Subversion Repositories wimsdev

Rev

Rev 3717 | Rev 8068 | Go to most recent revision | Blame | Compare with Previous | Last modification | View Log | RSS feed

  1. /*    Copyright (C) 1998-2003 XIAO, Gang of Universite de Nice - Sophia Antipolis
  2.  *
  3.  *  This program is free software; you can redistribute it and/or modify
  4.  *  it under the terms of the GNU General Public License as published by
  5.  *  the Free Software Foundation; either version 2 of the License, or
  6.  *  (at your option) any later version.
  7.  *
  8.  *  This program is distributed in the hope that it will be useful,
  9.  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
  10.  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  11.  *  GNU General Public License for more details.
  12.  *
  13.  *  You should have received a copy of the GNU General Public License
  14.  *  along with this program; if not, write to the Free Software
  15.  *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  16.  */
  17.  
  18. char t_buf[4][MAX_LINELEN+1];
  19. char maskbuf[MAX_LINELEN+1];
  20.  
  21. /* internal routine. */
  22. void _text_cut(char *p, char *w)
  23. {
  24.     char *p1, *p2;
  25.     p1=wordchr(p,w); if(p1==NULL) error2("syntax_error");
  26.     *p1=0; p2=find_word_start(p1+strlen(w));
  27.     ovlstrcpy(t_buf[0],p); ovlstrcpy(t_buf[1],p2);
  28.     strip_trailing_spaces(t_buf[0]);
  29.     substitute(t_buf[0]); substitute(t_buf[1]);
  30. }
  31.  
  32. /* Extract characters in buf[0] which are identical to
  33.  * corresponding characters in buf[1]. */
  34. void text_common(char *p)
  35. {
  36.     int i,j,n1,n2;
  37.     _text_cut(p,"and");
  38.     n1=strlen(t_buf[0]);n2=strlen(t_buf[1]);
  39.     if(n2<n1) n1=n2;
  40.     for(i=j=0;i<n1;i++) {
  41.       if(t_buf[0][i]==t_buf[1][i] && maskbuf[i]!='0') p[j++]=t_buf[0][i];
  42.     }
  43.     p[j]=0;
  44. }
  45.  
  46. /* Returns a mask string composed of '0's and '1's, where
  47.  * '0' means corresponding positions of buf[0] and buf[1] are
  48.  * equal. */
  49. void text_compare(char *p)
  50. {
  51.     int min,max, i;
  52.     _text_cut(p,"and");
  53.     min=strlen(t_buf[0]); max=strlen(t_buf[1]);
  54.     if(min>max) {
  55.       i=min; min=max; max=i;
  56.     }
  57.     for(i=0; i<min; i++) {
  58.       if(t_buf[0][i]==t_buf[1][i]) p[i]='0'; else p[i]='1';
  59.     }
  60.     for(; i<max; i++) p[i]='1';
  61.     p[max]=0;
  62. }
  63.  
  64. /* copy text according to mask. */
  65. void text_copy(char *p)
  66. {
  67.     int i, j, n;
  68.  
  69.     snprintf(t_buf[0],MAX_LINELEN,"%s",p);
  70.     strip_trailing_spaces(t_buf[0]); substitute(t_buf[0]);
  71.     n=strlen(t_buf[0]);
  72.     for(i=j=0;i<n;i++) {
  73.       if(maskbuf[i]!='0') p[j++]=t_buf[0][i];
  74.     }
  75.     p[j]=0;
  76. }
  77.  
  78. /* returns count of characters in buf[1] which appear in buf[0]. */
  79. void text_count(char *p)
  80. {
  81.     int i, n, c;
  82.     _text_cut(p,"in");
  83.     n=strlen(t_buf[1]);
  84.     for(i=c=0;i<n;i++) {
  85.       if(strchr(t_buf[0],t_buf[1][i])!=NULL && maskbuf[i]!='0') c++;
  86.     }
  87.     snprintf(p,MAX_LINELEN,"%d",c);
  88. }
  89.  
  90. /* Extract characters in buf[0] which are different than
  91.  * corresponding characters in buf[1].
  92.  */
  93. void text_diff(char *p)
  94. {
  95.     int i,j,n1,n2;
  96.     _text_cut(p,"from");
  97.     n1=strlen(t_buf[0]);n2=strlen(t_buf[1]);
  98.     if(n2<n1) n1=n2;
  99.     for(i=j=0;i<n1;i++) {
  100.       if(t_buf[0][i]!=t_buf[1][i] && maskbuf[i]!='0') p[j++]=t_buf[0][i];
  101.     }
  102.     p[j]=0;
  103. }
  104.  
  105. /* put chars in buf[0] in a new string, into positions
  106.  * corresponding to '1's in the mask buf[1].
  107.  * Positions corresponding to '0's are filled by space.
  108.  * Fill stops at the end of buf[0]. If buf[1] is
  109.  * too short, it is reused from the start.
  110.  */
  111. void text_expand(char *p)
  112. {
  113.     int i,j,k,n1,n2;
  114.     _text_cut(p,"using");
  115.     n1=strlen(t_buf[0]);n2=strlen(t_buf[1]);
  116.     if(n2==0) {p[0]=0; return;}
  117.     for(i=j=k=0;i<n1 && j<MAX_LINELEN;j++,k=j%n2) {
  118.       if(t_buf[1][k]=='0') p[j]=' ';
  119.       else p[j]=t_buf[0][i++];
  120.     }
  121.     p[j]=0;
  122. }
  123.  
  124. /* character by character replacement of buf[1] by buf[0],
  125.  * replacing only mask-effective chars.
  126.  * The resulting string is as long as buf[1], and the replacement
  127.  * stops when chars buf[0] has run out.
  128.  */
  129. void text_insert(char *p)
  130. {
  131.     int i,j,n1,n2;
  132.     _text_cut(p,"into");
  133.     n1=strlen(t_buf[0]);n2=strlen(t_buf[1]);
  134.     for(i=j=0; i<n2 && j<n1; i++) {
  135.       if(maskbuf[i]!='0') t_buf[1][i]=t_buf[0][j++];
  136.     }
  137.     snprintf(p,MAX_LINELEN,"%s",t_buf[1]);
  138. }
  139.  
  140. #define MAX_TLEN 96
  141.  
  142. /* interact of two strings according to rules
  143.  * defined a table.
  144.  */
  145. void text_interact(char *p)
  146. {
  147.     char *table, *dline, *tline[MAX_TLEN];
  148.     char *p1, *p2;
  149.     int i,j1,j2,k,l,l2,n;
  150.  
  151.     table=wordchr(p,"table");
  152.     if(table==NULL) error2("syntax_error");
  153.     *table=0; strip_trailing_spaces(p);
  154.     table=find_word_start(table+strlen("table"));
  155.     snprintf(t_buf[2],MAX_LINELEN,"%s",table);
  156.     _text_cut(p,"and");
  157.     strip_trailing_spaces(t_buf[2]); substitute(t_buf[2]);
  158.     n=linenum(t_buf[2])-1;
  159.     if(n>=MAX_TLEN) error2("text_bad_table");
  160.     p2=strchr(t_buf[2],'\n'); if(p2!=NULL) *p2++=0;
  161.     if(strlen(t_buf[2])!=n) error2("text_bad_table");
  162.     dline=t_buf[2];
  163.     for(i=0,p1=p2;i<n;i++,p1=p2) {
  164.       if(p1==NULL) error2("text_bad_table");
  165.       p2=strchr(p1,'\n');
  166.       if(p2!=NULL) *p2++=0;
  167.       if(strlen(p1)!=n) error2("text_bad_table");
  168.       tline[i]=p1;
  169.     }
  170.     l=strlen(t_buf[0]); l2=strlen(t_buf[1]); if(l2<l) l=l2;
  171.     for(i=k=0;i<l;i++) {
  172.       if(maskbuf[i]!='0') {
  173.           p1=strchr(dline,t_buf[0][i]);
  174.           p2=strchr(dline,t_buf[1][i]);
  175.           if(p1==NULL || p2==NULL) continue;
  176.           j1=p1-dline; j2=p2-dline;
  177.           if(j1>=n || j2>=n) continue; /* should not occur */
  178.           p[k++]=tline[j1][j2];
  179.       }
  180.     }
  181.     p[k]=0;
  182. }
  183.  
  184. /* returns a mask string composed of '0's and '1's, where
  185.  * '0' means corresponding char in buf[1] appears in buf[0]. */
  186. void text_mark(char *p)
  187. {
  188.     int i, n;
  189.     _text_cut(p,"in");
  190.     n=strlen(t_buf[1]);
  191.     for(i=0;i<n;i++) {
  192.       if(strchr(t_buf[0],t_buf[1][i])!=NULL) p[i]='1';
  193.       else p[i]='0';
  194.     }
  195.     p[i]=0;
  196. }
  197.  
  198. /* Returns a string whose characters are the maximum
  199.  * of the two corresponding chars in buf[0] and buf[1].
  200.  * Length of the string is the longuest one.
  201.  */
  202. void text_max(char *p)
  203. {
  204.     int min,max, i, j, k;
  205.     _text_cut(p,"and");
  206.     min=strlen(t_buf[0]); max=strlen(t_buf[1]);
  207.     if(min>max) {
  208.       i=min; min=max; max=i; j=0;
  209.     }
  210.     else j=1;
  211.     for(i=k=0; i<min; i++) {
  212.       if(maskbuf[i]=='0') continue;
  213.       if((unsigned char)t_buf[0][i]>(unsigned char)t_buf[1][i])
  214.           p[k++]=t_buf[0][i];
  215.       else p[k++]=t_buf[1][i];
  216.     }
  217.     for(;i<max;i++) {
  218.       if(maskbuf[i]!='0') p[k++]=t_buf[j][i];
  219.     }
  220.     p[k]=0;
  221. }
  222.  
  223. /* Returns a string whose characters are the minimum
  224.  * of the two corresponding chars in buf[0] and buf[1].
  225.  * Length of the string is the shortest one.
  226.  */
  227. void text_min(char *p)
  228. {
  229.     int min,max, i,k;
  230.     _text_cut(p,"and");
  231.     min=strlen(t_buf[0]); max=strlen(t_buf[1]);
  232.     if(min>max) {
  233.       i=min; min=max; max=i;
  234.     }
  235.     for(i=k=0; i<min; i++) {
  236.       if(maskbuf[i]=='0') continue;
  237.       if((unsigned char)t_buf[0][i]< (unsigned char)t_buf[1][i])
  238.           p[k++]=t_buf[0][i];
  239.       else p[k++]=t_buf[1][i];
  240.     }
  241.     p[k]=0;
  242. }
  243.  
  244. /* extract chars in buf[0] which occur in buf[1]. */
  245. void text_occur(char *p)
  246. {
  247.     int i,j,n;
  248.     char buf[MAX_LINELEN+1];
  249.     memset(buf,0,sizeof(buf));
  250.     _text_cut(p,"in");
  251.     n=strlen(t_buf[1]);
  252.     for(i=0;i<n;i++) {
  253.       char *pp;
  254.       if(maskbuf[i]=='0') continue;
  255.       pp=strchr(t_buf[0],t_buf[1][i]);
  256.       if(pp!=NULL) buf[pp - t_buf[0]]=1;
  257.     }
  258.     n=strlen(t_buf[0]);
  259.     for(i=j=0;i<n;i++) {
  260.       if(buf[i]) p[j++]=t_buf[0][i];
  261.     }
  262.     p[j]=0;
  263. }
  264.  
  265. /* remove characters of buf[1] in buf[0]. */
  266. void text_remove(char *p)
  267. {
  268.     int i, j, n;
  269.     _text_cut(p,"in");
  270.     n=strlen(t_buf[1]);
  271.     for(i=j=0;i<n;i++) {
  272.       if(strchr(t_buf[0],t_buf[1][i])==NULL
  273.          && maskbuf[i]!='0') p[j++]=t_buf[1][i];
  274.     }
  275.     p[j]=0;
  276. }
  277.  
  278. /* Cyclic reordering of text. */
  279. void text_reorder(char *p)
  280. {
  281.     int i,j,k,l,n,t;
  282.     int list[10240];
  283.     char buf[MAX_LINELEN+1];
  284.     _text_cut(p,"by"); *p=0;
  285.     n=itemnum(t_buf[1]); if(n<=0 || n>=10240) return;
  286.     for(i=0;i<n;i++) {
  287.         buf[0]=0; fnd_item(t_buf[1],i+1,buf);
  288.       j=atoi(buf); if(j<=0 || j>n) return;
  289.       list[i]=j;
  290.     }
  291.     t=strlen(t_buf[0]);
  292.     for(i=l=0;l<t && i<t+n;i++) {
  293.       j=i/n; k=j*n+list[i%n];
  294.       if(k>t || k<=0) continue;
  295.       p[l++]=t_buf[0][k-1];
  296.     }
  297.     p[l]=0;
  298. }
  299.  
  300. /* repeat a string to a given length. */
  301. void text_repeat(char *p)
  302. {
  303.     int n,i,k;
  304.     _text_cut(p,"to");
  305.     n=strevalue(t_buf[1]); if(n>MAX_LINELEN) n=MAX_LINELEN;
  306.     if(n<0) n=0;
  307.     k=strlen(t_buf[0]); if(k<=0) {*p=0; return;}
  308.     for(i=0;i<n;i++) {
  309.       p[i]=t_buf[0][i%k];
  310.     }
  311.     p[i]=0;
  312. }
  313.  
  314. /* reverse a string */
  315. void text_reverse(char *p)
  316. {
  317.     int i,n;
  318.     char buf[MAX_LINELEN+1];
  319.     snprintf(t_buf[0],sizeof(t_buf[0]),"%s",p);
  320.     substitute(t_buf[0]);
  321.     n=strlen(t_buf[0]); if(n>MAX_LINELEN) n=MAX_LINELEN;
  322.     for(i=0;i<n;i++) buf[i]=t_buf[0][n-1-i];
  323.     buf[n]=0;
  324.     ovlstrcpy(p,buf);
  325. }
  326.  
  327. /* remove characters of buf[1] not in buf[0]. */
  328. void text_select(char *p)
  329. {
  330.     int i, j, n;
  331.     _text_cut(p,"in");
  332.     n=strlen(t_buf[1]);
  333.     for(i=j=0;i<n;i++) {
  334.       if(strchr(t_buf[0],t_buf[1][i])!=NULL
  335.          && maskbuf[i]!='0') p[j++]=t_buf[1][i];
  336.     }
  337.     p[j]=0;
  338. }
  339.  
  340. /* tag: bit 0 is mask. */
  341. struct {
  342.     char *name;
  343.     int tag;
  344.     void (*routine) (char *p);
  345. } text_proc[]={
  346.       {"appear",    1,  text_occur},
  347.       {"common",    1,  text_common},
  348.       {"compare",   0,  text_compare},
  349.       {"copy",      1,  text_copy},
  350.       {"count",     1,  text_count},
  351.       {"delete",    1,  text_remove},
  352.       {"diff",      1,  text_diff},
  353.       {"differ",    1,  text_diff},
  354.       {"drop",      1,  text_remove},
  355.       {"expand",    0,  text_expand},
  356.       {"extract",   1,  text_select},
  357.       {"insert",    1,  text_insert},
  358.       {"interact",  1,  text_interact},
  359.       {"mark",      0,  text_mark},
  360.       {"max",       1,  text_max},
  361.       {"min",       1,  text_min},
  362.       {"occur",     1,  text_occur},
  363.       {"occurrence",1,  text_occur},
  364.       {"pick",      1,  text_select},
  365.       {"pickup",    1,  text_select},
  366.       {"remove",    1,  text_remove},
  367.       {"reorder",   0,  text_reorder},
  368.       {"repeat",    0,  text_repeat},
  369.       {"reverse",   0,  text_reverse},
  370.       {"select",    1,  text_select}
  371. };
  372. #define TEXT_PROC_NO (sizeof(text_proc)/sizeof(text_proc[0]))
  373.  
  374. int textab_verify(void) {
  375.     return verify_order(text_proc,TEXT_PROC_NO,sizeof(text_proc[0]));
  376. }
  377.  
  378. /* main entry point for text routines */
  379. void text(char *p)
  380. {
  381.     int i,j,n;
  382.     char *p1, *p2;
  383.     char c,cc;
  384.     char buf[MAX_LINELEN+1];
  385.     p1=find_word_start(p); p2=find_word_end(p1);
  386.     if(p2<=p1 || *p2==0) error2("syntax_error");
  387.     *p2=0;
  388.     i=search_list(text_proc,TEXT_PROC_NO,sizeof(text_proc[0]),p1);
  389.     if(i<0) error2("syntax_error");
  390.     snprintf(buf,sizeof(buf),"%s",find_word_start(p2+1));
  391.     if((text_proc[i].tag&1)!=0 && (p1=wordchr(buf,"mask"))!=NULL) {
  392.       *p1=0; strip_trailing_spaces(buf);
  393.       p2=find_word_start(p1+strlen("mask"));
  394.       strip_trailing_spaces(p2);
  395.       snprintf(maskbuf,sizeof(maskbuf),"%s",p2);
  396.       substitute(maskbuf);
  397.       n=strlen(maskbuf); if(n==0) goto zeromask;
  398.       c=maskbuf[n-1]; cc=0;
  399.       if(c=='+') cc='1'; if(c=='-') cc='0';
  400.       if(cc!=0) memset(maskbuf+n-1,cc,sizeof(maskbuf)-n);
  401.       else for(j=n;j<MAX_LINELEN;j++) maskbuf[j]=maskbuf[j%n];
  402.       maskbuf[sizeof(maskbuf)-1]=0;
  403.     }
  404.     else zeromask: memset(maskbuf,0,sizeof(maskbuf));
  405.     text_proc[i].routine(buf);
  406.     buf[MAX_LINELEN]=0;ovlstrcpy(p,buf);
  407. }
  408.