static void
pr_append(pr_append_str *x,
const char *s)
{
if (pr_append_external(x, s))
out_of_memory_error();
}
int
main (int argc, const char *argv[])
{
unsigned int idx;
char *end;
int debug = 0;
/* sequence file name specified by the user, otherwise STDIN will be used */
const char *sequence_file_name = NULL;
const char *lists_file_name = NULL;
/* data structure for all k-mer lists used for masking */
unsigned int nlists = 0;
unsigned int nlist_parameters = 0;
unsigned int npos = 0;
unsigned int list_pos[MAX_VARIABLES], list_components[MAX_VARIABLES];
masker_parameters mp = {};
parameters_builder pbuilder = {};
input_sequence *input_seq = NULL;
pr_append_str parse_err;
pr_append_str warnings;
init_pr_append_str (&parse_err);
init_pr_append_str (&warnings);
/* fill mp with default parameters */
mp.mdir = DEFAULT_MASKING_DIRECTION;
mp.failure_rate = DEFAULT_FAILURE_RATE;
mp.abs_cutoff = DEFAULT_ABS_CUTOFF;
mp.nucl_masked_in_5p_direction = DEFAULT_M5P;
mp.nucl_masked_in_3p_direction = DEFAULT_M3P;
mp.print_sequence = PRINT_SEQUENCE;
mp.do_soft_masking = HARD_MASKING;
mp.masking_char = DEFAULT_MASK_CHAR;
mp.list_prefix = DEFAULT_LIST_FILE_PREFIX;
/* parsing and checking the commandline arguments */
for (idx = 1; (int)idx < argc; idx++) {
if (!strcmp (argv[idx], "-h") || !strcmp (argv[idx], "--help") || !strcmp (argv[idx], "-?")) {
print_help (0);
} else if ((int)idx == argc - 1 && argv[idx][0] != '-') {
sequence_file_name = argv[idx];
} else if (!strcmp (argv[idx], "-lf") || !strcmp (argv[idx], "--lists_file")) {
/* lists specified from the file */
if (!argv[idx + 1] || argv[idx + 1][0] == '-') {
pr_append_new_chunk_external (&warnings, "No lists file specified.");
idx += 1;
continue;
}
lists_file_name = argv[idx + 1];
idx += 1;
} else if (!strcmp (argv[idx], "-l") || !strcmp (argv[idx], "--list")) {
/* lists specified from the commandline */
if (nlist_parameters == MAX_VARIABLES) {
pr_append_new_chunk_external (&parse_err, "Maximum number of list variables reached.");
break;
}
if (!argv[idx + 1]) {
pr_append_new_chunk_external (&warnings, "No list name specified with -l parameter!.");
continue;
}
/* get the positions of list files */
list_pos[nlist_parameters] = idx;
while (argv[idx + 1]) {
if (argv[idx + 1][0] == '-') {
if (!argv[idx + 1][1]) break;
strtod (argv[idx + 1] + 1, &end);
if (*end != 0) break;
} else if (idx + 1 != list_pos[nlist_parameters] + 1 && idx + 1 != list_pos[nlist_parameters] + 4) {
strtod (argv[idx + 1], &end);
if (*end != 0) break;
} else if (idx + 1 == list_pos[nlist_parameters] + 4 && strcmp (argv[idx + 1], "sq")) {
break;
}
idx += 1;
}
list_components[nlist_parameters] = idx - list_pos[nlist_parameters];
nlist_parameters += 1;
npos += 1;
} else if (!strcmp (argv[idx], "-lp") || !strcmp (argv[idx], "--list_prefix")) {
/* lists specified by the prefix */
if (!argv[idx + 1] || argv[idx + 1][0] == '-') {
pr_append_new_chunk_external (&warnings, "No list prefix specified! Using the default value.");
idx += 1;
continue;
}
mp.list_prefix = (char *)argv[idx + 1];
idx += 1;
} else if (!strcmp (argv[idx], "-p") || !strcmp (argv[idx], "--probability_cutoff")) {
if (!argv[idx + 1]) {
pr_append_new_chunk_external (&warnings, "No cutoff value specified! Using the default value.");
idx += 1;
continue;
}
mp.failure_rate = strtod (argv[idx + 1], &end);
mp.abs_cutoff = 0;
if (*end != 0 || mp.failure_rate < 0 || mp.failure_rate > 1) {
pr_append_new_chunk_external (&parse_err, "Invalid cutoff value: ");
pr_append_external (&parse_err, argv[idx + 1]);
break;
}
idx += 1;
} else if (!strcmp (argv[idx], "-a") || !strcmp (argv[idx], "--absolute_value_cutoff")) {
if (!argv[idx + 1]) {
pr_append_new_chunk_external (&warnings, "No absolute cutoff value specified! Using the default value.");
idx += 1;
continue;
}
mp.abs_cutoff = strtod (argv[idx + 1], &end);
mp.failure_rate = 0.0;
if (*end != 0 || mp.abs_cutoff < 0) {
pr_append_new_chunk_external (&parse_err, "Invalid absolute cutoff value: ");
pr_append_external (&parse_err, argv[idx + 1]);
break;
}
idx += 1;
} else if (!strcmp (argv[idx], "-m5") || !strcmp (argv[idx], "--mask_5p")) {
if (!argv[idx + 1]) {
pr_append_new_chunk_external (&warnings, "Number of nucleotides masked in 5' direction not specified! Using the default value.");
idx += 1;
continue;
}
mp.nucl_masked_in_5p_direction = strtod (argv[idx + 1], &end);
if (*end != 0) {
pr_append_new_chunk_external (&parse_err, "Invalid number of nucleotides masked in 5' direction: ");
pr_append_external (&parse_err, argv[idx + 1]);
break;
}
idx += 1;
} else if (!strcmp (argv[idx], "-m3") || !strcmp (argv[idx], "--mask_3p")) {
if (!argv[idx + 1]) {
pr_append_new_chunk_external (&warnings, "Number of nucleotides masked in 3' direction not specified! Using the default value.");
idx += 1;
continue;
}
mp.nucl_masked_in_3p_direction = strtod (argv[idx + 1], &end);
if (*end != 0) {
pr_append_new_chunk_external (&parse_err, "Invalid number of nucleotides masked in 3' direction: ");
pr_append_external (&parse_err, argv[idx + 1]);
break;
}
idx += 1;
} else if (!strcmp (argv[idx], "-c") || !strcmp (argv[idx], "--masking_char")) {
if (!argv[idx + 1 || argv[idx + 1][0] == '-']) {
pr_append_new_chunk_external (&warnings, "Character for masking not specified! Using the default value.");
idx += 1;
continue;
}
mp.masking_char = argv[idx + 1][0];
if (strlen(argv[idx + 1]) > 1 || mp.masking_char < 33 || mp.masking_char > 126) {
pr_append_new_chunk_external (&parse_err, "Invalid character for masking: ");
pr_append_external (&parse_err, argv[idx + 1]);
break;
}
idx += 1;
} else if (!strcmp (argv[idx], "-d") || !strcmp (argv[idx], "--masking_direction")) {
if (!argv[idx + 1] || argv[idx + 1][0] == '-') {
pr_append_new_chunk_external (&warnings, "Masking direction not specified! Masking both strands by default.");
} else if (!strcmp (argv[idx + 1], "both")) {
mp.mdir = both_on_same;
} else if (!strcmp (argv[idx + 1], "fwd")) {
mp.mdir = fwd;
} else if (!strcmp (argv[idx + 1], "rev")) {
mp.mdir = rev;
} else {
pr_append_new_chunk_external (&warnings, "Unknown masking direction: ");
pr_append_external (&warnings, argv[idx + 1]);
pr_append_external (&warnings, ". Masking both strands by default.");
}
idx += 1;
} else if (!strcmp (argv[idx], "-s") || !strcmp (argv[idx], "--soft_mask")) {
mp.do_soft_masking = SOFT_MASKING;
} else if (!strcmp (argv[idx], "-D")) {
debug += 1;
} else {
pr_append_new_chunk_external (&parse_err, "Unknown parameter: ");
pr_append_external (&parse_err, argv[idx]);
break;
}
}
input_seq = create_input_sequence_from_file_name (sequence_file_name, &parse_err);
if (parse_err.data != NULL) {
fprintf(stderr, "%s -> parsing commandline: ERROR: %s\n", pr_programme_name, parse_err.data);
exit(-1);
}
if (warnings.data != NULL) {
fprintf(stderr, "%s -> parsing commandline: WARNING: %s\n", pr_programme_name, warnings.data);
}
if (lists_file_name) {
/* if lists are given in a text file */
mp.fp = read_formula_parameters_from_file (lists_file_name, &nlist_parameters, &pbuilder, &mp.formula_intercept, &parse_err);
nlists = pbuilder.nfp;
}
if (npos != 0) {
/* if lists are given by commandline arguments (can be added to the ones given in a file) */
pbuilder.fp_array = mp.fp;
for (idx = 0; idx < npos; idx++) {
unsigned int pos = list_pos[idx] + 1;
char *values[4];
unsigned int nvalues = list_components[idx];
char *end;
memcpy (&values, &argv[pos], nvalues * sizeof(*argv));
if (nvalues == 1) {
double ic;
double neg = 1.0;
if (values[0][0] == '-') {
values[0] += 1;
neg = -1.0;
}
ic = strtod (values[0], &end);
if (*end == 0) {
mp.formula_intercept = ic * neg;
continue;
}
}
add_variable_to_formula_parameters (values, nvalues, &pbuilder, &parse_err);
}
nlists = pbuilder.nfp;
mp.fp = pbuilder.fp_array;
} else if (nlists == 0 && !lists_file_name) {
/* if there are no lists specified use the default formula */
mp.fp = create_default_formula_parameters (mp.list_prefix, &parse_err);
mp.formula_intercept = DEFAULT_INTERCEPT;
nlists = DEFAULT_NLISTS;
nlist_parameters = DEFAULT_NLIST_PARAMETERS;
}
mp.nlists = nlists;
if (mp.abs_cutoff > 0 && nlist_parameters != 1) {
fprintf (stderr, "Error: Absolute value cutoff works with one list and one k-mer frequency parameter only. Currently you are using %u lists and %u parameters.\n", nlists, nlist_parameters);
print_help(1);
}
if (parse_err.data != NULL) {
fprintf(stderr, "%s -> building formula: ERROR: %s\n", pr_programme_name, parse_err.data);
delete_input_sequence (input_seq);
exit(-1);
}
if (debug > 0) print_parameters (&mp);
read_and_mask_sequence (input_seq, NULL, &mp, &parse_err, debug);
if (parse_err.data != NULL) {
fprintf(stderr, "%s -> masking sequence: ERROR: %s\n", pr_programme_name, parse_err.data);
delete_input_sequence (input_seq);
delete_formula_parameters (mp.fp, nlists);
exit(-1);
}
destroy_pr_append_str_data (&warnings);
destroy_pr_append_str_data (&parse_err);
delete_input_sequence (input_seq);
delete_formula_parameters (mp.fp, nlists);
if (debug > 0) fprintf (stderr, "Done!\n");
return 0;
}
static void
p3sl_append(pr_append_str *x, const char *s)
{
int r = pr_append_external(x, s);
if (r) longjmp(_jmp_buf, 1);
}
int
main(int argc, char *argv[])
{
/* Setup the input data structures handlers */
int format_output = 0;
int strict_tags = 0;
int echo_settings = 0;
int io_version = 4;
int default_version = 2;
int dump_args = 0;
p3_global_settings *global_pa;
seq_args *sarg;
read_boulder_record_results read_boulder_record_res = {0,0};
pr_append_str p3_settings_path;
pr_append_str output_path;
pr_append_str error_path;
pr_append_str fatal_parse_err;
pr_append_str nonfatal_parse_err;
pr_append_str warnings;
/* Some variables needed by getopt */
int opt, option_index = 0;
struct option long_options[] = {
{"about", no_argument, 0, 'a'},
{"format_output", no_argument, &format_output, 1},
{"strict_tags", no_argument, &strict_tags, 1},
{"p3_settings_file", required_argument, 0, 'p'},
{"echo_settings_file", no_argument, &echo_settings, 1},
{"io_version", required_argument, 0, 'i'},
{"default_version", required_argument, 0, 'd'},
{"Dump_args", no_argument, 0, 'D'},
{"2x_compat", no_argument, 0, '2'},
{"output", required_argument, 0, 'o'},
{"error", required_argument, 0, 'e'},
{0, 0, 0, 0}
};
int about = 0, compat = 0, invalid_flag = 0;
/* Retval will point to the return value from choose_primers(). */
p3retval *retval = NULL;
int input_found=0;
init_pr_append_str(&fatal_parse_err);
init_pr_append_str(&nonfatal_parse_err);
init_pr_append_str(&warnings);
init_pr_append_str(&p3_settings_path);
init_pr_append_str(&output_path);
init_pr_append_str(&error_path);
/* Get the program name for correct error messages */
pr_release = libprimer3_release();
pr_program_name = argv[0];
p3_set_program_name(pr_program_name);
/* We set up some signal handlers in case someone starts up the program
* from the command line, wonders why nothing is happening, and then kills
* the program. */
signal(SIGINT, sig_handler);
signal(SIGTERM, sig_handler);
/* Read in the flags provided with the program call */
opterr = 0;
while ((opt = getopt_long_only(argc, argv, "", long_options, &option_index)) != -1) {
switch (opt) {
case 'a':
about = 1;
break;
case 'p':
if (pr_append_external(&p3_settings_path, optarg)) {
exit(-2); /* Out of memory. */
}
break;
case 'i':
if (!strcmp(optarg, "4"))
io_version = 4;
else
io_version = -1;
break;
case 'd': /* default_version */
if (!strcmp(optarg, "1"))
default_version = 1;
else if (!strcmp(optarg, "2"))
default_version = 2;
else
default_version = -1;
break;
case 'D': /* Undocumented flag for testing; causes
values of arguments to be echoed to
stdout. */
dump_args = 1;
break;
case '2':
compat = 1;
break;
case 'o':
if (pr_append_external(&output_path, optarg)) {
exit(-2); /* Out of memory. */
}
break;
case 'e':
if (pr_append_external(&error_path, optarg)) {
exit(-2); /* Out of memory. */
}
break;
case '?':
invalid_flag = 1;
break;
}
}
/* Open the output and error files specified */
if (!pr_is_empty(&error_path)) {
/* reassign stderr */
if (freopen(pr_append_str_chars(&error_path), "w", stderr)
== NULL) {
fprintf(stderr, "Error creating file %s\n",
pr_append_str_chars(&error_path));
exit(-1);
}
destroy_pr_append_str_data(&error_path);
}
if (!pr_is_empty(&output_path)) {
/* reassign stdout */
if (freopen(pr_append_str_chars(&output_path), "w", stdout)
== NULL) {
fprintf(stderr, "Error creating file %s\n",
pr_append_str_chars(&output_path));
exit(-1);
}
destroy_pr_append_str_data(&output_path);
}
/* We do any printing after redirecting stdout and stderr */
if (about == 1) {
printf("%s\n", pr_release);
exit(0);
}
if ((io_version == -1) || (invalid_flag == 1) || (default_version == -1)) {
print_usage();
exit(-1);
}
if (compat == 1) {
printf("PRIMER_ERROR=flag -2x_compat is no longer supported\n=\n");
exit(-1);
}
/* Check if an input file has been specified on the command line. */
if (optind < argc) {
if (optind + 1 != argc) {
print_usage();
exit(-1);
}
if (freopen(argv[optind], "r", stdin) == NULL) {
fprintf(stderr, "Error opening file %s\n", argv[optind]);
exit(-1);
}
}
/* Allocate the space for global settings and fill in default parameters */
if (default_version == 1)
global_pa = p3_create_global_settings_default_version_1();
else if (default_version == 2)
global_pa = p3_create_global_settings();
else {
print_usage();
exit(-1);
}
if (!global_pa) {
exit(-2); /* Out of memory. */
}
global_pa->dump = dump_args ;
/* Settings files have to be read in just below, and
the functions need a temporary sarg */
if (!(sarg = create_seq_arg())) {
exit(-2);
}
/* Read data from the settings file until a "=" line occurs. Assign
parameter values for primer picking to pa and sa. */
if (!pr_is_empty(&p3_settings_path)) {
read_p3_file(pr_append_str_chars(&p3_settings_path),
settings,
echo_settings && !format_output,
strict_tags,
global_pa, sarg, &fatal_parse_err,
&nonfatal_parse_err, &warnings, &read_boulder_record_res);
destroy_pr_append_str_data(&p3_settings_path);
/* Check if any thermodynamical alignment flag was given */
if ((global_pa->thermodynamic_oligo_alignment == 1) ||
(global_pa->thermodynamic_template_alignment == 1))
read_thermodynamic_parameters();
}
/* We also need to print out errors here because the loop erases all
errors at start. If there are fatal errors, write the proper
message and exit */
if (fatal_parse_err.data != NULL) {
if (format_output) {
format_error(stdout, sarg->sequence_name, fatal_parse_err.data);
} else {
print_boulder_error(fatal_parse_err.data);
}
fprintf(stderr, "%s: %s\n",
pr_program_name, fatal_parse_err.data);
destroy_seq_args(sarg);
exit(-4);
}
/* If there are nonfatal errors, write the proper message
* and skip to the end of the loop */
if (!pr_is_empty(&nonfatal_parse_err)) {
if (format_output) {
format_error(stdout, sarg->sequence_name,
nonfatal_parse_err.data);
} else {
print_boulder_error(nonfatal_parse_err.data);
}
}
/* The temporary sarg is not needed any more */
destroy_seq_args(sarg);
sarg = NULL;
/* Read the data from input stream record by record and process it if
* there are no errors. This is where the work is done. */
while (1) {
/* Create and initialize a seq_args data structure. sa (seq_args *) is
* initialized here because Values are _not_ retained across different
* input records. */
if (!(sarg = create_seq_arg())) {
exit(-2);
}
/* Reset all errors handlers and the return structure */
pr_set_empty(&fatal_parse_err);
pr_set_empty(&nonfatal_parse_err);
pr_set_empty(&warnings);
retval = NULL;
/* See read_boulder.h for documentation on read_boulder_record().*/
if (!read_boulder_record(stdin,
&strict_tags,
&io_version,
!format_output,
all_parameters,
global_pa,
sarg,
&fatal_parse_err,
&nonfatal_parse_err,
&warnings,
&read_boulder_record_res)) {
break; /* There were no more boulder records */
}
/* Check if any thermodynamical alignment flag was given and the
path to the parameter files changed - we need to reread them */
if (((global_pa->thermodynamic_oligo_alignment == 1) ||
(global_pa->thermodynamic_template_alignment == 1))
&& (thermodynamic_path_changed == 1))
read_thermodynamic_parameters();
/* Check that we found the thermodynamic parameters in case any thermodynamic flag was set to 1. */
if (((global_pa->thermodynamic_oligo_alignment == 1) ||
(global_pa->thermodynamic_template_alignment == 1))
&& (thermodynamic_params_path == NULL)) {
/* no parameter directory found, error */
printf("PRIMER_ERROR=thermodynamic approach chosen, but path to thermodynamic parameters not specified\n=\n");
exit(-1);
}
input_found = 1;
if ((global_pa->primer_task == generic)
&& (global_pa->pick_internal_oligo == 1)){
PR_ASSERT(global_pa->pick_internal_oligo);
}
/* If there are fatal errors, write the proper message and exit */
if (fatal_parse_err.data != NULL) {
if (format_output) {
format_error(stdout, sarg->sequence_name, fatal_parse_err.data);
} else {
print_boulder_error(fatal_parse_err.data);
}
fprintf(stderr, "%s: %s\n",
pr_program_name, fatal_parse_err.data);
destroy_p3retval(retval);
destroy_seq_args(sarg);
exit(-4);
}
/* If there are nonfatal errors, write the proper message
* and skip to the end of the loop */
if (!pr_is_empty(&nonfatal_parse_err)) {
if (format_output) {
format_error(stdout, sarg->sequence_name,
nonfatal_parse_err.data);
} else {
print_boulder_error(nonfatal_parse_err.data);
}
goto loop_wrap_up;
}
/* Print any warnings and continue processing */
if (!pr_is_empty(&warnings)) {
if (format_output) {
format_warning(stdout, sarg->sequence_name,
warnings.data);
} else {
print_boulder_warning(warnings.data);
}
}
if (read_boulder_record_res.file_flag && sarg->sequence_name == NULL) {
/* We will not have a base name for the files */
if (format_output) {
format_error(stdout, NULL,
"Need PRIMER_SEQUENCE_ID if PRIMER_FILE_FLAG is not 0");
} else {
print_boulder_error("Need PRIMER_SEQUENCE_ID if PRIMER_FILE_FLAG is not 0");
}
goto loop_wrap_up;
}
/* Pick the primers - the central function */
p3_set_gs_primer_file_flag(global_pa,
read_boulder_record_res.file_flag);
retval = choose_primers(global_pa, sarg);
if (NULL == retval) exit(-2); /* Out of memory. */
/* If it was necessary to use a left_input, right_input,
or internal_oligo_input primer that was
unacceptable, then add warnings. */
if (global_pa->pick_anyway && format_output) {
if (sarg->left_input) {
add_must_use_warnings(&retval->warnings,
"Left primer", &retval->fwd.expl);
}
if (sarg->right_input) {
add_must_use_warnings(&retval->warnings,
"Right primer", &retval->rev.expl);
}
if (sarg->internal_input) {
add_must_use_warnings(&retval->warnings,
"Hybridization probe", &retval->intl.expl);
}
}
if (pr_is_empty(&retval->glob_err)
&& pr_is_empty(&retval->per_sequence_err)) {
/* We need to test for errors before we call
p3_print_oligo_lists. This function only works on retval as
returned when there were no errors. */
if (read_boulder_record_res.file_flag) {
/* Create files with left, right, and internal oligos. */
p3_print_oligo_lists(retval, sarg, global_pa,
&retval->per_sequence_err,
sarg->sequence_name);
}
}
if (format_output) {
print_format_output(stdout, &io_version, global_pa,
sarg, retval, pr_release,
read_boulder_record_res.explain_flag);
} else {
/* Use boulder output */
print_boulder(io_version, global_pa, sarg, retval,
read_boulder_record_res.explain_flag);
}
loop_wrap_up: /* Here the failed loops join in again */
if (NULL != retval) {
/* Check for errors and print them */
if (NULL != retval->glob_err.data) {
fprintf(stderr, "%s: %s\n", pr_program_name, retval->glob_err.data);
destroy_p3retval(retval);
destroy_seq_args(sarg);
exit(-4);
}
}
destroy_p3retval(retval); /* This works even if retval is NULL */
retval = NULL;
destroy_seq_args(sarg);
sarg = NULL;
} /* while (1) (processing boulder io records) ...
End of the primary working loop */
/* To avoid being distracted when looking for leaks: */
if ((global_pa->thermodynamic_oligo_alignment == 1) ||
(global_pa->thermodynamic_template_alignment == 1))
destroy_thal_structures();
p3_destroy_global_settings(global_pa);
global_pa = NULL;
destroy_seq_args(sarg);
destroy_pr_append_str_data(&nonfatal_parse_err);
destroy_pr_append_str_data(&fatal_parse_err);
destroy_pr_append_str_data(&warnings);
destroy_dpal_thal_arg_holder();
free(thermodynamic_params_path);
/* If it could not read input, then complain and die */
if (0 == input_found) {
print_usage();
exit(-3);
}
return 0;
}