pcre2/RunTest

#! /bin/sh

###############################################################################
# Run the PCRE2 tests using the pcre2test program. The appropriate tests are
# selected, depending on which build-time options were used.
#
# When JIT support is available, all appropriate tests are run with and without
# JIT, unless "nojit" is given on the command line. There are also two tests
# for JIT-specific features, one to be run when JIT support is available
# (unless "nojit" is specified), and one when it is not.
#
# Whichever of the 8-, 16- and 32-bit libraries exist are tested. It is also
# possible to select which to test by giving "-8", "-16" or "-32" on the
# command line.
#
# As well as "nojit", "-8", "-16", and "-32", arguments for this script are
# individual test numbers, ranges of tests such as 3-6 or 3- (meaning 3 to the
# end), or a number preceded by ~ to exclude a test. For example, "3-15 ~10"
# runs tests 3 to 15, excluding test 10, and just "~10" runs all the tests
# except test 10. Whatever order the arguments are in, the tests are always run
# in numerical order.
#
# The special argument "3S" runs test 3, stopping if it fails. Test 3 is the
# locale test, and failure usually means there's an issue with the locale
# rather than a bug in PCRE2, so normally subsequent tests are run. "3S" is
# useful when you want to debug or update the test.
#
# Inappropriate tests are automatically skipped (with a comment to say so): for
# example, if JIT support is not compiled, test 12 is skipped, whereas if JIT
# support is compiled, test 13 is skipped.
#
# Other arguments can be one of the words "valgrind", "valgrind-log", or "sim"
# followed by an argument to run cross-compiled executables under a simulator,
# for example:
#
# RunTest 3 sim "qemu-arm -s 8388608"
#
# There are two special cases where only one argument is allowed:
#
# If the first and only argument is "ebcdic", the script runs the special
# EBCDIC test that can be useful for checking certain EBCDIC features, even
# when run in an ASCII environment.
#
# If the script is obeyed as "RunTest list", a list of available tests is
# output, but none of them are run.
###############################################################################

# Define test titles in variables so that they can be output as a list. Some
# of them are modified (e.g. with -8 or -16) when used in the actual tests.

title1="Test 1: Main non-UTF, non-UCP functionality (compatible with Perl >= 5.10)"
title2="Test 2: API, errors, internals, and non-Perl stuff"
title3="Test 3: Locale-specific features"
title4A="Test 4: UTF"
title4B=" and Unicode property support (compatible with Perl >= 5.10)"
title5A="Test 5: API, internals, and non-Perl stuff for UTF"
title5B=" and UCP support"
title6="Test 6: DFA matching main non-UTF, non-UCP functionality"
title7A="Test 7: DFA matching with UTF"
title7B=" and Unicode property support"
title8="Test 8: Internal offsets and code size tests"
title9="Test 9: Specials for the basic 8-bit library"
title10="Test 10: Specials for the 8-bit library with UTF-8 and UCP support"
title11="Test 11: Specials for the basic 16-bit and 32-bit libraries"
title12="Test 12: Specials for the 16-bit and 32-bit libraries UTF and UCP support"
title13="Test 13: DFA specials for the basic 16-bit and 32-bit libraries"
title14="Test 14: Non-JIT limits tests"
title15="Test 15: JIT-specific features when JIT is not available"
title16="Test 16: JIT-specific features when JIT is available"
title17="Test 17: Tests of the POSIX interface, excluding UTF/UCP"
title18="Test 18: Tests of the POSIX interface with UTF/UCP"

maxtest=18

if [ $# -eq 1 -a "$1" = "list" ]; then
  echo $title1
  echo $title2 "(not UTF or UCP)"
  echo $title3
  echo $title4A $title4B
  echo $title5A $title5B
  echo $title6
  echo $title7A $title7B
  echo $title8
  echo $title9
  echo $title10
  echo $title11
  echo $title12
  echo $title13
  echo $title14
  echo $title15
  echo $title16
  echo $title17
  echo $title18
  exit 0
fi

# Set up a suitable "diff" command for comparison. Some systems
# have a diff that lacks a -u option. Try to deal with this.

cf="diff"
diff -u /dev/null /dev/null 2>/dev/null && cf="diff -u"

# Find the test data

if [ -n "$srcdir" -a -d "$srcdir" ] ; then
  testdata="$srcdir/testdata"
elif [ -d "./testdata" ] ; then
  testdata=./testdata
elif [ -d "../testdata" ] ; then
  testdata=../testdata
else
  echo "Cannot find the testdata directory"
  exit 1
fi


# ------ Special EBCDIC Test -------

if [ $# -eq 1 -a "$1" = "ebcdic" ]; then
  ./pcre2test -C ebcdic >/dev/null
  ebcdic=$?
  if [ $ebcdic -ne 1 ] ; then
    echo "Cannot run EBCDIC tests: EBCDIC support not compiled"
    exit 1
  fi

  for opt in "" "-dfa"; do
    ./pcre2test -q $opt $testdata/testinputEBC >testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutputEBC testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    if [ "$opt" = "-dfa" ] ; then echo "  OK using DFA"
    else echo "  OK"
    fi
  done

exit 0
fi


# ------ Normal Tests ------

# Default values

arg8=
arg16=
arg32=
nojit=
sim=
skip=
valgrind=

# This is in case the caller has set aliases (as I do - PH)
unset cp ls mv rm

# Process options and select which tests to run; for those that are explicitly
# requested, check that the necessary optional facilities are available.

do1=no
do2=no
do3=no
do4=no
do5=no
do6=no
do7=no
do8=no
do9=no
do10=no
do11=no
do12=no
do13=no
do14=no
do15=no
do16=no
do17=no
do18=no

while [ $# -gt 0 ] ; do
  case $1 in
    1) do1=yes;;
    2) do2=yes;;
    3) do3=yes;;
    4) do4=yes;;
    5) do5=yes;;
    6) do6=yes;;
    7) do7=yes;;
    8) do8=yes;;
    9) do9=yes;;
   10) do10=yes;;
   11) do11=yes;;
   12) do12=yes;;
   13) do13=yes;;
   14) do14=yes;;
   15) do15=yes;;
   16) do16=yes;;
   17) do17=yes;;
   18) do18=yes;;
   -8) arg8=yes;;
  -16) arg16=yes;;
  -32) arg32=yes;;
   nojit) nojit=yes;;
   sim) shift; sim=$1;;
   valgrind) valgrind="valgrind --tool=memcheck -q --smc-check=all";;
   valgrind-log) valgrind="valgrind --tool=memcheck --num-callers=30 --leak-check=no --error-limit=no --smc-check=all --log-file=report.%p ";;
   ~*)
     if expr "$1" : '~[0-9][0-9]*$' >/dev/null; then
       skip="$skip `expr "$1" : '~\([0-9]*\)*$'`"
     else
       echo "Unknown option or test selector '$1'"; exit 1
     fi
   ;;
   *-*)
     if expr "$1" : '[0-9][0-9]*-[0-9]*$' >/dev/null; then
       tf=`expr "$1" : '\([0-9]*\)'`
       tt=`expr "$1" : '.*-\([0-9]*\)'`
       if [ "$tt" = "" ] ; then tt=$maxtest; fi
       if expr \( "$tf" "<" 1 \) \| \( "$tt" ">" "$maxtest" \) >/dev/null; then
         echo "Invalid test range '$1'"; exit 1
       fi
       while expr "$tf" "<=" "$tt" >/dev/null; do
         eval do${tf}=yes
         tf=`expr $tf + 1`
       done
     else
       echo "Invalid test range '$1'"; exit 1
     fi
   ;;
   *) echo "Unknown option or test selector '$1'"; exit 1;;
  esac
  shift
done

# Find which optional facilities are available.

$sim ./pcre2test -C linksize >/dev/null
link_size=$?
if [ $link_size -lt 2 ] ; then
  echo "Failed to find internal link size"
  exit 1
fi
if [ $link_size -gt 4 ] ; then
  echo "Failed to find internal link size"
  exit 1
fi

# All of 8-bit, 16-bit, and 32-bit character strings may be supported, but only
# one need be.

$sim ./pcre2test -C pcre8 >/dev/null
support8=$?
$sim ./pcre2test -C pcre16 >/dev/null
support16=$?
$sim ./pcre2test -C pcre32 >/dev/null
support32=$?

# Initialize all bitsizes skipped

test8=skip
test16=skip
test32=skip

# If no bitsize arguments, select all that are available

if [ "$arg8$arg16$arg32" = "" ] ; then
  if [ $support8 -ne 0 ] ; then
    test8=
  fi
  if [ $support16 -ne 0 ] ; then
    test16=-16
  fi
  if [ $support32 -ne 0 ] ; then
    test32=-32
  fi

# Select requested bit sizes

else
  if [ "$arg8" = yes ] ; then
    if [ $support8 -eq 0 ] ; then
      echo "Cannot run 8-bit library tests: 8-bit library not compiled"
      exit 1
    fi
    test8=
  fi
  if [ "$arg16" = yes ] ; then
    if [ $support16 -eq 0 ] ; then
      echo "Cannot run 16-bit library tests: 16-bit library not compiled"
      exit 1
    fi
    test16=-16
  fi
  if [ "$arg32" = yes ] ; then
    if [ $support32 -eq 0 ] ; then
      echo "Cannot run 32-bit library tests: 32-bit library not compiled"
      exit 1
    fi
    test32=-32
  fi
fi

# UTF support is implied by Unicode support, and it always applies to all bit
# sizes if both are supported; we can't have UTF-8 support without UTF-16 or
# UTF-32 support.

$sim ./pcre2test -C unicode >/dev/null
utf=$?

jitopt=
$sim ./pcre2test -C jit >/dev/null
jit=$?
if [ $jit -ne 0 -a "$nojit" != "yes" ] ; then
  jitopt=-jit
fi

# If no specific tests were requested, select all. Those that are not
# relevant will be automatically skipped.

if [ $do1  = no -a $do2  = no -a $do3  = no -a $do4  = no -a \
     $do5  = no -a $do6  = no -a $do7  = no -a $do8  = no -a \
     $do9  = no -a $do10 = no -a $do11 = no -a $do12 = no -a \
     $do13 = no -a $do14 = no -a $do15 = no -a $do16 = no -a \
     $do17 = no -a $do18 = no \
   ]; then
  do1=yes
  do2=yes
  do3=yes
  do4=yes
  do5=yes
  do6=yes
  do7=yes
  do8=yes
  do9=yes
  do10=yes
  do11=yes
  do12=yes
  do13=yes
  do14=yes
  do15=yes
  do16=yes
  do17=yes
  do18=yes
fi

# Handle any explicit skips at this stage, so that an argument list may consist
# only of explicit skips.

for i in $skip; do eval do$i=no; done

# Show which release and which test data

echo ""
echo PCRE2 C library tests using test data from $testdata
$sim ./pcre2test /dev/null
echo ""

for bmode in "$test8" "$test16" "$test32"; do
  case "$bmode" in
    skip) continue;;
    -16)  if [ "$test8$test32" != "skipskip" ] ; then echo ""; fi
          bits=16; echo "---- Testing 16-bit library ----"; echo "";;
    -32)  if [ "$test8$test16" != "skipskip" ] ; then echo ""; fi
          bits=32; echo "---- Testing 32-bit library ----"; echo "";;
    *)    bits=8; echo "---- Testing 8-bit library ----"; echo "";;
  esac

# Primary non-UTF test, compatible with JIT and all versions of Perl >= 5.8

if [ $do1 = yes ] ; then
  echo $title1
  for opt in "" $jitopt; do
    $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput1 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput1 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
    else echo "  OK"
    fi
  done
fi

# PCRE2 tests that are not JIT or Perl-compatible: API, errors, internals

if [ $do2 = yes ] ; then
  echo $title2 "(excluding UTF-$bits)"
  for opt in "" $jitopt; do
    $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput2 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput2 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else
      echo " "
      echo "** Test 2 requires a lot of stack. If it has crashed with a"
      echo "** segmentation fault, it may be that you do not have enough"
      echo "** stack available by default. Please see the 'pcre2stack' man"
      echo "** page for a discussion of PCRE2's stack usage."
      echo " "
      exit 1
    fi
    if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
    else echo "  OK"
    fi
  done
fi

# Locale-specific tests, provided that either the "fr_FR" or the "french"
# locale is available. The former is the Unix-like standard; the latter is
# for Windows. Another possibility is "fr". Unfortunately, different versions
# of the French locale give different outputs for some items. This test passes
# if the output matches any one of the alternative output files.

if [ $do3 = yes ] ; then
  locale -a | grep '^fr_FR$' >/dev/null
  if [ $? -eq 0 ] ; then
    locale=fr_FR
    infile=$testdata/testinput3
    outfile=$testdata/testoutput3
    outfile2=$testdata/testoutput3A
    outfile3=$testdata/testoutput3B
  else
    infile=test3input
    outfile=test3output
    outfile2=test3outputA
    outfile3=test3outputB
    locale -a | grep '^french$' >/dev/null
    if [ $? -eq 0 ] ; then
      locale=french
      sed 's/fr_FR/french/' $testdata/testinput3 >test3input
      sed 's/fr_FR/french/' $testdata/testoutput3 >test3output
      sed 's/fr_FR/french/' $testdata/testoutput3A >test3outputA
      sed 's/fr_FR/french/' $testdata/testoutput3B >test3outputB
    else
      locale -a | grep '^fr$' >/dev/null
      if [ $? -eq 0 ] ; then
        locale=fr
        sed 's/fr_FR/fr/' $testdata/intestinput3 >test3input
        sed 's/fr_FR/fr/' $testdata/intestoutput3 >test3output
        sed 's/fr_FR/fr/' $testdata/intestoutput3A >test3outputA
        sed 's/fr_FR/fr/' $testdata/intestoutput3B >test3outputB
      else
        locale=
      fi
    fi
  fi

  if [ "$locale" != "" ] ; then
    echo $title3 "(using '$locale' locale)"
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $infile testtry
      if [ $? = 0 ] ; then
        if $cf $outfile testtry >teststdout || \
           $cf $outfile2 testtry >teststdout || \
           $cf $outfile3 testtry >teststdout
        then
          if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
          else echo "  OK"
          fi
        else
          echo "** Locale test did not run successfully. The output did not match"
          echo "   $outfile, $outfile2 or $outfile3."
          echo "   This may mean that there is a problem with the locale settings rather"
          echo "   than a bug in PCRE2."
          exit 1
        fi
      else exit 1
      fi
    done
  else
    echo "Cannot test locale-specific features - none of the 'fr_FR', 'fr' or"
    echo "'french' locales exist, or the \"locale\" command is not available"
    echo "to check for them."
    echo " "
  fi
fi

# Tests for UTF and Unicode property support

if [ $do4 = yes ] ; then
  echo ${title4A}-${bits}${title4B}
  if [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput4 testtry
      if [ $? = 0 ] ; then
        $cf $testdata/testoutput4 testtry
        if [ $? != 0 ] ; then exit 1; fi
      else exit 1
      fi
      if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
      else echo "  OK"
      fi
    done
  fi
fi

if [ $do5 = yes ] ; then
  echo ${title5A}-${bits}$title5B
  if [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput5 testtry
      if [ $? = 0 ] ; then
        $cf $testdata/testoutput5 testtry
        if [ $? != 0 ] ; then exit 1; fi
      else exit 1
      fi
      if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
      else echo "  OK"
      fi
    done
  fi
fi

# Tests for DFA matching support

if [ $do6 = yes ] ; then
  echo $title6
  $sim $valgrind ./pcre2test -q $bmode $testdata/testinput6 testtry
  if [ $? = 0 ] ; then
    $cf $testdata/testoutput6 testtry
    if [ $? != 0 ] ; then exit 1; fi
  else exit 1
  fi
  echo "  OK"
fi

if [ $do7 = yes ] ; then
  echo ${title7A}-${bits}$title7B
  if [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput7 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput7 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# Test of internal offsets and code sizes. This test is run only when there
# is UTF/UCP support and the link size is 2. The actual tests are
# mostly the same as in some of the above, but in this test we inspect some
# offsets and sizes that require a known link size. This is a doublecheck for
# the maintainer, just in case something changes unexpectely. The output from
# this test is different in 8-bit, 16-bit, and 32-bit modes, so there are
# mode-specific output files.

if [ $do8 = yes ] ; then
  echo $title8
  if [ $link_size -ne 2 ] ; then
    echo "  Skipped because link size is not 2"
  elif [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput8 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput8-$bits testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# Tests for 8-bit-specific features

if [ "$do9" = yes ] ; then
  echo $title9
  if [ "$bits" = "16" -o "$bits" = "32" ] ; then
    echo "  Skipped when running 16/32-bit tests"
  else
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput9 testtry
      if [ $? = 0 ] ; then
        $cf $testdata/testoutput9 testtry
        if [ $? != 0 ] ; then exit 1; fi
      else exit 1
      fi
      if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
      else echo "  OK"
      fi
    done
  fi
fi

# Tests for UTF-8 and UCP 8-bit-specific features

if [ "$do10" = yes ] ; then
  echo $title10
  if [ "$bits" = "16" -o "$bits" = "32" ] ; then
    echo "  Skipped when running 16/32-bit tests"
  elif [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput10 testtry
      if [ $? = 0 ] ; then
        $cf $testdata/testoutput10 testtry
        if [ $? != 0 ] ; then exit 1; fi
      else exit 1
      fi
      if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
      else echo "  OK"
      fi
    done
  fi
fi

# Tests for 16-bit and 32-bit features. Output is different for the two widths.

if [ $do11 = yes ] ; then
  echo $title11
  if [ "$bits" = "8" ] ; then
    echo "  Skipped when running 8-bit tests"
  else
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput11 testtry
      if [ $? = 0 ] ; then
        $cf $testdata/testoutput11-$bits testtry
        if [ $? != 0 ] ; then exit 1; fi
      else exit 1
      fi
      if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
      else echo "  OK"
      fi
    done
  fi
fi

# Tests for 16-bit and 32-bit features with UTF-16/32 and UCP support. Output
# is different for the two widths.

if [ $do12 = yes ] ; then
  echo $title12
  if [ "$bits" = "8" ] ; then
    echo "  Skipped when running 8-bit tests"
  elif [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    for opt in "" $jitopt; do
      $sim $valgrind ./pcre2test -q $bmode $opt $testdata/testinput12 testtry
      if [ $? = 0 ] ; then
        $cf $testdata/testoutput12-$bits testtry
        if [ $? != 0 ] ; then exit 1; fi
      else exit 1
      fi
      if [ "$opt" = "-jit" ] ; then echo "  OK with JIT"
      else echo "  OK"
      fi
    done
  fi
fi

# Tests for 16/32-bit-specific features in DFA non-UTF modes

if [ $do13 = yes ] ; then
  echo $title13
  if [ "$bits" = "8" ] ; then
    echo "  Skipped when running 8-bit tests"
  else
    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput13 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput13 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# Test non-JIT match and recursion limits

if [ $do14 = yes ] ; then
  echo $title14
  $sim $valgrind ./pcre2test -q $bmode $testdata/testinput14 testtry
  if [ $? = 0 ] ; then
    $cf $testdata/testoutput14 testtry
    if [ $? != 0 ] ; then exit 1; fi
  else exit 1
  fi
  echo "  OK"
fi

# Test JIT-specific features when JIT is not available

if [ $do15 = yes ] ; then
  echo $title15
  if [ $jit -ne 0 ] ; then
    echo "  Skipped because JIT is available"
  else
    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput15 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput15 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# Test JIT-specific features when JIT is available

if [ $do16 = yes ] ; then
  echo $title16
  if [ $jit -eq 0 -o "$nojit" = "yes" ] ; then
    echo "  Skipped because JIT is not available or not usable"
  else
    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput16 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput16 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# Tests for the POSIX interface without UTF/UCP (8-bit only)

if [ $do17 = yes ] ; then
  echo $title17
  if [ "$bits" = "16" -o "$bits" = "32" ] ; then
    echo "  Skipped when running 16/32-bit tests"
  else
    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput17 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput17 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# Tests for the POSIX interface with UTF/UCP (8-bit only)

if [ $do18 = yes ] ; then
  echo $title18
  if [ "$bits" = "16" -o "$bits" = "32" ] ; then
    echo "  Skipped when running 16/32-bit tests"
  elif [ $utf -eq 0 ] ; then
    echo "  Skipped because UTF-$bits support is not available"
  else
    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput18 testtry
    if [ $? = 0 ] ; then
      $cf $testdata/testoutput18 testtry
      if [ $? != 0 ] ; then exit 1; fi
    else exit 1
    fi
    echo "  OK"
  fi
fi

# End of loop for 8/16/32-bit tests
done

# Clean up local working files
rm -f test3input test3output test3outputA testNinput testsaved* teststderr teststdout testtry

# End