Re: [PATCH] Attempt at speeding up Config.pm. Didn't work.

[perl5.git] / configpm

#!./miniperl -w

# commonly used names to put first (and hence lookup fastest)
my %Common = map {($_,$_)}
             qw(archname osname osvers prefix libs libpth
                dynamic_ext static_ext dlsrc so
                cc ccflags cppflags
                privlibexp archlibexp installprivlib installarchlib
                sharpbang startsh shsharp
               );

# names of things which may need to have slashes changed to double-colons
my %Extensions = map {($_,$_)}
                 qw(dynamic_ext static_ext extensions known_extensions);

# allowed opts as well as specifies default and initial values
my %Allowed_Opts = (
    'cross'    => '', # --cross=PALTFORM - crosscompiling for PLATFORM
    'glossary' => 1,  # --no-glossary  - no glossary file inclusion, 
                      #                  for compactness
);

sub opts {
    # user specified options
    my %given_opts = (
        # --opt=smth
        (map {/^--([\-_\w]+)=(.*)$/} @ARGV),
        # --opt --no-opt --noopt
        (map {/^no-?(.*)$/i?($1=>0):($_=>1)} map {/^--([\-_\w]+)$/} @ARGV),
    );

    my %opts = (%Allowed_Opts, %given_opts);

    for my $opt (grep {!exists $Allowed_Opts{$_}} keys %given_opts) {
        die "option '$opt' is not recognized";
    }
    @ARGV = grep {!/^--/} @ARGV;

    return %opts;
}


my %Opts = opts();

my $Config_PM;
my $Glossary = $ARGV[1] || 'Porting/Glossary';

if ($Opts{cross}) {
  # creating cross-platform config file
  mkdir "xlib";
  mkdir "xlib/$Opts{cross}";
  $Config_PM = $ARGV[0] || "xlib/$Opts{cross}/Config.pm";
}
else {
  $Config_PM = $ARGV[0] || 'lib/Config.pm';
}


open CONFIG, ">$Config_PM" or die "Can't open $Config_PM: $!\n";

my $myver = sprintf "v%vd", $^V;

printf CONFIG <<'ENDOFBEG', ($myver) x 3;
# This file was created by configpm when Perl was built. Any changes
# made to this file will be lost the next time perl is built.

package Config;
use Exporter ();
@EXPORT = qw(%%Config);
@EXPORT_OK = qw(myconfig config_sh config_vars);

# Define our own import method to avoid pulling in the full Exporter:
sub import {
  my $pkg = shift;
  @_ = @EXPORT unless @_;

  my @func = grep {$_ ne '%%Config'} @_;
  local $Exporter::ExportLevel = 1;
  Exporter::import('Config', @func) if @func;

  return if @func == @_;

  my $callpkg = caller(0);
  *{"$callpkg\::Config"} = \%%Config;
}

die "Perl lib version (%s) doesn't match executable version ($])"
    unless $^V;

$^V eq %s
  or die "Perl lib version (%s) doesn't match executable version (" .
    (sprintf "v%vd",$^V) . ")";

ENDOFBEG


my @non_v    = ();
my @v_fast   = ();
my %v_fast   = ();
my @v_others = ();
my $in_v     = 0;
my %Data     = ();

# This is somewhat grim, but I want the code for parsing config.sh here and
# now so that I can expand $Config{ivsize} and $Config{ivtype}

my $fetch_string = <<'EOT';

# Search for it in the big string 
sub fetch_string {
    my($self, $key) = @_;

    my $quote_type = "'";
    my $marker = "$key=";

    # Check for the common case, ' delimeted
    my $start = index($Config_SH, "\n$marker$quote_type");
    # If that failed, check for " delimited
    if ($start == -1) {
        $quote_type = '"';
        $start = index($Config_SH, "\n$marker$quote_type");
    }
    return undef if ( ($start == -1) &&  # in case it's first 
                      (substr($Config_SH, 0, length($marker)) ne $marker) );
    if ($start == -1) { 
        # It's the very first thing we found. Skip $start forward
        # and figure out the quote mark after the =.
        $start = length($marker) + 1;
        $quote_type = substr($Config_SH, $start - 1, 1);
    } 
    else { 
        $start += length($marker) + 2;
    }

    my $value = substr($Config_SH, $start, 
                       index($Config_SH, "$quote_type\n", $start) - $start);

    # If we had a double-quote, we'd better eval it so escape
    # sequences and such can be interpolated. Since the incoming
    # value is supposed to follow shell rules and not perl rules,
    # we escape any perl variable markers
    if ($quote_type eq '"') {
	$value =~ s/\$/\\\$/g;
	$value =~ s/\@/\\\@/g;
	eval "\$value = \"$value\"";
    }

    # So we can say "if $Config{'foo'}".
    $value = undef if $value eq 'undef';
    $self->{$key} = $value; # cache it
}
EOT

eval $fetch_string;
die if $@;

open(CONFIG_SH, 'config.sh') || die "Can't open config.sh: $!";
while (<CONFIG_SH>) {
    next if m:^#!/bin/sh:;

    # Catch PERL_CONFIG_SH=true and PERL_VERSION=n line from Configure.
    s/^(\w+)=(true|\d+)\s*$/$1='$2'\n/ or m/^(\w+)='(.*)'$/;
    my($k, $v) = ($1, $2);

    # grandfather PATCHLEVEL and SUBVERSION and CONFIG
    if ($k) {
	if ($k eq 'PERL_VERSION') {
	    push @v_others, "PATCHLEVEL='$v'\n";
	}
	elsif ($k eq 'PERL_SUBVERSION') {
	    push @v_others, "SUBVERSION='$v'\n";
	}
	elsif ($k eq 'PERL_CONFIG_SH') {
	    push @v_others, "CONFIG='$v'\n";
	}
    }

    # We can delimit things in config.sh with either ' or ". 
    unless ($in_v or m/^(\w+)=(['"])(.*\n)/){
	push(@non_v, "#$_"); # not a name='value' line
	next;
    }
    $quote = $2;
    if ($in_v) { 
        $val .= $_;
    }
    else { 
        ($name,$val) = ($1,$3); 
    }
    $in_v = $val !~ /$quote\n/;
    next if $in_v;

    s,/,::,g if $Extensions{$name};

    $val =~ s/$quote\n?\z//;

    my $line = "$name=$quote$val$quote\n";
    if (!$Common{$name}){
        push(@v_others, $line);
    }
    else {
        push(@v_fast, $line);
        $v_fast{$name} = "'$name' => $quote$val$quote";
    }
}
close CONFIG_SH;

print CONFIG @non_v, "\n";

# copy config summary format from the myconfig.SH script
print CONFIG "my \$summary = <<'!END!';\n";
open(MYCONFIG,"<myconfig.SH") || die "open myconfig.SH failed: $!";
1 while defined($_ = <MYCONFIG>) && !/^Summary of/;
do { print CONFIG $_ } until !defined($_ = <MYCONFIG>) || /^\s*$/;
close(MYCONFIG);

print CONFIG "\n!END!\n", <<'EOT';
my $summary_expanded = 0;

sub myconfig {
	return $summary if $summary_expanded;
	$summary =~ s{\$(\w+)}
		     { my $c = $Config{$1}; defined($c) ? $c : 'undef' }ge;
	$summary_expanded = 1;
	$summary;
}

our $Config_SH : unique = <<'!END!';
EOT

print CONFIG join("", @v_fast, sort @v_others);

print CONFIG "!END!\n", $fetch_string;

print CONFIG <<'ENDOFEND';

sub fetch_virtual {
    my($self, $key) = @_;

    my $value;

    if ($key =~ /^((?:cc|ld)flags|libs(?:wanted)?)_nolargefiles/) {
	# These are purely virtual, they do not exist, but need to
	# be computed on demand for largefile-incapable extensions.
	my $new_key = "${1}_uselargefiles";
	$value = $Config{$1};
	my $withlargefiles = $Config{$new_key};
	if ($new_key =~ /^(?:cc|ld)flags_/) {
	    $value =~ s/\Q$withlargefiles\E\b//;
	} elsif ($new_key =~ /^libs/) {
	    my @lflibswanted = split(' ', $Config{libswanted_uselargefiles});
	    if (@lflibswanted) {
		my %lflibswanted;
		@lflibswanted{@lflibswanted} = ();
		if ($new_key =~ /^libs_/) {
		    my @libs = grep { /^-l(.+)/ &&
                                      not exists $lflibswanted{$1} }
		                    split(' ', $Config{libs});
		    $Config{libs} = join(' ', @libs);
		} elsif ($new_key =~ /^libswanted_/) {
		    my @libswanted = grep { not exists $lflibswanted{$_} }
		                          split(' ', $Config{libswanted});
		    $Config{libswanted} = join(' ', @libswanted);
		}
	    }
	}
    }

    $self->{$key} = $value;
}

sub FETCH { 
    my($self, $key) = @_;

    # check for cached value (which may be undef so we use exists not defined)
    return $self->{$key} if exists $self->{$key};

    $self->fetch_string($key);
    return $self->{$key} if exists $self->{$key};
    $self->fetch_virtual($key);

    # Might not exist, in which undef is correct.
    return $self->{$key};
}
 
my $prevpos = 0;

sub FIRSTKEY {
    $prevpos = 0;
    substr($Config_SH, 0, index($Config_SH, '=') );
}

sub NEXTKEY {
    # Find out how the current key's quoted so we can skip to its end.
    my $quote = substr($Config_SH, index($Config_SH, "=", $prevpos)+1, 1);
    my $pos = index($Config_SH, qq($quote\n), $prevpos) + 2;
    my $len = index($Config_SH, "=", $pos) - $pos;
    $prevpos = $pos;
    $len > 0 ? substr($Config_SH, $pos, $len) : undef;
}

sub EXISTS { 
    return 1 if exists($_[0]->{$_[1]});

    return(index($Config_SH, "\n$_[1]='") != -1 or
           substr($Config_SH, 0, length($_[1])+2) eq "$_[1]='" or
           index($Config_SH, "\n$_[1]=\"") != -1 or
           substr($Config_SH, 0, length($_[1])+2) eq "$_[1]=\"" or
           $_[1] =~ /^(?:(?:cc|ld)flags|libs(?:wanted)?)_nolargefiles$/
          );
}

sub STORE  { die "\%Config::Config is read-only\n" }
*DELETE = \&STORE;
*CLEAR  = \&STORE;


sub config_sh {
    $Config_SH
}

sub config_re {
    my $re = shift;
    my @matches = grep /^$re=/, split /^/, $Config_SH;
    @matches ? (print @matches) : print "$re: not found\n";
}

sub config_vars {
    foreach(@_){
	config_re($_), next if /\W/;
	my $v=(exists $Config{$_}) ? $Config{$_} : 'UNKNOWN';
	$v='undef' unless defined $v;
	print "$_='$v';\n";
    }
}

ENDOFEND

if ($^O eq 'os2') {
  print CONFIG <<'ENDOFSET';
my %preconfig;
if ($OS2::is_aout) {
    my ($value, $v) = $Config_SH =~ m/^used_aout='(.*)'\s*$/m;
    for (split ' ', $value) {
        ($v) = $Config_SH =~ m/^aout_$_='(.*)'\s*$/m;
        $preconfig{$_} = $v eq 'undef' ? undef : $v;
    }
}
$preconfig{d_fork} = undef unless $OS2::can_fork; # Some funny cases can't
sub TIEHASH { bless {%preconfig} }
ENDOFSET
  # Extract the name of the DLL from the makefile to avoid duplication
  my ($f) = grep -r, qw(GNUMakefile Makefile);
  my $dll;
  if (open my $fh, '<', $f) {
    while (<$fh>) {
      $dll = $1, last if /^PERL_DLL_BASE\s*=\s*(\S*)\s*$/;
    }
  }
  print CONFIG <<ENDOFSET if $dll;
\$preconfig{dll_name} = '$dll';
ENDOFSET
} else {
  print CONFIG <<'ENDOFSET';
sub TIEHASH {
    bless $_[1], $_[0];
}
ENDOFSET
}


# Calculation for the keys for byteorder
# This is somewhat grim, but I need to run fetch_string here.
our $Config_SH = join "\n", @v_fast, @v_others;

my $t = fetch_string ({}, 'ivtype');
my $s = fetch_string ({}, 'ivsize');

# byteorder does exist on its own but we overlay a virtual
# dynamically recomputed value.

# However, ivtype and ivsize will not vary for sane fat binaries

my $f = $t eq 'long' ? 'L!' : $s == 8 ? 'Q': 'I';

my $byteorder_code;
if ($s == 4 || $s == 8) {

  my $list = join ',', reverse(2..$s);
  my $format = 'a'x$s;
  $byteorder_code = <<"EOT";
my \$i = 0;
foreach my \$c ($list) { \$i |= ord(\$c); \$i <<= 8 }
\$i |= ord(1);
my \$value = join('', unpack('$format', pack('$f', \$i)));
EOT
} else {
  $byteorder_code = "\$value = '?'x$s;\n";
}

my $fast_config = join '', map { "    $_,\n" }
  values (%v_fast), 'byteorder => $value' ;

print CONFIG sprintf <<'ENDOFTIE', $byteorder_code, $fast_config;

# avoid Config..Exporter..UNIVERSAL search for DESTROY then AUTOLOAD
sub DESTROY { }

%s

tie %%Config, 'Config', {
%s
};

1;
ENDOFTIE


open(CONFIG_POD, ">lib/Config.pod") or die "Can't open lib/Config.pod: $!";
print CONFIG_POD <<'ENDOFTAIL';
=head1 NAME

Config - access Perl configuration information

=head1 SYNOPSIS

    use Config;
    if ($Config{'cc'} =~ /gcc/) {
	print "built by gcc\n";
    } 

    use Config qw(myconfig config_sh config_vars);

    print myconfig();

    print config_sh();

    config_vars(qw(osname archname));


=head1 DESCRIPTION

The Config module contains all the information that was available to
the C<Configure> program at Perl build time (over 900 values).

Shell variables from the F<config.sh> file (written by Configure) are
stored in the readonly-variable C<%Config>, indexed by their names.

Values stored in config.sh as 'undef' are returned as undefined
values.  The perl C<exists> function can be used to check if a
named variable exists.

=over 4

=item myconfig()

Returns a textual summary of the major perl configuration values.
See also C<-V> in L<perlrun/Switches>.

=item config_sh()

Returns the entire perl configuration information in the form of the
original config.sh shell variable assignment script.

=item config_vars(@names)

Prints to STDOUT the values of the named configuration variable. Each is
printed on a separate line in the form:

  name='value';

Names which are unknown are output as C<name='UNKNOWN';>.
See also C<-V:name> in L<perlrun/Switches>.

=back

=head1 EXAMPLE

Here's a more sophisticated example of using %Config:

    use Config;
    use strict;

    my %sig_num;
    my @sig_name;
    unless($Config{sig_name} && $Config{sig_num}) {
	die "No sigs?";
    } else {
	my @names = split ' ', $Config{sig_name};
	@sig_num{@names} = split ' ', $Config{sig_num};
	foreach (@names) {
	    $sig_name[$sig_num{$_}] ||= $_;
	}   
    }

    print "signal #17 = $sig_name[17]\n";
    if ($sig_num{ALRM}) { 
	print "SIGALRM is $sig_num{ALRM}\n";
    }   

=head1 WARNING

Because this information is not stored within the perl executable
itself it is possible (but unlikely) that the information does not
relate to the actual perl binary which is being used to access it.

The Config module is installed into the architecture and version
specific library directory ($Config{installarchlib}) and it checks the
perl version number when loaded.

The values stored in config.sh may be either single-quoted or
double-quoted. Double-quoted strings are handy for those cases where you
need to include escape sequences in the strings. To avoid runtime variable
interpolation, any C<$> and C<@> characters are replaced by C<\$> and
C<\@>, respectively. This isn't foolproof, of course, so don't embed C<\$>
or C<\@> in double-quoted strings unless you're willing to deal with the
consequences. (The slashes will end up escaped and the C<$> or C<@> will
trigger variable interpolation)

=head1 GLOSSARY

Most C<Config> variables are determined by the C<Configure> script
on platforms supported by it (which is most UNIX platforms).  Some
platforms have custom-made C<Config> variables, and may thus not have
some of the variables described below, or may have extraneous variables
specific to that particular port.  See the port specific documentation
in such cases.

ENDOFTAIL

if ($Opts{glossary}) {
  open(GLOS, "<$Glossary") or die "Can't open $Glossary: $!";
}
%seen = ();
$text = 0;
$/ = '';

sub process {
  if (s/\A(\w*)\s+\(([\w.]+)\):\s*\n(\t?)/=item C<$1>\n\nFrom F<$2>:\n\n/m) {
    my $c = substr $1, 0, 1;
    unless ($seen{$c}++) {
      print CONFIG_POD <<EOF if $text;
=back

EOF
      print CONFIG_POD <<EOF;
=head2 $c

=over 4

EOF
     $text = 1;
    }
  }
  elsif (!$text || !/\A\t/) {
    warn "Expected a Configure variable header",
      ($text ? " or another paragraph of description" : () );
  }
  s/n't/n\00t/g;		# leave can't, won't etc untouched
  s/^\t\s+(.*)/\n$1/gm;		# Indented lines ===> new paragraph
  s/^(?<!\n\n)\t(.*)/$1/gm;	# Not indented lines ===> text
  s{([\'\"])(?=[^\'\"\s]*[./][^\'\"\s]*\1)([^\'\"\s]+)\1}(F<$2>)g; # '.o'
  s{([\'\"])([^\'\"\s]+)\1}(C<$2>)g; # "date" command
  s{\'([A-Za-z_\- *=/]+)\'}(C<$1>)g; # 'ln -s'
  s{
     (?<! [\w./<\'\"] )		# Only standalone file names
     (?! e \. g \. )		# Not e.g.
     (?! \. \. \. )		# Not ...
     (?! \d )			# Not 5.004
     (?! read/ )		# Not read/write
     (?! etc\. )		# Not etc.
     (?! I/O )			# Not I/O
     (
	\$ ?			# Allow leading $
	[\w./]* [./] [\w./]*	# Require . or / inside
     )
     (?<! \. (?= [\s)] ) )	# Do not include trailing dot
     (?! [\w/] )		# Include all of it
   }
   (F<$1>)xg;			# /usr/local
  s/((?<=\s)~\w*)/F<$1>/g;	# ~name
  s/(?<![.<\'\"])\b([A-Z_]{2,})\b(?![\'\"])/C<$1>/g;	# UNISTD
  s/(?<![.<\'\"])\b(?!the\b)(\w+)\s+macro\b/C<$1> macro/g; # FILE_cnt macro
  s/n[\0]t/n't/g;		# undo can't, won't damage
}

if ($Opts{glossary}) {
    <GLOS>;				# Skip the "DO NOT EDIT"
    <GLOS>;				# Skip the preamble
  while (<GLOS>) {
    process;
    print CONFIG_POD;
  }
}

print CONFIG_POD <<'ENDOFTAIL';

=back

=head1 NOTE

This module contains a good example of how to use tie to implement a
cache and an example of how to make a tied variable readonly to those
outside of it.

=cut

ENDOFTAIL

close(CONFIG);
close(GLOS);
close(CONFIG_POD);

# Now create Cross.pm if needed
if ($Opts{cross}) {
  open CROSS, ">lib/Cross.pm" or die "Can not open >lib/Cross.pm: $!";
  my $cross = <<'EOS';
# typical invocation:
#   perl -MCross Makefile.PL
#   perl -MCross=wince -V:cc
package Cross;

sub import {
  my ($package,$platform) = @_;
  unless (defined $platform) {
    # if $platform is not specified, then use last one when
    # 'configpm; was invoked with --cross option
    $platform = '***replace-marker***';
  }
  @INC = map {/\blib\b/?(do{local $_=$_;s/\blib\b/xlib\/$platform/;$_},$_):($_)} @INC;
  $::Cross::platform = $platform;
}

1;
EOS
  $cross =~ s/\*\*\*replace-marker\*\*\*/$Opts{cross}/g;
  print CROSS $cross;
  close CROSS;
}

# Now do some simple tests on the Config.pm file we have created
unshift(@INC,'lib');
require $Config_PM;
import Config;

die "$0: $Config_PM not valid"
	unless $Config{'PERL_CONFIG_SH'} eq 'true';

die "$0: error processing $Config_PM"
	if defined($Config{'an impossible name'})
	or $Config{'PERL_CONFIG_SH'} ne 'true' # test cache
	;

die "$0: error processing $Config_PM"
	if eval '$Config{"cc"} = 1'
	or eval 'delete $Config{"cc"}'
	;


exit 0;