#!/usr/local/bin/perl5
# Name: html-include		Author: Joe Smith <js-cgi@inwap.com> 04-Dec-97
# Purpose: Includes subfiles into a non-parsed html page.
#	This avoids the overhead caused by x-server-parsed-html.

$usage = <<EOM;
usage: $0 include_dir \$HOME/public_html
  where 'include_dir' is the directory containing the files to be included
  and '\$HOME/public_html' is the directory with HTML files to be updated.

  For each *.html file, this program looks for lines like this:
  <!-- html-include file=footer prev=1.html next=2.html up=index.html -->
  several lines that will be replaced with contents of include_dir/footer
  <!-- html-include-end -->

  Each file that has any changes will be renamed to have a *.bak extension.
  The names of these *.bak files are sent to STDOUT so that they can
  be deleted if there are no errors.
EOM

chop($cwd = `pwd`);
$include_dir = $_ = shift;	# First arg is where to find the included files
$include_abs = m%^/% ? $_ : "$cwd/$_";	# Absolute pathname

die($usage) unless @ARGV;	# Must have at least one more argument
$error_count = 0; $| = 1;
require "find.pl";
&find(@ARGV);			# Call &wanted() for each file in directory
exit($error_count ? 1 : 0);

############################################################################

sub wanted {			# Called from &find with $name = "$dir/$_"
  $mtime = (stat $_)[9];	# File's modification time
  -f _ && /^.*\.(html|htm|shtml|sht)$/ && &process($_,$name);
}
$_ = $name;			# For "perl -w" ($name is defined in find.pl)

############################################################################

sub process {			# Process an HTML file
  local($file,$fullname) = @_;
  print STDERR "$fullname - ";
  open(IN,$file) || return(warn("can't read: $!\n"));
  $/ = ">";			# Allow for \n inside the tags
  @lines = <IN>; close(IN);
  @output = ();
  $replacing = 0;		# Nonzero when replacing old lines
  $changes = $includes = 0;
  foreach $line (@lines) {
    if ($line =~ /<!-- html-include (.*)-->/m) {
      &parse_args($1);			# Look for 'file="name" back="x.html"'
      push(@output,$line);		# Copy this line verbatim
      $replacing = 1;			# But no other until html-include-end
      @old = ();
      $includes++;
    } elsif ($line =~ /<!-- html-include-end\s.*-->/m) {
      push(@old,$`);			# Lines just before this directive
      $include_end = $&;
      $new = &include_file;		# Use %args to locate file to include
      $old = join("",@old);		# Usually starts and ends with "\n"
      $changes++ if $new ne $old;
      push(@output,$new,$include_end);	# Replacement text
      $replacing = 0;
    } else {
      push(@output,$line) unless $replacing;
      push(@old,$line)    if     $replacing;
    }
  }
  if ($includes) {
    print STDERR ($changes == 1 ? "1 change" : "$changes changes"),"\n";
  } else {
    print STDERR "no <!-- html-include file="..." -->\n";
  }
  return unless $changes;
  $oldf = "$file.bak"; $oldfullname = "$fullname.bak";
  $newf = "$file.tmp";
  if (open(OUT,">$newf") && print(OUT @output) && close(OUT)) {
    utime($mtime,$mtime+60,$newf);	# Set file modification time
    unlink($oldf);			# Remove .bak from last time
    rename($file,$oldf) || warn("Could not rename $file to $oldf: $!\n");
    rename($newf,$file) || warn("Could not rename $newf to $file: $!\n");
    print "$oldfullname\n";		# This file can be deleted
  } else {
    warn("Could not rewrite $file: $!\n");
    $error_count++;
  }
}

############################################################################

sub parse_args {	# Search for key=val key="val" key='val'
  local($args) = @_;
  local($original) = $args;
  local($_);
  %args = ();
  while (($key,$_) = $args =~ /(\w+)=(.*)/) {
    ( /^"([^"]*)"\s*(.*)/ || /^'([^']*)'\s*(.*)/ || /(\S+)\s*(.*)/ )
	&& (($val,$_) = ($1,$2));
    $args{$key} = $val;
    $args = $_;
  }
  $args{"file"} || warn("Could not find 'file=\"...\"' in $original\n");
}

############################################################################

sub include_file {	# Include contents of file, with variable substitution
  local($inc_abs) = "$include_abs/" . $args{"file"};
  local($inc_rel) = "$include_dir/" . $args{"file"};
  unless($file_contents = $files{$inc_abs}) {
    if(open(INC,$inc_abs)) {
      local($/) = undef;
      $file_contents = "\n\t<!-- Contents of file $inc_rel -->\n";
      $file_contents .= <INC>;		# Read entire file as a single string
      $files{$inc_abs} = $file_contents;
    } else {
      warn("Could not read include file $inc_abs: $!\n");
    }
  }
  $file_contents =~ s/\$(\w+)/$args{$1}/eg;	# Insert args in result
  $file_contents;
}