GT_INLINE size_t gt_input_file_next_record(
gt_input_file* const input_file,gt_vector* const buffer_dst,gt_string* const first_field,
uint64_t* const num_blocks,uint64_t* const num_tabs) {
GT_INPUT_FILE_CHECK(input_file);
GT_VECTOR_CHECK(buffer_dst);
GT_INPUT_FILE_CHECK_BUFFER__DUMP(input_file,buffer_dst);
if (input_file->eof) return GT_INPUT_FILE_EOF;
// Read line
uint64_t const begin_line_pos_at_file = input_file->buffer_pos;
uint64_t const begin_line_pos_at_buffer = gt_vector_get_used(buffer_dst);
uint64_t current_pfield = 0, length_first_field = 0;
while (gt_expect_true(!input_file->eof &&
GT_INPUT_FILE_CURRENT_CHAR(input_file)!=EOL &&
GT_INPUT_FILE_CURRENT_CHAR(input_file)!=DOS_EOL)) {
if (current_pfield==0) {
if (gt_expect_false(GT_INPUT_FILE_CURRENT_CHAR(input_file)==TAB)) {
++current_pfield;
++(*num_tabs);
} else {
++length_first_field;
}
} else if (current_pfield==1) {
if (gt_expect_false(GT_INPUT_FILE_CURRENT_CHAR(input_file)==SPACE)) {
++(*num_blocks);
} else if (gt_expect_false(GT_INPUT_FILE_CURRENT_CHAR(input_file)==TAB)) {
++current_pfield;
++(*num_tabs);
++(*num_blocks);
}
} else {
if (gt_expect_false(GT_INPUT_FILE_CURRENT_CHAR(input_file)==TAB)) {
++current_pfield;
++(*num_tabs);
}
}
GT_INPUT_FILE_NEXT_CHAR__DUMP(input_file,buffer_dst);
}
// Handle EOL
GT_INPUT_FILE_HANDLE_EOL(input_file,buffer_dst);
// Set first field (from the input_file_buffer or the buffer_dst)
if (first_field) {
char* first_field_begin;
if (input_file->buffer_pos <= begin_line_pos_at_file) {
gt_input_file_dump_to_buffer(input_file,buffer_dst); // Forced to dump to buffer
first_field_begin = gt_vector_get_elm(buffer_dst,begin_line_pos_at_buffer,char);
} else {
/*
* Basic line functions
*/
GT_INLINE size_t gt_input_file_dump_to_buffer(gt_input_file* const input_file,gt_vector* const buffer_dst) { // FIXME: If mmap file, internal buffer is just pointers to mem
GT_INPUT_FILE_CHECK(input_file);
// Copy internal file buffer to buffer_dst
const uint64_t chunk_size = input_file->buffer_pos-input_file->buffer_begin;
if (gt_expect_false(chunk_size==0)) return 0;
gt_vector_reserve_additional(buffer_dst,chunk_size);
memcpy(gt_vector_get_mem(buffer_dst,uint8_t)+gt_vector_get_used(buffer_dst),
input_file->file_buffer+input_file->buffer_begin,chunk_size);
gt_vector_add_used(buffer_dst,chunk_size);
// Update position
input_file->buffer_begin=input_file->buffer_pos;
// Return number of written bytes
return chunk_size;
}
GT_INLINE void gt_cdna_string_append_string(gt_compact_dna_string* const cdna_string,char* const string,const uint64_t length) {
GT_COMPACT_DNA_STRING_CHECK(cdna_string);
// Check allocated bitmaps
register const uint64_t total_chars = cdna_string->length+length-1;
if (total_chars >= cdna_string->allocated) {
gt_cdna_string_resize(cdna_string,total_chars);
}
// Copy string
register uint64_t block_num, block_pos, i;
GT_CDNA_GET_BLOCK_POS(cdna_string->length,block_num,block_pos);
register uint64_t* block_mem = GT_CDNA_GET_MEM_BLOCK(cdna_string->bitmaps,block_num);
for (i=0; i<length; ++i,++block_pos) {
if (gt_expect_false(block_pos==GT_CDNA_BLOCK_CHARS)) {
block_pos=0;
block_mem+=GT_CDNA_BLOCK_BITMAPS;
}
register const uint8_t enc_char = gt_cdna_encode(string[i]);
GT_CDNA_SET_CHAR(block_mem,block_pos,enc_char);
}
// Update total length
cdna_string->length = total_chars+1;
}