use Carp 'cluck', 'croak';
use B qw(class main_root main_start main_cv svref_2object opnumber cstring
OPf_WANT OPf_WANT_VOID OPf_WANT_SCALAR OPf_WANT_LIST
- OPf_KIDS OPf_REF OPf_STACKED OPf_SPECIAL
+ OPf_KIDS OPf_REF OPf_STACKED OPf_SPECIAL OPf_MOD
OPpLVAL_INTRO OPpOUR_INTRO OPpENTERSUB_AMPER OPpSLICE OPpCONST_BARE
OPpTRANS_SQUASH OPpTRANS_DELETE OPpTRANS_COMPLEMENT OPpTARGET_MY
OPpCONST_ARYBASE OPpEXISTS_SUB OPpSORT_NUMERIC OPpSORT_INTEGER
CVf_METHOD CVf_LOCKED CVf_LVALUE
PMf_KEEP PMf_GLOBAL PMf_CONTINUE PMf_EVAL PMf_ONCE PMf_SKIPWHITE
PMf_MULTILINE PMf_SINGLELINE PMf_FOLD PMf_EXTENDED);
-$VERSION = 0.60;
+$VERSION = 0.62;
use strict;
use warnings ();
# - package declarations using cop_stash
# - subs, formats and code sorted by cop_seq
# Changes between 0.51 and 0.52:
-# - added pp_threadsv (special variables under USE_THREADS)
+# - added pp_threadsv (special variables under USE_5005THREADS)
# - added documentation
# Changes between 0.52 and 0.53:
# - many changes adding precedence contexts and associativity
# - separate recognition of constant subs
# - rewrote continue block handling, now recoginizing for loops
# - added more control of expanding control structures
+# Changes between 0.60 and 0.61 (mostly by Robin Houston)
+# - many bug-fixes
+# - support for pragmas and 'use'
+# - support for the little-used $[ variable
+# - support for __DATA__ sections
+# - UTF8 support
+# - BEGIN, CHECK, INIT and END blocks
+# - scoping of subroutine declarations fixed
+# - compile-time output from the input program can be suppressed, so that the
+# output is just the deparsed code. (a change to O.pm in fact)
+# - our() declarations
+# - *all* the known bugs are now listed in the BUGS section
+# - comprehensive test mechanism (TEST -deparse)
# Todo:
# (See also BUGS section at the end of this file)
return $self;
}
-sub WARN_MASK () {
- # Mask out the bits that C<use vars> uses
- $warnings::Bits{all} | $warnings::DeadBits{all};
+{
+ # Mask out the bits that L<warnings::register> uses
+ my $WARN_MASK;
+ BEGIN {
+ $WARN_MASK = $warnings::Bits{all} | $warnings::DeadBits{all};
+ }
+ sub WARN_MASK () {
+ return $WARN_MASK;
+ }
}
# Initialise the contextual information, either from
my(@args) = @_;
return sub {
my $self = B::Deparse->new(@args);
+ # First deparse command-line args
+ if (defined $^I) { # deparse -i
+ print q(BEGIN { $^I = ).cstring($^I).qq(; }\n);
+ }
+ if ($^W) { # deparse -w
+ print qq(BEGIN { \$^W = $^W; }\n);
+ }
+ if ($/ ne "\n" or defined $O::savebackslash) { # deparse -l and -0
+ my $fs = cstring($/) || 'undef';
+ my $bs = cstring($O::savebackslash) || 'undef';
+ print qq(BEGIN { \$/ = $fs; \$\\ = $bs; }\n);
+ }
my @BEGINs = B::begin_av->isa("B::AV") ? B::begin_av->ARRAY : ();
my @INITs = B::init_av->isa("B::AV") ? B::init_av->ARRAY : ();
my @ENDs = B::end_av->isa("B::AV") ? B::end_av->ARRAY : ();
elsif ($name eq 're') {
require re;
if ($val eq 'none') {
- $hint_bits &= ~re::bits(qw/taint eval asciirange/);
+ $hint_bits &= ~re::bits(qw/taint eval/);
next();
}
my @names;
if ($val eq 'all') {
- @names = qw/taint eval asciirange/;
+ @names = qw/taint eval/;
}
elsif (ref $val) {
@names = @$val;
elsif ($name eq 'warnings') {
if ($val eq 'none') {
- $warning_bits = "\0"x12;
+ $warning_bits = $warnings::NONE;
next();
}
@names = split/\s+/, $val;
}
- $warning_bits = "\0"x12 if !defined ($warning_bits);
+ $warning_bits = $warnings::NONE if !defined ($warning_bits);
$warning_bits |= warnings::bits(@names);
}
my @text;
local($self->{'curcv'}) = $form;
local($self->{'curcvlex'});
+ local($self->{'in_format'}) = 1;
local(@$self{qw'curstash warnings hints'})
- = @$self{'curstash warnings hints'};
+ = @$self{qw'curstash warnings hints'};
my $op = $form->ROOT;
my $kid;
+ return "\f." if $op->first->name eq 'stub';
$op = $op->first->first; # skip leavewrite, lineseq
while (not null $op) {
$op = $op->sibling; # skip nextstate
return "XXX";
}
+sub pp_method_named {
+ cluck "unexpected OP_METHOD_NAMED";
+ return "XXX";
+}
+
sub pp_flip { # see also flop
cluck "unexpected OP_FLIP";
return "XXX";
}
my $body = join(";\n", grep {length} @exprs);
my $subs = "";
- if (defined $root && defined $limit_seq) {
+ if (defined $root && defined $limit_seq && !$self->{'in_format'}) {
$subs = join "\n", $self->seq_subs($limit_seq);
}
return join(";\n", grep {length} $body, $subs);
my $warnings = $op->warnings;
my $warning_bits;
if ($warnings->isa("B::SPECIAL") && $$warnings == 4) {
- $warning_bits = $warnings::Bits{"all"};
+ $warning_bits = $warnings::Bits{"all"} & WARN_MASK;
}
elsif ($warnings->isa("B::SPECIAL") && $$warnings == 5) {
- $warning_bits = "\0"x12;
+ $warning_bits = $warnings::NONE;
}
elsif ($warnings->isa("B::SPECIAL")) {
$warning_bits = undef;
($left, $right) = ($right, $left);
}
$left = $self->deparse_binop_left($op, $left, $prec);
- $left = "($left)" if $flags & LIST_CONTEXT && $left =~ /^\$/;
+ $left = "($left)" if $flags & LIST_CONTEXT
+ && $left !~ /^(my|our|local|)[\@\(]/;
$right = $self->deparse_binop_right($op, $right, $prec);
return $self->maybe_parens("$left $opname$eq $right", $cx, $prec);
}
BEGIN { eval "sub OP_CONST () {" . opnumber("const") . "}" }
BEGIN { eval "sub OP_STRINGIFY () {" . opnumber("stringify") . "}" }
+BEGIN { eval "sub OP_RV2SV () {" . opnumber("rv2sv") . "}" }
+BEGIN { eval "sub OP_LIST () {" . opnumber("list") . "}" }
sub pp_null {
my $self = shift;
sub rv2x {
my $self = shift;
my($op, $cx, $type) = @_;
+
+ if (class($op) eq 'NULL' || !$op->can("first")) {
+ Carp::cluck("Unexpected op in pp_rv2x");
+ return 'XXX';
+ }
my $kid = $op->first;
my $str = $self->deparse($kid, 0);
return $self->stash_variable($type, $str) if is_scalar($kid);
}
# skip down to the old, ex-rv2cv
-sub pp_rv2cv { $_[0]->rv2x($_[1]->first->first->sibling, $_[2], "&") }
+sub pp_rv2cv {
+ my ($self, $op, $cx) = @_;
+ if (!null($op->first) && $op->first->name eq 'null' &&
+ $op->first->targ eq OP_LIST)
+ {
+ return $self->rv2x($op->first->first->sibling, $cx, "&")
+ }
+ else {
+ return $self->rv2x($op, $cx, "")
+ }
+}
sub pp_rv2av {
my $self = shift;
#
$idx =~ s/^\((.*)\)$/$1/ if $self->{'parens'};
+ # Hash-element braces will autoquote a bareword inside themselves.
+ # We need to make sure that C<$hash{warn()}> doesn't come out as
+ # C<$hash{warn}>, which has a quite different meaning. Currently
+ # B::Deparse will always quote strings, even if the string was a
+ # bareword in the original (i.e. the OPpCONST_BARE flag is ignored
+ # for constant strings.) So we can cheat slightly here - if we see
+ # a bareword, we know that it is supposed to be a function call.
+ #
+ $idx =~ s/^([A-Za-z_]\w*)$/$1()/;
+
return "\$" . $array . $left . $idx . $right;
}
} else {
$obj = $kid;
$kid = $kid->sibling;
- for (; not null $kid->sibling; $kid = $kid->sibling) {
+ for (; !null ($kid->sibling) && $kid->name ne "method_named";
+ $kid = $kid->sibling) {
push @exprs, $self->deparse($kid, 6);
}
$meth = $kid;
}
my $args = join(", ", @exprs);
$kid = $obj . "->" . $meth;
- if ($args) {
+ if (length $args) {
return $kid . "(" . $args . ")"; # parens mandatory
} else {
return $kid;
my $prefix = "";
my $amper = "";
my($kid, @exprs);
- if ($op->flags & OPf_SPECIAL) {
+ if ($op->flags & OPf_SPECIAL && !($op->flags & OPf_MOD)) {
$prefix = "do ";
} elsif ($op->private & OPpENTERSUB_AMPER) {
$amper = "&";
}
$simple = 1; # only calls of named functions can be prototyped
$kid = $self->deparse($kid, 24);
- } elsif (is_scalar $kid->first) {
+ } elsif (is_scalar ($kid->first) && $kid->first->name ne 'rv2cv') {
$amper = "&";
$kid = $self->deparse($kid, 24);
} else {
}
} else {
# glob() invocations can be translated into calls of
- # CORE::GLOBAL::glob with an second parameter, a number.
+ # CORE::GLOBAL::glob with a second parameter, a number.
# Reverse this.
if ($kid eq "CORE::GLOBAL::glob") {
$kid = "glob";
return $str;
}
-# the same, but treat $|, $), $( and $ at the end of the string differently
-sub re_uninterp {
- my($str) = @_;
-
+{
+my $bal;
+BEGIN {
use re "eval";
# Matches any string which is balanced with respect to {braces}
- my $bal = qr(
+ $bal = qr(
(?:
[^\\{}]
| \\\\
| \{(??{$bal})\}
)*
)x;
+}
+
+# the same, but treat $|, $), $( and $ at the end of the string differently
+sub re_uninterp {
+ my($str) = @_;
$str =~ s/
( ^|\G # $1
return $str;
}
+# This is for regular expressions with the /x modifier
+# We have to leave comments unmangled.
+sub re_uninterp_extended {
+ my($str) = @_;
+
+ $str =~ s/
+ ( ^|\G # $1
+ | [^\\]
+ )
+
+ ( # $2
+ (?:\\\\)*
+ )
+
+ ( # $3
+ ( \(\?\??\{$bal\}\) # $4 (skip over (?{}) and (??{}) blocks)
+ | \#[^\n]* # (skip over comments)
+ )
+ | [\$\@]
+ (?!\||\)|\(|$)
+ | \\[uUlLQE]
+ )
+
+ /length($4) ? "$1$2$4" : "$1$2\\$3"/xeg;
+
+ return $str;
+}
+}
+
# character escapes, but not delimiters that might need to be escaped
sub escape_str { # ASCII, UTF8
my($str) = @_;
return $str;
}
+# For regexes with the /x modifier.
+# Leave whitespace unmangled.
+sub escape_extended_re {
+ my($str) = @_;
+ $str =~ s/(.)/ord($1)>255 ? sprintf("\\x{%x}", ord($1)) : $1/eg;
+ $str =~ s/([\0\033-\037\177-\377])/'\\' . sprintf("%03o", ord($1))/ge;
+ $str =~ s/\n/\n\f/g;
+ return $str;
+}
+
# Don't do this for regexen
sub unback {
my($str) = @_;
} elsif ($sv->FLAGS & SVf_IOK) {
return $sv->int_value;
} elsif ($sv->FLAGS & SVf_NOK) {
- return $sv->NV;
+ # try the default stringification
+ my $r = "".$sv->NV;
+ if ($r =~ /e/) {
+ # If it's in scientific notation, we might have lost information
+ return sprintf("%.20e", $sv->NV);
+ }
+ return $r;
} elsif ($sv->FLAGS & SVf_ROK && $sv->can("RV")) {
return "\\(" . const($sv->RV) . ")"; # constant folded
} elsif ($sv->FLAGS & SVf_POK) {
} elsif ($type eq "concat") {
my $first = $self->dq($op->first);
my $last = $self->dq($op->last);
+
# Disambiguate "${foo}bar", "${foo}{bar}", "${foo}[1]"
- if ($last =~ /^[A-Z\\\^\[\]_?]/) {
- $first =~ s/([\$@])\^$/${1}{^}/; # "${^}W" etc
- }
- elsif ($last =~ /^[{\[\w]/) {
- $first =~ s/([\$@])([A-Za-z_]\w*)$/${1}{$2}/;
- }
+ ($last =~ /^[A-Z\\\^\[\]_?]/ &&
+ $first =~ s/([\$@])\^$/${1}{^}/) # "${^}W" etc
+ || ($last =~ /^[{\[\w_]/ &&
+ $first =~ s/([\$@])([A-Za-z_]\w*)$/${1}{$2}/);
+
return $first . $last;
} elsif ($type eq "uc") {
return '\U' . $self->dq($op->first->sibling) . '\E';
sub tr_decode_byte {
my($table, $flags) = @_;
my(@table) = unpack("s*", $table);
- splice @table, 0x100, 1; # Just flags presence of element 0x101
+ splice @table, 0x100, 1; # Number of subsequent elements
my($c, $tr, @from, @to, @delfrom, $delhyphen);
if ($table[ord "-"] != -1 and
$table[ord("-") - 1] == -1 || $table[ord("-") + 1] == -1)
# Like dq(), but different
sub re_dq {
my $self = shift;
- my $op = shift;
+ my ($op, $extended) = @_;
+
my $type = $op->name;
if ($type eq "const") {
return '$[' if $op->private & OPpCONST_ARYBASE;
- return re_uninterp(escape_str(re_unback($self->const_sv($op)->as_string)));
+ my $unbacked = re_unback($self->const_sv($op)->as_string);
+ return re_uninterp_extended(escape_extended_re($unbacked))
+ if $extended;
+ return re_uninterp(escape_str($unbacked));
} elsif ($type eq "concat") {
- my $first = $self->re_dq($op->first);
- my $last = $self->re_dq($op->last);
+ my $first = $self->re_dq($op->first, $extended);
+ my $last = $self->re_dq($op->last, $extended);
+
# Disambiguate "${foo}bar", "${foo}{bar}", "${foo}[1]"
- if ($last =~ /^[A-Z\\\^\[\]_?]/) {
- $first =~ s/([\$@])\^$/${1}{^}/;
- }
- elsif ($last =~ /^[{\[\w]/) {
- $first =~ s/([\$@])([A-Za-z_]\w*)$/${1}{$2}/;
- }
+ ($last =~ /^[A-Z\\\^\[\]_?]/ &&
+ $first =~ s/([\$@])\^$/${1}{^}/) # "${^}W" etc
+ || ($last =~ /^[{\[\w_]/ &&
+ $first =~ s/([\$@])([A-Za-z_]\w*)$/${1}{$2}/);
+
return $first . $last;
} elsif ($type eq "uc") {
- return '\U' . $self->re_dq($op->first->sibling) . '\E';
+ return '\U' . $self->re_dq($op->first->sibling, $extended) . '\E';
} elsif ($type eq "lc") {
- return '\L' . $self->re_dq($op->first->sibling) . '\E';
+ return '\L' . $self->re_dq($op->first->sibling, $extended) . '\E';
} elsif ($type eq "ucfirst") {
- return '\u' . $self->re_dq($op->first->sibling);
+ return '\u' . $self->re_dq($op->first->sibling, $extended);
} elsif ($type eq "lcfirst") {
- return '\l' . $self->re_dq($op->first->sibling);
+ return '\l' . $self->re_dq($op->first->sibling, $extended);
} elsif ($type eq "quotemeta") {
- return '\Q' . $self->re_dq($op->first->sibling) . '\E';
+ return '\Q' . $self->re_dq($op->first->sibling, $extended) . '\E';
} elsif ($type eq "join") {
return $self->deparse($op->last, 26); # was join($", @ary)
} else {
}
}
-sub pp_regcomp {
+sub pure_string {
+ my ($self, $op) = @_;
+ my $type = $op->name;
+
+ if ($type eq 'const') {
+ return 1;
+ }
+ elsif ($type =~ /^[ul]c(first)?$/ || $type eq 'quotemeta') {
+ return $self->pure_string($op->first->sibling);
+ }
+ elsif ($type eq 'join') {
+ my $join_op = $op->first->sibling; # Skip pushmark
+ return 0 unless $join_op->name eq 'null' && $join_op->targ eq OP_RV2SV;
+
+ my $gvop = $join_op->first;
+ return 0 unless $gvop->name eq 'gvsv';
+ return 0 unless '"' eq $self->gv_name($self->gv_or_padgv($gvop));
+
+ return 0 unless ${$join_op->sibling} eq ${$op->last};
+ return 0 unless $op->last->name =~ /^(rv2|pad)av$/;
+ }
+ elsif ($type eq 'concat') {
+ return $self->pure_string($op->first)
+ && $self->pure_string($op->last);
+ }
+ elsif (is_scalar($op) || $type =~ /^[ah]elem(fast)?$/) {
+ return 1;
+ }
+ else {
+ return 0;
+ }
+
+ return 1;
+}
+
+sub regcomp {
my $self = shift;
- my($op, $cx) = @_;
+ my($op, $cx, $extended) = @_;
my $kid = $op->first;
$kid = $kid->first if $kid->name eq "regcmaybe";
$kid = $kid->first if $kid->name eq "regcreset";
- return $self->re_dq($kid);
+ return ($self->re_dq($kid, $extended), 1) if $self->pure_string($kid);
+ return ($self->deparse($kid, $cx), 0);
+}
+
+sub pp_regcomp {
+ my ($self, $op, $cx) = @_;
+ return (($self->regcomp($op, $cx, 0))[0]);
}
# osmic acid -- see osmium tetroxide
$var = $self->deparse($kid, 20);
$kid = $kid->sibling;
}
+ my $quote = 1;
+ my $extended = ($op->pmflags & PMf_EXTENDED);
if (null $kid) {
- $re = re_unback($op->precomp);
- $re = re_uninterp(escape_str($re)) unless $op->pmflags & PMf_EXTENDED;
+ my $unbacked = re_unback($op->precomp);
+ if ($extended) {
+ $re = re_uninterp_extended(escape_extended_re($unbacked));
+ } else {
+ $re = re_uninterp(escape_str(re_unback($op->precomp)));
+ }
+ } elsif ($kid->name ne 'regcomp') {
+ Carp::cluck("found ".$kid->name." where regcomp expected");
} else {
- $re = $self->deparse($kid, 1);
+ ($re, $quote) = $self->regcomp($kid, 1, $extended);
}
my $flags = "";
$flags .= "c" if $op->pmflags & PMf_CONTINUE;
if ($op->pmflags & PMf_ONCE) { # only one kind of delimiter works here
$re =~ s/\?/\\?/g;
$re = "?$re?";
- } else {
+ } elsif ($quote) {
$re = single_delim($name, $delim, $re);
}
- $re = $re . $flags;
+ $re = $re . $flags if $quote;
if ($binop) {
return $self->maybe_parens("$var =~ $re", $cx, 20);
} else {
$repl = $self->dq($repl);
}
}
+ my $extended = ($op->pmflags & PMf_EXTENDED);
if (null $kid) {
- $re = re_uninterp(escape_str(re_unback($op->precomp)));
+ my $unbacked = re_unback($op->precomp);
+ if ($extended) {
+ $re = re_uninterp_extended(escape_extended_re($unbacked));
+ }
+ else {
+ $re = re_uninterp(escape_str($unbacked));
+ }
} else {
- $re = $self->deparse($kid, 1);
+ ($re) = $self->regcomp($kid, 1, $extended);
}
$flags .= "e" if $op->pmflags & PMf_EVAL;
$flags .= "g" if $op->pmflags & PMf_GLOBAL;
$flags .= "m" if $op->pmflags & PMf_MULTILINE;
$flags .= "o" if $op->pmflags & PMf_KEEP;
$flags .= "s" if $op->pmflags & PMf_SINGLELINE;
- $flags .= "x" if $op->pmflags & PMf_EXTENDED;
+ $flags .= "x" if $extended;
$flags = $substwords{$flags} if $substwords{$flags};
if ($binop) {
return $self->maybe_parens("$var =~ s"
=item *
-Lvalue method calls are not yet fully supported. (Ordinary lvalue
-subroutine calls ought to be okay though.)
+In fact, the above is a specific instance of a more general problem:
+we can't guarantee to produce BEGIN blocks or C<use> declarations in
+exactly the right place. So if you use a module which affects compilation
+(such as by over-riding keywords, overloading constants or whatever)
+then the output code might not work as intended.
-=item *
-
-If you have a regex which is anything other than a literal of some
-kind, B::Deparse will produce incorrect output.
-e.g. C<$foo =~ give_me_a_regex()> will come back as
-C<$foo =~ /give_me_a_regex()/>
-
-=item *
-
- m{ #foo
- bar }x
-
-comes out as
-
- m/#foo\n bar/x)
-
-which isn't right.
+This is the most serious outstanding problem, and will be very hard
+to fix.
=item *
If a keyword is over-ridden, and your program explicitly calls
the built-in version by using CORE::keyword, the output of B::Deparse
-will not reflect this.
+will not reflect this. If you run the resulting code, it will call
+the over-ridden version rather than the built-in one. (Maybe there
+should be an option to B<always> print keyword calls as C<CORE::name>.)
=item *
=item *
Constants (other than simple strings or numbers) don't work properly.
-Examples that fail include:
+Pathological examples that fail (and probably always will) include:
use constant E2BIG => ($!=7);
use constant x=>\$x; print x
+The following could (and should) be made to work:
+
+ use constant regex => qr/blah/;
+ print regex;
+
=item *
An input file that uses source filtering probably won't be deparsed into
https://perl5.git.perl.org / perl5.git / blobdiff