3 perltrap - Perl traps for the unwary
7 The biggest trap of all is forgetting to C<use warnings> or use the B<-w>
8 switch; see L<perllexwarn> and L<perlrun>. The second biggest trap is not
9 making your entire program runnable under C<use strict>. The third biggest
10 trap is not reading the list of changes in this version of Perl; see
15 Accustomed B<awk> users should take special note of the following:
21 The English module, loaded via
25 allows you to refer to special variables (like C<$/>) with names (like
26 $RS), as though they were in B<awk>; see L<perlvar> for details.
30 Semicolons are required after all simple statements in Perl (except
31 at the end of a block). Newline is not a statement delimiter.
35 Curly brackets are required on C<if>s and C<while>s.
39 Variables begin with "$", "@" or "%" in Perl.
43 Arrays index from 0. Likewise string positions in substr() and
48 You have to decide whether your array has numeric or string indices.
52 Hash values do not spring into existence upon mere reference.
56 You have to decide whether you want to use string or numeric
61 Reading an input line does not split it for you. You get to split it
62 to an array yourself. And the split() operator has different
63 arguments than B<awk>'s.
67 The current input line is normally in $_, not $0. It generally does
68 not have the newline stripped. ($0 is the name of the program
69 executed.) See L<perlvar>.
73 $<I<digit>> does not refer to fields--it refers to substrings matched
74 by the last match pattern.
78 The print() statement does not add field and record separators unless
79 you set C<$,> and C<$\>. You can set $OFS and $ORS if you're using
84 You must open your files before you print to them.
88 The range operator is "..", not comma. The comma operator works as in
93 The match operator is "=~", not "~". ("~" is the one's complement
98 The exponentiation operator is "**", not "^". "^" is the XOR
99 operator, as in C. (You know, one could get the feeling that B<awk> is
100 basically incompatible with C.)
104 The concatenation operator is ".", not the null string. (Using the
105 null string would render C</pat/ /pat/> unparsable, because the third slash
106 would be interpreted as a division operator--the tokenizer is in fact
107 slightly context sensitive for operators like "/", "?", and ">".
108 And in fact, "." itself can be the beginning of a number.)
112 The C<next>, C<exit>, and C<continue> keywords work differently.
117 The following variables work differently:
120 ARGC scalar @ARGV (compare with $#ARGV)
124 FS (whatever you like)
125 NF $#Fld, or some such
137 You cannot set $RS to a pattern, only a string.
141 When in doubt, run the B<awk> construct through B<a2p> and see what it
148 Cerebral C and C++ programmers should take note of the following:
154 Curly brackets are required on C<if>'s and C<while>'s.
158 You must use C<elsif> rather than C<else if>.
162 The C<break> and C<continue> keywords from C become in Perl C<last>
163 and C<next>, respectively. Unlike in C, these do I<not> work within a
164 C<do { } while> construct. See L<perlsyn/"Loop Control">.
168 There's no switch statement. (But it's easy to build one on the fly,
169 see L<perlsyn/"Basic BLOCKs and Switch Statements">)
173 Variables begin with "$", "@" or "%" in Perl.
177 Comments begin with "#", not "/*".
181 You can't take the address of anything, although a similar operator
182 in Perl is the backslash, which creates a reference.
186 C<ARGV> must be capitalized. C<$ARGV[0]> is C's C<argv[1]>, and C<argv[0]>
191 System calls such as link(), unlink(), rename(), etc. return nonzero for
192 success, not 0. (system(), however, returns zero for success.)
196 Signal handlers deal with signal names, not numbers. Use C<kill -l>
197 to find their names on your system.
203 Seasoned B<sed> programmers should take note of the following:
209 Backreferences in substitutions use "$" rather than "\".
213 The pattern matching metacharacters "(", ")", and "|" do not have backslashes
218 The range operator is C<...>, rather than comma.
224 Sharp shell programmers should take note of the following:
230 The backtick operator does variable interpolation without regard to
231 the presence of single quotes in the command.
235 The backtick operator does no translation of the return value, unlike B<csh>.
239 Shells (especially B<csh>) do several levels of substitution on each
240 command line. Perl does substitution in only certain constructs
241 such as double quotes, backticks, angle brackets, and search patterns.
245 Shells interpret scripts a little bit at a time. Perl compiles the
246 entire program before executing it (except for C<BEGIN> blocks, which
247 execute at compile time).
251 The arguments are available via @ARGV, not $1, $2, etc.
255 The environment is not automatically made available as separate scalar
262 Practicing Perl Programmers should take note of the following:
268 Remember that many operations behave differently in a list
269 context than they do in a scalar one. See L<perldata> for details.
273 Avoid barewords if you can, especially all lowercase ones.
274 You can't tell by just looking at it whether a bareword is
275 a function or a string. By using quotes on strings and
276 parentheses on function calls, you won't ever get them confused.
280 You cannot discern from mere inspection which builtins
281 are unary operators (like chop() and chdir())
282 and which are list operators (like print() and unlink()).
283 (Unless prototyped, user-defined subroutines can B<only> be list
284 operators, never unary ones.) See L<perlop> and L<perlsub>.
288 People have a hard time remembering that some functions
289 default to $_, or @ARGV, or whatever, but that others which
290 you might expect to do not.
294 The <FH> construct is not the name of the filehandle, it is a readline
295 operation on that handle. The data read is assigned to $_ only if the
296 file read is the sole condition in a while loop:
299 while (defined($_ = <FH>)) { }..
300 <FH>; # data discarded!
304 Remember not to use C<=> when you need C<=~>;
305 these two constructs are quite different:
312 The C<do {}> construct isn't a real loop that you can use
317 Use C<my()> for local variables whenever you can get away with
318 it (but see L<perlform> for where you can't).
319 Using C<local()> actually gives a local value to a global
320 variable, which leaves you open to unforeseen side-effects
325 If you localize an exported variable in a module, its exported value will
326 not change. The local name becomes an alias to a new value but the
327 external name is still an alias for the original.
331 =head2 Perl4 to Perl5 Traps
333 Practicing Perl4 Programmers should take note of the following
334 Perl4-to-Perl5 specific traps.
336 They're crudely ordered according to the following list:
340 =item Discontinuance, Deprecation, and BugFix traps
342 Anything that's been fixed as a perl4 bug, removed as a perl4 feature
343 or deprecated as a perl4 feature with the intent to encourage usage of
344 some other perl5 feature.
348 Traps that appear to stem from the new parser.
350 =item Numerical Traps
352 Traps having to do with numerical or mathematical operators.
354 =item General data type traps
356 Traps involving perl standard data types.
358 =item Context Traps - scalar, list contexts
360 Traps related to context within lists, scalar statements/declarations.
362 =item Precedence Traps
364 Traps related to the precedence of parsing, evaluation, and execution of
367 =item General Regular Expression Traps using s///, etc.
369 Traps related to the use of pattern matching.
371 =item Subroutine, Signal, Sorting Traps
373 Traps related to the use of signals and signal handlers, general subroutines,
374 and sorting, along with sorting subroutines.
382 Traps specific to the use of C<dbmopen()>, and specific dbm implementations.
384 =item Unclassified Traps
390 If you find an example of a conversion trap that is not listed here,
391 please submit it to <F<perlbug@perl.org>> for inclusion.
392 Also note that at least some of these can be caught with the
393 C<use warnings> pragma or the B<-w> switch.
395 =head2 Discontinuance, Deprecation, and BugFix traps
397 Anything that has been discontinued, deprecated, or fixed as
402 =item * Discontinuance
404 Symbols starting with "_" are no longer forced into package main, except
405 for C<$_> itself (and C<@_>, etc.).
411 print "\$_legacy is ",$_legacy,"\n";
413 # perl4 prints: $_legacy is 1
414 # perl5 prints: $_legacy is
418 Double-colon is now a valid package separator in a variable name. Thus these
419 behave differently in perl4 vs. perl5, because the packages don't exist.
421 $a=1;$b=2;$c=3;$var=4;
423 print "$var::abc::xyz\n";
425 # perl4 prints: 1::2::3 4::abc::xyz
428 Given that C<::> is now the preferred package delimiter, it is debatable
429 whether this should be classed as a bug or not.
430 (The older package delimiter, ' ,is used here)
436 # perl5 prints: Can't find string terminator "'" anywhere before EOF
438 You can avoid this problem, and remain compatible with perl4, if you
439 always explicitly include the package name:
442 print "x=${main'x}\n" ;
444 Also see precedence traps, for parsing C<$:>.
448 The second and third arguments of C<splice()> are now evaluated in scalar
449 context (as the Camel says) rather than list context.
451 sub sub1{return(0,2) } # return a 2-element list
452 sub sub2{ return(1,2,3)} # return a 3-element list
453 @a1 = ("a","b","c","d","e");
454 @a2 = splice(@a1,&sub1,&sub2);
455 print join(' ',@a2),"\n";
458 # perl5 prints: c d e
460 =item * Discontinuance
462 You can't do a C<goto> into a block that is optimized away. Darn.
468 print "Here I is!\n";
471 # perl4 prints: Here I is!
472 # perl5 errors: Can't "goto" into the middle of a foreach loop
474 =item * Discontinuance
476 It is no longer syntactically legal to use whitespace as the name
477 of a variable, or as a delimiter for any kind of quote construct.
482 print "a is $a, b is $b\n";
484 # perl4 prints: a is foo bar, b is baz
485 # perl5 errors: Bareword found where operator expected
487 =item * Discontinuance
489 The archaic while/if BLOCK BLOCK syntax is no longer supported.
498 # perl4 prints: True!
499 # perl5 errors: syntax error at test.pl line 1, near "if {"
503 The C<**> operator now binds more tightly than unary minus.
504 It was documented to work this way before, but didn't.
511 =item * Discontinuance
513 The meaning of C<foreach{}> has changed slightly when it is iterating over a
514 list which is not an array. This used to assign the list to a
515 temporary array, but no longer does so (for efficiency). This means
516 that you'll now be iterating over the actual values, not over copies of
517 the values. Modifications to the loop variable can change the original
520 @list = ('ab','abc','bcd','def');
521 foreach $var (grep(/ab/,@list)){
524 print (join(':',@list));
526 # perl4 prints: ab:abc:bcd:def
527 # perl5 prints: 1:1:bcd:def
529 To retain Perl4 semantics you need to assign your list
530 explicitly to a temporary array and then iterate over that. For
531 example, you might need to change
533 foreach $var (grep(/ab/,@list)){
537 foreach $var (@tmp = grep(/ab/,@list)){
539 Otherwise changing $var will clobber the values of @list. (This most often
540 happens when you use C<$_> for the loop variable, and call subroutines in
541 the loop that don't properly localize C<$_>.)
543 =item * Discontinuance
545 C<split> with no arguments now behaves like C<split ' '> (which doesn't
546 return an initial null field if $_ starts with whitespace), it used to
547 behave like C<split /\s+/> (which does).
550 print join(':', split);
552 # perl4 prints: :hi:mom
553 # perl5 prints: hi:mom
557 Perl 4 would ignore any text which was attached to an B<-e> switch,
558 always taking the code snippet from the following arg. Additionally, it
559 would silently accept an B<-e> switch without a following arg. Both of
560 these behaviors have been fixed.
562 perl -e'print "attached to -e"' 'print "separate arg"'
564 # perl4 prints: separate arg
565 # perl5 prints: attached to -e
570 # perl5 dies: No code specified for -e.
572 =item * Discontinuance
574 In Perl 4 the return value of C<push> was undocumented, but it was
575 actually the last value being pushed onto the target list. In Perl 5
576 the return value of C<push> is documented, but has changed, it is the
577 number of elements in the resulting list.
580 print push(@x, 'first new', 'second new');
582 # perl4 prints: second new
587 Some error messages will be different.
589 =item * Discontinuance
591 In Perl 4, if in list context the delimiters to the first argument of
592 C<split()> were C<??>, the result would be placed in C<@_> as well as
593 being returned. Perl 5 has more respect for your subroutine arguments.
595 =item * Discontinuance
597 Some bugs may have been inadvertently removed. :-)
603 Perl4-to-Perl5 traps from having to do with parsing.
609 Note the space between . and =
611 $string . = "more string";
614 # perl4 prints: more string
615 # perl5 prints: syntax error at - line 1, near ". ="
619 Better parsing in perl 5
623 print("hello, world\n");
625 # perl4 prints: hello, world
626 # perl5 prints: syntax error
630 "if it looks like a function, it is a function" rule.
633 ($foo == 1) ? "is one\n" : "is zero\n";
635 # perl4 prints: is zero
636 # perl5 warns: "Useless use of a constant in void context" if using -w
640 String interpolation of the C<$#array> construct differs when braces
641 are to used around the name.
647 # perl5 fails with syntax error
657 When perl sees C<map {> (or C<grep {>), it has to guess whether the C<{>
658 starts a BLOCK or a hash reference. If it guesses wrong, it will report
659 a syntax error near the C<}> and the missing (or unexpected) comma.
661 Use unary C<+> before C<{> on a hash reference, and unary C<+> applied
662 to the first thing in a BLOCK (after C<{>), for perl to guess right all
663 the time. (See L<perlfunc/map>.)
667 =head2 Numerical Traps
669 Perl4-to-Perl5 traps having to do with numerical operators,
670 operands, or output from same.
676 Formatted output and significant digits. In general, Perl 5
677 tries to be more precise. For example, on a Solaris Sparc:
679 print 7.373504 - 0, "\n";
680 printf "%20.18f\n", 7.373504 - 0;
683 7.3750399999999996141
690 Notice how the first result looks better in Perl 5.
692 Your results may vary, since your floating point formatting routines
693 and even floating point format may be slightly different.
697 This specific item has been deleted. It demonstrated how the auto-increment
698 operator would not catch when a number went over the signed int limit. Fixed
699 in version 5.003_04. But always be wary when using large integers.
706 Assignment of return values from numeric equality tests
707 does not work in perl5 when the test evaluates to false (0).
708 Logical tests now return a null, instead of 0
716 Also see L<"General Regular Expression Traps using s///, etc.">
717 for another example of this new feature...
719 =item * Bitwise string ops
721 When bitwise operators which can operate upon either numbers or
722 strings (C<& | ^ ~>) are given only strings as arguments, perl4 would
723 treat the operands as bitstrings so long as the program contained a call
724 to the C<vec()> function. perl5 treats the string operands as bitstrings.
725 (See L<perlop/Bitwise String Operators> for more details.)
729 $betty = $fred & $barney;
731 # Uncomment the next line to change perl4's behavior
732 # ($dummy) = vec("dummy", 0, 0);
740 # If vec() is used anywhere in the program, both print:
745 =head2 General data type traps
747 Perl4-to-Perl5 traps involving most data-types, and their usage
748 within certain expressions and/or context.
754 Negative array subscripts now count from the end of the array.
756 @a = (1, 2, 3, 4, 5);
757 print "The third element of the array is $a[3] also expressed as $a[-2] \n";
759 # perl4 prints: The third element of the array is 4 also expressed as
760 # perl5 prints: The third element of the array is 4 also expressed as 4
764 Setting C<$#array> lower now discards array elements, and makes them
765 impossible to recover.
768 print "Before: ",join('',@a);
770 print ", After: ",join('',@a);
772 print ", Recovered: ",join('',@a),"\n";
774 # perl4 prints: Before: abcde, After: ab, Recovered: abcd
775 # perl5 prints: Before: abcde, After: ab, Recovered: ab
779 Hashes get defined before use
782 die "scalar \$s defined" if defined($s);
783 die "array \@a defined" if defined(@a);
784 die "hash \%h defined" if defined(%h);
787 # perl5 dies: hash %h defined
789 Perl will now generate a warning when it sees defined(@a) and
794 glob assignment from variable to variable will fail if the assigned
795 variable is localized subsequent to the assignment
797 @a = ("This is Perl 4");
802 # perl4 prints: This is Perl 4
807 Assigning C<undef> to a glob has no effect in Perl 5. In Perl 4
808 it undefines the associated scalar (but may have other side effects
809 including SEGVs). Perl 5 will also warn if C<undef> is assigned to a
810 typeglob. (Note that assigning C<undef> to a typeglob is different
811 than calling the C<undef> function on a typeglob (C<undef *foo>), which
812 has quite a few effects.
819 # perl4 warns: "Use of uninitialized variable" if using -w
821 # perl5 warns: "Undefined value assigned to typeglob" if using -w
823 =item * (Scalar String)
825 Changes in unary negation (of strings)
826 This change effects both the return value and what it
827 does to auto(magic)increment.
834 # perl4 prints: aab : -0 : 1
835 # perl5 prints: aab : -aab : aac
839 perl 4 lets you modify constants:
843 for ($x = 0; $x < 3; $x++) {
847 print "before: $_[0]";
849 print " after: $_[0]\n";
860 # Modification of a read-only value attempted at foo.pl line 12.
865 The behavior is slightly different for:
867 print "$x", defined $x
870 # perl 5: <no output, $x is not called into existence>
872 =item * (Variable Suicide)
874 Variable suicide behavior is more consistent under Perl 5.
875 Perl5 exhibits the same behavior for hashes and scalars,
876 that perl4 exhibits for only scalars.
878 $aGlobal{ "aKey" } = "global value";
879 print "MAIN:", $aGlobal{"aKey"}, "\n";
884 local( *theArgument ) = @_;
885 local( %aNewLocal ); # perl 4 != 5.001l,m
886 $aNewLocal{"aKey"} = "this should never appear";
887 print "SUB: ", $theArgument{"aKey"}, "\n";
888 $aNewLocal{"aKey"} = "level $GlobalLevel"; # what should print
890 if( $GlobalLevel<4 ) {
905 # SUB: this should never appear
906 # SUB: this should never appear
907 # SUB: this should never appear
911 =head2 Context Traps - scalar, list contexts
915 =item * (list context)
917 The elements of argument lists for formats are now evaluated in list
918 context. This means you can interpolate list values now.
920 @fmt = ("foo","bar","baz");
927 # perl4 errors: Please use commas to separate fields in file
928 # perl5 prints: foo bar baz
930 =item * (scalar context)
932 The C<caller()> function now returns a false value in a scalar context
933 if there is no caller. This lets library files determine if they're
936 caller() ? (print "You rang?\n") : (print "Got a 0\n");
938 # perl4 errors: There is no caller
939 # perl5 prints: Got a 0
941 =item * (scalar context)
943 The comma operator in a scalar context is now guaranteed to give a
944 scalar context to its arguments.
950 # Perl4 prints: x = c # Thinks list context interpolates list
951 # Perl5 prints: x = 3 # Knows scalar uses length of list
953 =item * (list, builtin)
955 C<sprintf()> is prototyped as ($;@), so its first argument is given scalar
956 context. Thus, if passed an array, it will probably not do what you want,
959 @z = ('%s%s', 'foo', 'bar');
963 # perl4 prints: foobar
966 C<printf()> works the same as it did in Perl 4, though:
968 @z = ('%s%s', 'foo', 'bar');
971 # perl4 prints: foobar
972 # perl5 prints: foobar
976 =head2 Precedence Traps
978 Perl4-to-Perl5 traps involving precedence order.
980 Perl 4 has almost the same precedence rules as Perl 5 for the operators
981 that they both have. Perl 4 however, seems to have had some
982 inconsistencies that made the behavior differ from what was documented.
988 LHS vs. RHS of any assignment operator. LHS is evaluated first
989 in perl4, second in perl5; this can affect the relationship
990 between side-effects in sub-expressions.
992 @arr = ( 'left', 'right' );
993 $a{shift @arr} = shift @arr;
994 print join( ' ', keys %a );
997 # perl5 prints: right
1001 These are now semantic errors because of precedence:
1003 @list = (1,2,3,4,5);
1004 %map = ("a",1,"b",2,"c",3,"d",4);
1005 $n = shift @list + 2; # first item in list plus 2
1007 $m = keys %map + 2; # number of items in hash plus 2
1010 # perl4 prints: n is 3, m is 6
1011 # perl5 errors and fails to compile
1015 The precedence of assignment operators is now the same as the precedence
1016 of assignment. Perl 4 mistakenly gave them the precedence of the associated
1017 operator. So you now must parenthesize them in expressions like
1019 /foo/ ? ($a += 2) : ($a -= 2);
1023 /foo/ ? $a += 2 : $a -= 2
1025 would be erroneously parsed as
1027 (/foo/ ? $a += 2 : $a) -= 2;
1031 $a += /foo/ ? 1 : 2;
1033 now works as a C programmer would expect.
1039 is now incorrect. You need parentheses around the filehandle.
1040 Otherwise, perl5 leaves the statement as its default precedence:
1044 # perl4 opens or dies
1045 # perl5 opens FOO, dying only if 'FOO' is false, i.e. never
1049 perl4 gives the special variable, C<$:> precedence, where perl5
1050 treats C<$::> as main C<package>
1052 $a = "x"; print "$::a";
1054 # perl 4 prints: -:a
1059 perl4 had buggy precedence for the file test operators vis-a-vis
1060 the assignment operators. Thus, although the precedence table
1061 for perl4 leads one to believe C<-e $foo .= "q"> should parse as
1062 C<((-e $foo) .= "q")>, it actually parses as C<(-e ($foo .= "q"))>.
1063 In perl5, the precedence is as documented.
1067 # perl4 prints: no output
1068 # perl5 prints: Can't modify -e in concatenation
1072 In perl4, keys(), each() and values() were special high-precedence operators
1073 that operated on a single hash, but in perl5, they are regular named unary
1074 operators. As documented, named unary operators have lower precedence
1075 than the arithmetic and concatenation operators C<+ - .>, but the perl4
1076 variants of these operators actually bind tighter than C<+ - .>.
1083 # perl5 prints: Type of arg 1 to keys must be hash (not subtraction)
1085 The perl4 behavior was probably more useful, if less consistent.
1089 =head2 General Regular Expression Traps using s///, etc.
1091 All types of RE traps.
1095 =item * Regular Expression
1097 C<s'$lhs'$rhs'> now does no interpolation on either side. It used to
1098 interpolate $lhs but not $rhs. (And still does not match a literal
1102 $string = '1 2 $a $b';
1103 $string =~ s'$a'$b';
1106 # perl4 prints: $b 2 $a $b
1107 # perl5 prints: 1 2 $a $b
1109 =item * Regular Expression
1111 C<m//g> now attaches its state to the searched string rather than the
1112 regular expression. (Once the scope of a block is left for the sub, the
1113 state of the searched string is lost)
1119 sub doit{local($_) = shift; print "Got $_ "}
1121 # perl4 prints: Got blah Got blah Got blah Got blah
1122 # perl5 prints: infinite loop blah...
1124 =item * Regular Expression
1126 Currently, if you use the C<m//o> qualifier on a regular expression
1127 within an anonymous sub, I<all> closures generated from that anonymous
1128 sub will use the regular expression as it was compiled when it was used
1129 the very first time in any such closure. For instance, if you say
1132 my($left,$right) = @_;
1133 return sub { $_[0] =~ /$left stuff $right/o; };
1135 $good = build_match('foo','bar');
1136 $bad = build_match('baz','blarch');
1137 print $good->('foo stuff bar') ? "ok\n" : "not ok\n";
1138 print $bad->('baz stuff blarch') ? "ok\n" : "not ok\n";
1139 print $bad->('foo stuff bar') ? "not ok\n" : "ok\n";
1141 For most builds of Perl5, this will print:
1146 build_match() will always return a sub which matches the contents of
1147 $left and $right as they were the I<first> time that build_match()
1148 was called, not as they are in the current call.
1150 =item * Regular Expression
1152 If no parentheses are used in a match, Perl4 sets C<$+> to
1153 the whole match, just like C<$&>. Perl5 does not.
1158 # perl4 prints: bcde
1161 =item * Regular Expression
1163 substitution now returns the null string if it fails
1166 $value = ($string =~ s/foo//);
1172 Also see L<Numerical Traps> for another example of this new feature.
1174 =item * Regular Expression
1176 C<s`lhs`rhs`> (using backticks) is now a normal substitution, with no
1180 $string =~ s`^`hostname`;
1181 print $string, "\n";
1183 # perl4 prints: <the local hostname>
1184 # perl5 prints: hostname
1186 =item * Regular Expression
1188 Stricter parsing of variables used in regular expressions
1190 s/^([^$grpc]*$grpc[$opt$plus$rep]?)//o;
1192 # perl4: compiles w/o error
1193 # perl5: with Scalar found where operator expected ..., near "$opt$plus"
1195 an added component of this example, apparently from the same script, is
1196 the actual value of the s'd string after the substitution.
1197 C<[$opt]> is a character class in perl4 and an array subscript in perl5
1202 s/^([^$grpc]*$grpc[$opt]?)/foo/;
1206 # perl5 prints: foobar
1208 =item * Regular Expression
1210 Under perl5, C<m?x?> matches only once, like C<?x?>. Under perl4, it matched
1211 repeatedly, like C</x/> or C<m!x!>.
1214 sub match { $test =~ m?once?; }
1217 # m?x? matches more then once
1220 # m?x? matches only once
1224 # perl4 prints: perl4
1225 # perl5 prints: perl5
1227 =item * Regular Expression
1229 Unlike in Ruby, failed matches in Perl do not reset the match variables
1230 ($1, $2, ..., C<$`>, ...).
1234 =head2 Subroutine, Signal, Sorting Traps
1236 The general group of Perl4-to-Perl5 traps having to do with
1237 Signals, Sorting, and their related subroutines, as well as
1238 general subroutine traps. Includes some OS-Specific traps.
1244 Barewords that used to look like strings to Perl will now look like subroutine
1245 calls if a subroutine by that name is defined before the compiler sees them.
1247 sub SeeYa { warn"Hasta la vista, baby!" }
1248 $SIG{'TERM'} = SeeYa;
1249 print "SIGTERM is now $SIG{'TERM'}\n";
1251 # perl4 prints: SIGTERM is now main'SeeYa
1252 # perl5 prints: SIGTERM is now main::1 (and warns "Hasta la vista, baby!")
1254 Use B<-w> to catch this one
1256 =item * (Sort Subroutine)
1258 reverse is no longer allowed as the name of a sort subroutine.
1260 sub reverse{ print "yup "; $a <=> $b }
1261 print sort reverse (2,1,3);
1263 # perl4 prints: yup yup 123
1265 # perl5 warns (if using -w): Ambiguous call resolved as CORE::reverse()
1267 =item * warn() won't let you specify a filehandle.
1269 Although it _always_ printed to STDERR, warn() would let you specify a
1270 filehandle in perl4. With perl5 it does not.
1274 # perl4 prints: Foo!
1275 # perl5 prints: String found where operator expected
1285 Under HPUX, and some other SysV OSes, one had to reset any signal handler,
1286 within the signal handler function, each time a signal was handled with
1287 perl4. With perl5, the reset is now done correctly. Any code relying
1288 on the handler _not_ being reset will have to be reworked.
1290 Since version 5.002, Perl uses sigaction() under SysV.
1295 $SIG{'INT'} = 'gotit';
1304 while (1) {sleep(10);}
1307 # perl4 (HPUX) prints: Got INT...
1308 # perl5 (HPUX) prints: Got INT... Got INT...
1312 Under SysV OSes, C<seek()> on a file opened to append C<<< >> >>> now does
1313 the right thing w.r.t. the fopen() manpage. e.g., - When a file is opened
1314 for append, it is impossible to overwrite information already in
1317 open(TEST,">>seek.test");
1318 $start = tell TEST ;
1323 seek(TEST,$start,0);
1324 print TEST "18 characters here";
1326 # perl4 (solaris) seek.test has: 18 characters here
1327 # perl5 (solaris) seek.test has: 1 2 3 4 5 6 7 8 9 18 characters here
1333 =head2 Interpolation Traps
1335 Perl4-to-Perl5 traps having to do with how things get interpolated
1336 within certain expressions, statements, contexts, or whatever.
1340 =item * Interpolation
1342 @ now always interpolates an array in double-quotish strings.
1344 print "To: someone@somewhere.com\n";
1346 # perl4 prints: To:someone@somewhere.com
1347 # perl < 5.6.1, error : In string, @somewhere now must be written as \@somewhere
1348 # perl >= 5.6.1, warning : Possible unintended interpolation of @somewhere in string
1350 =item * Interpolation
1352 Double-quoted strings may no longer end with an unescaped $.
1355 print "foo is $foo\n";
1357 # perl4 prints: foo is foo$
1358 # perl5 errors: Final $ should be \$ or $name
1360 Note: perl5 DOES NOT error on the terminating @ in $bar
1362 =item * Interpolation
1364 Perl now sometimes evaluates arbitrary expressions inside braces that occur
1365 within double quotes (usually when the opening brace is preceded by C<$>
1371 sub foo { return "bar" };
1372 print "|@{w.w.w}|${main'foo}|";
1374 # perl4 prints: |@{w.w.w}|foo|
1375 # perl5 prints: |buz|bar|
1377 Note that you can C<use strict;> to ward off such trappiness under perl5.
1379 =item * Interpolation
1381 The construct "this is $$x" used to interpolate the pid at that point, but
1382 now tries to dereference $x. C<$$> by itself still works fine, however.
1386 print "this is $$x\n";
1388 # perl4 prints: this is XXXx (XXX is the current pid)
1389 # perl5 prints: this is a reference
1391 =item * Interpolation
1393 Creation of hashes on the fly with C<eval "EXPR"> now requires either both
1394 C<$>'s to be protected in the specification of the hash name, or both curlies
1395 to be protected. If both curlies are protected, the result will be compatible
1396 with perl4 and perl5. This is a very common practice, and should be changed
1397 to use the block form of C<eval{}> if possible.
1399 $hashname = "foobar";
1402 eval "\$$hashname{'$key'} = q|$value|";
1403 (defined($foobar{'baz'})) ? (print "Yup") : (print "Nope");
1406 # perl5 prints: Nope
1410 eval "\$$hashname{'$key'} = q|$value|";
1414 eval "\$\$hashname{'$key'} = q|$value|";
1416 causes the following result:
1418 # perl4 prints: Nope
1423 eval "\$$hashname\{'$key'\} = q|$value|";
1425 causes the following result:
1429 # and is compatible for both versions
1432 =item * Interpolation
1434 perl4 programs which unconsciously rely on the bugs in earlier perl versions.
1436 perl -e '$bar=q/not/; print "This is $foo{$bar} perl5"'
1438 # perl4 prints: This is not perl5
1439 # perl5 prints: This is perl5
1441 =item * Interpolation
1443 You also have to be careful about array and hash brackets during
1449 perl 5 prints: syntax error
1454 perl 5 prints: syntax error
1456 Perl 5 is expecting to find an index or key name following the respective
1457 brackets, as well as an ending bracket of the appropriate type. In order
1458 to mimic the behavior of Perl 4, you must escape the bracket like so.
1463 =item * Interpolation
1465 Similarly, watch out for:
1468 print "\$$foo{bar}\n";
1470 # perl4 prints: $baz{bar}
1473 Perl 5 is looking for C<$foo{bar}> which doesn't exist, but perl 4 is
1474 happy just to expand $foo to "baz" by itself. Watch out for this
1475 especially in C<eval>'s.
1477 =item * Interpolation
1479 C<qq()> string passed to C<eval>
1482 foreach \$y (keys %\$x\) {
1487 # perl4 runs this ok
1488 # perl5 prints: Can't find string terminator ")"
1500 Existing dbm databases created under perl4 (or any other dbm/ndbm tool)
1501 may cause the same script, run under perl5, to fail. The build of perl5
1502 must have been linked with the same dbm/ndbm as the default for C<dbmopen()>
1503 to function properly without C<tie>'ing to an extension dbm implementation.
1505 dbmopen (%dbm, "file", undef);
1509 # perl5 prints: ok (IFF linked with -ldbm or -lndbm)
1514 Existing dbm databases created under perl4 (or any other dbm/ndbm tool)
1515 may cause the same script, run under perl5, to fail. The error generated
1516 when exceeding the limit on the key/value size will cause perl5 to exit
1519 dbmopen(DB, "testdb",0600) || die "couldn't open db! $!";
1520 $DB{'trap'} = "x" x 1024; # value too large for most dbm/ndbm
1524 dbm store returned -1, errno 28, key "trap" at - line 3.
1528 dbm store returned -1, errno 28, key "trap" at - line 3.
1532 =head2 Unclassified Traps
1538 =item * C<require>/C<do> trap using returned value
1540 If the file doit.pl has:
1548 And the do.pl file has the following single line:
1552 Running doit.pl gives the following:
1554 # perl 4 prints: 3 (aborts the subroutine early)
1557 Same behavior if you replace C<do> with C<require>.
1559 =item * C<split> on empty string with LIMIT specified
1562 @list = split(/foo/, $string, 2)
1564 Perl4 returns a one element list containing the empty string but Perl5
1565 returns an empty list.
1569 As always, if any of these are ever officially declared as bugs,
1570 they'll be fixed and removed.