⭐ 欢迎来到虫虫下载站! | 📦 资源下载 📁 资源专辑 ℹ️ 关于我们
⭐ 虫虫下载站

📄 constant.t

📁 source of perl for linux application,
💻 T
📖 第 1 页 / 共 2 页
字号:
push @args, [PROXYSUBS => 1] if $] > 5.009002;foreach my $args (@args){  # Simple tests  start_tests();  my $parent_rfc1149 =    'A Standard for the Transmission of IP Datagrams on Avian Carriers';  # Test the code that generates 1 and 2 letter name comparisons.  my %compass = (                 N => 0, 'NE' => 45, E => 90, SE => 135,                 S => 180, SW => 225, W => 270, NW => 315                );  my $header = << "EOT";#define FIVE 5#define OK6 "ok 6\\n"#define OK7 1#define FARTHING 0.25#define NOT_ZERO 1#define Yes 0#define No 1#define Undef 1#define RFC1149 "$parent_rfc1149"#undef NOTDEF#define perl "rules"EOT  while (my ($point, $bearing) = each %compass) {    $header .= "#define $point $bearing\n"  }  my @items = ("FIVE", {name=>"OK6", type=>"PV",},               {name=>"OK7", type=>"PVN",                value=>['"not ok 7\\n\\0ok 7\\n"', 15]},               {name => "FARTHING", type=>"NV"},               {name => "NOT_ZERO", type=>"UV", value=>"~(UV)0"},               {name => "OPEN", type=>"PV", value=>'"/*"', macro=>1},               {name => "CLOSE", type=>"PV", value=>'"*/"',                macro=>["#if 1\n", "#endif\n"]},               {name => "ANSWER", default=>["UV", 42]}, "NOTDEF",               {name => "Yes", type=>"YES"},               {name => "No", type=>"NO"},               {name => "Undef", type=>"UNDEF"},  # OK. It wasn't really designed to allow the creation of dual valued  # constants.  # It was more for INADDR_ANY INADDR_BROADCAST INADDR_LOOPBACK INADDR_NONE               {name=>"RFC1149", type=>"SV", value=>"sv_2mortal(temp_sv)",                pre=>"SV *temp_sv = newSVpv(RFC1149, 0); "                . "(void) SvUPGRADE(temp_sv,SVt_PVIV); SvIOK_on(temp_sv); "                . "SvIV_set(temp_sv, 1149);"},              );  push @items, $_ foreach keys %compass;  # Automatically compile the list of all the macro names, and make them  # exported constants.  my @export_names = map {(ref $_) ? $_->{name} : $_} @items;  # Exporter::Heavy (currently) isn't able to export the last 3 of these:  push @items, @common_items;  my $test_body = <<"EOT";my \$test = $dummytest;EOT  $test_body .= <<'EOT';# What follows goes to the temporary file.# IVmy $five = FIVE;if ($five == 5) {  print "ok $test\n";} else {  print "not ok $test # \$five\n";}$test++;# PVif (OK6 eq "ok 6\n") {  print "ok $test\n";} else {  print "not ok $test # \$five\n";}$test++;# PVN containing embedded \0s$_ = OK7;s/.*\0//s;s/7/$test/;$test++;print;# NVmy $farthing = FARTHING;if ($farthing == 0.25) {  print "ok $test\n";} else {  print "not ok $test # $farthing\n";}$test++;# UVmy $not_zero = NOT_ZERO;if ($not_zero > 0 && $not_zero == ~0) {  print "ok $test\n";} else {  print "not ok $test # \$not_zero=$not_zero ~0=" . (~0) . "\n";}$test++;# Value includes a "*/" in an attempt to bust out of a C comment.# Also tests custom cpp #if clausesmy $close = CLOSE;if ($close eq '*/') {  print "ok $test\n";} else {  print "not ok $test # \$close='$close'\n";}$test++;# Default values if macro not defined.my $answer = ANSWER;if ($answer == 42) {  print "ok $test\n";} else {  print "not ok $test # What do you get if you multiply six by nine? '$answer'\n";}$test++;# not defined macromy $notdef = eval { NOTDEF; };if (defined $notdef) {  print "not ok $test # \$notdef='$notdef'\n";} elsif ($@ !~ /Your vendor has not defined ExtTest macro NOTDEF/) {  print "not ok $test # \$@='$@'\n";} else {  print "ok $test\n";}$test++;# not a macromy $notthere = eval { &ExtTest::NOTTHERE; };if (defined $notthere) {  print "not ok $test # \$notthere='$notthere'\n";} elsif ($@ !~ /NOTTHERE is not a valid ExtTest macro/) {  chomp $@;  print "not ok $test # \$@='$@'\n";} else {  print "ok $test\n";}$test++;# Truthmy $yes = Yes;if ($yes) {  print "ok $test\n";} else {  print "not ok $test # $yes='\$yes'\n";}$test++;# Falsehoodmy $no = No;if (defined $no and !$no) {  print "ok $test\n";} else {  print "not ok $test # \$no=" . defined ($no) ? "'$no'\n" : "undef\n";}$test++;# Undefmy $undef = Undef;unless (defined $undef) {  print "ok $test\n";} else {  print "not ok $test # \$undef='$undef'\n";}$test++;# invalid macro (chosen to look like a mix up between No and SW)$notdef = eval { &ExtTest::So };if (defined $notdef) {  print "not ok $test # \$notdef='$notdef'\n";} elsif ($@ !~ /^So is not a valid ExtTest macro/) {  print "not ok $test # \$@='$@'\n";} else {  print "ok $test\n";}$test++;# invalid defined macro$notdef = eval { &ExtTest::EW };if (defined $notdef) {  print "not ok $test # \$notdef='$notdef'\n";} elsif ($@ !~ /^EW is not a valid ExtTest macro/) {  print "not ok $test # \$@='$@'\n";} else {  print "ok $test\n";}$test++;my %compass = (EOTwhile (my ($point, $bearing) = each %compass) {  $test_body .= "'$point' => $bearing, "}$test_body .= <<'EOT';);my $fail;while (my ($point, $bearing) = each %compass) {  my $val = eval $point;  if ($@) {    print "# $point: \$@='$@'\n";    $fail = 1;  } elsif (!defined $bearing) {    print "# $point: \$val=undef\n";    $fail = 1;  } elsif ($val != $bearing) {    print "# $point: \$val=$val, not $bearing\n";    $fail = 1;  }}if ($fail) {  print "not ok $test\n";} else {  print "ok $test\n";}$test++;EOT$test_body .= <<"EOT";my \$rfc1149 = RFC1149;if (\$rfc1149 ne "$parent_rfc1149") {  print "not ok \$test # '\$rfc1149' ne '$parent_rfc1149'\n";} else {  print "ok \$test\n";}\$test++;if (\$rfc1149 != 1149) {  printf "not ok \$test # %d != 1149\n", \$rfc1149;} else {  print "ok \$test\n";}\$test++;EOT$test_body .= <<'EOT';# test macro=>1my $open = OPEN;if ($open eq '/*') {  print "ok $test\n";} else {  print "not ok $test # \$open='$open'\n";}$test++;EOT$dummytest+=18;  end_tests("Simple tests", \@items, \@export_names, $header, $test_body,	    $args);}if ($do_utf_tests) {  # utf8 tests  start_tests();  my ($inf, $pound_bytes, $pound_utf8);  $inf = chr 0x221E;  # Check that we can distiguish the pathological case of a string, and the  # utf8 representation of that string.  $pound_utf8 = $pound . '1';  if ($better_than_56) {    $pound_bytes = $pound_utf8;    utf8::encode ($pound_bytes);  } else {    # Must have that "U*" to generate a zero length UTF string that forces    # top bit set chars (such as the pound sign) into UTF8, so that the    # unpack 'C*' then gets the byte form of the UTF8.    $pound_bytes =  pack 'C*', unpack 'C*', $pound_utf8 . pack "U*";  }  my @items = (@common_items,               {name=>$inf, type=>"PV", value=>'"Infinity"', macro=>1},               {name=>$pound_utf8, type=>"PV", value=>'"1 Pound"', macro=>1},               {name=>$pound_bytes, type=>"PV", value=>'"1 Pound (as bytes)"',                macro=>1},              );=podThe above set of names seems to produce a suitably bad set of compileproblems on a Unicode naive version of ExtUtils::Constant (ie 0.11):nick@thinking-cap 15439-32-utf$ PERL_CORE=1 ./perl lib/ExtUtils/t/Constant.t1..33# perl=/stuff/perl5/15439-32-utf/perl# ext-30370 being created...Wide character in print at lib/ExtUtils/t/Constant.t line 140.ok 1ok 2# make = 'make'ExtTest.xs: In function `constant_1':ExtTest.xs:80: warning: multi-character character constantExtTest.xs:80: warning: case value out of rangeok 3=cut# Grr `  # Do this in 7 bit in case someone is testing with some settings that cause  # 8 bit files incapable of storing this character.  my @values    = map {"'" . join (",", unpack "U*", $_ . pack "U*") . "'"}      ($pound, $inf, $pound_bytes, $pound_utf8);  # Values is a list of strings, such as ('194,163,49', '163,49')  my $test_body .= "my \$test = $dummytest;\n";  $dummytest += 7 * 3; # 3 tests for each of the 7 things:  $test_body .= << 'EOT';use utf8;my $better_than_56 = $] > 5.007;my ($pound, $inf, $pound_bytes, $pound_utf8) = map {eval "pack 'U*', $_"}EOT  $test_body .= join ",", @values;  $test_body .= << 'EOT';;foreach (["perl", "rules", "rules"],	 ["/*", "OPEN", "OPEN"],	 ["*/", "CLOSE", "CLOSE"],	 [$pound, 'Sterling', []],         [$inf, 'Infinity', []],	 [$pound_utf8, '1 Pound', '1 Pound (as bytes)'],	 [$pound_bytes, '1 Pound (as bytes)', []],        ) {  # Flag an expected error with a reference for the expect string.  my ($string, $expect, $expect_bytes) = @$_;  (my $name = $string) =~ s/([^ !"#\$%&'()*+,\-.\/0-9:;<=>?\@A-Z[\\\]^_`a-z{|}~])/sprintf '\x{%X}', ord $1/ges;  print "# \"$name\" => \'$expect\'\n";  # Try to force this to be bytes if possible.  if ($better_than_56) {    utf8::downgrade ($string, 1);  } else {    if ($string =~ tr/0-\377// == length $string) {      # No chars outside range 0-255      $string = pack 'C*', unpack 'U*', ($string . pack 'U*');    }  }EOT  $test_body .=  "my (\$error, \$got) = ${package}::constant (\$string);\n";  $test_body .= <<'EOT';  if ($error or $got ne $expect) {    print "not ok $test # error '$error', got '$got'\n";  } else {    print "ok $test\n";  }  $test++;  print "# Now upgrade '$name' to utf8\n";  if ($better_than_56) {    utf8::upgrade ($string);  } else {    $string = pack ('U*') . $string;  }EOT  $test_body .=  "my (\$error, \$got) = ${package}::constant (\$string);\n";  $test_body .= <<'EOT';  if ($error or $got ne $expect) {    print "not ok $test # error '$error', got '$got'\n";  } else {    print "ok $test\n";  }  $test++;  if (defined $expect_bytes) {    print "# And now with the utf8 byte sequence for name\n";    # Try the encoded bytes.    if ($better_than_56) {      utf8::encode ($string);    } else {      $string = pack 'C*', unpack 'C*', $string . pack "U*";    }EOT    $test_body .= "my (\$error, \$got) = ${package}::constant (\$string);\n";    $test_body .= <<'EOT';    if (ref $expect_bytes) {      # Error expected.      if ($error) {        print "ok $test # error='$error' (as expected)\n";      } else {        print "not ok $test # expected error, got no error and '$got'\n";      }    } elsif ($got ne $expect_bytes) {      print "not ok $test # error '$error', expect '$expect_bytes', got '$got'\n";    } else {      print "ok $test\n";    }    $test++;  }}EOT  end_tests("utf8 tests", \@items, [], "#define perl \"rules\"\n", $test_body);}# XXX I think that I should merge this into the utf8 test above.sub explict_call_constant {  my ($string, $expect) = @_;  # This does assume simple strings suitable for ''  my $test_body = <<"EOT";{  my (\$error, \$got) = ${package}::constant ('$string');\n;EOT  if (defined $expect) {    # No error expected    $test_body .= <<"EOT";  if (\$error or \$got ne "$expect") {    print "not ok $dummytest # error '\$error', expect '$expect', got '\$got'\n";  } else {    print "ok $dummytest\n";    }  }EOT  } else {    # Error expected.    $test_body .= <<"EOT";  if (\$error) {    print "ok $dummytest # error='\$error' (as expected)\n";  } else {    print "not ok $dummytest # expected error, got no error and '\$got'\n";  }EOT  }  $dummytest++;  return $test_body . <<'EOT';}EOT}# Simple tests to verify bits of the switch generation system work.sub simple {  start_tests();  # Deliberately leave $name in @_, so that it is indexed from 1.  my ($name, @items) = @_;  my $test_header;  my $test_body = "my \$value;\n";  foreach my $counter (1 .. $#_) {    my $thisname = $_[$counter];    $test_header .= "#define $thisname $counter\n";    $test_body .= <<"EOT";\$value = $thisname;if (\$value == $counter) {  print "ok $dummytest\n";} else {  print "not ok $dummytest # $thisname gave \$value\n";}EOT    ++$dummytest;    # Yes, the last time round the loop appends a z to the string.    for my $i (0 .. length $thisname) {      my $copyname = $thisname;      substr ($copyname, $i, 1) = 'z';      $test_body .= explict_call_constant ($copyname,                                           $copyname eq $thisname                                             ? $thisname : undef);    }  }  # Ho. This seems to be buggy in 5.005_03:  # # Now remove $name from @_:  # shift @_;  end_tests($name, \@items, \@items, $test_header, $test_body);}# Check that the memeq clauses work correctly when there isn't a switch# statement to bump off a charactersimple ("Singletons", "A", "AB", "ABC", "ABCD", "ABCDE");# Check the three code.simple ("Three start", qw(Bea kea Lea lea nea pea rea sea tea Wea yea Zea));# There were 162 2 letter words in /usr/share/dict/words on FreeBSD 4.6, which# I felt was rather too many. So I used words with 2 vowels.simple ("Twos and three middle", qw(aa ae ai ea eu ie io oe era eta));# Given the choice go for the end, else the earliest pointsimple ("Three end and four symetry", qw(ean ear eat barb marm tart));# Need this if the single test below is rolled into @tests :# --$dummytest;print "1..$dummytest\n";write_and_run_extension @$_ foreach @tests;# This was causing an assertion failure (a C<confess>ion)# Any single byte > 128 should do it.C_constant ($package, undef, undef, undef, undef, undef, chr 255);print "ok $realtest\n"; $realtest++;print STDERR "# You were running with \$keep_files set to $keep_files\n"  if $keep_files;

⌨️ 快捷键说明

复制代码 Ctrl + C
搜索代码 Ctrl + F
全屏模式 F11
切换主题 Ctrl + Shift + D
显示快捷键 ?
增大字号 Ctrl + =
减小字号 Ctrl + -