X-Git-Url: https://git.donarmstrong.com/?p=rsem.git;a=blobdiff_plain;f=parseIt.cpp;h=3373ef280979428230b74d77439d61bff3a96d47;hp=55dd954d3d1e2059911987ed5a8521748b16e9a1;hb=d13cdd443afbefff6f7c8c0be818e1edcbc9cb8d;hpb=58f823f5be6dfbe00896fc44ac3aac5e881e9c5c

diff --git a/parseIt.cpp b/parseIt.cpp
index 55dd954..3373ef2 100644
--- a/parseIt.cpp
+++ b/parseIt.cpp
@@ -13,8 +13,6 @@
 #include "utils.h"
 
 #include "GroupInfo.h"
-
-#include "Transcript.h"
 #include "Transcripts.h"
 
 #include "SingleRead.h"
@@ -30,12 +28,11 @@
 using namespace std;
 
 int read_type; // 0 SingleRead, 1 SingleReadQ, 2 PairedEndRead, 3 PairedEndReadQ
-int N[3]; // note, N = N0 + N1 + N2 , but may not be equal to the total number of reads in data
-int nHits; // # of hits
-int nUnique, nMulti, nIsoMulti;
+READ_INT_TYPE N[3]; // note, N = N0 + N1 + N2 , but may not be equal to the total number of reads in data
+HIT_INT_TYPE nHits; // # of hits
+READ_INT_TYPE nUnique, nMulti, nIsoMulti;
 char fn_list[STRLEN];
 char groupF[STRLEN], tiF[STRLEN];
-char imdName[STRLEN];
 char datF[STRLEN], cntF[STRLEN];
 
 GroupInfo gi;
@@ -48,8 +45,8 @@ int n_os; // number of ostreams
 ostream *cat[3][2]; // cat : category  1-dim 0 N0 1 N1 2 N2; 2-dim  0 mate1 1 mate2
 char readOutFs[3][2][STRLEN];
 
-map<int, int> counter;
-map<int, int>::iterator iter;
+map<int, READ_INT_TYPE> counter;
+map<int, READ_INT_TYPE>::iterator iter;
 
 void init(const char* imdName, char alignFType, const char* alignF) {
 
@@ -87,7 +84,7 @@ void parseIt(SamParser *parser) {
 	nUnique = nMulti = nIsoMulti = 0;
 	memset(N, 0, sizeof(N));
 
-	long long cnt = 0;
+	READ_INT_TYPE cnt = 0;
 
 	record_val = -2; //indicate no recorded read now
 	while ((val = parser->parseNext(read, hit)) >= 0) {
@@ -96,6 +93,9 @@ void parseIt(SamParser *parser) {
 			if (record_val >= 0) {
 				record_read.write(n_os, cat[record_val]);
 				++N[record_val];
+			}
+			// flush out previous read's hits if the read is alignable reads
+			if (record_val == 1) {
 				hits.updateRI();
 				nHits += hits.getNHits();
 				nMulti += hits.calcNumGeneMultiReads(gi);
@@ -121,12 +121,15 @@ void parseIt(SamParser *parser) {
 		}
 
 		++cnt;
-		if (verbose && (cnt % 1000000 == 0)) { printf("Parsed %lld entries\n", cnt); }
+		if (verbose && (cnt % 1000000 == 0)) { cout<< "Parsed "<< cnt<< " entries"<< endl; }
 	}
 
 	if (record_val >= 0) {
 		record_read.write(n_os, cat[record_val]);
 		++N[record_val];
+	}
+
+	if (record_val == 1) {
 		hits.updateRI();
 		nHits += hits.getNHits();
 		nMulti += hits.calcNumGeneMultiReads(gi);
@@ -163,7 +166,7 @@ int main(int argc, char* argv[]) {
 	bool quiet = false;
 
 	if (argc < 6) {
-		printf("Usage : rsem-parse-alignments refName sampleName sampleToken alignFType('s' for sam, 'b' for bam) alignF [-t Type] [-l fn_list] [-tag tagName] [-q]\n");
+		printf("Usage : rsem-parse-alignments refName imdName statName alignFType('s' for sam, 'b' for bam) alignF [-t Type] [-l fn_list] [-tag tagName] [-q]\n");
 		exit(-1);
 	}
 
@@ -191,15 +194,14 @@ int main(int argc, char* argv[]) {
 	sprintf(tiF, "%s.ti", argv[1]);
 	transcripts.readFrom(tiF);
 
-	sprintf(imdName, "%s.temp/%s", argv[2], argv[3]);
-	sprintf(datF, "%s.dat", imdName);
-	sprintf(cntF, "%s.stat/%s.cnt", argv[2], argv[3]);
+	sprintf(datF, "%s.dat", argv[2]);
+	sprintf(cntF, "%s.cnt", argv[3]);
 
-	init(imdName, argv[4][0], argv[5]);
+	init(argv[2], argv[4][0], argv[5]);
 
 	hit_out.open(datF);
 
-	string firstLine(59, ' ');
+	string firstLine(99, ' ');
 	firstLine.append(1, '\n');		//May be dangerous!
 	hit_out<<firstLine;
 
@@ -220,9 +222,11 @@ int main(int argc, char* argv[]) {
 	fout<<N[0]<<" "<<N[1]<<" "<<N[2]<<" "<<(N[0] + N[1] + N[2])<<endl;
 	fout<<nUnique<<" "<<nMulti<<" "<<nIsoMulti<<endl;
 	fout<<nHits<<" "<<read_type<<endl;
+	fout<<"0\t"<<N[0]<<endl;
 	for (iter = counter.begin(); iter != counter.end(); iter++) {
 		fout<<iter->first<<'\t'<<iter->second<<endl;
 	}
+	fout<<"Inf\t"<<N[2]<<endl;
 	fout.close();
 
 	release();