Imported Upstream version 1.2.17

[rsem.git] / parseIt.cpp

diff --git a/parseIt.cpp b/parseIt.cpp
index 8620432c2840140561451c211208b560b7e48363..f40f82295302c48b16923d6655f3757d4057e09b 100644 (file)
--- a/parseIt.cpp
+++ b/parseIt.cpp
@@ -28,12 +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;
@@ -46,14 +45,14 @@ 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) {
 
        char* aux = 0;
        if (strcmp(fn_list, "")) aux = fn_list;
-       parser = new SamParser(alignFType, alignF, transcripts, aux);
+       parser = new SamParser(alignFType, alignF, aux, transcripts, imdName);
 
        memset(cat, 0, sizeof(cat));
        memset(readOutFs, 0, sizeof(readOutFs));
@@ -85,7 +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) {
@@ -95,6 +94,9 @@ void parseIt(SamParser *parser) {
                                record_read.write(n_os, cat[record_val]);
                                ++N[record_val];
                        }
+
+                       general_assert(record_val == 1 || hits.getNHits() == 0, "Read " + record_read.getName() + " is both unalignable and alignable according to the input SAM/BAM file!");
+
                        // flush out previous read's hits if the read is alignable reads
                        if (record_val == 1) {
                                hits.updateRI();
@@ -122,7 +124,7 @@ 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) {
@@ -167,7 +169,7 @@ 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);
        }
 
@@ -195,15 +197,14 @@ 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;