|
|
1.1 root 1: %{
2:
3: #ifndef lint
4: static char sccsid[] = "@(#)part.l 4.2 (Berkeley) 82/11/06";
5: #endif not lint
6:
7: #include "style.h"
8: #include "names.h"
9: #include "conp.h"
10: FILE *deb;
11: int nosave = 1;
12: int part = 0;
13: int barebones = 0;
14: int topic = 0;
15: int style = 1;
16: int pastyle = 0;
17: int pstyle = 0;
18: int lstyle = 0;
19: int rstyle = 0;
20: int estyle = 0;
21: int nstyle = 0;
22: int Nstyle = 0;
23: int lthresh;
24: int rthresh;
25: int nomin;
26: char s[SCHAR];
27: char *sptr = s;
28: struct ss sent[SLENG];
29: struct ss *sentp = sent;
30: float wperc();
31: float sperc();
32: float typersent();
33: float vperc();
34: int numsent = 0;
35: int qcount = 0;
36: int icount = 0;
37: long vowel = 0;
38: long numwds = 0;
39: long twds = 0;
40: long numnonf = 0;
41: long letnonf = 0;
42: int maxsent = 0;
43: int maxindex = 0;
44: int minsent = 30;
45: int minindex = 0;
46: int simple = 0;
47: int compound = 0;
48: int compdx = 0;
49: int prepc = 0;
50: int conjc = 0;
51: int complex = 0;
52: int tobe = 0;
53: int adj = 0;
54: int infin = 0;
55: int pron = 0;
56: int passive = 0;
57: int aux = 0;
58: int adv = 0;
59: int verbc = 0;
60: int tverbc = 0;
61: int noun = 0;
62: long numlet = 0;
63: int beg[15] = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 };
64: int sleng[50];
65: int nsleng = 0;
66: int j,jj,i;
67: int comma = 0;
68: int cflg;
69: int question;
70: int quote = 0;
71: char *st;
72: int initf = 0;
73: int over = 1;
74: int nroff = 0;
75: int nrofflg = 0;
76: int leng[MAXPAR];
77: int sentno= 0;
78: %}
79: C [A-DF-Za-z]
80: %%
81: ^{C}:.+ {
82: collect:
83: sentp->cc = sentp->ic = yytext[0];
84: if(sentp->cc == NOM)
85: sentp->cc = NOUN;
86: collect1:
87: nsleng++;
88: sentp->leng = yyleng-2;
89: sentp++->sp = sptr;
90: if(sentp >= &sent[SLENG-1]){
91: if(over)fprintf(stderr,"sentence %d too many words\n",numsent+2);
92: over=0;
93: sentp--;
94: }
95: if(sptr+yyleng >= &s[SCHAR-1]){
96: if(over)fprintf(stderr,"sentence %d too many characters\n",numsent+2);
97: over=0;
98: }
99: else {
100: for(i=2;i<yyleng;i++)*sptr++=yytext[i];
101: *sptr++ = '\0';
102: }
103: }
104: ^";:".+ {
105: sentp->cc=END;
106: sentp->ic = ';';
107: goto collect1;
108: }
109: ^",:".+ {
110: comma++;
111: goto collect;
112: }
113: ^",:"[:,-]+ {
114: comma++;
115: goto collect;
116: }
117: [\n] ;
118: ^"\":".+ {
119: goto collect;
120: }
121: ^E:"/." {
122: cflg = 1;
123: goto sdone;
124: }
125: ^E:.+ {
126: cflg = 0;
127: sdone:
128: over=1;
129: sentp->cc=sentp->ic=END;
130: sentp++->sp = sptr;
131: for(i=2;i<yyleng;i++)*sptr++=yytext[i];
132: *sptr++='\0';
133: if(yytext[2]=='?')question=1;
134: else question=0;
135:
136: fragment:
137: jj=0;
138: if(quote == 1 && sent[jj].cc == ED){
139: sent[jj].cc = VERB;
140: quote = 0;
141: }
142: if(sent[jj].cc=='"')jj++;
143: if(sent[jj].cc==SUBCONJ){
144: if(sent[jj+1].cc == ','){
145: sent[jj].cc=ADV;
146: jj += 2;
147: comma--;
148: }
149: else {
150: jj=scan(1,',',0);
151: if(jj != -1)jj++;
152: comma--;
153: }
154: }
155: if(jj != -1){
156: if(sent[jj].cc==CONJ || sent[jj].cc=='"')jj++;
157: while((jj=scan(jj,END,cflg)) != -1){
158: jj++;
159: if(sent[jj].cc == SUBCONJ && sent[jj+1].cc == ','){
160: sent[jj].cc=ADV;
161: jj += 2;
162: comma--;
163: }
164: }
165: }
166: st = sent[i].sp;
167: if(*(st+1) == '"')
168: if(*st == '?' || *st == '!')quote = 1;
169: outp();
170: nsleng = 0;
171: if(nroff){
172: if(sentno > 0){
173: printf(".SL \"");
174: for(i=0;i<sentno;i++)
175: printf(" %d",leng[i]);
176: printf("\"\n");
177: sentno = 0;
178: }
179: printf("%s",&yytext[1]);
180: nroff = 0;
181: }
182: sptr=s;
183: sentp=sent;
184: comma=0;
185: }
186: ;.+ {
187: if(style){
188: nomin = atoi(&yytext[1]);
189: }
190: }
191: ^:.+[\n] {
192: nrofflg=1;
193: if(sentp != sent){
194: sentp->cc = sentp->ic = END;
195: sentp++->sp = sptr;
196: *sptr++ = '.';
197: *sptr++ = '\0';
198: over = 1;
199: nroff = 1;
200: goto fragment;
201: }
202: if(sentno > 0){
203: printf(".SL \"");
204: for(i=0;i<sentno;i++)
205: printf(" %d",leng[i]);
206: printf("\"\n");
207: sentno = 0;
208: }
209: printf("%s",&yytext[1]);
210: }
211: %%
212: yywrap(){
213: int ii;
214: int ml,mg,lsum,gsum;
215: float aindex, avl, avw;
216: float cindex,kindex,findex,fgrad;
217: float syl, avsy, adjs,snonf;
218: FILE *io;
219:
220: if(style){
221: if(numwds == 0 || numsent == 0)exit(0);
222: avw = (float)(numwds)/(float)(numsent);
223: avl = (float)(numlet)/(float)(numwds);
224: aindex = 4.71*avl + .5*avw -21.43;
225: syl = .9981*vowel-.3432*twds;
226: avsy = syl/twds;
227: kindex = 11.8*avsy+.39*avw-15.59;
228: findex = 206.835-84.6*avsy-1.015*avw;
229: if(findex < 30.)fgrad = 17.;
230: else if(findex > 100.) fgrad = 4.;
231: else if(findex > 70.)fgrad=(100.-findex)/10 +5.;
232: else if(findex > 60.)fgrad =(70.-findex)/10+8.;
233: else if(findex >50.)fgrad=(60.-findex)/5+10;
234: else fgrad=(50.-findex)/6.66 +13.;
235: adjs = 100 * (float)numsent/numwds;
236: cindex = 5.89*avl-.3*adjs-15.8;
237: 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);
238: printf("sentence info:\n");
239: printf(" no. sent %d no. wds %ld\n",numsent,numwds);
240: printf(" av sent leng %4.1f av word leng %4.2f\n",avw,avl);
241: printf(" no. questions %d no. imperatives %d\n",qcount,icount);
242: if(numnonf != 0){
243: snonf = (float)(letnonf)/(float)(numnonf);
244: }
245: printf(" no. nonfunc wds %ld %4.1f%% av leng %4.2f\n",numnonf,(float)(numnonf)*100/numwds,snonf);
246: mg = avw + 10.5;
247: if(mg > 49)mg = 49;
248: ml = avw - 4.5;
249: if(ml <= 0)ml = 1;
250: else if(ml > 49)ml=48;
251: gsum = lsum = 0;
252: for(ii=0;ii<50;ii++){
253: if(ii < ml)lsum += sleng[ii];
254: else if(ii > mg)gsum+= sleng[ii];
255: }
256: printf(" short sent (<%d)%3.0f%% (%d) long sent (>%d) %3.0f%% (%d)\n",ml,sperc(lsum),lsum,mg,sperc(gsum),gsum);
257: printf(" longest sent %d wds at sent %d; shortest sent %d wds at sent %d\n",maxsent,maxindex,minsent,minindex);
258: printf("sentence types:\n");
259: printf(" simple %3.0f%% (%d) complex %3.0f%% (%d)\n",sperc(simple),simple,sperc(complex),complex);
260: printf(" compound %3.0f%% (%d) compound-complex %3.0f%% (%d)\n",sperc(compound),compound,sperc(compdx),compdx);
261: printf("word usage:\n");
262: printf(" verb types as %% of total verbs\n");
263: printf(" tobe %3.0f%% (%d) aux %3.0f%% (%d) inf %3.0f%% (%d)\n",vperc(tobe),tobe,vperc(aux),aux,vperc(infin),infin);
264: if(verbc != 0)adjs = (float)(passive)*100/(float)(verbc);
265: else adjs=0;
266: printf(" passives as %% of non-inf verbs %3.0f%% (%d)\n",adjs,passive);
267: printf(" types as %% of total\n");
268: printf(" prep %3.1f%% (%d) conj %3.1f%% (%d) adv %3.1f%% (%d)\n",wperc(prepc),prepc,wperc(conjc),conjc,wperc(adv),adv);
269: printf(" noun %3.1f%% (%d) adj %3.1f%% (%d) pron %3.1f%% (%d)\n",wperc(noun),noun,
270: wperc(adj),adj,wperc(pron),pron);
271: printf(" nominalizations %3.0f %% (%d)\n",wperc(nomin),nomin);
272: printf("sentence beginnings:\n");
273: ii=beg[0]+beg[7]+beg[6]+beg[3]+beg[8];
274: printf(" subject opener: noun (%d) pron (%d) pos (%d) adj (%d) art (%d) tot %3.0f%%\n",
275: beg[0],beg[7],beg[6],beg[3],beg[8],sperc(ii));
276: printf(" prep %3.0f%% (%d) adv %3.0f%% (%d) \n",sperc(beg[9]),beg[9],sperc(beg[4]),beg[4]);
277: printf(" verb %3.0f%% (%d) ",sperc(beg[1]+beg[10]+beg[11]),beg[1]+beg[10]+beg[11]);
278: printf(" sub_conj %3.0f%% (%d) conj %3.0f%% (%d)\n",sperc(beg[13]),beg[13],sperc(beg[5]),beg[5]);
279: printf(" expletives %3.0f%% (%d)\n",sperc(beg[14]),beg[14]);
280: #ifdef SCATCH
281: if(nosave && (fopen(SCATCH,"r")) != NULL){
282: if(((io=fopen(SCATCH,"a")) != NULL)){
283: fprintf(io," read %4.1f %4.1f %4.1f %4.1f %4.1f\n",kindex, aindex, cindex, findex, fgrad);
284: fprintf(io," sentl %d %ld %4.2f %4.2f %d %d %ld %4.2f\n",numsent,numwds,avw,avl,qcount,icount,numnonf,snonf);
285: fprintf(io," l var %d %d %d %d %d\n",ml,lsum,mg,gsum,maxsent);
286: fprintf(io," t var %d %d %d %d\n",simple,complex,compound,compdx);
287: fprintf(io," verbs %d %d %d %d %d %d\n",tverbc,verbc,tobe,aux,infin,passive);
288: fprintf(io," ty %d %d %d %d %d %d %d\n",prepc,conjc,adv,noun,adj,pron,nomin);
289: fprintf(io," beg %d %d %d %d %d %d\n",beg[0],beg[7],beg[6],beg[3],beg[8],ii);
290: 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]);
291: }
292: }
293: #endif
294: }
295: return(1);
296: }
297: float
298: wperc(a)
299: {
300: return((float)(a)*100/numwds);
301: }
302: float
303: sperc(a)
304: {
305: return((float)(a)*100/numsent);
306: }
307: float
308: typersent(a)
309: {
310: return((float)(a)/numsent);
311: }
312: float
313: vperc(a)
314: {
315: if(tverbc == 0)return(0);
316: return((float)(a)*100/tverbc);
317: }
318: main(argc,argv)
319: char **argv;
320: {
321: while(--argc > 0 && (++argv)[0][0] == '-' ){
322: switch(argv[0][1]){
323: case 'd': nosave = 0;
324: continue;
325: case 's': style=1;
326: continue;
327: case 'p': pastyle=style=1;
328: continue;
329: case 'a': pstyle=style=1;
330: continue;
331: case 'e': estyle = style = 1;
332: continue;
333: case 'n': nstyle = style = 1;
334: continue;
335: case 'N': Nstyle = style = 1;
336: continue;
337: case 'l': style=lstyle=1;
338: lthresh = atoi(*(++argv));
339: argc--;
340: continue;
341: case 'r':
342: style=rstyle=1;
343: rthresh = atoi(*(++argv));
344: argc--;
345: continue;
346: case 'P':
347: part = 1;
348: style = 0;
349: continue;
350: case 'b': /* print bare bones info rje */
351: barebones = 1;
352: style = 0;
353: continue;
354: case 'T': /*topic*/
355: style = 0;
356: topic = 1;
357: continue;
358: default:
359: fprintf(stderr,"unknown flag to part %s\n",*argv);
360: exit(1);
361: }
362: argv++;
363: }
364: #ifdef SNOM
365: if(fopen(SNOM,"r") != NULL){
366: deb = fopen(SNOM,"a"); /* SAVE NOM*/
367: }
368: #else
369: deb = NULL;
370: #endif
371: yylex();
372: if(nrofflg && sentno > 0){
373: printf(".SL \"");
374: for(i=0;i<sentno;i++)
375: printf(" %d",leng[i]);
376: printf("\"\n");
377: }
378: }
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.