use fread instead of getline, lower memory, static allocation, better performance (by a lot). all win win

2 files changed, 27 insertions, 26 deletions

diff --git a/count_nucleobases b/count_nucleobases
deleted file mode 100755
index 8c1b122..0000000
--- a/count_nucleobases
+++ /dev/null
diff --git a/count_nucleobases.c b/count_nucleobases.c
index c2aafe5..4bc3ca1 100755
--- a/count_nucleobases.c
+++ b/count_nucleobases.c
@@ -3,35 +3,36 @@
 #include <stdlib.h>
 #include <unistd.h>
 #include <stdint.h>
-int main(int argc, char **argv) {
+#include <stdbool.h>
+int main() {
 
-  char *line = NULL;
-  size_t len = 0;
-  ssize_t read;
-
-  if(argc != 2) {
-    printf("Please supply a filename, and only a filename\n");
-    exit(EXIT_FAILURE);
-  }
-
-  FILE *fh = fopen(argv[1], "r" );
-  if(fh == NULL) {
-    fprintf(stderr, "Couldn't open: %s\n", argv[1]);
-    exit(EXIT_FAILURE);
-  }
+  ssize_t len = 0;
 
   unsigned long long counts[256] = {0};
-
-  while ((read = getline(&line, &len, fh)) != -1) {
-    if(line[0] != '>')   {
-      unsigned int i = 0;
-      for(i = 0; i < strlen(line); i++) {
-        counts[line[i]]++;
-      }
-    }
-  }
-
-  free(line);
+	char buffer[4096];
+	bool header = false;
+
+	len =	fread(&buffer, 1, 1, stdin);
+
+	if(len != NULL) {
+		if(buffer[0] == '>') {
+			header = true;
+			while((len = fread(&buffer, 1, 4096, stdin)) != NULL) {
+				unsigned int i = 0;
+				for(i = 0; i < len; i++) {
+ 					if(buffer[i] == '>') {
+						header = true;
+						continue;
+					}
+					else if(buffer[i] == '\n' && header == true) 
+						header = false;
+					if(header == false ) {
+						counts[(int)buffer[i]]++;
+					}
+				}
+			}
+		}  else { fprintf(stderr, "this does not look like a fasta file\n"); }
+	} else { fprintf(stderr, "could not read file\n"); }
 
   printf("A:%llu\nC:%llu\nG:%llu\nT:%llu\n",
          counts['a'] + counts['A'],