-// Functions which handle translation between HTML and plain text
-// Copyright (c) 2000-2022 by the citadel.org team
+// This is an HTML to plain text converter.
+// Copyright (c) 2000-2024 by the citadel.org team (Art Cancro et al.)
//
-// This program is open source software. Use, duplication, or disclosure
-// is subject to the terms of the GNU General Public License, version 3.
+// This program is open source software. Use, duplication, or disclosure is subject to the GNU General Public License version 3.
#include <stdlib.h>
#include <unistd.h>
#include <sys/stat.h>
#include <errno.h>
#include <limits.h>
+#include <time.h>
+#include "libcitadel.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
+int u8_wc_toutf8(char *dest, u_int32_t ch) {
+ if (ch < 0x80) {
+ dest[0] = (char)ch;
+ return 1;
+ }
+ if (ch < 0x800) {
+ dest[0] = (ch>>6) | 0xC0;
+ dest[1] = (ch & 0x3F) | 0x80;
+ return 2;
+ }
+ if (ch < 0x10000) {
+ dest[0] = (ch>>12) | 0xE0;
+ dest[1] = ((ch>>6) & 0x3F) | 0x80;
+ dest[2] = (ch & 0x3F) | 0x80;
+ return 3;
+ }
+ if (ch < 0x110000) {
+ dest[0] = (ch>>18) | 0xF0;
+ dest[1] = ((ch>>12) & 0x3F) | 0x80;
+ dest[2] = ((ch>>6) & 0x3F) | 0x80;
+ dest[3] = (ch & 0x3F) | 0x80;
+ return 4;
+ }
+ return 0;
+}
+
+
+// Try to embed an image in the display stream.
+// out = the StrBuf to which we are writing the display stream
+// url = the URL of the image (warning: it might be a data: URL)
+// display_protocol = currently only H2A_SIXEL is supported
+void h2a_embed_image(StrBuf *out, char *url, int display_protocol) {
+
+ char buf[4096];
+ snprintf(buf, sizeof(buf), "curl -s '%s' | img2sixel -", url);
+
+ FILE *cmd = popen(buf, "r");
+ if (!cmd) {
+ return;
+ }
+
+ size_t bytes;
+ while (bytes = fread(buf, 1, sizeof(buf), cmd), bytes>0) {
+ StrBufAppendBufPlain(out, buf, bytes, 0);
+ }
+ pclose(cmd);
+}
-#include "libcitadel.h"
-
// Convert HTML to plain text.
//
-// inputmsg = pointer to raw HTML message
-// msglen = stop reading after this many bytes
-// screenwidth = desired output screenwidth
-// ansi = if nonzero, assume output is to a terminal that supports ANSI escape codes
+// inputmsg = pointer to raw HTML message
+// msglen = stop reading after this many bytes
+// screenwidth = desired output screenwidth
+// flags = Flags that can be set:
+// H2A_ANSI = Output ANSI terminal escape sequences
+// H2A_SIXEL = Output Sixel graphics (not yet implemented)
//
-char *html_to_ascii(const char *inputmsg, int msglen, int screenwidth, int ansi) {
- char inbuf[SIZ];
+char *html_to_ascii(const char *inputmsg, int msglen, int screenwidth, unsigned int flags) {
+ char *inbuf = NULL;
int inbuf_len = 0;
- char outbuf[SIZ];
char tag[1024];
- int done_reading = 0;
- const char *inptr;
+ char *tag_start = NULL;
+ char *tag_end = NULL;
+ StrBuf *out;
char *outptr;
- size_t outptr_buffer_size;
- size_t output_len = 0;
- int i, j, ch, did_out, rb, scanch;
- int nest = 0; // Bracket nesting level
+ int j;
+ char ch;
+ int tag_nesting_level = 0; // angle bracket nesting level
int blockquote = 0; // BLOCKQUOTE nesting level
int styletag = 0; // STYLE tag nesting level
- int styletag_start = 0;
- int bytes_processed = 0;
- char nl[128];
+ char nl[128]; // The current value of what a "newline" looks like (changes during blockquotes)
+
+ int ansi = (flags & H2A_ANSI) ? 1 : 0; // Output to a terminal that can accept ANSI escape sequences
+ int sixel = (flags & H2A_SIXEL) ? 1 : 0; // Output to a terminal that can accept Sixel graphics
+
+ out = NewStrBuf();
+ if (!out) {
+ return(NULL);
+ }
tag[0] = '\0';
strcpy(nl, "\n");
- inptr = inputmsg;
- strcpy(inbuf, "");
- strcpy(outbuf, "");
- if (msglen == 0) msglen = strlen(inputmsg);
-
- outptr_buffer_size = strlen(inptr) + SIZ;
- outptr = malloc(outptr_buffer_size);
- if (outptr == NULL) return NULL;
- strcpy(outptr, "");
- output_len = 0;
-
- do {
- // Fill the input buffer
- inbuf_len = strlen(inbuf);
- if ( (done_reading == 0) && (inbuf_len < (SIZ-128)) ) {
-
- ch = *inptr++;
- if (ch != 0) {
- inbuf[inbuf_len++] = ch;
- inbuf[inbuf_len] = 0;
- }
- else {
- done_reading = 1;
- }
+ if (msglen == 0) {
+ msglen = strlen(inputmsg);
+ }
- ++bytes_processed;
- if (bytes_processed > msglen) {
- done_reading = 1;
- }
+ inbuf = strdup(inputmsg);
+ if (!inbuf) {
+ return NULL;
+ }
+
+ // "inbuf" ingests the unparsed HTML while we work with it.
+ inbuf_len = strlen(inbuf);
+ if (inbuf_len > msglen) {
+ inbuf[msglen] = 0;
+ inbuf_len = msglen;
+ }
+ // Do some parsing
+ if (!IsEmptyStr(inbuf)) {
+
+ // Convert newlines, carriage returns, and tabs to spaces
+ char *sp;
+ while ( (sp = strchr(inbuf, '\r'))
+ || (sp = strchr(inbuf, '\n'))
+ || (sp = strchr(inbuf, '\t'))
+ ) {
+ *sp = ' ';
}
- // Do some parsing
- if (!IsEmptyStr(inbuf)) {
-
- // Fold in all the spacing
- for (i=0; !IsEmptyStr(&inbuf[i]); ++i) {
- if (inbuf[i]==10) inbuf[i]=32;
- if (inbuf[i]==13) inbuf[i]=32;
- if (inbuf[i]==9) inbuf[i]=32;
- }
- for (i=0; !IsEmptyStr(&inbuf[i]); ++i) {
- while ((inbuf[i]==32)&&(inbuf[i+1]==32)) {
- strcpy(&inbuf[i], &inbuf[i+1]);
- }
- }
+ // Convert multiple spaces to a single space.
+ while (sp = strstr(inbuf, " "), sp!=NULL) {
+ strcpy(sp, sp+1);
+ }
- for (i=0; !IsEmptyStr(&inbuf[i]); ++i) {
+ // Run through the markup performing the conversion.
+ char *inptr = inbuf;
+ int linelen = 0;
+ while (ch = inptr[0], ch != 0) {
- ch = inbuf[i];
+ // Keep track of how many angle brackets were found in case someone is sloppy with them
+ // or tries to nest tags. If nest is 0 then we are within text; if it is nonzero then we
+ // are within a tag.
- if (ch == '<') {
- ++nest;
+ if (ch == '<') { // We have hit the beginning of a tag.
+ ++tag_nesting_level;
+ tag_start = inptr + 1;
strcpy(tag, "");
}
- else if (ch == '>') { // We have a tag.
- if (nest > 0) --nest;
-
- // Unqualify the tag (truncate at first space)
- if (strchr(tag, ' ') != NULL) {
- strcpy(strchr(tag, ' '), "");
- }
-
- if (!strcasecmp(tag, "P")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
+ else if (ch == '>') { // We have hit the end of a tag.
+ if (tag_nesting_level > 0) {
+ --tag_nesting_level;
}
+ if (tag_nesting_level == 0) {
+ tag_end = inptr;
- if (!strcasecmp(tag, "/DIV")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
- }
-
- if (!strcasecmp(tag, "LI")) {
- strcat(outbuf, nl);
- strcat(outbuf, " * ");
- }
-
- else if (!strcasecmp(tag, "/UL")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "H1")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "H2")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "H3")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "H4")) {
- strcat(outbuf, nl);
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "/H1")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "/H2")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "/H3")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "/H4")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "HR")) {
- strcat(outbuf, nl);
- strcat(outbuf, " ");
- for (j=0; j<screenwidth-2; ++j)
- strcat(outbuf, "-");
- strcat(outbuf, nl);
- }
-
- else if (
- (!strcasecmp(tag, "B"))
- || (!strcasecmp(tag, "STRONG"))
- ) {
- if (ansi) {
- strcat(outbuf, "\033[1m");
+ size_t tag_len = tag_end - tag_start;
+ if (tag_len >= sizeof(tag)) {
+ tag_len = sizeof(tag);
}
- }
- else if (
- (!strcasecmp(tag, "/B"))
- || (!strcasecmp(tag, "/STRONG"))
- ) {
- if (ansi) {
- strcat(outbuf, "\033[22m");
- }
- }
+ strncpy(tag, tag_start, tag_len);
+ tag[tag_len] = 0;
- else if (
- (!strcasecmp(tag, "I"))
- || (!strcasecmp(tag, "EM"))
- ) {
- if (ansi) {
- strcat(outbuf, "\033[3m");
+ // Unqualify the tag (truncate at first space)
+ char *tagsp = strchr(tag, ' ');
+ if (tagsp) {
+ *tagsp = 0;
}
- }
- else if (
- (!strcasecmp(tag, "/I"))
- || (!strcasecmp(tag, "/EM"))
- ) {
- if (ansi) {
- strcat(outbuf, "\033[23m");
+ // IMG tag on sixel terminals -- try to display the image
+ if ( (!strcasecmp(tag, "img")) && sixel) {
+ char *q1, *q2;
+
+ // look for src attribute
+ char *src = bmstrcasestr(tag_start, "src=");
+ q1 = q2 = NULL;
+ if (src && src<tag_end) {
+ if (q1 = strchr(src, '"'), q1 && q1<tag_end) { // in double quotes
+ ++q1;
+ q2 = strchr(q1, '"');
+ }
+ else if (q1 = strchr(src, '\''), q1 && q1<tag_end) { // in single quotes
+ ++q1;
+ q2 = strchr(q1, '\'');
+ }
+ if (q1 && q1<q2 && q2<tag_end) {
+ char url[SIZ];
+ memcpy(url, q1, q2-q1);
+ url[q2-q1] = 0;
+ h2a_embed_image(out, url, H2A_SIXEL); // try to display
+ linelen = 0;
+ }
+ }
}
- }
- else if (!strcasecmp(tag, "U")) {
- if (ansi) {
- strcat(outbuf, "\033[4m");
+ // IMG tag on non-sixel terminals -- we can display the alt text
+ if ( (!strcasecmp(tag, "img")) && !sixel) {
+ char *q1, *q2;
+
+ // look for alt text
+ char *alt = bmstrcasestr(tag_start, "alt=");
+ q1 = q2 = NULL;
+ if (alt && alt<tag_end) {
+ if (q1 = strchr(alt, '"'), q1 && q1<tag_end) { // in double quotes
+ ++q1;
+ q2 = strchr(q1, '"');
+ }
+ else if (q1 = strchr(alt, '\''), q1 && q1<tag_end) { // in single quotes
+ ++q1;
+ q2 = strchr(q1, '\'');
+ }
+ if (q1 && q1<q2 && q2<tag_end) {
+ StrBufAppendBufPlain(out, q1, (long)(q2-q1), 0);
+ }
+ }
}
- }
- else if (!strcasecmp(tag, "/U")) {
- if (ansi) {
- strcat(outbuf, "\033[24m");
+ if (!strcasecmp(tag, "P")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
}
- }
- else if (!strcasecmp(tag, "BR")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "TR")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "/TABLE")) {
- strcat(outbuf, nl);
- }
-
- else if (!strcasecmp(tag, "BLOCKQUOTE")) {
- ++blockquote;
- strcpy(nl, "\n");
- if ( (blockquote == 1) && (ansi) ) {
- strcat(nl, "\033[2m\033[3m");
+ if (!strcasecmp(tag, "/DIV")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
}
- for (j=0; j<blockquote; ++j) strcat(nl, ">");
- strcat(outbuf, nl);
- }
- else if (!strcasecmp(tag, "/BLOCKQUOTE")) {
- strcat(outbuf, "\n");
- --blockquote;
- if ( (blockquote == 0) && (ansi) ) {
- strcat(outbuf, "\033[22m\033[23m");
+ if (!strcasecmp(tag, "LI")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, HKEY(" * "), 0);
+ linelen = 3;
}
- strcpy(nl, "\n");
- for (j=0; j<blockquote; ++j) strcat(nl, ">");
- strcat(outbuf, nl);
- }
- else if (!strcasecmp(tag, "STYLE")) {
- ++styletag;
- if (styletag == 1) {
- styletag_start = strlen(outbuf);
+ else if (!strcasecmp(tag, "/UL")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
}
- }
- else if (!strcasecmp(tag, "/STYLE")) {
- --styletag;
- if (styletag == 0) {
- outbuf[styletag_start] = 0;
+ else if (!strcasecmp(tag, "H1")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
}
- }
-
- }
-
- else if ((nest > 0) && (strlen(tag)<(sizeof(tag)-1))) {
- tag[strlen(tag)+1] = 0;
- tag[strlen(tag)] = ch;
- }
-
- else if ((!nest) && (styletag == 0)) {
- outbuf[strlen(outbuf)+1] = 0;
- outbuf[strlen(outbuf)] = ch;
- }
- }
- strcpy(inbuf, &inbuf[i]);
- }
-
- // Convert &; tags to the forbidden characters
- if (!IsEmptyStr(outbuf)) for (i=0; !IsEmptyStr(&outbuf[i]); ++i) {
-
- // Character entity references
- if (!strncasecmp(&outbuf[i], " ", 6)) {
- outbuf[i] = ' ';
- strcpy(&outbuf[i+1], &outbuf[i+6]);
- }
-
- if (!strncasecmp(&outbuf[i], " ", 6)) {
- outbuf[i] = ' ';
- strcpy(&outbuf[i+1], &outbuf[i+6]);
- }
-
- if (!strncasecmp(&outbuf[i], " ", 6)) {
- outbuf[i] = ' ';
- strcpy(&outbuf[i+1], &outbuf[i+6]);
- }
- if (!strncasecmp(&outbuf[i], " ", 8)) {
- outbuf[i] = ' ';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
-
- else if (!strncasecmp(&outbuf[i], "<", 4)) {
- outbuf[i] = '<';
- strcpy(&outbuf[i+1], &outbuf[i+4]);
- }
-
- else if (!strncasecmp(&outbuf[i], ">", 4)) {
- outbuf[i] = '>';
- strcpy(&outbuf[i+1], &outbuf[i+4]);
- }
-
- else if (!strncasecmp(&outbuf[i], "&", 5)) {
- strcpy(&outbuf[i+1], &outbuf[i+5]);
- }
-
- else if (!strncasecmp(&outbuf[i], """, 6)) {
- outbuf[i] = '\"';
- strcpy(&outbuf[i+1], &outbuf[i+6]);
- }
-
- else if (!strncasecmp(&outbuf[i], "‘", 7)) {
- outbuf[i] = '`';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
-
- else if (!strncasecmp(&outbuf[i], "’", 7)) {
- outbuf[i] = '\'';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
-
- 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], "•", 6)) {
- outbuf[i] = ' ';
- outbuf[i+1] = '*';
- outbuf[i+2] = ' ';
- strcpy(&outbuf[i+3], &outbuf[i+6]);
- }
-
- else if (!strncasecmp(&outbuf[i], "…", 8)) {
- outbuf[i] = '.';
- outbuf[i+1] = '.';
- outbuf[i+2] = '.';
- strcpy(&outbuf[i+3], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "H2")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "™", 7)) {
- outbuf[i] = '(';
- outbuf[i+1] = 't';
- outbuf[i+2] = 'm';
- outbuf[i+3] = ')';
- strcpy(&outbuf[i+4], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "H3")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "®", 5)) {
- outbuf[i] = '(';
- outbuf[i+1] = 'r';
- outbuf[i+2] = ')';
- strcpy(&outbuf[i+3], &outbuf[i+5]);
- }
+ else if (!strcasecmp(tag, "H4")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "¼", 8)) {
- outbuf[i] = '1';
- outbuf[i+1] = '/';
- outbuf[i+2] = '4';
- strcpy(&outbuf[i+3], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "/H1")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "½", 8)) {
- outbuf[i] = '1';
- outbuf[i+1] = '/';
- outbuf[i+2] = '2';
- strcpy(&outbuf[i+3], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "/H2")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "¾", 8)) {
- outbuf[i] = '3';
- outbuf[i+1] = '/';
- outbuf[i+2] = '4';
- strcpy(&outbuf[i+3], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "/H3")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "–", 7)) {
- outbuf[i] = '-';
- outbuf[i+1] = '-';
- strcpy(&outbuf[i+2], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "/H4")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "—", 7)) {
- outbuf[i] = '-';
- outbuf[i+1] = '-';
- outbuf[i+2] = '-';
- strcpy(&outbuf[i+3], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "HR")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ StrBufAppendBufPlain(out, HKEY(" "), 0);
+ for (j = 0; j < screenwidth - 2; ++j) {
+ StrBufAppendBufPlain(out, HKEY("-"), 0);
+ }
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncmp(&outbuf[i], "Ç", 8)) {
- outbuf[i] = 'C';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if ( (!strcasecmp(tag, "B"))
+ || (!strcasecmp(tag, "STRONG"))
+ ) {
+ if (ansi) {
+ StrBufAppendBufPlain(out, HKEY("\033[1m"), 0);
+ }
+ }
+ else if ( (!strcasecmp(tag, "/B"))
+ || (!strcasecmp(tag, "/STRONG"))
+ ) {
+ if (ansi) {
+ StrBufAppendBufPlain(out, HKEY("\033[22m"), 0);
+ }
+ }
- else if (!strncasecmp(&outbuf[i], "ç", 8)) {
- outbuf[i] = 'c';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if ( (!strcasecmp(tag, "I"))
+ || (!strcasecmp(tag, "EM"))
+ ) {
+ if (ansi) {
+ StrBufAppendBufPlain(out, HKEY("\033[3m"), 0);
+ }
+ }
- else if (!strncmp(&outbuf[i], "È", 8)) {
- outbuf[i] = 'E';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if ( (!strcasecmp(tag, "/I"))
+ || (!strcasecmp(tag, "/EM"))
+ ) {
+ if (ansi) {
+ StrBufAppendBufPlain(out, HKEY("\033[23m"), 0);
+ }
+ }
- else if (!strncasecmp(&outbuf[i], "è", 8)) {
- outbuf[i] = 'e';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "U")) {
+ if (ansi) {
+ StrBufAppendBufPlain(out, HKEY("\033[4m"), 0);
+ }
+ }
- else if (!strncmp(&outbuf[i], "Ê", 7)) {
- outbuf[i] = 'E';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "/U")) {
+ if (ansi) {
+ StrBufAppendBufPlain(out, HKEY("\033[24m"), 0);
+ }
+ }
- else if (!strncasecmp(&outbuf[i], "ê", 7)) {
- outbuf[i] = 'e';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "BR")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncmp(&outbuf[i], "É", 8)) {
- outbuf[i] = 'E';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "TR")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "é", 8)) {
- outbuf[i] = 'e';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "/TABLE")) {
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncmp(&outbuf[i], "À", 8)) {
- outbuf[i] = 'A';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "BLOCKQUOTE")) {
+ ++blockquote;
+ strcpy(nl, "\n");
+ if ((blockquote == 1) && (ansi)) {
+ strcat(nl, "\033[2m\033[2m");
+ }
+ for (j = 0; j < blockquote; ++j) {
+ strcat(nl, ">");
+ }
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "à", 8)) {
- outbuf[i] = 'a';
- strcpy(&outbuf[i+1], &outbuf[i+8]);
- }
+ else if (!strcasecmp(tag, "/BLOCKQUOTE")) {
+ StrBufAppendBufPlain(out, HKEY("\n"), 0);
+ --blockquote;
+ if ((blockquote == 0) && (ansi)) {
+ StrBufAppendBufPlain(out, HKEY("\033[22m\033[22m"), 0);
+ }
+ strcpy(nl, "\n");
+ for (j = 0; j < blockquote; ++j) {
+ strcat(nl, ">");
+ }
+ StrBufAppendBufPlain(out, nl, -1, 0);
+ linelen = 0;
+ }
- else if (!strncasecmp(&outbuf[i], "“", 7)) {
- outbuf[i] = '\"';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "STYLE")) {
+ ++styletag;
+ }
- else if (!strncasecmp(&outbuf[i], "”", 7)) {
- outbuf[i] = '\"';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
+ else if (!strcasecmp(tag, "/STYLE")) {
+ --styletag;
+ }
+ }
- else if (!strncasecmp(&outbuf[i], "´", 7)) {
- outbuf[i] = '\'';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
}
- else if (!strncasecmp(&outbuf[i], "’", 7)) {
- outbuf[i] = '\'';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
+ // copy non-tag text to the output buffer
+ else if ((!tag_nesting_level) && (styletag == 0)) {
+ StrBufAppendBufPlain(out, &ch, 1, 0);
+ ++linelen;
}
- else if (!strncasecmp(&outbuf[i], "–", 7)) {
- outbuf[i] = '-';
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
+ // Handle numeric entities
+ if (ch == ';') {
- // two-digit decimal equivalents
- else if (outbuf[i] == '&' &&
- outbuf[i + 1] == '#' &&
- isdigit(outbuf[i + 2]) &&
- isdigit(outbuf[i + 3]) &&
- (outbuf[i+4] == ';') )
- {
- scanch = 0;
- sscanf(&outbuf[i+2], "%02d", &scanch);
- outbuf[i] = scanch;
- strcpy(&outbuf[i+1], &outbuf[i+5]);
- }
+ u_int32_t scanch = 0;
+ int elen = 0;
- // three-digit decimal equivalents
- else if (outbuf[i] == '&' &&
- outbuf[i + 1] == '#' &&
- isdigit(outbuf[i + 2]) &&
- isdigit(outbuf[i + 3]) &&
- isdigit(outbuf[i + 4]) &&
- (outbuf[i + 5] == ';') )
- {
- scanch = 0;
- sscanf(&outbuf[i+2], "%03d", &scanch);
- outbuf[i] = scanch;
- strcpy(&outbuf[i+1], &outbuf[i+6]);
- }
+ if ( (linelen >= 5) && (*(inptr-4) == '&') && (*(inptr-3) == '#') ) {
+ sscanf(inptr-2, "%02d", &scanch);
+ elen = 5;
+ }
+ else if ( (linelen >= 6) && (*(inptr-5) == '&') && (*(inptr-4) == '#') ) {
+ sscanf(inptr-3, "%03d", &scanch);
+ elen = 6;
+ }
+ else if ( (linelen >= 7) && (*(inptr-6) == '&') && (*(inptr-5) == '#') ) {
+ sscanf(inptr-3, "%04d", &scanch);
+ elen = 7;
+ }
+ else if ( (linelen >= 8) && (*(inptr-7) == '&') && (*(inptr-6) == '#') ) {
+ sscanf(inptr-4, "%05d", &scanch);
+ elen = 8;
+ }
- // four-digit decimal equivalents
- else if (outbuf[i] == '&' &&
- outbuf[i + 1] == '#' &&
- isdigit(outbuf[i + 2]) &&
- isdigit(outbuf[i + 3]) &&
- isdigit(outbuf[i + 4]) &&
- isdigit(outbuf[i + 5]) &&
- (outbuf[i + 6] == ';') )
- {
- scanch = 0;
- sscanf(&outbuf[i+2], "%04d", &scanch);
- outbuf[i] = scanch;
- strcpy(&outbuf[i+1], &outbuf[i+7]);
- }
+ if (scanch) {
+ StrBufCutRight(out, elen);
+ linelen -= elen;
- }
+ char utf[5];
+ int ulen = u8_wc_toutf8(utf, scanch);
+ utf[ulen] = 0;
+ StrBufAppendBufPlain(out, utf, ulen, 0);
+ linelen += elen;
+ }
- // Make sure the output buffer is big enough
- if ((output_len + strlen(outbuf) + SIZ) > outptr_buffer_size) {
- outptr_buffer_size += SIZ;
- outptr = realloc(outptr, outptr_buffer_size);
- if (outptr == NULL) {
- abort();
}
- }
-
- // 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')) {
-
- strncpy(&outptr[output_len], outbuf, i+1);
- output_len += (i+1);
- strcpy(outbuf, &outbuf[i+1]);
- i = 0;
- did_out = 1;
+ // Add soft line breaks when necessary
+ if (linelen > (screenwidth - 8)) {
+ char *ptr = (char *)ChrPtr(out) + StrLength(out) - linelen;
+ char *rightmost_space = strrchr(ptr, ' ');
+ if (rightmost_space && rightmost_space > ptr) {
+ int space_pos = rightmost_space - ChrPtr(out);
+ StrBufReplaceToken(out, (long)space_pos, 1, nl, strlen(nl));
+ linelen = strlen(rightmost_space) - 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) {
- strncpy(&outptr[output_len], outbuf, rb);
- output_len += rb;
- strcpy(&outptr[output_len], nl);
- output_len += strlen(nl);
- strcpy(outbuf, &outbuf[rb+1]);
- }
- else {
- strncpy(&outptr[output_len], outbuf, screenwidth-2);
- output_len += (screenwidth-2);
- strcpy(&outptr[output_len], nl);
- output_len += strlen(nl);
- strcpy(outbuf, &outbuf[screenwidth-2]);
- }
- }
-
- } while (done_reading == 0);
- strcpy(&outptr[output_len], outbuf);
- output_len += strlen(outbuf);
-
- // Strip leading/trailing whitespace.
+ // Advance to the next byte of input.
+ inptr++;
+ }
+ }
+ free(inbuf);
+
+ // Convert entity tags to printable characters
+ StrBufReplaceAllOccurrences(out, " ", " ");
+ StrBufReplaceAllOccurrences(out, " ", " ");
+ StrBufReplaceAllOccurrences(out, " ", " ");
+ StrBufReplaceAllOccurrences(out, " ", " ");
+ StrBufReplaceAllOccurrences(out, "<", "<");
+ StrBufReplaceAllOccurrences(out, ">", ">");
+ StrBufReplaceAllOccurrences(out, "&", "&");
+ StrBufReplaceAllOccurrences(out, """, "\"");
+ StrBufReplaceAllOccurrences(out, "‘", "`");
+ StrBufReplaceAllOccurrences(out, "’", "'");
+ StrBufReplaceAllOccurrences(out, "•", " * ");
+ StrBufReplaceAllOccurrences(out, "…", "…");
+ StrBufReplaceAllOccurrences(out, "©", "©");
+ StrBufReplaceAllOccurrences(out, "™", "™");
+ StrBufReplaceAllOccurrences(out, "®", "®");
+ StrBufReplaceAllOccurrences(out, "¼", "¼");
+ StrBufReplaceAllOccurrences(out, "½", "½");
+ StrBufReplaceAllOccurrences(out, "¾", "¾");
+ StrBufReplaceAllOccurrences(out, "–", "–");
+ StrBufReplaceAllOccurrences(out, "—", "—");
+ StrBufReplaceAllOccurrences(out, "Ç", "Ç");
+ StrBufReplaceAllOccurrences(out, "ç", "ç");
+ StrBufReplaceAllOccurrences(out, "È", "È");
+ StrBufReplaceAllOccurrences(out, "è", "è");
+ StrBufReplaceAllOccurrences(out, "Ê", "Ê");
+ StrBufReplaceAllOccurrences(out, "ê", "ê");
+ StrBufReplaceAllOccurrences(out, "É", "É");
+ StrBufReplaceAllOccurrences(out, "é", "é");
+ StrBufReplaceAllOccurrences(out, "À", "À");
+ StrBufReplaceAllOccurrences(out, "à", "à");
+ StrBufReplaceAllOccurrences(out, "“", "\"");
+ StrBufReplaceAllOccurrences(out, "”", "\"");
+ StrBufReplaceAllOccurrences(out, "´", "'");
+ StrBufReplaceAllOccurrences(out, "’", "'");
+ StrBufReplaceAllOccurrences(out, "–", "-");
+
+ // Convert from a StrBuf to a plain C string
+ int output_len = StrLength(out);
+ outptr = SmashStrBuf(&out);
+
+ // Strip leading whitespace
while ((output_len > 0) && (isspace(outptr[0]))) {
strcpy(outptr, &outptr[1]);
--output_len;
}
- while ((output_len > 0) && (isspace(outptr[output_len-1]))) {
- outptr[output_len-1] = 0;
+
+ // Strip trailing whitespace
+ while ((output_len > 0) && (isspace(outptr[output_len - 1]))) {
+ outptr[output_len - 1] = 0;
--output_len;
}
// Make sure the final line ends with a newline character.
- if ((output_len > 0) && (outptr[output_len-1] != '\n')) {
+ if ((output_len > 0) && (outptr[output_len - 1] != '\n')) {
strcat(outptr, "\n");
++output_len;
}
return outptr;
-
}
projects / citadel.git / blobdiff