my_bool
my_parse_charset_xml(MY_CHARSET_LOADER *loader, const char *buf, size_t len)
{
MY_XML_PARSER p;
struct my_cs_file_info info;
my_bool rc;
my_charset_file_init(&info);
my_xml_parser_create(&p);
my_xml_set_enter_handler(&p,cs_enter);
my_xml_set_value_handler(&p,cs_value);
my_xml_set_leave_handler(&p,cs_leave);
info.loader= loader;
my_xml_set_user_data(&p, (void *) &info);
rc= (my_xml_parse(&p,buf,len) == MY_XML_OK) ? FALSE : TRUE;
my_xml_parser_free(&p);
my_charset_file_free(&info);
if (rc != MY_XML_OK)
{
const char *errstr= my_xml_error_string(&p);
if (sizeof(loader->error) > 32 + strlen(errstr))
{
/* We cannot use my_snprintf() here. See previous comment. */
sprintf(loader->error, "at line %d pos %d: %s",
my_xml_error_lineno(&p)+1,
(int) my_xml_error_pos(&p),
my_xml_error_string(&p));
}
}
return rc;
}
my_bool my_parse_charset_xml(const char *buf, uint len,
int (*add_collation)(CHARSET_INFO *cs))
{
MY_XML_PARSER p;
struct my_cs_file_info i;
my_bool rc;
my_xml_parser_create(&p);
my_xml_set_enter_handler(&p,cs_enter);
my_xml_set_value_handler(&p,cs_value);
my_xml_set_leave_handler(&p,cs_leave);
i.add_collation= add_collation;
my_xml_set_user_data(&p,(void*)&i);
rc= (my_xml_parse(&p,buf,len) == MY_XML_OK) ? FALSE : TRUE;
my_xml_parser_free(&p);
return rc;
}
int main(int ac, char **av)
{
char str[1024*64]="";
const char *fn;
int f;
uint len;
MY_XML_PARSER p;
if (ac<2)
{
usage(av[0]);
return 0;
}
fn=av[1]?av[1]:"test.xml";
if ((f=open(fn,O_RDONLY))<0)
{
fprintf(stderr,"Err '%s'\n",fn);
return 1;
}
len=read(f,str,sizeof(str)-1);
str[len]='\0';
my_xml_parser_create(&p);
my_xml_set_enter_handler(&p,bstr);
my_xml_set_value_handler(&p,dstr);
my_xml_set_leave_handler(&p,estr);
if (MY_XML_OK!=(f=my_xml_parse(&p,str,len)))
{
printf("ERROR at line %d pos %d '%s'\n",
my_xml_error_lineno(&p)+1,
my_xml_error_pos(&p),
my_xml_error_string(&p));
}
my_xml_parser_free(&p);
return 0;
}
bool aggregator_parse_feed( string data, map <string, string> &feed )
{
channel.clear();
image.clear();
items.clear();
item = 0;
XML_Parser xml_parser = my_xml_parser_create( data );
XML_SetUserData(xml_parser, &xml_parser);
XML_SetElementHandler(xml_parser, aggregator_element_start, aggregator_element_end);
XML_SetCharacterDataHandler(xml_parser, aggregator_element_data);
if( !XML_Parse(xml_parser, data.c_str(), (int) data.length(), 1) )
{
string error = "The feed from "+feed["title"]+" seems to be broken, because of error " +
XML_ErrorString( XML_GetErrorCode( xml_parser ) ) + " on line " +
str( XML_GetCurrentLineNumber( xml_parser ) ) + ".";
set_page_message( error, "error");
return false;
}
XML_ParserFree(xml_parser);
for( int i = item; i>=0; i-- )
{
string title, link, guid, date;
for( map <string, string>::iterator curr = items[i].begin(), end = items[i].end(); curr != end; curr++ )
{
trim( curr->second );
items[i][curr->first] = curr->second;
}
// Resolve the item's title. If no title is found, we use up to 40
// characters of the description ending at a word boundary but not
// splitting potential entities.
if (isset(items[i]["TITLE"]))
{
title = items[i]["TITLE"];
}
else if (isset(items[i]["DESCRIPTION"]))
{
title = items[i]["DESCRIPTION"].substr(0, 40);
}
else
{
title = "";
}
// Resolve the items link.
if (isset(items[i]["LINK"]))
{
link = items[i]["LINK"];
}
else
{
link = feed["link"];
}
guid = isset(items[i]["GUID"]) ? items[i]["GUID"] : "";
// Atom feeds have a CONTENT and/or SUMMARY tag instead of a DESCRIPTION tag.
if (isset(items[i]["CONTENT:ENCODED"])) {
items[i]["DESCRIPTION"] = items[i]["CONTENT:ENCODED"];
}
else if (isset(items[i]["SUMMARY"])) {
items[i]["DESCRIPTION"] = items[i]["SUMMARY"];
}
else if (isset(items[i]["CONTENT"])) {
items[i]["DESCRIPTION"] = items[i]["CONTENT"];
}
if (isset(items[i]["DC:CREATOR"])) {
items[i]["AUTHOR"] = items[i]["DC:CREATOR"];
}
// Try to resolve and parse the item's publication date. If no date is
// found, we use the current date instead.
date = "now";
if (isset(items[i]["PUBDATE"])) { date = items[i]["PUBDATE"]; }
if (isset(items[i]["DC:DATE"])) { date = items[i]["DC:DATE"]; }
if (isset(items[i]["DCTERMS:ISSUED"])) { date = items[i]["DCTERMS:ISSUED"]; }
if (isset(items[i]["DCTERMS:CREATED"])) { date = items[i]["DCTERMS:CREATED"]; }
if (isset(items[i]["DCTERMS:MODIFIED"])) { date = items[i]["DCTERMS:MODIFIED"]; }
if (isset(items[i]["ISSUED"])) { date = items[i]["ISSUED"]; }
if (isset(items[i]["CREATED"])) { date = items[i]["CREATED"]; }
if (isset(items[i]["MODIFIED"])) { date = items[i]["MODIFIED"]; }
if (isset(items[i]["PUBLISHED"])) { date = items[i]["PUBLISHED"]; }
if (isset(items[i]["UPDATED"])) { date = items[i]["UPDATED"]; }
string timestamp = strtotime( date );
if( timestamp == "") timestamp = str( time() );
map <string, string> entry;
if(DB_TYPE==1)
{
if( isset( guid ) )
{
map <string, string> item;
REDIS_RES *result = redis_query_fields( redis_arg("SORT aggregator_item:fid:%d BY nosort", intval(feed["fid"])), "GET aggregator_item:*->", "#iid,guid");
while( redis_fetch_fields( result, item ) ) {
if( item["guid"] == guid) {
redis_free(result);
entry["iid"] = item["iid"];
break;
}
}
}
else if (isset(link) && link != feed["link"] && link != feed["url"])
{
map <string, string> item;
REDIS_RES *result = redis_query_fields( redis_arg("SORT aggregator_item:fid:%d BY nosort", intval(feed["fid"])), "GET aggregator_item:*->", "#iid,link");
while( redis_fetch_fields( result, item ) ) {
if( entry["link"] == link) {
redis_free(result);
entry["iid"] = item["iid"];
break;
}
}
}
else
{
map <string, string> item;
REDIS_RES *result = redis_query_fields( redis_arg("SORT aggregator_item:fid:%d BY nosort", intval(feed["fid"])), "GET aggregator_item:*->", "#iid,title");
while( redis_fetch_fields( result, item ) ) {
if( entry["title"] == title) {
redis_free(result);
entry["iid"] = item["iid"];
break;
}
}
}
}
if(DB_TYPE==2)
{
if( isset( guid ) )
{
db_fetch(db_querya("SELECT iid FROM aggregator_item WHERE fid = %d AND guid = '%s'", feed["fid"].c_str(), guid.c_str()), entry );
}
else if (isset(link) && link != feed["link"] && link != feed["url"])
{
db_fetch(db_querya("SELECT iid FROM aggregator_item WHERE fid = %d AND link = '%s'", feed["fid"].c_str(), link.c_str()), entry );
}
else {
db_fetch(db_querya("SELECT iid FROM aggregator_item WHERE fid = %d AND title = '%s'", feed["fid"].c_str(), title.c_str()), entry );
}
}
entry["fid"] = feed["fid"];
entry["livetime"] = feed["livetime"];
entry["timestamp"] = timestamp;
entry["title"] = title;
entry["link"] = link;
entry["author"] = items[i]["AUTHOR"];
entry["description"] = items[i]["DESCRIPTION"];
entry["guid"] = guid;
aggregator_save_item( entry );
}
// Remove all items that are older than flush item timer.
if(DB_TYPE==1)
{
}
if(DB_TYPE==2)
{
db_querya("DELETE FROM aggregator_item WHERE fid = %d AND livetime>0 AND timestamp < unix_timestamp()-livetime", feed["fid"].c_str() );
}
return true;
}
