void AsmPass1(Assembler *a, char *text) { // 第一階段的組譯
int i, address = 0, number;
Array* lines = split(text, "\r\n", REMOVE_SPLITER); // 將組合語言分割成一行一行
ArrayEach(lines, strPrintln); // 印出以便觀察
printf("=================PASS1================\n");
for (i=0; i<lines->count; i++) { // 對於每一行
strReplace(lines->item[i], ";", '\0');
strReplace(lines->item[i], "\t", ' ');
AsmCode *code = AsmCodeNew(lines->item[i]); // 建立指令物件
if (code == NULL) continue;
code->address = address; // 設定該行的位址
Op *op = NULL;
if (code->op != NULL) {
op = HashTableGet(opTable, code->op); // 查詢運算碼
if (op != NULL) { // 如果查到
code->opCode = op->code; // 設定運算碼
code->type = op->type; // 設定型態
}
}
if (code->label != NULL) // 如果有標記符號
HashTablePut(a->symTable, code->label, code); // 加入符號表中
ArrayAdd(a->codes, code); // 建構指令物件陣列
AsmCodePrintln(code); // 印出觀察
code->size = AsmCodeSize(code); // 計算指令大小
address += code->size; // 計算下一個指令位址
}
ArrayFree(lines, strFree); // 釋放記憶體
}
int transfer_SAM_to_position_table(char * sam_file)
{
char linebuf[max(2*MAX_READ_LENGTH+300,3000)];
int allreads=0,mapped=0;
char mate_chro[MAX_CHROMOSOME_NAME_LEN+1];
unsigned int mate_pos;
int flags_mate;
char cigar_mate[EXON_MAX_CIGAR_LEN+1];
gene_input_t input_file;
HashTable * local_reassembly_pileup_files;
HashTable * chromosome_size_table;
chromosome_size_table = HashTableCreate(100);
HashTableSetDeallocationFunctions(chromosome_size_table, NULL, NULL);
HashTableSetKeyComparisonFunction(chromosome_size_table, my_strcmp);
HashTableSetHashFunction(chromosome_size_table, HashTableStringHashFunction);
local_reassembly_pileup_files = HashTableCreate(100);
HashTableSetDeallocationFunctions(local_reassembly_pileup_files, NULL, NULL);
HashTableSetKeyComparisonFunction(local_reassembly_pileup_files, my_strcmp);
HashTableSetHashFunction(local_reassembly_pileup_files ,HashTableStringHashFunction);
geinput_open_sam(sam_file, &input_file,0);
while(1)
{
unsigned int read_pos =0xffffffff;
char read_name[MAX_READ_NAME_LEN+1];
int flags;
char chro_name[MAX_CHROMOSOME_NAME_LEN+1];
unsigned int pos, pair_dist;
char cigar[EXON_MAX_CIGAR_LEN+1];
char read_text[MAX_READ_LENGTH+1];
char qual_text[MAX_READ_LENGTH+1];
int read_len, is_repeated, mapping_quality, is_anchor_certain=1;
int pos_delta = 0;
int is_paired_end_reads = 0;
unsigned long long int file_offset = ftello(input_file.input_fp);
if(feof(input_file.input_fp))break;
read_text[0]=0;
qual_text[0]=0;
geinput_readline(&input_file, linebuf,0);
int res = parse_SAM_line(linebuf, read_name,& flags, chro_name, & pos, cigar, & mapping_quality, & pair_dist, read_text , qual_text, & read_len, & is_repeated);
int cigar_cursor = 0;
int firstM = 1,xx=0;
is_paired_end_reads = flags &1;
if(res==0)
{
for(; cigar[cigar_cursor]; cigar_cursor++)
{
char nch = cigar[cigar_cursor];
if(nch>='0'&&nch<='9')
{
xx=xx*10+(nch-'0');
}else
{
if(nch=='M') firstM=0;
else if(nch=='S')
{
if(firstM)
pos_delta = xx;
}
xx=0;
}
}
pos -= pos_delta;
}
if(res == 1) {chro_name[0]='*'; chro_name[1]=0;}
//printf("MAPPED=%d\n", res);
if(res == 0) // mapped
{
read_pos = pos - 1;
mapped++;
}
else if(res == 1 && is_paired_end_reads) // unmapped
{
is_anchor_certain=0;
if(mate_chro[0])
{
if(mate_chro[0]!='*')
{
read_pos = mate_pos ;
strcpy(chro_name, mate_chro);
}
//printf("RECOVERED 1: %u - %s ; LEN=%d ; YOU_ARE_FIRST=%d\n%s\n%s\n", read_anchor_position, read_name, read_len, flags_mate & SAM_FLAG_FIRST_READ_IN_PAIR, read_text, qual_text);
}
else
{
char read_text_null[MAX_READ_LENGTH+1];
char qual_text_null[MAX_READ_LENGTH+1];
geinput_readline_back(&input_file, linebuf);
res = parse_SAM_line(linebuf, read_name,& flags_mate, mate_chro, & mate_pos, cigar_mate, & mapping_quality, & pair_dist, read_text_null , qual_text_null, & read_len, & is_repeated);
if(res==0)
{
read_pos = mate_pos - 1;
strcpy(chro_name, mate_chro);
}
//printf("RECOVERED 2: %u - %s ; LEN=%d ; YOU_ARE_FIRST=%d\n%s\n%s\n", read_anchor_position, read_name, read_len, flags_mate & SAM_FLAG_FIRST_READ_IN_PAIR, read_text, qual_text);
}
}
if(read_pos<0xffff0000)
{
unsigned int read_span = calculate_read_span(cigar);
write_read_pos(chro_name, read_pos, read_len, flags, file_offset, cigar, read_span, local_reassembly_pileup_files);
unsigned int old_max_pos = HashTableGet(chromosome_size_table, chro_name) - NULL;
if(old_max_pos==0)
{
char * chro_name_new = malloc(strlen(chro_name)+1);
strcpy(chro_name_new , chro_name);
HashTablePut(chromosome_size_table, chro_name_new, NULL+read_pos + read_span+1);
}
else if(read_pos + read_span +1 > old_max_pos )
HashTablePutReplace(chromosome_size_table, chro_name, NULL+read_pos + read_span+1, 0);
}
if(allreads %2==1)
mate_chro[0] = 0;
else
{
strcpy(mate_chro, chro_name);
mate_pos = pos - 1;
flags_mate = flags;
}
allreads++;
}
SUBREADprintf("Processed %d reads; %d mapped.\n", allreads, mapped);
destroy_pileup_table(local_reassembly_pileup_files);
finalise_sam_index(chromosome_size_table, sam_file);
HashTableDestroy(chromosome_size_table);
return 0;
}
int load_junc_table(char * file_name, HashTable * tab, HashTable * edge_table_l, HashTable * edge_table_r, HashTable * bin_table)
{
FILE * fp=f_subr_open(file_name, "r");
char new_fl[200];
if(!fp) return -1;
while(1)
{
char * ll = fgets(new_fl, 199, fp);
char * tmp_tok=NULL;
if(!ll || strlen(ll)<4) break;
strtok_r(ll, "\t", &tmp_tok); //name
char * chrostr = strtok_r(NULL, "\t", &tmp_tok); //chro
char * pos1str = strtok_r(NULL, "\t", &tmp_tok); //pos1
char * pos2str = strtok_r(NULL, "\t", &tmp_tok); //pos2
unsigned int pos1 = (unsigned int)atoll(pos1str);
unsigned int pos2 = (unsigned int)atoll(pos2str);
if(pos1==0||pos2==0)continue;
char * chro_mem = malloc(30);
chro_mem[0]=0;
if(strlen(chrostr)<3)strcpy(chro_mem, "chr");
strcat(chro_mem, chrostr);
long long int pos_key;
if(pos2<pos1){
HashTablePut(edge_table_l, NULL+pos2, chro_mem);
HashTablePut(edge_table_r, NULL+pos1, chro_mem);
pos_key = (pos2*1LLU<<32)|pos1;
}
else{
HashTablePut(edge_table_l, NULL+pos1, chro_mem);
HashTablePut(edge_table_r, NULL+pos2, chro_mem);
pos_key = (pos1*1LLU<<32)|pos2;
}
HashTablePut(bin_table, NULL+pos1/50,NULL+pos1);
HashTablePut(bin_table, NULL+pos2/50,NULL+pos2);
HashTablePut(bin_table, NULL+pos1/50-1,NULL+pos1);
HashTablePut(bin_table, NULL+pos2/50-1,NULL+pos2);
HashTablePut(bin_table, NULL+pos1/50+1,NULL+pos1);
HashTablePut(bin_table, NULL+pos2/50+1,NULL+pos2);
HashTablePut(tab, (void *)pos_key, chro_mem);
//printf("NEW JUNC: %s: %u - %u\n", chro_mem, pos1, pos2);
}
fclose(fp);
return 0;
}
