static void initialise_rbt(GtHuffman *huffman,
const void *distr,
GtDistrFunc distrfunc)
{
GtHuffmanTree *huffptr, GT_UNUSED *huffptr2;
unsigned long i;
bool nodecreated = false;
huffman->numofsymbols = 0;
huffman->rbt_root = gt_rbtree_new(huffman_tree_cmp,
huffman_tree_delete,
NULL);
gt_assert(huffman->rbt_root);
for (i = 0; i < huffman->totalnumofsymbols; i++) {
if (distrfunc(distr, i) > 0) {
huffptr = huffman_tree_new(i, distrfunc(distr, i));
huffptr2 = (GtHuffmanTree*)gt_rbtree_search(huffman->rbt_root,
huffptr,
&nodecreated);
gt_assert(nodecreated && huffptr2);
huffman->numofsymbols++;
}
}
}
static GtRBTree *seq_decoder_init_file_info(FastqFileInfo *fileinfos,
GtUword num_off_files)
{
GtRBTree *tree;
bool nodecreated = false;
GtUword i;
tree = gt_rbtree_new(hcr_cmp_FastqFileInfo, NULL, NULL);
for (i = 0; i < num_off_files; i++) {
(void) gt_rbtree_search(tree, &fileinfos[i], &nodecreated);
}
gt_assert(nodecreated);
return tree;
}
int gt_rbtree_unit_test(GtError *err)
{
int had_err = 0;
GtRBTree *tree = NULL;
GtUword i, j, k, *v;
gt_error_check (err);
gt_rbtree_xtab = gt_malloc(GT_RBTREE_SIZE * sizeof (*gt_rbtree_xtab));
gt_rbtree_ytab = gt_malloc(2*GT_RBTREE_SIZE * sizeof (*gt_rbtree_ytab));
gt_rbtree_ztab = gt_malloc(GT_RBTREE_SIZE * sizeof (*gt_rbtree_ztab));
gt_rbtree_depths = gt_malloc(GT_RBTREE_SIZE * sizeof (*gt_rbtree_depths));
tree = gt_rbtree_new(nrbt_cmp_fn, NULL, NULL);
gt_ensure(tree != NULL);
for (i = 0; i < (GtUword) GT_RBTREE_SIZE; ++i) {
gt_rbtree_xtab[i] = i;
}
/* Do this loop several times to get different permutations for the random
case. */
for (i = 0; i < (GtUword) GT_RBTREE_PASSES; ++i) {
NRBT_MANGLECHECK(GT_RBTREE_ASCENDING, GtRBTreeBuild, 0);
NRBT_MANGLECHECK(GT_RBTREE_ASCENDING, GtRBTreeFind, 0);
NRBT_MANGLECHECK(GT_RBTREE_DESCENDING, GtRBTreeFind, 0);
NRBT_MANGLECHECK(GT_RBTREE_RANDOMORDER, GtRBTreeFind, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeDelete, 0);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeBuild, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeDelete, 0);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeBuild, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeDelete, 0);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeBuild, 0);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeFind, 0);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeFind, 0);
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeFind, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeDelete, 0);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING,GtRBTreeBuild, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeDelete, 0);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeBuild, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeDelete, 0);
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeBuild, 0);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeFind, 0);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeFind, 0);
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeFind, 0);
NRBT_WALKCHECK;
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeDelete, 0);
for (j = 1UL; j < (GtUword) GT_RBTREE_SIZE; j *= 2) {
NRBT_MANGLECHECK (GT_RBTREE_RANDOMORDER, GtRBTreeBuildDelete, j);
}
}
for (i = 1UL; i < (GtUword) GT_RBTREE_SIZE; i *= 2) {
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_ASCENDING, GtRBTreeBuildDelete, i);
NRBT_MANGLECHECK (GT_RBTREE_DESCENDING, GtRBTreeBuildDelete, i);
}
gt_rbtree_delete(tree);
gt_free(gt_rbtree_xtab);
gt_free(gt_rbtree_ytab);
gt_free(gt_rbtree_ztab);
gt_free(gt_rbtree_depths);
i = 0; j = 1UL, k = 2UL;
tree = gt_rbtree_new(nrbt_cmp_fn, NULL, NULL);
v = gt_rbtree_root_key(tree);
gt_ensure(!v);
(void) gt_rbtree_insert(tree, &i);
v = gt_rbtree_root_key(tree);
gt_ensure(v == &i);
gt_ensure(*v == i);
(void) gt_rbtree_insert(tree, &j);
(void) gt_rbtree_insert(tree, &k);
v = gt_rbtree_root_key(tree);
gt_ensure(v == &j);
gt_ensure(*v == j);
gt_rbtree_delete(tree);
return had_err;
}
static int hmmsearch_process_coarse_hits(
char *table_filename,
GtCondenseq *ces,
GtCondenseqHmmsearchArguments *arguments,
GtLogger *logger,
GtError *err) {
int had_err = 0;
GtStr *line = gt_str_new();
FILE *table = NULL;
GtSplitter *splitter = gt_splitter_new();
GtStr *query = gt_str_new(),
*fine_fasta_filename = gt_str_new_cstr("condenseq");
GtRBTree *sequences = NULL;
GtUword filecount = (GtUword) 1;
unsigned int querycount = 0;
const GtUword fine_fasta_name_length = gt_str_length(fine_fasta_filename);
const GtUword table_name_length = gt_str_length(arguments->outtable_filename);
table = gt_xfopen(table_filename, "r");
sequences = gt_rbtree_new(hmmsearch_cmp_seqnum,
hmmsearch_tree_free_node, NULL);
while (!had_err && gt_str_read_next_line(line, table) == 0) {
char *c_line = gt_str_get(line);
GtUword uid;
const GtUword target_column = 0,
query_column = (GtUword) 3;
if (c_line[0] != '#') {
gt_splitter_split_non_empty(splitter, c_line, gt_str_length(line), ' ');
gt_assert(gt_splitter_size(splitter) == (GtUword) 23);
if (sscanf(gt_splitter_get_token(splitter, target_column),
GT_WU, &uid) != 1) {
gt_error_set(err, "couldn't parse target number: %s",
gt_splitter_get_token(splitter, target_column));
had_err = -1;
}
if (gt_str_length(query) == 0 ||
strcmp(gt_str_get(query),
gt_splitter_get_token(splitter, query_column)) != 0) {
gt_str_set(query, gt_splitter_get_token(splitter, query_column));
gt_logger_log(logger, "new query: %s", gt_str_get(query));
querycount++;
}
if (!had_err && querycount == arguments->max_queries) {
hmmsearch_create_fine_fas(fine_fasta_filename, sequences, ces);
if (table_name_length != 0)
gt_str_append_uword(arguments->outtable_filename, filecount++);
had_err =
hmmsearch_call_fine_search(table_name_length != 0 ?
arguments->outtable_filename :
NULL,
gt_str_get(fine_fasta_filename),
gt_str_get(arguments->hmmsearch_path),
gt_str_get(arguments->hmm),
logger, err);
gt_rbtree_clear(sequences);
gt_str_set_length(fine_fasta_filename, fine_fasta_name_length);
if (table_name_length != 0)
gt_str_set_length(arguments->outtable_filename, table_name_length);
querycount = 0;
}
if (!had_err) {
if (gt_condenseq_each_redundant_seq(ces, uid,
hmmsearch_process_seq,
sequences, err) == 0) {
had_err = -1;
}
}
gt_splitter_reset(splitter);
}
gt_str_reset(line);
}
gt_splitter_delete(splitter);
gt_str_delete(line);
gt_str_delete(query);
gt_xfclose(table);
if (!had_err) {
hmmsearch_create_fine_fas(fine_fasta_filename, sequences, ces);
if (table_name_length != 0)
gt_str_append_uword(arguments->outtable_filename, filecount++);
had_err =
hmmsearch_call_fine_search(table_name_length != 0 ?
arguments->outtable_filename :
NULL,
gt_str_get(fine_fasta_filename),
gt_str_get(arguments->hmmsearch_path),
gt_str_get(arguments->hmm),
logger, err);
}
gt_log_log("created " GT_WU " files", filecount);
gt_rbtree_delete(sequences);
gt_str_delete(fine_fasta_filename);
return had_err;
}