// Set the initial state
static state_t* init(parsed_opts_t* opts)
{
state_t* retval = calloc(sizeof(state_t), 1);
if (!retval) {
fprintf(pysamerr, "Out of memory");
return NULL;
}
retval->merged_input_file = sam_open(opts->merged_input_name, "rb");
if (!retval->merged_input_file) {
fprintf(pysamerr, "Could not open input file (%s)\n", opts->merged_input_name);
free(retval);
return NULL;
}
retval->merged_input_header = sam_hdr_read(retval->merged_input_file);
if (opts->unaccounted_name) {
if (opts->unaccounted_header_name) {
samFile* hdr_load = sam_open(opts->unaccounted_header_name, "r");
if (!hdr_load) {
fprintf(pysamerr, "Could not open unaccounted header file (%s)\n", opts->unaccounted_header_name);
cleanup_state(retval);
return NULL;
}
retval->unaccounted_header = sam_hdr_read(hdr_load);
sam_close(hdr_load);
} else {
retval->unaccounted_header = bam_hdr_dup(retval->merged_input_header);
}
retval->unaccounted_file = sam_open(opts->unaccounted_name, "wb");
if (retval->unaccounted_file == NULL) {
fprintf(pysamerr, "Could not open unaccounted output file: %s\n", opts->unaccounted_name);
cleanup_state(retval);
return NULL;
}
}
// Open output files for RGs
if (!count_RG(retval->merged_input_header, &retval->output_count, &retval->rg_id)) return NULL;
if (opts->verbose) fprintf(pysamerr, "@RG's found %zu\n",retval->output_count);
retval->rg_output_file = (samFile**)calloc(retval->output_count, sizeof(samFile*));
retval->rg_output_header = (bam_hdr_t**)calloc(retval->output_count, sizeof(bam_hdr_t*));
retval->rg_hash = kh_init_c2i();
if (!retval->rg_output_file || !retval->rg_output_header) {
fprintf(pysamerr, "Could not allocate memory for output file array. Out of memory?");
cleanup_state(retval);
return NULL;
}
char* dirsep = strrchr(opts->merged_input_name, '/');
char* input_base_name = strdup(dirsep? dirsep+1 : opts->merged_input_name);
if (!input_base_name) {
fprintf(pysamerr, "Out of memory\n");
cleanup_state(retval);
return NULL;
}
char* extension = strrchr(input_base_name, '.');
if (extension) *extension = '\0';
size_t i;
for (i = 0; i < retval->output_count; i++) {
char* output_filename = NULL;
if ( ( output_filename = expand_format_string(opts->output_format_string, input_base_name, retval->rg_id[i], i) ) == NULL) {
fprintf(pysamerr, "Error expanding output filename format string.\r\n");
cleanup_state(retval);
free(input_base_name);
return NULL;
}
retval->rg_output_file[i] = sam_open(output_filename, "wb");
if (retval->rg_output_file[i] == NULL) {
fprintf(pysamerr, "Could not open output file: %s\r\n", output_filename);
cleanup_state(retval);
free(input_base_name);
return NULL;
}
// Record index in hash
int ret;
khiter_t iter = kh_put_c2i(retval->rg_hash, retval->rg_id[i], &ret);
kh_val(retval->rg_hash,iter) = i;
// Set and edit header
retval->rg_output_header[i] = bam_hdr_dup(retval->merged_input_header);
if ( !filter_header_rg(retval->rg_output_header[i], retval->rg_id[i]) ) {
fprintf(pysamerr, "Could not rewrite header for file: %s\r\n", output_filename);
cleanup_state(retval);
free(output_filename);
free(input_base_name);
return NULL;
}
free(output_filename);
}
free(input_base_name);
return retval;
}
int main(int argc, char**argv)
{
// test state
const int NUM_TESTS = 2;
int verbose = 0;
int success = 0;
int failure = 0;
int getopt_char;
while ((getopt_char = getopt(argc, argv, "v")) != -1) {
switch (getopt_char) {
case 'v':
++verbose;
break;
default:
printf(
"usage: test_filter_header_rg [-v]\n\n"
" -v verbose output\n"
);
break;
}
}
// Setup pysamerr redirect
kstring_t res = { 0, 0, NULL };
FILE* orig_pysamerr = fdopen(dup(STDERR_FILENO), "a"); // Save pysamerr
char* tempfname = (optind < argc)? argv[optind] : "test_count_rg.tmp";
FILE* check = NULL;
// setup
if (verbose) printf("BEGIN test 1\n"); // test eliminating a tag that isn't there
bam_hdr_t* hdr1;
const char* id_to_keep_1 = "1#2.3";
setup_test_1(&hdr1);
if (verbose > 1) {
printf("hdr1\n");
dump_hdr(hdr1);
}
if (verbose) printf("RUN test 1\n");
// test
xfreopen(tempfname, "w", pysamerr); // Redirect pysamerr to pipe
bool result_1 = filter_header_rg(hdr1, id_to_keep_1);
fclose(pysamerr);
if (verbose) printf("END RUN test 1\n");
if (verbose > 1) {
printf("hdr1\n");
dump_hdr(hdr1);
}
// check result
res.l = 0;
check = fopen(tempfname, "r");
if ( result_1
&& check_test_1(hdr1)
&& kgetline(&res, (kgets_func *)fgets, check) < 0
&& (feof(check) || res.l == 0)) {
++success;
} else {
++failure;
if (verbose) printf("FAIL test 1\n");
}
fclose(check);
// teardown
bam_hdr_destroy(hdr1);
if (verbose) printf("END test 1\n");
if (verbose) printf("BEGIN test 2\n"); // test eliminating a tag that is there
bam_hdr_t* hdr2;
const char* id_to_keep_2 = "fish";
setup_test_2(&hdr2);
if (verbose > 1) {
printf("hdr2\n");
dump_hdr(hdr2);
}
if (verbose) printf("RUN test 2\n");
// test
xfreopen(tempfname, "w", pysamerr); // Redirect pysamerr to pipe
bool result_2 = filter_header_rg(hdr2, id_to_keep_2);
fclose(pysamerr);
if (verbose) printf("END RUN test 2\n");
if (verbose > 1) {
printf("hdr2\n");
dump_hdr(hdr2);
}
// check result
res.l = 0;
check = fopen(tempfname, "r");
if ( result_2
&& check_test_2(hdr2)
&& kgetline(&res, (kgets_func *)fgets, check) < 0
&& (feof(check) || res.l == 0)) {
++success;
} else {
++failure;
if (verbose) printf("FAIL test 2\n");
}
fclose(check);
// teardown
bam_hdr_destroy(hdr2);
if (verbose) printf("END test 2\n");
// Cleanup
free(res.s);
remove(tempfname);
if (failure > 0)
fprintf(orig_pysamerr, "%d failures %d successes\n", failure, success);
fclose(orig_pysamerr);
return (success == NUM_TESTS)? EXIT_SUCCESS : EXIT_FAILURE;
}
