1#!/usr/local/bin/perl -w 2# asm_count - count physical lines of code in Assembly programs. 3# Usage: asm_count [-f file] [list_of_files] 4# file: file with a list of files to count (if "-", read list from stdin) 5# list_of_files: list of files to count 6# -f file or list_of_files can be used, or both 7# This is a trivial/naive program. 8 9# For each file, it looks at the contents to heuristically determine 10# if C comments are permitted and what the "comment" character is. 11# If /* and */ are in the file, then C comments are permitted. 12# The punctuation mark that starts the most lines must be the comment 13# character (but ignoring "/" if C comments are allowed, and 14# ignoring '#' if cpp commands appear to be used) 15 16# This is part of SLOCCount, a toolsuite that counts 17# source lines of code (SLOC). 18# Copyright (C) 2001-2004 David A. Wheeler. 19# 20# This program is free software; you can redistribute it and/or modify 21# it under the terms of the GNU General Public License as published by 22# the Free Software Foundation; either version 2 of the License, or 23# (at your option) any later version. 24# 25# This program is distributed in the hope that it will be useful, 26# but WITHOUT ANY WARRANTY; without even the implied warranty of 27# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 28# GNU General Public License for more details. 29# 30# You should have received a copy of the GNU General Public License 31# along with this program; if not, write to the Free Software 32# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 33# 34# To contact David A. Wheeler, see his website at: 35# http://www.dwheeler.com. 36 37 38 39$total_sloc = 0; 40 41# Do we have "-f" (read list of files from second argument)? 42if (($#ARGV >= 1) && ($ARGV[0] eq "-f")) { 43 # Yes, we have -f 44 if ($ARGV[1] eq "-") { 45 # The list of files is in STDIN 46 while (<STDIN>) { 47 chomp ($_); 48 &count_file ($_); 49 } 50 } else { 51 # The list of files is in the file $ARGV[1] 52 open (FILEWITHLIST, $ARGV[1]) || die "Error: Could not open $ARGV[1]\n"; 53 while (<FILEWITHLIST>) { 54 chomp ($_); 55 &count_file ($_); 56 } 57 close FILEWITHLIST; 58 } 59 shift @ARGV; shift @ARGV; 60} 61# Process all (remaining) arguments as file names 62while ($file = shift @ARGV) { 63 &count_file ($file); 64} 65 66print "Total:\n"; 67print "$total_sloc\n"; 68 69sub count_file { 70 my ($file) = @_; 71 # First, use heuristics to determine the comment char and if it uses C comments 72 $found_c_start = 0; 73 $found_c_end = 0; 74 $cpp_suspicious = 0; 75 $cpp_likely = 0; 76 $cpp_used = 0; 77 %count = (); 78 if ($file eq "") { 79 *CURRENTFILE = *STDIN 80 } else { 81 open(CURRENTFILE, "<$file"); 82 } 83 while (<CURRENTFILE>) { 84 if (m!\/\*!) { $found_c_start++;} 85 if (m!\*\/!) { $found_c_end++;} 86 if ( (m!^#\s*define\s!) || (m!^#\s*else!)) {$cpp_suspicious++;} 87 if ( (m!^#\s*ifdef\s!) || (m!^#\s*endif!) || (m!#\s*include!)) {$cpp_likely++;} 88 if (m/^\s*([;!\/#\@\|\*])/) { $count{$1}++; } # Found a likely comment char. 89 } 90 # Done examing file, let's figure out the parameters. 91 if ($found_c_start && $found_c_end) { 92 $ccomments = 1; 93 $count{'/'} = 0; 94 # $count{'*'} = 0; # Do this to ignore '*' if C comments are used. 95 } else { 96 $ccomments = 0; 97 } 98 if (($cpp_suspicious > 2) || ($cpp_likely >= 1)) { 99 $cpp_used = 1; 100 $count{'#'} = 0; 101 } else { 102 $cpp_used = 0; 103 } 104 $likeliest = ';'; 105 $likeliest_count = 0; 106 foreach $i (keys(%count)) { 107 # print "DEBUG: key=$i count=$count{$i}\n"; 108 if ($count{$i} > $likeliest_count) { 109 $likeliest = $i; 110 $likeliest_count = $count{$i}; 111 } 112 } 113 # print "DEBUG: likeliest = $likeliest\n"; 114 $commentchar=$likeliest; 115 close(CURRENTFILE); 116 117 # Now count SLOC. 118 $sloc = 0; 119 $isincomment = 0; 120 open(CURRENTFILE, "<$file"); 121 while (<CURRENTFILE>) { 122 # We handle C comments first, so that if an EOL-comment 123 # occurs inside a C comment, it's ignored. 124 if ($ccomments) { 125 # Handle C /* */ comments; this will get fooled if they're in strings, 126 # but that would be rare in assembly. 127 while ( (m!\/\*!) || (m!\*\/!)) { # While unprocessed C comment. 128 if ($isincomment) { 129 s!.*?\*\/.*!!; 130 $isincomment = 0; 131 } else { # Not in C comment, but have end comment marker. 132 if (! m/\/\*/) { # Whups, there's no starting marker! 133 print STDERR "Warning: file $file line $. has unmatched comment end\n"; 134 # Get us back to a plausible state: 135 s/.*//; # Destroy everything 136 $isincomment = 0; 137 } else { 138 if (! s!\/\*.*?\*\/!!) { # Try to delete whole comment. 139 # We couldn't delete whole comment. Delete what's there. 140 s!\/\*.*!!; 141 $isincomment = 1; 142 } 143 } 144 } 145 } 146 } # End of handling C comments. 147 # This requires $[ be unchanged. 148 $locate_comment = index($_, $commentchar); 149 if ($locate_comment >= 0) { # We found a comment character, delete comment 150 $_ = substr($_, 0, $locate_comment); 151 # print "DEBUG New text: @",$_,"@\n"; 152 } 153 # old: s/${commentchar}.*//; # Delete leading comments. 154 155 # FOR DEBUG: print "Finally isincomment=$isincomment line=$_\n"; 156 if ((! $isincomment) && (m/\S/)) {$sloc++;} 157 } 158 159 # End-of-file processing 160 print "$sloc (commentchar=$commentchar C-comments=$ccomments) $file\n"; 161 $total_sloc += $sloc; 162 $sloc = 0; 163 if ($isincomment) { 164 print STDERR "Missing comment close in file $file\n"; 165 } 166} 167