From ef1b3015c71ba8bdf90ae64fef043174f24dd82c Mon Sep 17 00:00:00 2001
From: elserj <elserj@localhost>
Date: Thu, 23 Sep 2010 00:24:05 +0000
Subject: [PATCH] Moved the "clusters" table select statement outside the loop.
  Fixes random stalls and improves performance (not by much)

svn path=/; revision=47
---
 .../inparanoid_output_parse.pl                | 27 ++++++++++---------
 1 file changed, 14 insertions(+), 13 deletions(-)

diff --git a/interactome_scripts/inparanoid_output_parse.pl b/interactome_scripts/inparanoid_output_parse.pl
index bd58591..bb8de08 100755
--- a/interactome_scripts/inparanoid_output_parse.pl
+++ b/interactome_scripts/inparanoid_output_parse.pl
@@ -66,12 +66,24 @@ my $screen = Term::Screen::ReadLine->new();
 my $clust_table = "clusters";
 my $safe_clust_table = $dbh->quote_identifier($clust_table);
 
+my %id_hash;
 if($new) {
 	$dbh->do("create table $safe_clust_table (
 		`id` INT( 11 ) NOT NULL AUTO_INCREMENT,
 		`clust_id` VARCHAR( 255 ) NOT NULL ,
 		INDEX ( `id` )
 		) TYPE = MYISAM");
+}else{ # if not new load, grab the data from the clusters table
+	my $sth_id_dump = $dbh->prepare("select * from clusters");
+	# put the clusters table data into a hash
+	print "before dump\n";
+	$sth_id_dump->execute();
+	print "after dump\n";
+	while(my @ids = $sth_id_dump->fetchrow_array) {
+		if(!defined($id_hash{$ids[1]})) {
+			$id_hash{$ids[1]} = $ids[0];
+		}
+	}
 }
 		
 my $tot_file_count = scalar @files;
@@ -91,14 +103,12 @@ foreach my $input_file (@files) {
 	my $species_table = $species_1."_".$species_2;
 	
 
-	my %id_hash;
 	
 	print "Working on $species_table: file $file_counter/$tot_file_count\n";
 	$file_counter++;
 
 
 	
-		
 	my $safe_species_table = $dbh->quote_identifier($species_table);
 	$dbh->do("drop table if exists $safe_species_table");
 	$dbh->do("CREATE TABLE $safe_species_table (
@@ -113,17 +123,11 @@ foreach my $input_file (@files) {
 
 	
 		
-	my $sth_id_dump = $dbh->prepare("select * from clusters");
+	
 	my $sth_id_add = $dbh->prepare("insert into $safe_clust_table (id, clust_id) values (?,?)");
 	my $sth_get_id = $dbh->prepare("select last_insert_id(id) from clusters order by id desc limit 1");
 
-	# put the clusters table data into a hash
-	$sth_id_dump->execute();
-	while(my @ids = $sth_id_dump->fetchrow_array) {
-		if(!defined($id_hash{$ids[1]})) {
-			$id_hash{$ids[1]} = $ids[0];
-		}
-	}
+	
 	
 	my $clust_id_prev = "";
 
@@ -140,7 +144,6 @@ foreach my $input_file (@files) {
 		$clust_id = "$species_1"."___".$species_2."___".$clust_id;
 		$species = find_species($species);
 		my $gene = find_gene($gene_header,$species);
-		
 		# skip isomers that are not .1
 		next if ($gene =~ /\.[2-9]$/);
 		next if ($gene =~ /\.1[0-9]$/);
@@ -156,7 +159,6 @@ foreach my $input_file (@files) {
 		if(!defined($id_hash{$clust_id})) {
 			if ($clust_id ne $clust_id_prev) {
 				$sth_id_add->execute(undef, $clust_id);
-			
 				$sth_get_id->execute();
 				my @new_id = $sth_get_id->fetchrow_array;
 				$id = $new_id[0];
@@ -168,7 +170,6 @@ foreach my $input_file (@files) {
 	
 		
 		$sth->execute($id, $bit_score, $species, $score, $gene);
-	
 		$clust_id_prev = $clust_id;
 
 		
-- 
2.34.1