enc2xs

source of perl for linux application,

         }, @inc);
    warn join("\n", keys %e2x_dir), "\n";
    for my $d (sort {$e2x_dir{$a} <=> $e2x_dir{$b}} keys %e2x_dir){
    $_E2X = $d;
    # warn "$_E2X => ", scalar localtime($e2x_dir{$d});
    return $_E2X;
    }
}

sub make_makefile_pl
{
    eval { require Encode; };
    $@ and die "You need to install Encode to use enc2xs -M\nerror: $@\n";
    # our used for variable expanstion
    $_Enc2xs = $0;
    $_Version = $VERSION;
    $_E2X = find_e2x();
    $_Name = shift;
    $_TableFiles = join(",", map {qq('$_')} @_);
    $_Now = scalar localtime();

    eval { require File::Spec; };
    _print_expand(File::Spec->catfile($_E2X,"Makefile_PL.e2x"),"Makefile.PL");
    _print_expand(File::Spec->catfile($_E2X,"_PM.e2x"),        "$_Name.pm");
    _print_expand(File::Spec->catfile($_E2X,"_T.e2x"),         "t/$_Name.t");
    _print_expand(File::Spec->catfile($_E2X,"README.e2x"),     "README");
    _print_expand(File::Spec->catfile($_E2X,"Changes.e2x"),    "Changes");
    exit;
}

use vars qw(
        $_ModLines
        $_LocalVer
        );

sub make_configlocal_pm {
    eval { require Encode; };
    $@ and die "Unable to require Encode: $@\n";
    eval { require File::Spec; };

    # our used for variable expanstion
    my %in_core = map { $_ => 1 } (
        'ascii',      'iso-8859-1', 'utf8',
        'ascii-ctrl', 'null',       'utf-8-strict'
    );
    my %LocalMod = ();
    # check @enc;
    use File::Find ();
    my $wanted = sub{
	-f $_ or return;
	$File::Find::name =~ /\A\./        and return;
	$File::Find::name =~ /\.pm\z/      or  return;
	$File::Find::name =~ m/\bEncode\b/ or  return;
	my $mod = $File::Find::name;
	$mod =~ s/.*\bEncode\b/Encode/o;
	$mod =~ s/\.pm\z//o;
	$mod =~ s,/,::,og;
	warn qq{ require $mod;\n};
	eval qq{ require $mod; };
	$@ and die "Can't require $mod: $@\n";
	for my $enc ( Encode->encodings() ) {
	    no warnings;
	    $in_core{$enc}                   and next;
	    $Encode::Config::ExtModule{$enc} and next;
	    $LocalMod{$enc} ||= $mod;
	}
    };
    File::Find::find({wanted => $wanted}, @INC);
    $_ModLines = "";
    for my $enc ( sort keys %LocalMod ) {
        $_ModLines .=
          qq(\$Encode::ExtModule{'$enc'} = "$LocalMod{$enc}";\n);
    }
    warn $_ModLines;
    $_LocalVer = _mkversion();
    $_E2X      = find_e2x();
    $_Inc      = $INC{"Encode.pm"};
    $_Inc =~ s/\.pm$//o;
    _print_expand( File::Spec->catfile( $_E2X, "ConfigLocal_PM.e2x" ),
        File::Spec->catfile( $_Inc, "ConfigLocal.pm" ), 1 );
    exit;
}

sub _mkversion{
    # v-string is now depreciated; use time() instead;
    #my ($ss,$mm,$hh,$dd,$mo,$yyyy) = localtime();
    #$yyyy += 1900, $mo +=1;
    #return sprintf("v%04d.%04d.%04d", $yyyy, $mo*100+$dd, $hh*100+$mm);
    return time();
}

sub _print_expand{
    eval { require File::Basename; };
    $@ and die "File::Basename needed.  Are you on miniperl?;\nerror: $@\n";
    File::Basename->import();
    my ($src, $dst, $clobber) = @_;
    if (!$clobber and -e $dst){
    warn "$dst exists. skipping\n";
    return;
    }
    warn "Generating $dst...\n";
    open my $in, $src or die "$src : $!";
    if ((my $d = dirname($dst)) ne '.'){
    -d $d or mkdir $d, 0755 or die  "mkdir $d : $!";
    }	   
    open my $out, ">$dst" or die "$!";
    my $asis = 0;
    while (<$in>){ 
    if (/^#### END_OF_HEADER/){
        $asis = 1; next;
    }	  
    s/(\$_[A-Z][A-Za-z0-9]+)_/$1/gee unless $asis;
    print $out $_;
    }
}
__END__

=head1 NAME

enc2xs -- Perl Encode Module Generator

=head1 SYNOPSIS

  enc2xs -[options]
  enc2xs -M ModName mapfiles...
  enc2xs -C

=head1 DESCRIPTION

F<enc2xs> builds a Perl extension for use by Encode from either
Unicode Character Mapping files (.ucm) or Tcl Encoding Files (.enc).
Besides being used internally during the build process of the Encode
module, you can use F<enc2xs> to add your own encoding to perl.
No knowledge of XS is necessary.

=head1 Quick Guide

If you want to know as little about Perl as possible but need to
add a new encoding, just read this chapter and forget the rest.

=over 4

=item 0.

Have a .ucm file ready.  You can get it from somewhere or you can write
your own from scratch or you can grab one from the Encode distribution
and customize it.  For the UCM format, see the next Chapter.  In the
example below, I'll call my theoretical encoding myascii, defined
in I<my.ucm>.  C<$> is a shell prompt.

  $ ls -F
  my.ucm

=item 1.

Issue a command as follows;

  $ enc2xs -M My my.ucm
  generating Makefile.PL
  generating My.pm
  generating README
  generating Changes

Now take a look at your current directory.  It should look like this.

  $ ls -F
  Makefile.PL   My.pm         my.ucm        t/

The following files were created.

  Makefile.PL - MakeMaker script
  My.pm       - Encode submodule
  t/My.t      - test file

=over 4

=item 1.1.

If you want *.ucm installed together with the modules, do as follows;

  $ mkdir Encode
  $ mv *.ucm Encode
  $ enc2xs -M My Encode/*ucm

=back

=item 2.

Edit the files generated.  You don't have to if you have no time AND no
intention to give it to someone else.  But it is a good idea to edit
the pod and to add more tests.

=item 3.

Now issue a command all Perl Mongers love:

  $ perl Makefile.PL
  Writing Makefile for Encode::My

=item 4.

Now all you have to do is make.

  $ make
  cp My.pm blib/lib/Encode/My.pm
  /usr/local/bin/perl /usr/local/bin/enc2xs -Q -O \
    -o encode_t.c -f encode_t.fnm
  Reading myascii (myascii)
  Writing compiled form
  128 bytes in string tables
  384 bytes (75%) saved spotting duplicates
  1 bytes (0.775%) saved using substrings
  ....
  chmod 644 blib/arch/auto/Encode/My/My.bs
  $

The time it takes varies depending on how fast your machine is and
how large your encoding is.  Unless you are working on something big
like euc-tw, it won't take too long.

=item 5.

You can "make install" already but you should test first.

  $ make test
  PERL_DL_NONLAZY=1 /usr/local/bin/perl -Iblib/arch -Iblib/lib \
    -e 'use Test::Harness  qw(&runtests $verbose); \
    $verbose=0; runtests @ARGV;' t/*.t
  t/My....ok
  All tests successful.
  Files=1, Tests=2,  0 wallclock secs
   ( 0.09 cusr + 0.01 csys = 0.09 CPU)

=item 6.

If you are content with the test result, just "make install"

=item 7.

If you want to add your encoding to Encode's demand-loading list
(so you don't have to "use Encode::YourEncoding"), run

  enc2xs -C

to update Encode::ConfigLocal, a module that controls local settings.
After that, "use Encode;" is enough to load your encodings on demand.

=back

=head1 The Unicode Character Map

Encode uses the Unicode Character Map (UCM) format for source character
mappings.  This format is used by IBM's ICU package and was adopted
by Nick Ing-Simmons for use with the Encode module.  Since UCM is
more flexible than Tcl's Encoding Map and far more user-friendly,
this is the recommended format for Encode now.

A UCM file looks like this.

  #
  # Comments
  #
  <code_set_name> "US-ascii" # Required
  <code_set_alias> "ascii"   # Optional
  <mb_cur_min> 1             # Required; usually 1
  <mb_cur_max> 1             # Max. # of bytes/char
  <subchar> \x3F             # Substitution char
  #
  CHARMAP
  <U0000> \x00 |0 # <control>
  <U0001> \x01 |0 # <control>
  <U0002> \x02 |0 # <control>
  ....
  <U007C> \x7C |0 # VERTICAL LINE
  <U007D> \x7D |0 # RIGHT CURLY BRACKET
  <U007E> \x7E |0 # TILDE
  <U007F> \x7F |0 # <control>
  END CHARMAP

=over 4

=item *

Anything that follows C<#> is treated as a comment.

=item *

The header section continues until a line containing the word
CHARMAP. This section has a form of I<E<lt>keywordE<gt> value>, one
pair per line.  Strings used as values must be quoted. Barewords are
treated as numbers.  I<\xXX> represents a byte.

Most of the keywords are self-explanatory. I<subchar> means
substitution character, not subcharacter.  When you decode a Unicode
sequence to this encoding but no matching character is found, the byte
sequence defined here will be used.  For most cases, the value here is
\x3F; in ASCII, this is a question mark.

=item *

CHARMAP starts the character map section.  Each line has a form as
follows:

  <UXXXX> \xXX.. |0 # comment
    ^     ^      ^
    |     |      +- Fallback flag
    |     +-------- Encoded byte sequence
    +-------------- Unicode Character ID in hex

The format is roughly the same as a header section except for the
fallback flag: | followed by 0..3.   The meaning of the possible
values is as follows:

=over 4

=item |0 

Round trip safe.  A character decoded to Unicode encodes back to the
same byte sequence.  Most characters have this flag.

=item |1

Fallback for unicode -> encoding.  When seen, enc2xs adds this
character for the encode map only.

=item |2 

Skip sub-char mapping should there be no code point.

=item |3 

Fallback for encoding -> unicode.  When seen, enc2xs adds this
character for the decode map only.

=back

=item *

And finally, END OF CHARMAP ends the section.

=back

When you are manually creating a UCM file, you should copy ascii.ucm
or an existing encoding which is close to yours, rather than write
your own from scratch.

When you do so, make sure you leave at least B<U0000> to B<U0020> as
is, unless your environment is EBCDIC.

B<CAVEAT>: not all features in UCM are implemented.  For example,
icu:state is not used.  Because of that, you need to write a perl
module if you want to support algorithmical encodings, notably
the ISO-2022 series.  Such modules include L<Encode::JP::2022_JP>,
L<Encode::KR::2022_KR>, and L<Encode::TW::HZ>.

=head2 Coping with duplicate mappings

When you create a map, you SHOULD make your mappings round-trip safe.
That is, C<encode('your-encoding', decode('your-encoding', $data)) eq
$data> stands for all characters that are marked as C<|0>.  Here is
how to make sure:

=over 4

=item * 

Sort your map in Unicode order.

=item *

When you have a duplicate entry, mark either one with '|1' or '|3'.
  
=item * 

And make sure the '|1' or '|3' entry FOLLOWS the '|0' entry.

=back

Here is an example from big5-eten.

  <U2550> \xF9\xF9 |0
  <U2550> \xA2\xA4 |3

Internally Encoding -> Unicode and Unicode -> Encoding Map looks like
this;

  E to U               U to E
  --------------------------------------
  \xF9\xF9 => U2550    U2550 => \xF9\xF9
  \xA2\xA4 => U2550
 
So it is round-trip safe for \xF9\xF9.  But if the line above is upside
down, here is what happens.

  E to U               U to E
  --------------------------------------
  \xA2\xA4 => U2550    U2550 => \xF9\xF9
  (\xF9\xF9 => U2550 is now overwritten!)

The Encode package comes with F<ucmlint>, a crude but sufficient
utility to check the integrity of a UCM file.  Check under the
Encode/bin directory for this.

When in doubt, you can use F<ucmsort>, yet another utility under
Encode/bin directory.

=head1 Bookmarks

=over 4

=item *

ICU Home Page 
L<http://oss.software.ibm.com/icu/>

=item *

ICU Character Mapping Tables
L<http://oss.software.ibm.com/icu/charset/>

=item *

ICU:Conversion Data
L<http://oss.software.ibm.com/icu/userguide/conversion-data.html>

=back

=head1 SEE ALSO

L<Encode>,
L<perlmod>,
L<perlpod>

=cut

# -Q to disable the duplicate codepoint test
# -S make mapping errors fatal
# -q to remove comments written to output files
# -O to enable the (brute force) substring optimiser
# -o <output> to specify the output file name (else it's the first arg)
# -f <inlist> to give a file with a list of input files (else use the args)
# -n <name> to name the encoding (else use the basename of the input file.

With %seen holding array refs:

      865.66 real        28.80 user         8.79 sys
      7904  maximum resident set size
      1356  average shared memory size
     18566  average unshared data size
       229  average unshared stack size
     46080  page reclaims
     33373  page faults

With %seen holding simple scalars:

      342.16 real        27.11 user         3.54 sys
      8388  maximum resident set size
      1394  average shared memory size
     14969  average unshared data size
       236  average unshared stack size
     28159  page reclaims
      9839  page faults

Yes, 5 minutes is faster than 15. Above is for CP936 in CN. Only difference is
how %seen is storing things its seen. So it is pathalogically bad on a 16M
RAM machine, but it's going to help even on modern machines.
Swapping is bad, m'kay :-)