#!perl -w

# Copyright (c) 2009 Symbian Foundation Ltd

# This component and the accompanying materials are made available

# under the terms of the License "Eclipse Public License v1.0"

# which accompanies this distribution, and is available

# at the URL "http://www.eclipse.org/legal/epl-v10.html".

#

# Initial Contributors:

# Symbian Foundation Ltd - initial contribution.

# 

# Contributors:

#

# Description:

#

use strict;

use File::Find;

use File::Copy;

use Cwd;

sub diffstat();

my $Logs_Dir = $ARGV[0];

my $dir_left = $ARGV[1];

my $dir_right = $ARGV[2];

my $dir_tmp_left = $ARGV[0].'\\'.$ARGV[1];

my $dir_tmp_right = $ARGV[0].'\\'.$ARGV[2];

print "left changeset $dir_left\n";

print "right chnageset $dir_right\n";

mkdir $dir_tmp_left;

mkdir $dir_tmp_right;

# default inclusions from churn.pl are "*.cpp", "*.c", "*.cxx", "*.h", "*.hpp", "*.inl" 

my @file_pattern=('\.cpp$','\.c$','\.hpp$','\.h$','\.inl$','\.cxx$','\.hrh$');

my $totallinecount=0;

my $countcomments=0;

if (! -d $Logs_Dir)

{

    die("$Logs_Dir does not exist \n");

}

$dir_left =~ m/^(\w+)\.[0-9a-fA-F]+/;

my $package_name = $1;

$dir_left =~ m/^\w+\.([0-9a-fA-F]+)/;

my $changeset_left = $1;

$dir_right =~ m/^\w+\.([0-9a-fA-F]+)/;

my $changeset_right = $1;

print "\nWorking on package: $package_name\n";

print "\nProcessing $dir_left\n";

find(\&process_files, $dir_left);

#DEBUG INFO:

print "\nTotal linecount for changed files in $dir_left is $totallinecount\n";

my $code_size_left = $totallinecount;

$totallinecount=0;

print "\nProcessing $dir_right\n";

find(\&process_files, $dir_right);

#DEBUG INFO:

print "\nTotal linecount for changed files in $dir_right is $totallinecount\n";    

my $code_size_right = $totallinecount;

my @diffs;

if (-d $dir_tmp_left && -d $dir_tmp_left)

{

	@diffs = `diff -r -N $dir_tmp_left $dir_tmp_right`;

}

my $changed_lines=@diffs;

my $diffsfile = $Logs_Dir.'\\'."dirdiffs.out";

open (DIFFS, ">$diffsfile");

print DIFFS @diffs;

close (DIFFS);

diffstat();

$dir_tmp_left =~ s{/}{\\}g;

$dir_tmp_right =~ s{/}{\\}g;

if (-d $dir_tmp_left)

{

	system("rmdir /S /Q $dir_tmp_left");

}

if (-d $dir_tmp_right)

{

system("rmdir /S /Q $dir_tmp_right");

}

unlink $diffsfile;

unlink "$Logs_Dir\\line_count_newdir.txt";

print "\n** Finished processing $package_name **\n\n\n\n\n";

exit(0);

sub diffstat()

{

open (DIFFSFILE,"$diffsfile");

my $curfile = "";

my %changes = ();

while (<DIFFSFILE>)

{

	my $line = $_;

				# diff -r -N D:/mirror\fbf_churn_output\commsfw.000000000000\serialserver\c32serialserver\Test\te_C32Performance\USB PC Side Code\resource.h 

				# diff -r <anything><changeset(12 chars)><slash><full_filename><optional_whitespace><EOL>

	if ($line =~ m/^diff -r.*\.[A-Fa-f0-9]{12}[\/\\](.*)\s*$/)

	{

		$curfile = $1;

		#DEBUG INFO:

		#print "\t$curfile\n";

		if (!defined $changes{$curfile})

		{

			$changes{$curfile} = {'a'=>0,'c'=>0,'d'=>0,'filetype'=>'unknown'};

		}

		$curfile =~ m/\.(\w+)$/g;

		#if filetype known...

		my $filetype = $+;

		$changes{$curfile}->{'filetype'}=uc($filetype);

	}

	elsif ($line =~ m/^(\d+)(,(\d+))?(d)\d+(,\d+)?/)

	{	

		if (defined $3)

		{

			$changes{$curfile}->{$4} += ($3-$1)+1;

		}

		else

		{

			$changes{$curfile}->{$4}++;

		}

	}

	elsif ($line =~ m/^\d+(,\d+)?([ac])(\d+)(,(\d+))?/)

	{	

		if (defined $5)

		{

			$changes{$curfile}->{$2} += ($5-$3)+1;

		}

		else

		{

			$changes{$curfile}->{$2}++;

		}	

	}

}

close (DIFFSFILE);

my %package_changes = ("CPP"=>0, "H"=>0, "HPP"=>0, "INL"=>0, "C"=>0, "CXX"=>0,"HRH"=>0,);

my %package_deletions = ("CPP"=>0, "H"=>0, "HPP"=>0, "INL"=>0, "C"=>0, "CXX"=>0,"HRH"=>0,);

my %package_additions = ("CPP"=>0, "H"=>0, "HPP"=>0, "INL"=>0, "C"=>0, "CXX"=>0,"HRH"=>0,);

my $package_churn = 0;

for my $file (keys %changes)

{

	$package_changes{$changes{$file}->{'filetype'}} += $changes{$file}->{'c'};

	$package_deletions{$changes{$file}->{'filetype'}} += $changes{$file}->{'d'};

	$package_additions{$changes{$file}->{'filetype'}} += $changes{$file}->{'a'};

}

#DEBUG INFO: For printing contents of hashes containing per filetype summary

#print "\n\n\n\n";

#print "package_changes:\n";

#print map { "$_ => $package_changes{$_}\n" } keys %package_changes;

#print "\n\n\n\n";

#print "package_deletions:\n";

#print map { "$_ => $package_deletions{$_}\n" } keys %package_deletions;

#print "\n\n\n\n";

#print "package_additions:\n";

#print map { "$_ => $package_additions{$_}\n" } keys %package_additions;

my $overall_changes = 0;

for my $filetype (keys %package_changes)

{

	$overall_changes += $package_changes{$filetype};

}

my $overall_deletions = 0;

for my $filetype (keys %package_deletions)

{

	$overall_deletions += $package_deletions{$filetype};

}

my $overall_additions = 0;

for my $filetype (keys %package_additions)

{

	$overall_additions += $package_additions{$filetype};

}

$package_churn = $overall_changes + $overall_additions;

print "\n\n\n\nSummary for Package: $package_name\n";

print "-------------------\n";

print "Changesets Compared: $dir_left and $dir_right\n";

#print "Code Size for $dir_left = $code_size_left lines\n";

#print "Code Size for $dir_right = $code_size_right lines\n";

print "Total Lines Changed = $overall_changes\n";

print "Total Lines Added = $overall_additions\n";

print "Total Lines Deleted = $overall_deletions\n";

print "Package Churn = $package_churn lines\n";

my @header = qw(filetype a c d);

my $outputfile = $Logs_Dir.'\\'."$package_name\_diffstat.csv";

open(PKGSTATCSV, ">$outputfile") or die "Coudln't open $outputfile";

print PKGSTATCSV " SF CODE-CHURN SUMMARY\n";

print PKGSTATCSV "Package: $package_name\n";

print PKGSTATCSV "Changesets Compared: $dir_left and $dir_right\n";

#print PKGSTATCSV "Code Size for $dir_left = $code_size_left lines\n";

#print PKGSTATCSV "Code Size for $dir_right = $code_size_right lines\n";

print PKGSTATCSV "Total Lines Changed = $overall_changes\n";

print PKGSTATCSV "Total Lines Added = $overall_additions\n";

print PKGSTATCSV "Total Lines Deleted = $overall_deletions\n";

print PKGSTATCSV "Package Churn = $package_churn lines\n\n\n\n\n";

# print the header

print PKGSTATCSV "FILENAME,";

foreach my $name (@header)

{

  if ($name eq 'filetype')

  {

	print PKGSTATCSV uc($name).",";

  }  

  elsif ($name eq 'a')

 {

	print PKGSTATCSV "LINES_ADDED,";

 }

  elsif ($name eq 'c')

 {

	print PKGSTATCSV "LINES_CHANGED,";

 }

  elsif ($name eq 'd')

 {

	print PKGSTATCSV "LINES_DELETED,";

 }

}

print PKGSTATCSV "\n";

foreach my $file (sort keys %changes)

{

  print PKGSTATCSV $file.",";

  foreach my $key (@header)

  {

    if(defined $changes{$file}->{$key})

    {

      print PKGSTATCSV $changes{$file}->{$key};

    }

    print PKGSTATCSV ",";

  }

  print PKGSTATCSV "\n";

}

close (PKGSTATCSV);

my $diffstat_summary = $Logs_Dir.'\\'."diffstat_summary.csv";

if (-e $diffstat_summary)

{ 

	open(DIFFSTATCSV, ">>$diffstat_summary") or die "Coudln't open $outputfile";

	print DIFFSTATCSV "$package_name,";

	print DIFFSTATCSV "$changeset_left,";

	print DIFFSTATCSV "$changeset_right,";

	#print DIFFSTATCSV ",";

	foreach my $filetype (sort keys %package_changes)

	{

		if(defined $package_changes{$filetype})

		{

		  print DIFFSTATCSV $package_changes{$filetype}.",";

		}

	}

	#print DIFFSTATCSV ",";

	foreach my $filetype (sort keys %package_additions)

	{

		if(defined $package_additions{$filetype})

		{

		  print DIFFSTATCSV $package_additions{$filetype}.",";

		}

	}

	#print DIFFSTATCSV ",";

	foreach my $filetype (sort keys %package_deletions)

	{

		if(defined $package_deletions{$filetype})

		{

		  print DIFFSTATCSV $package_deletions{$filetype}.",";

		  #print DIFFSTATCSV ",";

		}

	}

	#print DIFFSTATCSV ",";

	print DIFFSTATCSV "$overall_changes,";

	print DIFFSTATCSV "$overall_additions,";

	print DIFFSTATCSV "$overall_deletions,";

	print DIFFSTATCSV "$package_churn,";

	print DIFFSTATCSV "\n";

	close (DIFFSTATCSV);

}

else

{

	open(DIFFSTATCSV, ">$diffstat_summary") or die "Couldn't open $outputfile";

	# print the header

	print DIFFSTATCSV "PACKAGE_NAME,";

	print DIFFSTATCSV "LEFT_CHANGESET,";

	print DIFFSTATCSV "RIGHT_CHANGESET,";

	#print DIFFSTATCSV ",";

	foreach my $name (sort keys %package_changes)

	{

		print DIFFSTATCSV $name." CHANGES,";    

	}

	#print DIFFSTATCSV ",";

	foreach my $name (sort keys %package_additions)

	{

		print DIFFSTATCSV $name." ADDITIONS,";    

	}

	#print DIFFSTATCSV ",";

	foreach my $name (sort keys %package_deletions)

	{

		print DIFFSTATCSV $name." DELETIONS,";    

	}

	#print DIFFSTATCSV ",";

	print DIFFSTATCSV "PACKAGE_CHANGES,";

	print DIFFSTATCSV "PACKAGE_ADDITIONS,";

	print DIFFSTATCSV "PACKAGE_DELETIONS,";

	print DIFFSTATCSV "PACKAGE_CHURN,";

	print DIFFSTATCSV "\n";

	print DIFFSTATCSV "$package_name,";

	print DIFFSTATCSV "$changeset_left,";

	print DIFFSTATCSV "$changeset_right,";

	#print DIFFSTATCSV ",";

	foreach my $filetype (sort keys %package_changes)

	{

		if(defined $package_changes{$filetype})

		{

		  print DIFFSTATCSV $package_changes{$filetype}.",";

		}

	}

	#print DIFFSTATCSV ",";

	foreach my $filetype (sort keys %package_additions)

	{

		if(defined $package_additions{$filetype})

		{

		  print DIFFSTATCSV $package_additions{$filetype}.",";

		}

	}

	#print DIFFSTATCSV ",";

	foreach my $filetype (sort keys %package_deletions)

	{

		if(defined $package_deletions{$filetype})

		{

		  print DIFFSTATCSV $package_deletions{$filetype}.",";

		}

	}

	#print DIFFSTATCSV ",";

	print DIFFSTATCSV "$overall_changes,";

	print DIFFSTATCSV "$overall_additions,";

	print DIFFSTATCSV "$overall_deletions,";

	print DIFFSTATCSV "$package_churn,";

	print DIFFSTATCSV "\n";

	close (DIFFSTATCSV);

}

}

sub process_files() 

{

    my $lfile = $_;

    my $lfile_fullpath=$File::Find::name;

    $lfile_fullpath =~ s#\/#\\#g;

    #print "$lfile\t\tFull path $lfile_fullpath\n" ;

    if (-f $lfile)

    { 

        foreach my $regpat (@file_pattern)

        {

            if (lc($lfile) =~ m/$regpat/)

            {

                $lfile  =~ s#\/#\\#g;

                #print "Processing file $lfile (Matched $regpat) \n"; #ck

                #print `type $lfile`;

                # We copy mathching files to a separate temp directory

                # so that the final diff can simply diff the full dir

                # Note :  RemoveNoneLOC routine edits the file in-situ.

                my $lfile_abs = cwd().'\\'.$lfile;

                my $lfile_local = $Logs_Dir.'\\'.$lfile_fullpath;

                makepath($lfile_local);

                print "%";

                copy($lfile_abs,$lfile_local);

				$totallinecount += RemoveNonLOC( $lfile, $lfile_local, "newdir" );

            }

        }

    }   

}

sub makepath()

{

    my $absfile = shift; 

    $absfile =~ s#\\#\/#g;

    my @dirs = split /\//, $absfile;

    pop @dirs;  # throw away the filename

    my $path = "";

    foreach my $dir (@dirs)

    {

        $path = ($path eq "") ? $dir : "$path/$dir";

        if (!-d $path)

        {

#          print "making $path \n";

          mkdir $path;

        }

    }

}

sub RemoveNonLOC($$$) {

    # Gather arguments

    my $file = shift;

    my $original_file  = shift;

    my $type_of_dir = shift;

#    print("\nDebug: in ProcessFile, file is $file, full file + path is $original_file \n");

	# Remove comments...

    # Set up the temporary files that will be used to perform the processing steps

    my $temp1File = $original_file."temp1";

    my $temp2File = $original_file."temp2";

    open(TEMP1, "+>$temp1File");

    if (!($countcomments)) {

     	# Remove any comments from the file

		my $original_file_string;

     	open INPUT, "<", $original_file;

		{

			local $/ = undef;

			$original_file_string = <INPUT>;

		}

		close INPUT;

     	my $dbl = qr/"[^"\\]*(?:\\.[^"\\]*)*"/s;

        my $sgl = qr/'[^'\\]*(?:\\.[^'\\]*)*'/s;

        my $C   = qr{/\*.*?\*/}s; # C style comments /*  */

        my $CPP = qr{//.*}; # C+ style comments //

        my $com = qr{$C|$CPP};

        my $other = qr{.[^/"'\\]*}s; # all other '"

        my $keep = qr{$sgl|$dbl|$other};

     	#Remove the comments (need to turn off warnings on the next regexp for unititialised variable)

no warnings 'uninitialized';

        $original_file_string=~ s/$com|($keep)/$1/gom;  

        print TEMP1 "$original_file_string";

use warnings 'uninitialized';

    }

    else {

        print("\n option --CountComments specified so comments will be included in the count\n");

        #Just copy over original with comments still in it

		copy($original_file,$temp1File); 

    }

    close(TEMP1);

    # Remove blank lines...

#   print("\nDebug: Getting rid of blank lines in \n$temp1File to produce \n$temp2File \n");

    open (TEMP1, "+<$temp1File"); # include lines + pre-processed code

    open (TEMP2, "+>$temp2File"); 

    while (<TEMP1>) {