the failure is the following (one of the
new sub-tests)

p=A
cat <<EOF
${p+\%$p\%}
${p+%$p%}
EOF

which should output

\%A\%
%A%


as a here doc is treated as a double quoted string, except
that the " character is just a character.  In such a string,
the \ is only an escape character when the following character
is special, otherwise it represents a literal \ (which is the
case here).

An unfixed shell will omit the backslashes in the output.

It gets even more wrong if the % chars are replaced by "
(double quote) chars, which should make no difference, other
than the corresponding change, in the output.   But doesn't
(it doesn't even produce output broken in a similar way).

This one is a harder case to be specific about however,
as while the fixed (and expected in the test) output is what
is technically correct, only a few shells actually produce
it, most generate something different (but not all the same.)
2019-01-22 14:31:53 +00:00

605 lines
16 KiB
Bash
Raw Blame History

# $NetBSD: t_here.sh,v 1.7 2019/01/22 14:31:53 kre Exp $
#
# Copyright (c) 2007 The NetBSD Foundation, Inc.
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions
# are met:
# 1. Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.
# 2. Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
#
# THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
# ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
# TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
# BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
#
# the implementation of "sh" to test
: ${TEST_SH:="/bin/sh"}
nl='
'
reset()
{
TEST_NUM=0
TEST_FAILURES=''
TEST_FAIL_COUNT=0
TEST_ID="$1"
}
check()
{
fail=false
TEMP_FILE=$( mktemp OUT.XXXXXX )
TEST_NUM=$(( $TEST_NUM + 1 ))
# our local shell (ATF_SHELL) better do quoting correctly...
# some of the tests expect us to expand $nl internally...
CMD="nl='${nl}'; $1"
result="$( ${TEST_SH} -c "${CMD}" 2>"${TEMP_FILE}" )"
STATUS=$?
if [ "${STATUS}" -ne "$3" ]; then
echo >&2 "[$TEST_NUM] expected exit code $3, got ${STATUS}"
# don't actually fail just because of wrong exit code
# unless we either expected, or received "good"
case "$3/${STATUS}" in
(*/0|0/*) fail=true;;
esac
fi
if [ "$3" -eq 0 ]; then
if [ -s "${TEMP_FILE}" ]; then
echo >&2 \
"[$TEST_NUM] Messages produced on stderr unexpected..."
cat "${TEMP_FILE}" >&2
fail=true
fi
else
if ! [ -s "${TEMP_FILE}" ]; then
echo >&2 \
"[$TEST_NUM] Expected messages on stderr, nothing produced"
fail=true
fi
fi
rm -f "${TEMP_FILE}"
# Remove newlines (use local shell for this)
result="$(
IFS="$nl"
set -f
set -- $result
IFS=' '
printf %s "$*"
)"
if [ "$2" != "$result" ]
then
echo >&2 "[$TEST_NUM] Expected output '$2', received '$result'"
fail=true
fi
if $fail
then
echo >&2 "[$TEST_NUM] Full command: <<${CMD}>>"
fi
$fail && test -n "$TEST_ID" && {
TEST_FAILURES="${TEST_FAILURES}${TEST_FAILURES:+
}${TEST_ID}[$TEST_NUM]: test of '$1' failed";
TEST_FAIL_COUNT=$(( $TEST_FAIL_COUNT + 1 ))
return 0
}
$fail && atf_fail "Test[$TEST_NUM] of '$1' failed"
return 0
}
results()
{
test -z "${TEST_ID}" && return 0
test -z "${TEST_FAILURES}" && return 0
echo >&2 "=========================================="
echo >&2 "While testing '${TEST_ID}'"
echo >&2 " - - - - - - - - - - - - - - - - -"
echo >&2 "${TEST_FAILURES}"
atf_fail \
"Test ${TEST_ID}: $TEST_FAIL_COUNT subtests (of $TEST_NUM) failed - see stderr"
}
atf_test_case do_simple
do_simple_head() {
atf_set "descr" "Basic tests for here documents"
}
do_simple_body() {
y=x
reset 'simple'
IFS=' '
check 'x=`cat <<EOF'$nl'text'${nl}EOF$nl'`; echo $x' 'text' 0
check 'x=`cat <<\EOF'$nl'text'${nl}EOF$nl'`; echo $x' 'text' 0
check "y=${y};"'x=`cat <<EOF'$nl'te${y}t'${nl}EOF$nl'`; echo $x' \
'text' 0
check "y=${y};"'x=`cat <<\EOF'$nl'te${y}t'${nl}EOF$nl'`; echo $x' \
'te${y}t' 0
check "y=${y};"'x=`cat <<"EOF"'$nl'te${y}t'${nl}EOF$nl'`; echo $x' \
'te${y}t' 0
check "y=${y};"'x=`cat <<'"'EOF'"$nl'te${y}t'${nl}EOF$nl'`; echo $x' \
'te${y}t' 0
# check that quotes in the here doc survive and cause no problems
check "cat <<EOF${nl}te'xt${nl}EOF$nl" "te'xt" 0
check "cat <<\EOF${nl}te'xt${nl}EOF$nl" "te'xt" 0
check "cat <<'EOF'${nl}te'xt${nl}EOF$nl" "te'xt" 0
check "cat <<EOF${nl}te\"xt${nl}EOF$nl" 'te"xt' 0
check "cat <<\EOF${nl}te\"xt${nl}EOF$nl" 'te"xt' 0
check "cat <<'EOF'${nl}te\"xt${nl}EOF$nl" 'te"xt' 0
check "cat <<'EO'F${nl}te\"xt${nl}EOF$nl" 'te"xt' 0
check "y=${y};"'x=`cat <<EOF'$nl'te'"'"'${y}t'${nl}EOF$nl'`; echo $x' \
'te'"'"'xt' 0
check "y=${y};"'x=`cat <<EOF'$nl'te'"''"'${y}t'${nl}EOF$nl'`; echo $x' \
'te'"''"'xt' 0
# note that the blocks of empty space in the following must
# be entirely tab characters, no spaces.
check 'x=`cat <<EOF'"$nl text${nl}EOF$nl"'`; echo "$x"' \
' text' 0
check 'x=`cat <<-EOF'"$nl text${nl}EOF$nl"'`; echo $x' \
'text' 0
check 'x=`cat <<-EOF'"${nl}text${nl} EOF$nl"'`; echo $x' \
'text' 0
check 'x=`cat <<-\EOF'"$nl text${nl} EOF$nl"'`; echo $x' \
'text' 0
check 'x=`cat <<- "EOF"'"$nl text${nl}EOF$nl"'`; echo $x' \
'text' 0
check 'x=`cat <<- '"'EOF'${nl}text${nl} EOF$nl"'`; echo $x' \
'text' 0
results
}
atf_test_case end_markers
end_markers_head() {
atf_set "descr" "Tests for various end markers of here documents"
}
end_markers_body() {
reset 'end_markers'
for end in EOF 1 \! '$$$' "string " a\\\ a\\\ \ '&' '' ' ' ' ' \
--STRING-- . '~~~' ')' '(' '#' '()' '(\)' '(\/)' '--' '\' '{' '}' \
VERYVERYVERYVERYLONGLONGLONGin_fact_absurdly_LONG_LONG_HERE_DOCUMENT_TERMINATING_MARKER_THAT_goes_On_forever_and_ever_and_ever...
do
# check unquoted end markers
case "${end}" in
('' | *[' ()\$&#*~']* ) ;; # skip unquoted endmark test for these
(*) check \
'x=$(cat << '"${end}${nl}text${nl}${end}${nl}"'); printf %s "$x"' 'text' 0
;;
esac
# and quoted end markers
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end}${nl}"'); printf %s "$x"' 'text' 0
# and see what happens if we encounter "almost" an end marker
case "${#end}" in
(0|1) ;; # too short to try truncation tests
(*) check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end%?}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end%?}" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end#?}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end#?}" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end%?}+${nl}${end}${nl}"');printf %s "$x"' \
"text ${end%?}+" 0
;;
esac
# or something that is a little longer
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end}x${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end}x" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl}!${end}${nl}${end}${nl}"'); printf %s "$x"' \
"text !${end}" 0
# or which does not begin at start of line
check \
'x=$(cat <<'"'${end}'${nl}text${nl} ${end}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end}" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl} ${end}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end}" 0
# or end at end of line
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end} ${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end} " 0
# or something that is correct much of the way, but then...
case "${#end}" in
(0) ;; # cannot test this one
(1) check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end}${end}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end}${end}" 0
;;
(2-7) pfx="${end%?}"
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end}${pfx}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end}${pfx}" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${pfx}${end}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${pfx}${end}" 0
;;
(*) pfx=${end%??????}; sfx=${end#??????}
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${end}${sfx}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${end}${sfx}" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${pfx}${end}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${pfx}${end}" 0
check \
'x=$(cat <<'"'${end}'${nl}text${nl}${pfx}${sfx}${nl}${end}${nl}"'); printf %s "$x"' \
"text ${pfx}${sfx}" 0
;;
esac
done
# Add striptabs tests (in similar way) here one day...
results
}
atf_test_case incomplete
incomplete_head() {
atf_set "descr" "Basic tests for incomplete here documents"
}
incomplete_body() {
reset incomplete
check 'cat <<EOF' '' 2
check 'cat <<- EOF' '' 2
check 'cat <<\EOF' '' 2
check 'cat <<- \EOF' '' 2
check 'cat <<EOF'"${nl}" '' 2
check 'cat <<- EOF'"${nl}" '' 2
check 'cat <<'"'EOF'${nl}" '' 2
check 'cat <<- "EOF"'"${nl}" '' 2
check 'cat << EOF'"${nl}${nl}" '' 2
check 'cat <<-EOF'"${nl}${nl}" '' 2
check 'cat << '"'EOF'${nl}${nl}" '' 2
check 'cat <<-"EOF"'"${nl}${nl}" '' 2
check 'cat << EOF'"${nl}"'line 1'"${nl}" '' 2
check 'cat <<-EOF'"${nl}"' line 1'"${nl}" '' 2
check 'cat << EOF'"${nl}"'line 1'"${nl}"' line 2'"${nl}" '' 2
check 'cat <<-EOF'"${nl}"' line 1'"${nl}"'line 2'"${nl}" '' 2
check 'cat << EOF'"${nl}line 1${nl}${nl}line3${nl}${nl}5!${nl}" '' 2
results
}
atf_test_case lineends
lineends_head() {
atf_set "descr" "Tests for line endings in here documents"
}
lineends_body() {
reset lineends
# note that "check" removes newlines from stdout before comparing.
# (they become blanks, provided there is something before & after)
check 'cat << \echo'"${nl}"'\'"${nl}echo${nl}echo${nl}" '\' 0
check 'cat << echo'"${nl}"'\'"${nl}echo${nl}echo${nl}" 'echo' 0
check 'cat << echo'"${nl}"'\\'"${nl}echo${nl}echo${nl}" '\' 0
check 'X=3; cat << ec\ho'"${nl}"'$X\'"${nl}echo${nl}echo${nl}" \
'$X\' 0
check 'X=3; cat << echo'"${nl}"'$X'"${nl}echo${nl}echo${nl}" \
'3' 0
check 'X=3; cat << echo'"${nl}"'$X\'"${nl}echo${nl}echo${nl}" \
'' 0
check 'X=3; cat << echo'"${nl}"'${X}\'"${nl}echo${nl}echo${nl}" \
'3echo' 0
check 'X=3; cat << echo'"${nl}"'\$X\'"${nl}echo${nl}echo${nl}" \
'$Xecho' 0
check 'X=3; cat << echo'"${nl}"'\\$X \'"${nl}echo${nl}echo${nl}" \
'\3 echo' 0
check \
'cat << "echo"'"${nl}"'line1\'"${nl}"'line2\'"${nl}echo${nl}echo${nl}" \
'line1\ line2\' 0
check \
'cat << echo'"${nl}"'line1\'"${nl}"'line2\'"${nl}echo${nl}echo${nl}" \
'line1line2echo' 0
results
}
atf_test_case multiple
multiple_head() {
atf_set "descr" "Tests for multiple here documents on one cmd line"
}
multiple_body() {
reset multiple
check \
"(cat ; cat <&3) <<EOF0 3<<EOF3${nl}STDIN${nl}EOF0${nl}-3-${nl}EOF3${nl}" \
'STDIN -3-' 0
check "(read line; echo \"\$line\"; cat <<EOF1; echo \"\$line\") <<EOF2
The File
EOF1
The Line
EOF2
" 'The Line The File The Line' 0
check "(read line; echo \"\$line\"; cat <<EOF; echo \"\$line\") <<EOF
The File
EOF
The Line
EOF
" 'The Line The File The Line' 0
check "V=1; W=2; cat <<-1; cat <<2; cat <<- 3; cat <<'4';"' cat <<\5
$V
$W
3
4
5
1
2
5
4*$W+\$V
3
$W
1
2
3
4
7+$V
$W+6
5
' '1 2 3 4 5 5 4*2+$V $W 1 2 3 7+$V $W+6' 0
results
}
atf_test_case nested
nested_head() {
atf_set "descr" "Tests for nested here documents for one cmd"
}
nested_body() {
reset nested
check \
'cat << EOF1'"${nl}"'$(cat << EOF2'"${nl}LINE${nl}EOF2${nl}"')'"${nl}EOF1${nl}"\
'LINE' 0
# This next one fails ... and correctly, so we will omit it (bad test)
# Reasoning is that the correct data "$(cat << EOF2)\nLINE\nEOF2\n" is
# collected for the outer (EOF1) heredoc, when that is parsed, it looks
# like
# $(cat <<EOF2)
# LINE
# EOF2
# which looks like a good command - except it is being parsed in "heredoc"
# syntax, which means it is enclosed in double quotes, which means that
# the newline after the ')' in the first line is not a newline token, but
# just a character. The EOF2 heredoc cannot start until after the next
# newline token, of which there are none here... LINE and EOF2 are just
# more data in the outer EOF1 heredoc for its "cat" command to read & write.
#
# The previous sub-test works because there the \n comes inside the
# $( ), and in there, the outside quoting rules are suspended, and it
# all starts again - so that \n is a newline token, and the EOF2 heredoc
# is processed.
#
# check \
# 'cat << EOF1'"${nl}"'$(cat << EOF2 )'"${nl}LINE${nl}EOF2${nl}EOF1${nl}" \
# 'LINE' 0
L='cat << EOF1'"${nl}"'LINE1$(cat << EOF2'"${nl}"
L="${L}"'LINE2$(cat << EOF3'"${nl}"
L="${L}"'LINE3$(cat << EOF4'"${nl}"
L="${L}"'LINE4$(cat << EOF5'"${nl}"
L="${L}LINE5${nl}EOF5${nl})4${nl}EOF4${nl})3${nl}"
L="${L}EOF3${nl})2${nl}EOF2${nl})1${nl}EOF1${nl}"
# That mess is ...
#
# cat <<EOF1
# LINE1$(cat << EOF2
# LINE2$(cat << EOF3
# LINE3$(cat << EOF4
# LINE4$(cat << EOF5
# LINE5
# EOF5
# )4
# EOF4
# )3
# EOF3
# )2
# EOF2
# )1
# EOF1
check "${L}" 'LINE1LINE2LINE3LINE4LINE54321' 0
results
}
atf_test_case quoting
quoting_head() {
atf_set "descr" "Tests for use of quotes inside here documents"
}
quoting_body() {
reset quoting
check 'X=!; cat <<- E\0F
<'\''"'\'' \\$X\$X "'\''" \\>
E0F
' '<'\''"'\'' \\$X\$X "'\''" \\>' 0
check 'X=!; cat <<- E0F
<'\''"'\'' \\$X\$X "'\''" \\>
E0F
' '<'\''"'\'' \!$X "'\''" \>' 0
check 'cat <<- END
$( echo "'\''" ) $( echo '\''"'\'' ) $( echo \\ )
END
' "' \" \\" 0
check 'X=12345; Y="string1 line1?-line2"; Z=; unset W; cat <<-EOF
${#X}${Z:-${Y}}${W+junk}${Y%%l*}${Y#*\?}
"$Z"'\''$W'\'' ${Y%" "*} $(( X + 54321 ))
EOF
' '5string1 line1?-line2string1 -line2 ""'\'\'' string1 66666' 0
# check that \ only quotes the magic chars, otherwise is retained
check 'p=A; cat <<-EOF
${p+\%$p\%}
${p+%$p%}
EOF
' '\%A\% %A%' 0
# and check that " is not magic, so \ does not quote it
check 'p=A; cat <<-EOF
${p+\"$p\"}
${p+"$p"}
EOF
' '\"A\" "A"' 0
# except in a ${var%<word>} word, base syntax reapplies, and
# there quotes are magic again
check 'p=ABCD; cat <<-EOF
${p%B?D}
${p%B\?D}
${p%"BCD"}
"${p%??}"
${p#"${p%??}"}
"${p#"${p%?"?"}"}"
EOF
' 'A ABCD A "AB" CD ""' 0
check 'p=AB??; cat <<-EOF
${p%B?D}
${p%B\??}
${p%"B??"}
"${p%??}"
${p#"${p%??}"}
"${p#"${p%?"?"}"}"
EOF
' 'AB?? A A "AB" ?? "??"' 0
results
}
atf_test_case side_effects
side_effects_head() {
atf_set "descr" "Tests how side effects in here documents are handled"
}
side_effects_body() {
atf_check -s exit:0 -o inline:'2\n1\n' -e empty ${TEST_SH} -c '
unset X
cat <<-EOF
${X=2}
EOF
echo "${X-1}"
'
}
atf_test_case vicious
vicious_head() {
atf_set "descr" "Tests for obscure and obnoxious uses of here docs"
}
vicious_body() {
reset
cat <<- \END_SCRIPT > script
cat <<ONE && cat \
<<TWO
a
ONE
b
TWO
END_SCRIPT
atf_check -s exit:0 -o inline:'a\nb\n' -e empty ${TEST_SH} script
# This next one is causing discussion currently (late Feb 2016)
# amongst stds writers & implementors. Consequently we
# will not check what it produces. The eventual result
# seems unlikely to be what we currently output, which
# is:
# A:echo line 1
# B:echo line 2)" && prefix DASH_CODE <<DASH_CODE
# B:echo line 3
# line 4
# line 5
#
# The likely intended output is ...
#
# A:echo line 3
# B:echo line 1
# line 2
# DASH_CODE:echo line 4)"
# DASH_CODE:echo line 5
#
# The difference is explained by differing opinions on just
# when processing of a here doc should start
cat <<- \END_SCRIPT > script
prefix() { sed -e "s/^/$1:/"; }
DASH_CODE() { :; }
prefix A <<XXX && echo "$(prefix B <<XXX
echo line 1
XXX
echo line 2)" && prefix DASH_CODE <<DASH_CODE
echo line 3
XXX
echo line 4)"
echo line 5
DASH_CODE
END_SCRIPT
# we will just verify that the shell can parse the
# script somehow, and doesn't fall over completely...
atf_check -s exit:0 -o ignore -e empty ${TEST_SH} script
}
atf_init_test_cases() {
atf_add_test_case do_simple # not worthy of a comment
atf_add_test_case end_markers # the mundane, the weird, the bizarre
atf_add_test_case incomplete # where the end marker isn't...
atf_add_test_case lineends # test weird line endings in heredocs
atf_add_test_case multiple # multiple << operators on one cmd
atf_add_test_case nested # here docs inside here docs
atf_add_test_case quoting # stuff quoted inside
atf_add_test_case side_effects # here docs that modify environment
atf_add_test_case vicious # evil test from the austin-l list...
}
Reference in New Issue View Git Blame Copy Permalink