1 %{ 2 3 /*- 4 * %sccs.include.proprietary.c% 5 */ 6 7 #ifndef lint 8 static char sccsid[] = "@(#)part.l 8.2 (Berkeley) 01/28/94"; 9 #endif /* not lint */ 10 11 #include "style.h" 12 #include "names.h" 13 #include "conp.h" 14 15 #undef yywrap 16 17 FILE *deb; 18 int nosave = 1; 19 int part = 0; 20 int barebones = 0; 21 int topic = 0; 22 int style = 1; 23 int pastyle = 0; 24 int pstyle = 0; 25 int lstyle = 0; 26 int rstyle = 0; 27 int estyle = 0; 28 int nstyle = 0; 29 int Nstyle = 0; 30 int lthresh; 31 int rthresh; 32 int nomin; 33 char s[SCHAR]; 34 char *sptr = s; 35 struct ss sent[SLENG]; 36 struct ss *sentp = sent; 37 float wperc(); 38 float sperc(); 39 float typersent(); 40 float vperc(); 41 int numsent = 0; 42 int qcount = 0; 43 int icount = 0; 44 long vowel = 0; 45 long numwds = 0; 46 long twds = 0; 47 long numnonf = 0; 48 long letnonf = 0; 49 int maxsent = 0; 50 int maxindex = 0; 51 int minsent = 30; 52 int minindex = 0; 53 int simple = 0; 54 int compound = 0; 55 int compdx = 0; 56 int prepc = 0; 57 int conjc = 0; 58 int complex = 0; 59 int tobe = 0; 60 int adj = 0; 61 int infin = 0; 62 int pron = 0; 63 int passive = 0; 64 int aux = 0; 65 int adv = 0; 66 int verbc = 0; 67 int tverbc = 0; 68 int noun = 0; 69 long numlet = 0; 70 int beg[15] = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 }; 71 int sleng[50]; 72 int nsleng = 0; 73 int j,jj,i; 74 int comma = 0; 75 int cflg; 76 int question; 77 int quote = 0; 78 char *st; 79 int initf = 0; 80 int over = 1; 81 int nroff = 0; 82 int nrofflg = 0; 83 int leng[MAXPAR]; 84 int sentno= 0; 85 %} 86 C [A-DF-Za-z] 87 %% 88 ^{C}:.+ { 89 collect: 90 sentp->cc = sentp->ic = yytext[0]; 91 if(sentp->cc == NOM) 92 sentp->cc = NOUN; 93 collect1: 94 nsleng++; 95 sentp->leng = yyleng-2; 96 sentp++->sp = sptr; 97 if(sentp >= &sent[SLENG-1]){ 98 if(over)fprintf(stderr,"sentence %d too many words\n",numsent+2); 99 over=0; 100 sentp--; 101 } 102 if(sptr+yyleng >= &s[SCHAR-1]){ 103 if(over)fprintf(stderr,"sentence %d too many characters\n",numsent+2); 104 over=0; 105 } 106 else { 107 for(i=2;i<yyleng;i++)*sptr++=yytext[i]; 108 *sptr++ = '\0'; 109 } 110 } 111 ^";:".+ { 112 sentp->cc=END; 113 sentp->ic = ';'; 114 goto collect1; 115 } 116 ^",:".+ { 117 comma++; 118 goto collect; 119 } 120 [\n] ; 121 ^"\":".+ { 122 goto collect; 123 } 124 ^E:"/." { 125 cflg = 1; 126 goto sdone; 127 } 128 ^E:.+ { 129 cflg = 0; 130 sdone: 131 over=1; 132 sentp->cc=sentp->ic=END; 133 sentp++->sp = sptr; 134 for(i=2;i<yyleng;i++)*sptr++=yytext[i]; 135 *sptr++='\0'; 136 if(yytext[2]=='?')question=1; 137 else question=0; 138 139 fragment: 140 jj=0; 141 if(quote == 1 && sent[jj].cc == ED){ 142 sent[jj].cc = VERB; 143 quote = 0; 144 } 145 if(sent[jj].cc=='"')jj++; 146 if(sent[jj].cc==SUBCONJ){ 147 if(sent[jj+1].cc == ','){ 148 sent[jj].cc=ADV; 149 jj += 2; 150 comma--; 151 } 152 else { 153 jj=scan(1,',',0); 154 if(jj != -1)jj++; 155 comma--; 156 } 157 } 158 if(jj != -1){ 159 if(sent[jj].cc==CONJ || sent[jj].cc=='"')jj++; 160 while((jj=scan(jj,END,cflg)) != -1){ 161 jj++; 162 if(sent[jj].cc == SUBCONJ && sent[jj+1].cc == ','){ 163 sent[jj].cc=ADV; 164 jj += 2; 165 comma--; 166 } 167 } 168 } 169 st = sent[i].sp; 170 if(*(st+1) == '"') 171 if(*st == '?' || *st == '!')quote = 1; 172 outp(); 173 nsleng = 0; 174 if(nroff){ 175 if(sentno > 0){ 176 printf(".SL \""); 177 for(i=0;i<sentno;i++) 178 printf(" %d",leng[i]); 179 printf("\"\n"); 180 sentno = 0; 181 } 182 printf("%s",&yytext[1]); 183 nroff = 0; 184 } 185 sptr=s; 186 sentp=sent; 187 comma=0; 188 } 189 ;.+ { 190 if(style){ 191 nomin = atoi(&yytext[1]); 192 } 193 } 194 ^:.+[\n] { 195 nrofflg=1; 196 if(sentp != sent){ 197 sentp->cc = sentp->ic = END; 198 sentp++->sp = sptr; 199 *sptr++ = '.'; 200 *sptr++ = '\0'; 201 over = 1; 202 nroff = 1; 203 goto fragment; 204 } 205 if(sentno > 0){ 206 printf(".SL \""); 207 for(i=0;i<sentno;i++) 208 printf(" %d",leng[i]); 209 printf("\"\n"); 210 sentno = 0; 211 } 212 printf("%s",&yytext[1]); 213 } 214 %% 215 yywrap(){ 216 int ii; 217 int ml,mg,lsum,gsum; 218 float aindex, avl, avw; 219 float cindex,kindex,findex,fgrad; 220 float syl, avsy, adjs,snonf; 221 FILE *io; 222 223 if(style){ 224 if(numwds == 0 || numsent == 0)exit(0); 225 avw = (float)(numwds)/(float)(numsent); 226 avl = (float)(numlet)/(float)(numwds); 227 aindex = 4.71*avl + .5*avw -21.43; 228 syl = .9981*vowel-.3432*twds; 229 avsy = syl/twds; 230 kindex = 11.8*avsy+.39*avw-15.59; 231 findex = 206.835-84.6*avsy-1.015*avw; 232 if(findex < 30.)fgrad = 17.; 233 else if(findex > 100.) fgrad = 4.; 234 else if(findex > 70.)fgrad=(100.-findex)/10 +5.; 235 else if(findex > 60.)fgrad =(70.-findex)/10+8.; 236 else if(findex >50.)fgrad=(60.-findex)/5+10; 237 else fgrad=(50.-findex)/6.66 +13.; 238 adjs = 100 * (float)numsent/numwds; 239 cindex = 5.89*avl-.3*adjs-15.8; 240 printf("readability grades:\n (Kincaid) %4.1f (auto) %4.1f (Coleman-Liau) %4.1f (Flesch) %4.1f (%4.1f)\n",kindex,aindex,cindex,fgrad,findex); 241 printf("sentence info:\n"); 242 printf(" no. sent %d no. wds %ld\n",numsent,numwds); 243 printf(" av sent leng %4.1f av word leng %4.2f\n",avw,avl); 244 printf(" no. questions %d no. imperatives %d\n",qcount,icount); 245 if(numnonf != 0){ 246 snonf = (float)(letnonf)/(float)(numnonf); 247 } 248 printf(" no. nonfunc wds %ld %4.1f%% av leng %4.2f\n",numnonf,(float)(numnonf)*100/numwds,snonf); 249 mg = avw + 10.5; 250 if(mg > 49)mg = 49; 251 ml = avw - 4.5; 252 if(ml <= 0)ml = 1; 253 else if(ml > 49)ml=48; 254 gsum = lsum = 0; 255 for(ii=0;ii<50;ii++){ 256 if(ii < ml)lsum += sleng[ii]; 257 else if(ii > mg)gsum+= sleng[ii]; 258 } 259 printf(" short sent (<%d)%3.0f%% (%d) long sent (>%d) %3.0f%% (%d)\n",ml,sperc(lsum),lsum,mg,sperc(gsum),gsum); 260 printf(" longest sent %d wds at sent %d; shortest sent %d wds at sent %d\n",maxsent,maxindex,minsent,minindex); 261 printf("sentence types:\n"); 262 printf(" simple %3.0f%% (%d) complex %3.0f%% (%d)\n",sperc(simple),simple,sperc(complex),complex); 263 printf(" compound %3.0f%% (%d) compound-complex %3.0f%% (%d)\n",sperc(compound),compound,sperc(compdx),compdx); 264 printf("word usage:\n"); 265 printf(" verb types as %% of total verbs\n"); 266 printf(" tobe %3.0f%% (%d) aux %3.0f%% (%d) inf %3.0f%% (%d)\n",vperc(tobe),tobe,vperc(aux),aux,vperc(infin),infin); 267 if(verbc != 0)adjs = (float)(passive)*100/(float)(verbc); 268 else adjs=0; 269 printf(" passives as %% of non-inf verbs %3.0f%% (%d)\n",adjs,passive); 270 printf(" types as %% of total\n"); 271 printf(" prep %3.1f%% (%d) conj %3.1f%% (%d) adv %3.1f%% (%d)\n",wperc(prepc),prepc,wperc(conjc),conjc,wperc(adv),adv); 272 printf(" noun %3.1f%% (%d) adj %3.1f%% (%d) pron %3.1f%% (%d)\n",wperc(noun),noun, 273 wperc(adj),adj,wperc(pron),pron); 274 printf(" nominalizations %3.0f %% (%d)\n",wperc(nomin),nomin); 275 printf("sentence beginnings:\n"); 276 ii=beg[0]+beg[7]+beg[6]+beg[3]+beg[8]; 277 printf(" subject opener: noun (%d) pron (%d) pos (%d) adj (%d) art (%d) tot %3.0f%%\n", 278 beg[0],beg[7],beg[6],beg[3],beg[8],sperc(ii)); 279 printf(" prep %3.0f%% (%d) adv %3.0f%% (%d) \n",sperc(beg[9]),beg[9],sperc(beg[4]),beg[4]); 280 printf(" verb %3.0f%% (%d) ",sperc(beg[1]+beg[10]+beg[11]),beg[1]+beg[10]+beg[11]); 281 printf(" sub_conj %3.0f%% (%d) conj %3.0f%% (%d)\n",sperc(beg[13]),beg[13],sperc(beg[5]),beg[5]); 282 printf(" expletives %3.0f%% (%d)\n",sperc(beg[14]),beg[14]); 283 #ifdef SCATCH 284 if(nosave && (fopen(SCATCH,"r")) != NULL){ 285 if(((io=fopen(SCATCH,"a")) != NULL)){ 286 fprintf(io," read %4.1f %4.1f %4.1f %4.1f %4.1f\n",kindex, aindex, cindex, findex, fgrad); 287 fprintf(io," sentl %d %ld %4.2f %4.2f %d %d %ld %4.2f\n",numsent,numwds,avw,avl,qcount,icount,numnonf,snonf); 288 fprintf(io," l var %d %d %d %d %d\n",ml,lsum,mg,gsum,maxsent); 289 fprintf(io," t var %d %d %d %d\n",simple,complex,compound,compdx); 290 fprintf(io," verbs %d %d %d %d %d %d\n",tverbc,verbc,tobe,aux,infin,passive); 291 fprintf(io," ty %d %d %d %d %d %d %d\n",prepc,conjc,adv,noun,adj,pron,nomin); 292 fprintf(io," beg %d %d %d %d %d %d\n",beg[0],beg[7],beg[6],beg[3],beg[8],ii); 293 fprintf(io," sbeg %d %d %d %d %d %d\n",beg[9],beg[4],beg[1]+beg[10]+beg[11],beg[13],beg[5],beg[14]); 294 } 295 } 296 #endif 297 } 298 return(1); 299 } 300 float 301 wperc(a) 302 { 303 return((float)(a)*100/numwds); 304 } 305 float 306 sperc(a) 307 { 308 return((float)(a)*100/numsent); 309 } 310 float 311 typersent(a) 312 { 313 return((float)(a)/numsent); 314 } 315 float 316 vperc(a) 317 { 318 if(tverbc == 0)return(0); 319 return((float)(a)*100/tverbc); 320 } 321 main(argc,argv) 322 char **argv; 323 { 324 while(--argc > 0 && (++argv)[0][0] == '-' ){ 325 switch(argv[0][1]){ 326 case 'd': nosave = 0; 327 continue; 328 case 's': style=1; 329 continue; 330 case 'p': pastyle=style=1; 331 continue; 332 case 'a': pstyle=style=1; 333 continue; 334 case 'e': estyle = style = 1; 335 continue; 336 case 'n': nstyle = style = 1; 337 continue; 338 case 'N': Nstyle = style = 1; 339 continue; 340 case 'l': style=lstyle=1; 341 lthresh = atoi(*(++argv)); 342 argc--; 343 continue; 344 case 'r': 345 style=rstyle=1; 346 rthresh = atoi(*(++argv)); 347 argc--; 348 continue; 349 case 'P': 350 part = 1; 351 style = 0; 352 continue; 353 case 'b': /* print bare bones info rje */ 354 barebones = 1; 355 style = 0; 356 continue; 357 case 'T': /*topic*/ 358 style = 0; 359 topic = 1; 360 continue; 361 default: 362 fprintf(stderr,"unknown flag to part %s\n",*argv); 363 exit(1); 364 } 365 argv++; 366 } 367 #ifdef SNOM 368 if(fopen(SNOM,"r") != NULL){ 369 deb = fopen(SNOM,"a"); /* SAVE NOM*/ 370 } 371 #else 372 deb = NULL; 373 #endif 374 yylex(); 375 if(nrofflg && sentno > 0){ 376 printf(".SL \""); 377 for(i=0;i<sentno;i++) 378 printf(" %d",leng[i]); 379 printf("\"\n"); 380 } 381 } 382