ppcrcd/trunk/doc/gendoc.pl

Fri Mar 24 14:04:25 CET 2006

Author: sparky
Date: Fri Mar 24 14:02:33 2006
New Revision: 7304

Modified:
   ppcrcd/trunk/doc/gendoc.pl
Log:
- code to cleanup html iteration


Modified: ppcrcd/trunk/doc/gendoc.pl
==============================================================================

--- ppcrcd/trunk/doc/gendoc.pl	(original)
+++ ppcrcd/trunk/doc/gendoc.pl	Fri Mar 24 14:02:33 2006
@@ -123,6 +123,7 @@
 }
 
 my $content = adddoc(shift @ARGV);
+$content = cleanhtml($content);
 print <<EOF;
 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
    "http://www.w3.org/TR/html4/loose.dtd">
@@ -144,9 +145,116 @@
 		</style>
 	</head>
 	<body>
+	
 $content
 	</body>
 </html>
 EOF
 
+sub paragraph($$);
+my $text_to_clean;
+sub cleanhtml {
+	$text_to_clean = "$_[0]</end>";
+	return paragraph("","end");
+}
+
+sub gettag() {
+	$text_to_clean =~ s/^\s*<(.*?)>//;
+	return $1;
+}
+sub getword() {
+	$text_to_clean =~ s/^\s*([^<\s]*)//;
+	return $1;
+}
+
+sub paragraph($$) {
+	my ($step, $end) = @_;
+	my $clean="";
+	my $line=$step;
+	while (length $text_to_clean) {
+		my $tag = gettag();
+		my $word ="";
+		if ( $tag ) {
+			( my $line_nosp = $line ) =~ s/\s*//g;
+			if ( $tag eq "/$end" ) {
+				$clean .= $line."\n" if length $line_nosp;
+				return $clean;
+			}
+			if ( $tag eq "div" or $tag eq "p" ) {
+				$clean .= $line."\n" if length $line_nosp;
+				$line = $step;
+				$clean .= $step."<$tag>\n";
+				$clean .= paragraph($step."\t", $tag);
+				$clean .= $step."</$tag>\n";
+				next;
+			}
+			if ( $tag =~ /^[biu]$/ ) {
+				$word = "<$tag>" . getword();
+				$tag = "";
+			}
+			if ( $tag =~ /^\/[biu]$/ ) {
+				$word = "<$tag>";
+				$tag = "";
+			}
+			if ( $tag =~ /^h[1-6]$/ ) {
+				$clean .= $line."\n" if length $line_nosp;
+				$clean .= "\n" unless length $line;
+				$line = paragraph($step."     ", $tag);
+				$line =~ s/^\s*//;
+				chomp $line;
+				$clean .= $step . "<$tag>" . $line . "</$tag>\n";
+				$line = $step;
+				next;
+			}
+			if ( $tag eq "pre" ) {
+				$text_to_clean =~ s#^(.*?)</pre>##s;
+				my $pre = $1;
+				if ( $pre =~ /\n/ ) { #multiline
+					$clean .= $line."\n" if length $line_nosp;
+					$clean .= $step . "<pre>\n";
+					$clean .= $pre . "\n";
+					$clean .= $step . "</pre>\n";
+					$line = $step;
+					next;
+				} else {
+					$clean .= $line."\n" if length $line_nosp;
+					$clean .= $step. "<pre>" . $pre . "</pre>\n";
+					$line = $step;
+					next;
+				}
+			}
+			if ( $tag =~ /^hr\s*\/$/ ) {
+				$clean .= $line . "\n" if length $line_nosp;
+				$line = $step;
+				$clean .= $step . "<hr />\n";
+				next;
+			}
+			if ( $tag =~ /^br\s*\/$/ ) {
+				$clean .= $line if length $line_nosp;
+				$clean .= "<br />\n";
+				$line = $step;
+				next;
+			}
+			if ( length $tag ) {
+				warn "Unknown tag: <$tag>\n";
+				$word = "<$tag>";
+			}
+		}
+		$word = getword() unless length $word;
+		last unless $word;
+		( my $tmpline = $line ) =~ s/\t/        /g;
+		if ( length ($tmpline . " " . $word) >= 80 ) {
+			$clean .= $line . "\n";
+			$line = $step . $word;
+		} else {
+			$line .= " " . $word;
+		}
+	}
+	warn "should never reach this point";
+	if ( length $line ) {
+		$clean .= $line . "\n";
+	}
+	return $clean;
+}
+
 # vim: ts=4:sw=4
