citadel/html.c

   1 /*
   2  * html.c -- Functions which handle translation between HTML and plain text
   3  * $Id$
   4  */
   5
   6 #include <stdlib.h>
   7 #include <unistd.h>
   8 #include <stdio.h>
   9 #include <ctype.h>
  10 #include <string.h>
  11
  12
  13 /*
  14  * Convert HTML to plain text.
  15  */
  16 void html_to_ascii(int screenwidth) {
  17         char inbuf[256];
  18         char outbuf[256];
  19         char tag[1024];
  20         int done_reading = 0;
  21         char *ptr;
  22         int i, ch;
  23         int nest = 0;           /* Bracket nesting level */
  24
  25         strcpy(inbuf, "");
  26         strcpy(outbuf, "");
  27
  28         do {
  29                 /* Fill the input buffer */
  30                 if ( (done_reading == 0) && (strlen(inbuf) < 128) ) {
  31                         /* FIX ... genericize this */
  32                         ptr = fgets(&inbuf[strlen(inbuf)], 127, stdin);
  33                         if (ptr == NULL) done_reading = 1;
  34                 }
  35
  36                 /* Do some parsing */
  37                 if (strlen(inbuf)>0) {
  38
  39                     /* Fold in all the spacing */
  40                     for (i=0; i<strlen(inbuf); ++i) {
  41                         if (inbuf[i]==10) inbuf[i]=32;
  42                         if (inbuf[i]==13) inbuf[i]=32;
  43                         if (inbuf[i]==9) inbuf[i]=32;
  44                         if ((inbuf[i]<32) || (inbuf[i]>126))
  45                                 strcpy(&inbuf[i], &inbuf[i+1]);
  46                         while ((inbuf[i]==32)&&(inbuf[i+1]==32))
  47                                 strcpy(&inbuf[i], &inbuf[i+1]);
  48                     }
  49
  50                     for (i=0; i<strlen(inbuf); ++i) {
  51
  52                         ch = inbuf[i];
  53
  54                         if (ch == '<') {
  55                                 ++nest;
  56                                 strcpy(tag, "");
  57                         }
  58
  59                         else if (ch == '>') {
  60                                 if (nest > 0) --nest;
  61
  62                                 if (!strcasecmp(tag, "P")) {
  63                                         strcat(outbuf, "\n\n");
  64                                 }
  65
  66                                 if (!strcasecmp(tag, "HR")) {
  67                                         strcat(outbuf, "\n ----- \n");
  68                                 }
  69
  70                                 if (!strcasecmp(tag, "BR")) {
  71                                         strcat(outbuf, "\n");
  72                                 }
  73
  74                                 if (!strcasecmp(tag, "TR")) {
  75                                         strcat(outbuf, "\n");
  76                                 }
  77
  78                                 if (!strcasecmp(tag, "/TABLE")) {
  79                                         strcat(outbuf, "\n");
  80                                 }
  81
  82                         }
  83
  84                         else if ((nest > 0) && (strlen(tag)<(sizeof(tag)-1))) {
  85                                 tag[strlen(tag)+1] = 0;
  86                                 tag[strlen(tag)] = ch;
  87                         }
  88
  89                         else if (!nest) {
  90                                 outbuf[strlen(outbuf)+1] = 0;
  91                                 outbuf[strlen(outbuf)] = ch;
  92                         }
  93                     }
  94                     strcpy(inbuf, &inbuf[i]);
  95                 }
  96
  97                 /* Convert &; tags to the forbidden characters */
  98                 if (strlen(outbuf)>0) for (i=0; i<strlen(outbuf); ++i) {
  99
 100                         if (!strncasecmp(&outbuf[i], "&nbsp;", 6)) {
 101                                 outbuf[i] = ' ';
 102                                 strcpy(&outbuf[i+1], &outbuf[i+6]);
 103                         }
 104
 105                         else if (!strncasecmp(&outbuf[i], "&lb;", 4)) {
 106                                 outbuf[i] = '<';
 107                                 strcpy(&outbuf[i+1], &outbuf[i+4]);
 108                         }
 109
 110                         else if (!strncasecmp(&outbuf[i], "&rb;", 4)) {
 111                                 outbuf[i] = '>';
 112                                 strcpy(&outbuf[i+1], &outbuf[i+4]);
 113                         }
 114
 115                         else if (!strncasecmp(&outbuf[i], "&amp;", 5)) {
 116                                 strcpy(&outbuf[i+1], &outbuf[i+5]);
 117                         }
 118
 119                         else if (!strncasecmp(&outbuf[i], "&quot;", 6)) {
 120                                 outbuf[i] = '\"';
 121                                 strcpy(&outbuf[i+1], &outbuf[i+6]);
 122                         }
 123
 124                 }
 125
 126                 /* Output our finely-crafted plain ASCII */
 127                 printf("%s", outbuf);   /* FIX ... genericize this */
 128                 strcpy(outbuf, "");
 129
 130         } while (done_reading == 0);
 131
 132 }
 133
 134
 135 /*
 136  * Temporary main loop for testing
 137  */
 138 int main() {
 139         html_to_ascii(80);
 140         return 0;
 141 }