use Carp;
$open::hint_bits = 0x20000;
-# layers array and hash mainly manipulated by C code in perlio.c
-use vars qw(%layers @layers);
+our $VERSION = '1.01';
-# Populate hash in non-PerlIO case
-%layers = (crlf => 1, raw => 0) unless (@layers);
+my $locale_encoding;
-# warn join(',',keys %layers);
+sub in_locale { $^H & $locale::hint_bits }
-our $VERSION = '1.00';
+sub _get_locale_encoding {
+ unless (defined $locale_encoding) {
+ # I18N::Langinfo isn't available everywhere
+ eval {
+ require I18N::Langinfo;
+ I18N::Langinfo->import(qw(langinfo CODESET));
+ $locale_encoding = langinfo(CODESET());
+ };
+ my $country_language;
+
+ no warnings 'uninitialized';
+
+ if (not $locale_encoding && in_locale()) {
+ if ($ENV{LC_ALL} =~ /^([^.]+)\.([^.]+)$/) {
+ ($country_language, $locale_encoding) = ($1, $2);
+ } elsif ($ENV{LANG} =~ /^([^.]+)\.([^.]+)$/) {
+ ($country_language, $locale_encoding) = ($1, $2);
+ }
+ } elsif (not $locale_encoding) {
+ if ($ENV{LC_ALL} =~ /\butf-?8\b/i ||
+ $ENV{LANG} =~ /\butf-?8\b/i) {
+ $locale_encoding = 'utf8';
+ }
+ # Could do more heuristics based on the country and language
+ # parts of LC_ALL and LANG (the parts before the dot (if any)),
+ # since we have Locale::Country and Locale::Language available.
+ # TODO: get a database of Language -> Encoding mappings
+ # (the Estonian database at http://www.eki.ee/letter/
+ # would be excellent!) --jhi
+ }
+ if (defined $locale_encoding &&
+ $locale_encoding eq 'euc' &&
+ defined $country_language) {
+ if ($country_language =~ /^ja_JP|japan(?:ese)?$/i) {
+ $locale_encoding = 'euc-jp';
+ } elsif ($country_language =~ /^ko_KR|korean?$/i) {
+ $locale_encoding = 'euc-kr';
+ } elsif ($country_language =~ /^zh_CN|chin(?:a|ese)?$/i) {
+ $locale_encoding = 'euc-cn';
+ } elsif ($country_language =~ /^zh_TW|taiwan(?:ese)?$/i) {
+ $locale_encoding = 'big5';
+ }
+ croak "Locale encoding 'euc' too ambiguous"
+ if $locale_encoding eq 'euc';
+ }
+ }
+}
sub import {
my ($class,@args) = @_;
croak("`use open' needs explicit list of disciplines") unless @args;
$^H |= $open::hint_bits;
- my ($in,$out) = split(/\0/,(${^OPEN} || '\0'));
- my @in = split(/\s+/,$in);
- my @out = split(/\s+/,$out);
+ my ($in,$out) = split(/\0/,(${^OPEN} || "\0"), -1);
while (@args) {
my $type = shift(@args);
- my $discp = shift(@args);
+ my $dscp;
+ if ($type =~ /^:?(utf8|locale|encoding\(.+\))$/) {
+ $type = 'IO';
+ $dscp = ":$1";
+ } else {
+ $dscp = shift(@args) || '';
+ }
my @val;
- foreach my $layer (split(/\s+/,$discp)) {
+ foreach my $layer (split(/\s+/,$dscp)) {
$layer =~ s/^://;
- unless(exists $layers{$layer}) {
- carp("Unknown discipline layer '$layer'");
+ if ($layer eq 'locale') {
+ use Encode;
+ _get_locale_encoding()
+ unless defined $locale_encoding;
+ (carp("Cannot figure out an encoding to use"), last)
+ unless defined $locale_encoding;
+ if ($locale_encoding =~ /^utf-?8$/i) {
+ $layer = "utf8";
+ } else {
+ $layer = "encoding($locale_encoding)";
+ }
+ } else {
+ unless(PerlIO::Layer::->find($layer)) {
+ carp("Unknown discipline layer '$layer'");
+ }
}
push(@val,":$layer");
if ($layer =~ /^(crlf|raw)$/) {
$^H{"open_$type"} = $layer;
}
}
+ # print "# type = $type, val = @val\n";
if ($type eq 'IN') {
$in = join(' ',@val);
}
elsif ($type eq 'OUT') {
$out = join(' ',@val);
}
+ elsif ($type eq 'IO') {
+ $in = $out = join(' ',@val);
+ }
else {
croak "Unknown discipline class '$type'";
}
}
- ${^OPEN} = join('\0',$in,$out);
+ ${^OPEN} = join("\0",$in,$out) if $in or $out;
}
1;
=head1 SYNOPSIS
- use open IN => ":crlf", OUT => ":raw";
+ use open IN => ":crlf", OUT => ":raw";
+ use open OUT => ':utf8';
+ use open IO => ":encoding(iso-8859-7)";
+
+ use open IO => ':locale';
+
+ use open ':utf8';
+ use open ':locale';
+ use open ':encoding(iso-8859-7)';
=head1 DESCRIPTION
operators found within the lexical scope of this pragma will use the
declared defaults.
+With the C<IN> subpragma you can declare the default layers
+of input streams, and with the C<OUT> subpragma you can declare
+the default layers of output streams. With the C<IO> subpragma
+you can control both input and output streams simultaneously.
+
+If you have a legacy encoding, you can use the C<:encoding(...)> tag.
+
+if you want to set your encoding disciplines based on your
+locale environment variables, you can use the C<:locale> tag.
+For example:
+
+ $ENV{LANG} = 'ru_RU.KOI8-R';
+ # the :locale will probe the locale environment variables like LANG
+ use open OUT => ':locale';
+ open(O, ">koi8");
+ print O chr(0x430); # Unicode CYRILLIC SMALL LETTER A = KOI8-R 0xc1
+ close O;
+ open(I, "<koi8");
+ printf "%#x\n", ord(<I>), "\n"; # this should print 0xc1
+ close I;
+
+These are equivalent
+
+ use open ':utf8';
+ use open IO => ':utf8';
+
+as are these
+
+ use open ':locale';
+ use open IO => ':locale';
+
+and these
+
+ use open ':encoding(iso-8859-7)';
+ use open IO => ':encoding(iso-8859-7)';
+
When open() is given an explicit list of layers they are appended to
the list declared using this pragma.
=head1 IMPLEMENTATION DETAILS
-There are two package variables C<%layers> and C<@layers> which are
-mainly manipulated by C code in F<perlio.c>, but are visible to the
-nosy:
-
- print "Have ",join(',',keys %open::layers),"\n";
- print "Using ",join(',',@open::layers),"\n";
-
-The C<%open::layers> hash is a record of the available "layers" that
-may be pushed onto a C<PerlIO> stream. The values of the hash are Perl
-objects, of class C<PerlIO::Layer> which are created by the C code in
-F<perlio.c>. As yet there is nothing useful you can do with the
-objects at the perl level.
-
-The C<@open::layers> array is the current set of layers and their
-arguments. The array consists of layer => argument pairs and I<must>
-always have even number of entries and the even entries I<must> be
-C<PerlIO::Layer> objects or Perl will "die" when it attempts to open a
-filehandle. In most cases the odd entry will be C<undef>, but in the
-case of (say) ":encoding(iso-8859-1)" it will be 'iso-8859-1'. These
-argument entries are currently restricted to being strings.
+There is a class method in C<PerlIO::Layer> C<find> which is
+implemented as XS code. It is called by C<import> to validate the
+layers:
-When a new C<PerlIO> stream is opened, the C code looks at the array
-to determine the default layers to be pushed. So with care it is
-possible to manipulate the default layer "stack":
+ PerlIO::Layer::->find("perlio")
- splice(@PerlIO::layers,-2,2);
- push(@PerlIO::layers,$PerlIO::layers{'stdio'} => undef);
+The return value (if defined) is a Perl object, of class
+C<PerlIO::Layer> which is created by the C code in F<perlio.c>. As
+yet there is nothing useful you can do with the object at the perl
+level.
=head1 SEE ALSO
-L<perlfunc/"binmode">, L<perlfunc/"open">, L<perlunicode>, L<PerlIO>
+L<perlfunc/"binmode">, L<perlfunc/"open">, L<perlunicode>, L<PerlIO>,
+L<encoding>
=cut
projects / p5sagit/p5-mst-13.2.git / blobdiff