Blame view
tools/sctk-2.4.10/src/sclite/stm.c
7.76 KB
8dcb6dfcb
first commit
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 |
#include "sctk.h"
STM *alloc_STM(int n){
STM *ts;
alloc_singarr(ts,1,STM);
ts->max = n;
ts->num = 0;
ts->s = 0;
alloc_singarr(ts->seg,ts->max,STM_SEG);
return(ts);
}
void expand_STM(STM *stm){
expand_singarr(stm->seg,stm->num,stm->max,2,STM_SEG);
}
void free_STM_SEG(STM_SEG *seg){
if (seg == (STM_SEG *)0)
return;
TEXT_free(seg->file);
TEXT_free(seg->spkr);
TEXT_free(seg->chan);
TEXT_free(seg->text);
if (seg->labels != (TEXT *)0) TEXT_free(seg->labels);
}
void free_STM(STM *stm){
int i;
for (i=0; i<stm->num; i++)
free_STM_SEG(&(stm->seg[i]));
free_singarr(stm->seg,STM_SEG);
free_singarr(stm,STM);
}
void read_stm_line(TEXT **buf, int *len, FILE *fp){
**buf = *(*buf+1) = COMMENT_CHAR;
while (!feof(fp) && (TEXT_is_comment(*buf) ||
TEXT_is_comment_info(*buf) ||
TEXT_is_empty(*buf))){
if (TEXT_ensure_fgets(buf, len, fp) == NULL)
**buf = '\0';
}
}
void parse_stm_line(STM_SEG *seg, TEXT **buf_ptr, int *buf_len, int case_sense, int dbg){
TEXT *rp, *buf;
int i, len;
if (**buf_ptr == '\0')
return;
if (!case_sense)
TEXT_str_case_change_with_mem_expand(buf_ptr, buf_len, 1);
buf = *buf_ptr;
len = TEXT_strlen(buf);
/* parse stm string */
rp = TEXT_strtok(buf,(TEXT *)" \t
"); seg->file = TEXT_strdup(rp);
rp = TEXT_strtok((TEXT *)0,(TEXT *)" \t
"); seg->chan = TEXT_strdup(rp);
rp = TEXT_strtok((TEXT *)0,(TEXT *)" \t
"); seg->spkr = TEXT_strdup(rp);
rp = TEXT_strtok((TEXT *)0,(TEXT *)" \t
"); seg->t1 = TEXT_atof(rp);
rp = TEXT_strtok((TEXT *)0,(TEXT *)" \t
"); seg->t2 = TEXT_atof(rp);
seg->flag1 = 0;
seg->labels = (TEXT *)0;
if (rp != (TEXT *)0){
/* snag the next token, if it's a set identifier, load it, */
/* if not, overwrite the ending null with a space provided */
/* there is data beyond the end of this token */
if (((rp = TEXT_strtok((TEXT *)0,(TEXT *)" \t
")) != (TEXT *)0) &&
(*rp == '<')) {
seg->labels = TEXT_strdup(rp);
rp = TEXT_strtok((TEXT *)0,(TEXT *)" \t
");
} else
seg->labels = (TEXT *)0;
if (rp != (TEXT *)0){
/* reclaim the first token from the text */
if (rp + TEXT_strlen(rp) < buf + len)
*(rp + TEXT_strlen(rp)) = ' ';
seg->text = TEXT_strdup(rp);
} else
seg->text = TEXT_strdup((TEXT *)"");
} else
seg->text = TEXT_strdup((TEXT *)"");
/* was seg->text = TEXT_strdup(rp + TEXT_strlen(rp) + 1); */
if (dbg){
printf("Parsed: file: %s
",seg->file);
printf(" chan: %s
",seg->chan);
printf(" spkr: %s
",seg->spkr);
printf(" t1 : %f
",seg->t1);
printf(" t2 : %f
",seg->t2);
printf(" text: %s
",seg->text);
}
i = TEXT_strlen(seg->text) - 1;
if (i > 0 && seg->text[i] == '
')
seg->text[i] = (TEXT)0;
}
void fill_STM(FILE *fp, STM *stm, char *fname, boolean *end_of_file, int case_sense, int *perr){
static int len=100;
int i;
TEXT *buf;
int dbg=0;
alloc_singZ(buf,len,TEXT,(TEXT)0);
*perr = 0;
/* if the data doesn't begin at 1, copy it down and go from there */
if (stm->s > 1){
/* first free the already alloc'd data */
for (i=0; i<stm->s; i++){
TEXT_free(stm->seg[i].file);
TEXT_free(stm->seg[i].spkr);
TEXT_free(stm->seg[i].chan);
TEXT_free(stm->seg[i].text);
if (stm->seg[i].labels != (TEXT *)0)
TEXT_free(stm->seg[i].labels);
}
/* then copy down the residual data */
for (i=stm->s; i<=stm->num; i++)
stm->seg[i - stm->s] = stm->seg[i];
stm->num = stm->num - stm->s;
stm->s=0;
}
/* now read in the data */
while (!feof(fp) && stm->num+1 < stm->max){
read_stm_line(&buf,&len,fp);
if (dbg) printf("STM Read %s
",buf);
if (*buf != NULL_TEXT){
parse_stm_line(&(stm->seg[stm->num]), &buf, &len, case_sense, dbg);
stm->num++;
}
}
*end_of_file = feof(fp);
free_singarr(buf,TEXT);
}
void locate_STM_boundary(STM *stm, int start, int by_file, int by_chan, int *end){
int w;
int limit=-1;
int tchg, cchg;
if (start == stm->num){
*end = start;
return;
}
for (w=start; w<stm->num && limit == -1; w++){
tchg = (!by_file) ? 1 :
(TEXT_strcmp(stm->seg[start].file,stm->seg[w].file) == 0);
cchg = (!by_chan) ? 1 :
(TEXT_strcmp(stm->seg[start].chan,stm->seg[w].chan) == 0);
if (!(tchg && cchg))
limit = w-1;
}
if (limit == -1)
limit = stm->num;
*end = limit;
}
void dump_STM_words(STM *stm,int s, int e, char *file){
int i;
FILE *fp = fopen(file,"w");
TEXT *ctext, *tbuf;
int tbuf_len=100;
if (fp == NULL){
fprintf(stderr,"Error: Can't open STM words file '%s'
",file);
exit(1);
}
alloc_singZ(tbuf,tbuf_len,TEXT,'\0');
for (i=s; i<=e; i++){
/* Write each individual word to a file */
if (TEXT_strlen(stm->seg[i].text) > tbuf_len+1){
free_singarr(tbuf,TEXT);
tbuf = TEXT_strdup(stm->seg[i].text);
tbuf_len = TEXT_strlen(tbuf);
} else
TEXT_strcpy(tbuf,stm->seg[i].text);
ctext = tokenize_TEXT_first_alt(tbuf,(TEXT *)" \t
");
while (ctext != NULL) {
fprintf(fp,"%s
",ctext);
ctext = tokenize_TEXT_first_alt(NULL,(TEXT *)" \t
");
}
}
fclose(fp);
free_singarr(tbuf,TEXT);
}
void dump_STM(STM *stm, int s, int e){
char *proc="dump_STM";
int i;
printf("%s: Range [%d,%d] of [%d,%d]
",proc,s,e,stm->s,stm->num);
for (i=s; i<e; i++)
printf("%d: %s chan: %s Spkr: %s T1: %f T2:%f Text: '%s'
",i,
stm->seg[i].file,stm->seg[i].chan,
stm->seg[i].spkr,stm->seg[i].t1,
stm->seg[i].t2,stm->seg[i].text);
printf("
");
}
void convert_stm_to_word_list(char *file, char *words, int case_sense, int *num_ref){
STM_SEG seg;
FILE *fp, *fp_out;
TEXT *buf, *ctext;
int buf_len=100;
if ((fp=fopen(file,"r")) == NULL){
fprintf(stderr,"Error: Unable to open input STM file '%s'
",file);
exit(1);
}
if ((fp_out=fopen(words,"w")) == NULL){
fprintf(stderr,"Error: Unable to open STM word file '%s'
",words);
exit(1);
}
(*num_ref) = 0;
alloc_singZ(buf,buf_len,TEXT,'\0');
/* for each stm, load the aligned text and create a path as before */
while (!feof(fp)){
/* read the next ref line */
read_stm_line(&buf,&buf_len,fp);
if (feof(fp)) break;
/* parse the reference transcript */
parse_stm_line(&seg,&buf,&buf_len,case_sense,0);
/* for each reference word, located it's matching diff output */
ctext = tokenize_TEXT_first_alt(seg.text,(TEXT *)" \t
");
while (ctext != NULL){
fprintf(fp_out,"%s
",ctext);
ctext = tokenize_TEXT_first_alt(NULL,(TEXT *)" \t
");
}
(*num_ref) ++;
free_STM_SEG(&seg);
}
fclose(fp);
fclose(fp_out);
free_singarr(buf,TEXT);
}
/****************************************************************/
/* File an stm structure, making sure all records have been */
/* loaded for each file in the designator. If not, additional */
/* records are read in. */
/****************************************************************/
void fill_STM_structure(STM *stm, FILE *fp_stm, char *stm_file, int *stm_file_end, int case_sense){
static int stm_eof=0, stm_err=0;
int just_read;
if (stm == (STM *)0){
/* Reset the static variables */
stm_eof = stm_err = 0;
return;
}
just_read = 0;
do {
locate_STM_boundary(stm, stm->s, 1, 0, stm_file_end);
if ((*stm_file_end == stm->num) && !stm_eof){
if (just_read == 1){
just_read = 0;
expand_STM(stm);
}
fill_STM(fp_stm, stm, stm_file, &stm_eof, case_sense,&stm_err);
if (stm_err != 0){
fprintf(stdout,"; *Err: Error detected in STM file '%s'
",
stm_file);
exit(1);
}
just_read = 1;
} else
just_read = 0;
} while (just_read);
}
|