Optimised sequence parsing (from previous commit)

bug

Makefile

@@ -0,0 +1,56 @@
+PREFIX=/usr/local
+
+SOURCES = libfasta/fasta_header_parser.c \
+		  libfasta/fasta_seq_writer.c \
+		  libfasta/fasta_header_handler.c \
+		  libfasta/header_mem_handler.c \
+          libfasta/sequence.c \
+          libfile/fileHandling.c \
+          liblcs/sse_banded_LCS_alignment.c \
+		  liblcs/upperband.c \
+		  libutils/utilities.c \
+          libutils/debug.c  
+                   
+SRCS=$(SOURCES)
+         
+OBJECTS= $(patsubst %.c,%.o,$(SOURCES))
+
+LIBFILE = libsuma.a
+
+RANLIB = ranlib
+
+CC=gcc
+
+LDFLAGS=
+
+CFLAGS = -O3 -w
+
+default: all
+
+all: $(LIBFILE)
+
+%.o: %.c
+	$(CC) $(CFLAGS) -c -o $@ $< $(LIB)
+
+libfasta/fasta_header_parser.c: libfasta/fasta_header_parser.l
+	flex -Pheader_yy -t $< > $@
+
+libfasta/dic_parser.c: libfasta/dic_parser.l
+	lex -Phashtable_yy -t $< > $@
+
+clean:
+	rm -rf $(OBJECTS) $(LIBFILE)
+
+$(LIBFILE): $(OBJECTS)
+	ar -cr $@ $?
+	$(RANLIB) $@
+
+install: all
+	install -d $(DESTDIR)$(PREFIX)/lib/
+	install -m 644 $(LIBFILE) $(DESTDIR)$(PREFIX)/lib/
+	install -d $(DESTDIR)$(PREFIX)/include/
+	for lib in libfasta liblcs libsse libutils libfile ; do \
+		install -d $(DESTDIR)$(PREFIX)/include/$$lib ; \
+		cp $$lib/*.h $(DESTDIR)$(PREFIX)/include/$$lib ; \
+	done
+

global.mk

@@ -1,10 +0,0 @@
-
-CC=gcc
-LDFLAGS=
-
-CFLAGS = -O3 -w
-
-default: all
-
-%.o: %.c
-	$(CC) $(CFLAGS) -c -o $@ $< $(LIB)

libfasta/Makefile

@@ -1,33 +0,0 @@
-
-SOURCES = fasta_header_parser.c \
-		  fasta_seq_writer.c \
-		  fasta_header_handler.c \
-		  header_mem_handler.c \
-          sequence.c 
-                   
-SRCS=$(SOURCES)
-
-         
-OBJECTS= $(patsubst %.c,%.o,$(SOURCES))
-
-LIBFILE = libfasta.a
-RANLIB = ranlib
-
-
-include ../global.mk
-
-all: $(LIBFILE)
-
-fasta_header_parser.c: fasta_header_parser.l
-	flex -Pheader_yy -t $< > $@
-
-dic_parser.c: dic_parser.l
-	lex -Phashtable_yy -t $< > $@
-
-clean:
-	rm -rf $(OBJECTS) $(LIBFILE)
-	rm -f *.a
-
-$(LIBFILE): $(OBJECTS)
-	ar -cr $@ $?
-	$(RANLIB) $@

libfasta/sequence.c

@@ -162,19 +162,24 @@ void seq_fillSeqOnlyATGC(char *seq, fastaSeqPtr seqElem, int seqLen)
 {
 	char* seqTemp;
 	char c;
-	int32_t index = 0, seqIndex = 0, len = strlen(seq);
+	int32_t index = 1, seqIndex = 0, len = strlen(seq);
 	char* seqAlphabets = "acgtACGT";
 	int notAllATGC = 0;
+	int goOnParsing = 1;
 
 	seqTemp = (char*) util_malloc(seqLen*sizeof(char), __FILE__, __LINE__);
 
-	while (index < len)
+	while (goOnParsing)
 	{
 		c = seq[index++];
 		if (strchr(seqAlphabets, c) != NULL)
 			seqTemp[seqIndex++] = tolower(c);
+		else if (seq[index+1]=='\0')
+			goOnParsing = 0; // end of the sequence has been reached.
 		else if (c != '\n')
 			notAllATGC = 1;
+		if (index == len)
+			goOnParsing = 0;
 	}
 
 	if (notAllATGC)

libfile/Makefile

@@ -1,25 +0,0 @@
-
-SOURCES = fileHandling.c
-
-         
-SRCS=$(SOURCES)
-
-         
-OBJECTS= $(patsubst %.c,%.o,$(SOURCES))
-
-LIBFILE= libfile.a
-RANLIB=ranlib
-
-
-include ../global.mk
-
-all: $(LIBFILE)
-
-clean:
-	rm -rf $(OBJECTS) $(LIBFILE)
-	rm -f *.P
-	rm -f *.a
-
-$(LIBFILE): $(OBJECTS)
-	ar -cr $@ $?
-	$(RANLIB) $@

liblcs/Makefile

@@ -1,25 +0,0 @@
-
-SOURCES = sse_banded_LCS_alignment.c  \
-		  upperband.c
-         
-SRCS=$(SOURCES)
-
-         
-OBJECTS= $(patsubst %.c,%.o,$(SOURCES))
-
-LIBFILE= liblcs.a
-RANLIB=ranlib
-
-
-include ../global.mk
-
-all: $(LIBFILE)
-
-clean:
-	rm -rf $(OBJECTS) $(LIBFILE)
-	rm -f *.P
-	rm -f *.a
-
-$(LIBFILE): $(OBJECTS)
-	ar -cr $@ $?
-	$(RANLIB) $@

liblcs/sse_banded_LCS_alignment.c

@@ -458,6 +458,9 @@ int calculateSizeToAllocate(int maxLen, int minLen, int LCSmin)
 
 	calculateBandLengths(maxLen, minLen, &notUsed, &size, LCSmin);		// max size = max left band length * 2
 
+	if (!size)	// Happens if there is no threshold (threshold is 100% similarity) and generates bugs if left to 0
+		size = 1;
+
 	//fprintf(stderr, "\nsize for address before %8 = %d", size);
 
 	size*= 2;

libutils/Makefile

@@ -1,25 +0,0 @@
-
-SOURCES = utilities.c  \
-          debug.c      
-         
-SRCS=$(SOURCES)
-
-         
-OBJECTS= $(patsubst %.c,%.o,$(SOURCES))
-
-LIBFILE= libutils.a
-RANLIB=ranlib
-
-
-include ../global.mk
-
-all: $(LIBFILE)
-
-clean:
-	rm -rf $(OBJECTS) $(LIBFILE)
-	rm -f *.P
-	rm -f *.a
-
-$(LIBFILE): $(OBJECTS)
-	ar -cr $@ $?
-	$(RANLIB) $@