static void sc_html_parse_special(SC_HTMLParser *parser)
{
gchar symbol_name[9];
gint n;
const gchar *val;
parser->state = SC_HTML_UNKNOWN;
cm_return_if_fail(*parser->bufp == '&');
/* &foo; */
for (n = 0; parser->bufp[n] != '\0' && parser->bufp[n] != ';'; n++)
;
if (n > 7 || parser->bufp[n] != ';') {
/* output literal `&' */
sc_html_append_char(parser, *parser->bufp++);
parser->state = SC_HTML_NORMAL;
return;
}
strncpy2(symbol_name, parser->bufp, n + 2);
parser->bufp += n + 1;
if ((val = g_hash_table_lookup(parser->symbol_table, symbol_name))
!= NULL) {
sc_html_append_str(parser, val, -1);
parser->state = SC_HTML_NORMAL;
return;
}
sc_html_append_str(parser, symbol_name, -1);
}
gchar *sc_html_parse(SC_HTMLParser *parser)
{
parser->state = SC_HTML_NORMAL;
g_string_truncate(parser->str, 0);
if (*parser->bufp == '\0') {
g_string_truncate(parser->buf, 0);
parser->bufp = parser->buf->str;
if (sc_html_read_line(parser) == SC_HTML_EOF)
return NULL;
}
while (*parser->bufp != '\0') {
switch (*parser->bufp) {
case '<': {
SC_HTMLState st;
st = sc_html_parse_tag(parser);
/* when we see an href, we need to flush the str
* buffer. Then collect all the chars until we
* see the end anchor tag
*/
if (SC_HTML_HREF_BEG == st || SC_HTML_HREF == st)
return parser->str->str;
}
break;
case '&':
sc_html_parse_special(parser);
break;
case ' ':
case '\t':
case '\r':
case '\n':
if (parser->bufp[0] == '\r' && parser->bufp[1] == '\n')
parser->bufp++;
if (!parser->pre) {
if (!parser->newline)
parser->space = TRUE;
parser->bufp++;
break;
}
/* fallthrough */
default:
sc_html_append_char(parser, *parser->bufp++);
}
}
return parser->str->str;
}
static void sc_html_parse_special(SC_HTMLParser *parser)
{
gchar *entity;
parser->state = SC_HTML_UNKNOWN;
cm_return_if_fail(*parser->bufp == '&');
entity = entity_decode(parser->bufp);
if (entity != NULL) {
sc_html_append_str(parser, entity, -1);
g_free(entity);
while (*parser->bufp++ != ';');
} else {
/* output literal `&' */
sc_html_append_char(parser, *parser->bufp++);
}
parser->state = SC_HTML_NORMAL;
}
static SC_HTMLState sc_html_parse_tag(SC_HTMLParser *parser)
{
gchar buf[SC_HTMLBUFSIZE];
SC_HTMLTag *tag;
sc_html_get_parenthesis(parser, buf, sizeof(buf));
tag = sc_html_get_tag(buf);
parser->state = SC_HTML_UNKNOWN;
if (!tag) return SC_HTML_UNKNOWN;
if (!strcmp(tag->name, "br") || !strcmp(tag->name, "br/")) {
parser->space = FALSE;
sc_html_append_char(parser, '\n');
parser->state = SC_HTML_BR;
} else if (!strcmp(tag->name, "a")) {
GList *cur;
if (parser->href != NULL) {
g_free(parser->href);
parser->href = NULL;
}
for (cur = tag->attr; cur != NULL; cur = cur->next) {
if (cur->data && !strcmp(((SC_HTMLAttr *)cur->data)->name, "href")) {
g_free(parser->href);
parser->href = g_strdup(((SC_HTMLAttr *)cur->data)->value);
decode_href(parser);
parser->state = SC_HTML_HREF_BEG;
break;
}
}
if (parser->href == NULL)
parser->href = g_strdup("");
parser->state = SC_HTML_HREF_BEG;
} else if (!strcmp(tag->name, "/a")) {
parser->state = SC_HTML_HREF;
} else if (!strcmp(tag->name, "p")) {
parser->space = FALSE;
if (!parser->empty_line) {
parser->space = FALSE;
if (!parser->newline) sc_html_append_char(parser, '\n');
sc_html_append_char(parser, '\n');
}
parser->state = SC_HTML_PAR;
} else if (!strcmp(tag->name, "pre")) {
parser->pre = TRUE;
parser->state = SC_HTML_PRE;
} else if (!strcmp(tag->name, "/pre")) {
parser->pre = FALSE;
parser->state = SC_HTML_NORMAL;
} else if (!strcmp(tag->name, "hr")) {
if (!parser->newline) {
parser->space = FALSE;
sc_html_append_char(parser, '\n');
}
sc_html_append_str(parser, HR_STR, -1);
sc_html_append_char(parser, '\n');
parser->state = SC_HTML_HR;
} else if (!strcmp(tag->name, "div") ||
!strcmp(tag->name, "ul") ||
!strcmp(tag->name, "li") ||
!strcmp(tag->name, "table") ||
!strcmp(tag->name, "dd") ||
!strcmp(tag->name, "tr")) {
if (!parser->newline) {
parser->space = FALSE;
sc_html_append_char(parser, '\n');
}
if (!strcmp(tag->name, "li")) {
sc_html_append_str(parser, LI_STR, -1);
}
parser->state = SC_HTML_NORMAL;
} else if (tag->name[0] == 'h' && g_ascii_isdigit(tag->name[1])) {
if (!parser->newline) {
parser->space = FALSE;
sc_html_append_char(parser, '\n');
}
sc_html_append_char(parser, '\n');
} else if (!strcmp(tag->name, "blockquote")) {
parser->state = SC_HTML_NORMAL;
parser->indent++;
} else if (!strcmp(tag->name, "/blockquote")) {
parser->state = SC_HTML_NORMAL;
parser->indent--;
} else if (!strcmp(tag->name, "/table") ||
(tag->name[0] == '/' &&
tag->name[1] == 'h' &&
g_ascii_isdigit(tag->name[2]))) {
if (!parser->empty_line) {
parser->space = FALSE;
if (!parser->newline) sc_html_append_char(parser, '\n');
sc_html_append_char(parser, '\n');
}
parser->state = SC_HTML_NORMAL;
} else if (!strcmp(tag->name, "/div") ||
!strcmp(tag->name, "/ul") ||
!strcmp(tag->name, "/li")) {
if (!parser->newline) {
parser->space = FALSE;
sc_html_append_char(parser, '\n');
}
parser->state = SC_HTML_NORMAL;
}
sc_html_free_tag(tag);
return parser->state;
}