/*
- * html.c -- Functions which handle translation between HTML and plain text
* $Id$
+ *
+ * Functions which handle translation between HTML and plain text
*/
+#include "sysdep.h"
#include <stdlib.h>
#include <unistd.h>
#include <stdio.h>
+#include <fcntl.h>
+#include <signal.h>
+
+#if TIME_WITH_SYS_TIME
+# include <sys/time.h>
+# include <time.h>
+#else
+# if HAVE_SYS_TIME_H
+# include <sys/time.h>
+# else
+# include <time.h>
+# endif
+#endif
+
#include <ctype.h>
#include <string.h>
-
+#include <errno.h>
+#include <limits.h>
+#include <syslog.h>
+#include "citadel.h"
+#include "server.h"
+#include "control.h"
+#include "sysdep_decls.h"
+#include "support.h"
+#include "config.h"
+#include "msgbase.h"
+#include "tools.h"
+#include "room_ops.h"
+#include "html.h"
+
/*
* Convert HTML to plain text.
+ *
+ * inputmsg = pointer to raw HTML message
+ * screenwidth = desired output screenwidth
+ * do_citaformat = set to 1 to indent newlines with spaces
*/
-void html_to_ascii(int screenwidth) {
- char inbuf[256];
- char outbuf[256];
+char *html_to_ascii(char *inputmsg, int screenwidth, int do_citaformat) {
+ char inbuf[SIZ];
+ char outbuf[SIZ];
char tag[1024];
int done_reading = 0;
- char *ptr;
- int i, ch;
+ char *inptr;
+ char *outptr;
+ size_t outlen;
+ int i, j, ch, did_out, rb;
int nest = 0; /* Bracket nesting level */
+ inptr = inputmsg;
strcpy(inbuf, "");
strcpy(outbuf, "");
+ outptr = mallok(strlen(inptr) + SIZ);
+ if (outptr == NULL) return NULL;
+ strcpy(outptr, "");
+ outlen = 0;
+
do {
/* Fill the input buffer */
if ( (done_reading == 0) && (strlen(inbuf) < 128) ) {
- /* FIX ... genericize this */
- ptr = fgets(&inbuf[strlen(inbuf)], 127, stdin);
- if (ptr == NULL) done_reading = 1;
+
+ ch = *inputmsg++;
+ if (ch > 0) {
+ inbuf[strlen(inbuf)+1] = 0;
+ inbuf[strlen(inbuf)] = ch;
+ }
+ else {
+ done_reading = 1;
+ }
+
}
/* Do some parsing */
strcat(outbuf, "\n\n");
}
- if (!strcasecmp(tag, "HR")) {
- strcat(outbuf, "\n ----- \n");
+ if (!strcasecmp(tag, "/DIV")) {
+ strcat(outbuf, "\n\n");
+ }
+
+ else if (!strcasecmp(tag, "H1")) {
+ strcat(outbuf, "\n\n");
+ }
+
+ else if (!strcasecmp(tag, "H2")) {
+ strcat(outbuf, "\n\n");
+ }
+
+ else if (!strcasecmp(tag, "H3")) {
+ strcat(outbuf, "\n\n");
+ }
+
+ else if (!strcasecmp(tag, "H4")) {
+ strcat(outbuf, "\n\n");
+ }
+
+ else if (!strcasecmp(tag, "/H1")) {
+ strcat(outbuf, "\n");
+ }
+
+ else if (!strcasecmp(tag, "/H2")) {
+ strcat(outbuf, "\n");
+ }
+
+ else if (!strcasecmp(tag, "/H3")) {
+ strcat(outbuf, "\n");
}
- if (!strcasecmp(tag, "BR")) {
+ else if (!strcasecmp(tag, "/H4")) {
strcat(outbuf, "\n");
}
- if (!strcasecmp(tag, "TR")) {
+ else if (!strcasecmp(tag, "HR")) {
+ strcat(outbuf, "\n ");
+ for (j=0; j<screenwidth-2; ++j)
+ strcat(outbuf, "-");
strcat(outbuf, "\n");
}
- if (!strcasecmp(tag, "/TABLE")) {
+ else if (!strcasecmp(tag, "BR")) {
+ strcat(outbuf, "\n");
+ }
+
+ else if (!strcasecmp(tag, "TR")) {
+ strcat(outbuf, "\n");
+ }
+
+ else if (!strcasecmp(tag, "/TABLE")) {
strcat(outbuf, "\n");
}
strcpy(&outbuf[i+1], &outbuf[i+6]);
}
- else if (!strncasecmp(&outbuf[i], "&lb;", 4)) {
+ else if (!strncasecmp(&outbuf[i], "<", 4)) {
outbuf[i] = '<';
strcpy(&outbuf[i+1], &outbuf[i+4]);
}
- else if (!strncasecmp(&outbuf[i], "&rb;", 4)) {
+ else if (!strncasecmp(&outbuf[i], ">", 4)) {
outbuf[i] = '>';
strcpy(&outbuf[i+1], &outbuf[i+4]);
}
strcpy(&outbuf[i+1], &outbuf[i+6]);
}
+ else if (!strncasecmp(&outbuf[i], "©", 6)) {
+ outbuf[i] = '(';
+ outbuf[i+1] = 'c';
+ outbuf[i+2] = ')';
+ strcpy(&outbuf[i+3], &outbuf[i+6]);
+ }
+
+ else if (!strncasecmp(&outbuf[i], "®", 5)) {
+ outbuf[i] = '(';
+ outbuf[i+1] = 'r';
+ outbuf[i+2] = ')';
+ strcpy(&outbuf[i+3], &outbuf[i+5]);
+ }
+
}
- /* Output our finely-crafted plain ASCII */
- printf("%s", outbuf); /* FIX ... genericize this */
- strcpy(outbuf, "");
+ /* Make sure the output buffer is big enough */
+ if ((strlen(outptr) + strlen(outbuf) + 128) > outlen) {
+ outlen += 128;
+ outptr = realloc(outptr, outlen);
+ }
+
+ /* Output any lines terminated with hard line breaks */
+ do {
+ did_out = 0;
+ if (strlen(outbuf)>0)
+ for (i = 0; i<strlen(outbuf); ++i) {
+ if ( (i<(screenwidth-2)) && (outbuf[i]=='\n')) {
+ strncat(outptr, outbuf, i+1);
+ strcat(outptr, "\n");
+ if (do_citaformat)
+ strcat(outptr, " ");
+ strcpy(outbuf, &outbuf[i+1]);
+ i = 0;
+ did_out = 1;
+ }
+ }
+ } while (did_out);
+
+ /* Add soft line breaks */
+ if (strlen(outbuf) > (screenwidth - 2)) {
+ rb = (-1);
+ for (i=0; i<(screenwidth-2); ++i) {
+ if (outbuf[i]==32) rb = i;
+ }
+ if (rb>=0) {
+ strncat(outptr, outbuf, rb);
+ strcat(outptr, "\n");
+ if (do_citaformat)
+ strcat(outptr, " ");
+ strcpy(outbuf, &outbuf[rb+1]);
+ } else {
+
+ strncat(outptr, outbuf, screenwidth-2);
+ strcat(outptr, "\n");
+ if (do_citaformat)
+ strcat(outptr, " ");
+ strcpy(outbuf, &outbuf[screenwidth-2]);
+ }
+ }
} while (done_reading == 0);
-}
+ strcat(outptr, outbuf);
+ strcat(outptr, "\n");
+ return outptr;
-/*
- * Temporary main loop for testing
- */
-int main() {
- html_to_ascii(80);
- return 0;
}
+