) {
# chomp;
#
# if ( /\sRepository revision:\s+(\d+\.\d+)\s/ ) {
# $filerevision = $1;
# }
# }
#
# close (CVSREV);
# name should be defined
unless (defined($species{$sfile})) {
$species{$sfile} = $sfile;
# STDOUT outs to logfile, STDERR goes to cron user
print "Found $sfile, no description for this file available.\n";
warn "Found $sfile, no description for this file available.\n";
}
if ($adata[2] == 0 ) {
# removed files with no annotations. This includes
# goa_arabidopsis, goa_rat, goa_mouse, GeneDB_Spombe, and several others
next;
}
if ( $filteredfile ) {
if ($adata[2] > 9 ) {
print POUT "| $species{$sfile} | \n";
print POUT "$adata[2] | \n";
print POUT "$adata[0]
($adata[1] non-IEA) | \n";
print POUT "$filemoddate | \n";
print POUT " |
\n";
}
} else {
print PUNFOUT "| $species{$sfile} | \n";
print PUNFOUT "$adata[2] | \n";
print PUNFOUT "$adata[0]
($adata[1] non-IEA) | \n";
print PUNFOUT "$filemoddate | \n";
print PUNFOUT " |
\n";
}
}
close POUT;
close PUNFOUT;
#
# check is project counts changed using sum, if yes update
#
open (SUMOUT, "/usr/bin/cksum " . PCOUNTFILE . " " . DOCDIR . PROJECT_COUNT_FILE . " |") || &logdie ("cksum failed: $!\n");
my $newsum = 0;
my $cachesum = 0;
while ( ) {
chomp;
my (@sumoutput) = split(/\s+/);
if ($sumoutput[2] eq PCOUNTFILE) {
$newsum = $sumoutput[0];
} elsif ($sumoutput[2] eq DOCDIR . PROJECT_COUNT_FILE) {
$cachesum = $sumoutput[0];
}
}
unless ($newsum == $cachesum) {
# clone LOG filehandle to STDOUT and STDERR to capture output from cvs commit
open (STDOUT, ">&LOG") || &logdie ("Count not dup LOG filehandle: $!\n");
open (STDERR, ">&LOG") || &logdie ("Count not dup LOG filehandle: $!\n");
# tmp files moved to production directory, a sandbox of the GO CVS
move(PCOUNTFILE, DOCDIR . PROJECT_COUNT_FILE) || &logdie ("Cannot move " . PROJECT_COUNT_FILE . " into place in " . DOCDIR . "\n");
move(PUNFCOUNTFILE, DOCDIR . PROJECT_UNFILTER_COUNT_FILE) || &logdie ("Cannot move " . PROJECT_UNFILTER_COUNT_FILE . " into place in " . DOCDIR . "\n");
# get time, localtime starts with 1900
my ($mon, $day, $year) = (localtime())[4,3,5];
$year += 1900;
# converts abbreviated month name to full month name
my $month = $mon2month[$mon];
my $dstr = "$year/$month/$day";
# overwrite file with current date
open (DATE, "/bin/date '+%B %e, %Y' |") || &logdie ("Cannot run date command: $!\n");
open (DATEFILE, ">" . DOCDIR . PROJECT_DATESTAMP) || &logdie ("Cannot open project-datestamp.html: $!\n");
while ( ) {
chomp;
print DATEFILE "$_\n";
}
close (DATEFILE);
close (DATE);
# commit files
my $cmd = "cd " . DOCDIR . " && /usr/local/bin/cvs -d /share/go/cvs commit -m $dstr " . PROJECT_COUNT_FILE . " " . PROJECT_UNFILTER_COUNT_FILE . " " . DOCDIR . PROJECT_DATESTAMP;
my $status = system($cmd);
if ($status) { &logdie ("CVS commit failed: $!\n"); }
} else {
print "No change needed. File not changed and CVS commit not run.\n";
# no changes -- remove tmp files
unlink PCOUNTFILE;
unlink PUNFCOUNTFILE;
}
close LOG;
exit 0;
sub dodir {
my ($dir,$nlink) = @_;
my ($dev,$ino,$mode,$subcount);
my @filenames;
# At the top level, we need to find nlink ourselves.
($dev,$ino,$mode,$nlink) = stat($dir) unless $nlink;
# Get the list of files in the current directory.
opendir(DIR, $dir) || &logdie ("Cannot open \"$dir\" for reading: $!\n");
@filenames = readdir(DIR);
closedir(DIR);
if ($nlink == 2) { # This dir has no subdirectories.
for (@filenames) {
next if $_ eq '.';
next if $_ eq '..';
next if -l $_;
next if $_ =~ /submission/;
next if $_ =~ /\/\.svn\//;
if ($_ =~ /gene_association/ && -f $_ ) {
push (@filelist, "$dir/$_");
}
}
} else { # This dir has subdirectories.
$subcount = $nlink - 2;
for (@filenames) {
next if $_ eq '.';
next if $_ eq '..';
next if -l $_;
next if $_ =~ /submission/;
if ( $_ =~ /gene_association./ && -f $_ ) {
push (@filelist, "$dir/$_");
}
next if $subcount == 0; # Seen all the subdirs?
# Get link count and check for directoriness.
($dev,$ino,$mode,$nlink) = lstat($_);
next unless -d $_;
# It really is a directory, so do it recursively.
chdir $_ || &logdie ("Cannot chdir to $_: $!");
&dodir("$dir/$_", $nlink);
chdir '..';
--$subcount;
}
}
}
#
# counts from each gene associations file
#
sub associationcounts {
# Annotation Counts
my $tcount = 0;
my $ieacount = 0;
# Total Gene Product Count
my %gphash = ();
my $gpcount = 0;
while ( ) {
chomp;
if (/^\!/) { next; }
my @cols = split(/\t/);
# simple check of an annotation, the aspect is stated
unless (defined($cols[8])) { next; }
if ($cols[6] ne 'IEA') {
$ieacount++;
}
$tcount++;
unless (defined($gphash{$cols[1]})) {
$gpcount++;
$gphash{$cols[1]} = 1;
}
}
return ($tcount, $ieacount, $gpcount);
}
# print to STDERR and to the LOG file
sub logdie
{
my $message = $_[0];
print $message;
die $message;
}