#!/usr/bin/perl -w
#
# Clean a patch file -- or directory of patch files -- of stealth whitespace.
# WARNING: this can be a highly destructive operation.  Use with caution.
#

use bytes;
use File::Basename;

# Default options
$max_width = 79;

# Clean up space-tab sequences, either by removing spaces or
# replacing them with tabs.
sub clean_space_tabs($)
{
    no bytes;			# Tab alignment depends on characters

    my($li) = @_;
    my($lo) = '';
    my $pos = 0;
    my $nsp = 0;
    my($i, $c);

    for ($i = 0; $i < length($li); $i++) {
	$c = substr($li, $i, 1);
	if ($c eq "\t") {
	    my $npos = ($pos+$nsp+8) & ~7;
	    my $ntab = ($npos >> 3) - ($pos >> 3);
	    $lo .= "\t" x $ntab;
	    $pos = $npos;
	    $nsp = 0;
	} elsif ($c eq "\n" || $c eq "\r") {
	    $lo .= " " x $nsp;
	    $pos += $nsp;
	    $nsp = 0;
	    $lo .= $c;
	    $pos = 0;
	} elsif ($c eq " ") {
	    $nsp++;
	} else {
	    $lo .= " " x $nsp;
	    $pos += $nsp;
	    $nsp = 0;
	    $lo .= $c;
	    $pos++;
	}
    }
    $lo .= " " x $nsp;
    return $lo;
}

# Compute the visual width of a string
sub strwidth($) {
    no bytes;			# Tab alignment depends on characters

    my($li) = @_;
    my($c, $i);
    my $pos = 0;
    my $mlen = 0;

    for ($i = 0; $i < length($li); $i++) {
	$c = substr($li,$i,1);
	if ($c eq "\t") {
	    $pos = ($pos+8) & ~7;
	} elsif ($c eq "\n") {
	    $mlen = $pos if ($pos > $mlen);
	    $pos = 0;
	} else {
	    $pos++;
	}
    }

    $mlen = $pos if ($pos > $mlen);
    return $mlen;
}

$name = basename($0);

@files = ();

while (defined($a = shift(@ARGV))) {
    if ($a =~ /^-/) {
	if ($a eq '-width' || $a eq '-w') {
	    $max_width = shift(@ARGV)+0;
	} else {
	    print STDERR "Usage: $name [-width #] files...\n";
	    exit 1;
	}
    } else {
	push(@files, $a);
    }
}

foreach $f ( @files ) {
    print STDERR "$name: $f\n";

    if (! -f $f) {
	print STDERR "$f: not a file\n";
	next;
    }

    if (!open(FILE, '+<', $f)) {
	print STDERR "$name: Cannot open file: $f: $!\n";
	next;
    }

    binmode FILE;

    # First, verify that it is not a binary file; consider any file
    # with a zero byte to be a binary file.  Is there any better, or
    # additional, heuristic that should be applied?
    $is_binary = 0;

    while (read(FILE, $data, 65536) > 0) {
	if ($data =~ /\0/) {
	    $is_binary = 1;
	    last;
	}
    }

    if ($is_binary) {
	print STDERR "$name: $f: binary file\n";
	next;
    }

    seek(FILE, 0, 0);

    $in_bytes = 0;
    $out_bytes = 0;
    $lineno = 0;

    @lines  = ();

    $in_hunk = 0;
    $err = 0;

    while ( defined($line = <FILE>) ) {
	$lineno++;
	$in_bytes += length($line);

	if (!$in_hunk) {
	    if ($line =~
		/^\@\@\s+\-([0-9]+),([0-9]+)\s+\+([0-9]+),([0-9]+)\s\@\@/) {
		$minus_lines = $2;
		$plus_lines = $4;
		if ($minus_lines || $plus_lines) {
		    $in_hunk = 1;
		    @hunk_lines = ($line);
		}
	    } else {
		push(@lines, $line);
		$out_bytes += length($line);
	    }
	} else {
	    # We're in a hunk

	    if ($line =~ /^\+/) {
		$plus_lines--;

		$text = substr($line, 1);
		$text =~ s/[ \t\r]*$//;		# Remove trailing spaces
		$text = clean_space_tabs($text);

		$l_width = strwidth($text);
		if ($max_width && $l_width > $max_width) {
		    print STDERR
			"$f:$lineno: adds line exceeds $max_width ",
			"characters ($l_width)\n";
		}

		push(@hunk_lines, '+'.$text);
	    } elsif ($line =~ /^\-/) {
		$minus_lines--;
		push(@hunk_lines, $line);
	    } elsif ($line =~ /^ /) {
		$plus_lines--;
		$minus_lines--;
		push(@hunk_lines, $line);
	    } else {
		print STDERR "$name: $f: malformed patch\n";
		$err = 1;
		last;
	    }

	    if ($plus_lines < 0 || $minus_lines < 0) {
		print STDERR "$name: $f: malformed patch\n";
		$err = 1;
		last;
	    } elsif ($plus_lines == 0 && $minus_lines == 0) {
		# End of a hunk.  Process this hunk.
		my $i;
		my $l;
		my @h = ();
		my $adj = 0;
		my $done = 0;

		for ($i = scalar(@hunk_lines)-1; $i > 0; $i--) {
		    $l = $hunk_lines[$i];
		    if (!$done && $l eq "+\n") {
			$adj++; # Skip this line
		    } elsif ($l =~ /^[ +]/) {
			$done = 1;
			unshift(@h, $l);
		    } else {
			unshift(@h, $l);
		    }
		}

		$l = $hunk_lines[0];  # Hunk header
		undef @hunk_lines;    # Free memory

		if ($adj) {
		    die unless
			($l =~ /^\@\@\s+\-([0-9]+),([0-9]+)\s+\+([0-9]+),([0-9]+)\s\@\@(.*)$/);
		    my $mstart = $1;
		    my $mlin = $2;
		    my $pstart = $3;
		    my $plin = $4;
		    my $tail = $5; # doesn't include the final newline

		    $l = sprintf("@@ -%d,%d +%d,%d @@%s\n",
				 $mstart, $mlin, $pstart, $plin-$adj,
				 $tail);
		}
		unshift(@h, $l);

		# Transfer to the output array
		foreach $l (@h) {
		    $out_bytes += length($l);
		    push(@lines, $l);
		}

		$in_hunk = 0;
	    }
	}
    }

    if ($in_hunk) {
	print STDERR "$name: $f: malformed patch\n";
	$err = 1;
    }

    if (!$err) {
	if ($in_bytes != $out_bytes) {
	    # Only write to the file if changed
	    seek(FILE, 0, 0);
	    print FILE @lines;

	    if ( !defined($where = tell(FILE)) ||
		 !truncate(FILE, $where) ) {
		die "$name: Failed to truncate modified file: $f: $!\n";
	    }
	}
    }

    close(FILE);
}
4'>64</a>
<a id='n65' href='#n65'>65</a>
<a id='n66' href='#n66'>66</a>
<a id='n67' href='#n67'>67</a>
<a id='n68' href='#n68'>68</a>
<a id='n69' href='#n69'>69</a>
<a id='n70' href='#n70'>70</a>
<a id='n71' href='#n71'>71</a>
<a id='n72' href='#n72'>72</a>
<a id='n73' href='#n73'>73</a>
<a id='n74' href='#n74'>74</a>
<a id='n75' href='#n75'>75</a>
<a id='n76' href='#n76'>76</a>
<a id='n77' href='#n77'>77</a>
<a id='n78' href='#n78'>78</a>
<a id='n79' href='#n79'>79</a>
<a id='n80' href='#n80'>80</a>
<a id='n81' href='#n81'>81</a>
<a id='n82' href='#n82'>82</a>
<a id='n83' href='#n83'>83</a>
<a id='n84' href='#n84'>84</a>
<a id='n85' href='#n85'>85</a>
<a id='n86' href='#n86'>86</a>
<a id='n87' href='#n87'>87</a>
<a id='n88' href='#n88'>88</a>
<a id='n89' href='#n89'>89</a>
<a id='n90' href='#n90'>90</a>
<a id='n91' href='#n91'>91</a>
<a id='n92' href='#n92'>92</a>
<a id='n93' href='#n93'>93</a>
<a id='n94' href='#n94'>94</a>
<a id='n95' href='#n95'>95</a>
<a id='n96' href='#n96'>96</a>
<a id='n97' href='#n97'>97</a>
<a id='n98' href='#n98'>98</a>
<a id='n99' href='#n99'>99</a>
<a id='n100' href='#n100'>100</a>
<a id='n101' href='#n101'>101</a>
<a id='n102' href='#n102'>102</a>
<a id='n103' href='#n103'>103</a>
<a id='n104' href='#n104'>104</a>
<a id='n105' href='#n105'>105</a>
<a id='n106' href='#n106'>106</a>
<a id='n107' href='#n107'>107</a>
<a id='n108' href='#n108'>108</a>
<a id='n109' href='#n109'>109</a>
<a id='n110' href='#n110'>110</a>
<a id='n111' href='#n111'>111</a>
<a id='n112' href='#n112'>112</a>
<a id='n113' href='#n113'>113</a>
<a id='n114' href='#n114'>114</a>
<a id='n115' href='#n115'>115</a>
<a id='n116' href='#n116'>116</a>
<a id='n117' href='#n117'>117</a>
<a id='n118' href='#n118'>118</a>
<a id='n119' href='#n119'>119</a>
<a id='n120' href='#n120'>120</a>
<a id='n121' href='#n121'>121</a>
<a id='n122' href='#n122'>122</a>
<a id='n123' href='#n123'>123</a>
<a id='n124' href='#n124'>124</a>
<a id='n125' href='#n125'>125</a>
<a id='n126' href='#n126'>126</a>
<a id='n127' href='#n127'>127</a>
<a id='n128' href='#n128'>128</a>
<a id='n129' href='#n129'>129</a>
<a id='n130' href='#n130'>130</a>
<a id='n131' href='#n131'>131</a>
<a id='n132' href='#n132'>132</a>
<a id='n133' href='#n133'>133</a>
<a id='n134' href='#n134'>134</a>
<a id='n135' href='#n135'>135</a>
<a id='n136' href='#n136'>136</a>
<a id='n137' href='#n137'>137</a>
<a id='n138' href='#n138'>138</a>
<a id='n139' href='#n139'>139</a>
<a id='n140' href='#n140'>140</a>
<a id='n141' href='#n141'>141</a>
<a id='n142' href='#n142'>142</a>
<a id='n143' href='#n143'>143</a>
<a id='n144' href='#n144'>144</a>
<a id='n145' href='#n145'>145</a>
<a id='n146' href='#n146'>146</a>
<a id='n147' href='#n147'>147</a>
<a id='n148' href='#n148'>148</a>
<a id='n149' href='#n149'>149</a>
<a id='n150' href='#n150'>150</a>
<a id='n151' href='#n151'>151</a>
<a id='n152' href='#n152'>152</a>
<a id='n153' href='#n153'>153</a>
<a id='n154' href='#n154'>154</a>
<a id='n155' href='#n155'>155</a>
<a id='n156' href='#n156'>156</a>
<a id='n157' href='#n157'>157</a>
<a id='n158' href='#n158'>158</a>
<a id='n159' href='#n159'>159</a>
<a id='n160' href='#n160'>160</a>
<a id='n161' href='#n161'>161</a>
<a id='n162' href='#n162'>162</a>
<a id='n163' href='#n163'>163</a>
<a id='n164' href='#n164'>164</a>
<a id='n165' href='#n165'>165</a>
<a id='n166' href='#n166'>166</a>
<a id='n167' href='#n167'>167</a>
<a id='n168' href='#n168'>168</a>
<a id='n169' href='#n169'>169</a>
<a id='n170' href='#n170'>170</a>
<a id='n171' href='#n171'>171</a>
<a id='n172' href='#n172'>172</a>
<a id='n173' href='#n173'>173</a>
</pre></td>
<td class='lines'><pre><code>