- improved parsing of html a-tags

git-svn-id: http://mirotr.googlecode.com/svn/trunk@15 eced67a3-f377-a0ae-92ae-d6de1850b05a

1 files changed, 15 insertions, 4 deletions

diff --git a/MirOTR/striphtml.cpp b/MirOTR/striphtml.cpp
index f4eebaf..3262f6d 100644
--- a/MirOTR/striphtml.cpp
+++ b/MirOTR/striphtml.cpp
@@ -35,14 +35,14 @@ void starttag_cb (void *cbdata, ekhtml_string_t *tag, ekhtml_attr_t *attrs) {
 		case 3:

 			if (_strnicmp(tag->str, "img", 3) == 0) {

 				ekhtml_attr_t *attr = attrs;

-				data->buffer.append("IMAGE:( ");

+				data->buffer.append("IMAGE [ ");

 					while (attr) {

 						if (_strnicmp(attr->name.str, "src", attr->name.len)==0) {

 							data->buffer.append(attr->val.str, attr->val.len);

 							break;

 						}

 					}

-				data->buffer.append(" )");

+				data->buffer.append(" ] ");

 			}

 			break;

 	}

@@ -55,11 +55,12 @@ void endtag_cb (void *cbdata, ekhtml_string_t *tag) {
 			switch (*(tag->str)) {

 				case 'a':

 				case 'A':

-					data->buffer.append(":( ");

+					if (data->stack.empty()) break;

+					data->buffer.append(" [ ");

 					data->buffer.append(data->stack.top());

 					mir_free(data->stack.top());

 					data->stack.pop();

-					data->buffer.append(" ) ");

+					data->buffer.append(" ] ");

 					break;

 				case 'i':

 				case 'I':

@@ -78,6 +79,16 @@ void data_cb (void *cbdata, ekhtml_string_t *text) {
 	STRIPHTML_DATA *data = (STRIPHTML_DATA *)cbdata;

 	char* s = (char*) mir_calloc(text->len+1);

 	decode_html_entities_utf8(s, text->str, text->len);

+

+	if (!data->stack.empty()) {

+		char *top = data->stack.top();

+		if (_stricmp(s, top)==0) {

+			mir_free(top);

+			data->stack.pop();

+		}

+	}

+	

+

 	data->buffer.append(s);

 	mir_free(s);

 }