chkspaces

#!/usr/local/bin/perl
# Check for lack of space before a start tag and
# after and end tag.
# Ian Jacobs
# $Id: chkspaces,v 1.10 2009-04-20 13:40:26 bbos Exp $

$PROGNAME = substr($0, rindex($0, "/") + 1);

# copy file in memory
sub readfile {
    $buf = "";
    if (!open(INPUT, $_[0])) {
	die "$PROGNAME Error: Cannot open file: $_[0]\n";
    }
    while (<INPUT>) {
	$buf .= $_;
    }
    close(INPUT);
}

# read reference database
$pattern1 = "(<\\/[^>]+>\\w+)";
$pattern2 = "(\\w(?:<(\\w+)\\s*[^>]*>)+)";
# Check for duplicate words.
$pattern3 = "(\\b([a-z]\\w*)(?:\\s+|\\s*(?:<[^>]+>)+\\s*|\\s*(?:<\\/[^>]+>)+\\s*)\\2\\b)";

foreach $file (@ARGV) {
    print STDOUT "Checking $file\n";
    if (readfile($file)) {
	$_ = $buf;
	while (/$pattern1/sio) {
	    $continue = $';
	    print STDOUT "\tMissing space?\t$1\n";
	    $_ = $continue;
	}
	while (/$pattern2/sio) {
	    $continue = $';
	    $match = $1;
	    $elem = $2;
	    if ( $2 =~ /br|th|td|sup|sub|ol|ul|li|dl|dt|dd/i ) {
		print STDOUT "\tMissing space?\t$match (Possibly harmless for $elem)\n";
	    } else {
		print STDOUT "\tMissing space?\t$match\n";
	    }
	    $_ = $continue;
	}
	while (/$pattern3/si) {
	    $continue = $';
	    $word = $2;
	    $context = $1;
	    $context =~ s/\n/\\n/g;
	    print STDOUT "\tDuplicate word?\t`$word' in `$context'\n";
	    $_ = $continue;
	}
	while (/((?:\w([\(\)])\w)|(?:\w([.?:\"'])[A-Z]))/s) {
	    $continue = $';
	    $word = "$2$3";
	    $context = $1;
	    $context =~ s/\n/\\n/g;
	    print STDOUT "\tMissing space around punc?\t`$word' in `$context'\n";
	    $_ = $continue;
	}
    }
}