// Reads a field ending with a comma, hyphen, newLine, semi-colon, or
// Opening or closing parentheses
// Delimiting character is not appended
// CurPos is incremented past the delimiting character
// Returns the delimiting character, or NULL if blank field
char CSVFile::getShortField (char *str, int maxlen)
{
int i = 0;
while (dataLine[curPos] != COMMA &&
dataLine[curPos] != '\0' &&
// dataLine[curPos] != HYPHEN &&
dataLine[curPos] != SEMI_COLON &&
dataLine[curPos] != OPEN_PAREN &&
dataLine[curPos] != CLOSE_PAREN &&
curPos < lineLength &&
i < maxlen)
str[i++] = dataLine[curPos++];
if (dataLine[curPos] != COMMA &&
dataLine[curPos] != '\0' &&
// dataLine[curPos] != HYPHEN &&
dataLine[curPos] != SEMI_COLON &&
dataLine[curPos] != OPEN_PAREN &&
dataLine[curPos] != CLOSE_PAREN &&
i == maxlen && i)
{
str[i-1] = NULL;
skipField (1);
}
else
{
str[i] = NULL;
curPos++;
}
// returns the delimiting character
return (dataLine[curPos-1])? dataLine[curPos-1]: '\1';
}
// Reads a field ending with a comma, newline, or at maxlen
// Delimiting character is not appended
// CurPos is incremented to the next field
// Returns FALSE if the current field is blank
int CSVFile::getField (char *str, int maxlen)
{
int i = 0;
while (dataLine[curPos] != COMMA && i < maxlen && curPos < lineLength)
str[i++] = dataLine[curPos++];
if (dataLine[curPos] != COMMA && i == maxlen)
{
str[--i] = NULL;
skipField (1);
}
else
{
str[i] = NULL;
curPos++;
}
CString csStr = str;
csStr.TrimLeft();
csStr.TrimRight();
strcpy( str, csStr );
// returns count of characters in field
return i;
}
void MongoSchema::process(){
//std::cout << "Processing " << m_dbname << "." << m_col << std::endl;
std::string querystr;
querystr.clear();
querystr.append(m_dbname);
querystr.append(".");
querystr.append(m_col);
int recordscount = m_conn->count(querystr);
//std::cout << "count:" << recordscount << std::endl;
std::auto_ptr<mongo::DBClientCursor> cursor = m_conn->query(querystr, mongo::Query());
//std::set<std::string> fields;
while(cursor->more()){
mongo::BSONObj bo = cursor->next();
for( BSONObj::iterator i = bo.begin(); i.more(); ) {
BSONElement e = i.next();
if(skipField(e.fieldName())){
continue;
}
if(e.isSimpleType()){
hashmap::const_iterator keyexsit = m_map.find(e.fieldName());
SchemaModel* sm = new SchemaModel();
if(keyexsit != m_map.end()){
sm = &m_map[e.fieldName()];
sm->count ++;
}else{
sm->count = 1;
sm->datatype = getType(e);
m_map[e.fieldName()] = *sm;
}
}else if(e.isABSONObj()){
int depth = 0;
std::string parent = e.fieldName();
extractBSON(e.Obj(), depth, parent);
}
}
}
BSONObjBuilder bOb;
BSONArrayBuilder bArr;
std::tr1::hash<std::string> hashfunc = m_map.hash_function();
for( hashmap::const_iterator i = m_map.begin(), e = m_map.end() ; i != e ; ++i ) {
SchemaModel sm = i->second;
float percentage = (float)sm.count / (float)recordscount * 100.0;
std::cout.precision(4);
BSONObj bo = BSON( "field" << i->first << "percent" << percentage << "datatype" << sm.datatype );
bArr.append(bo);
//std::cout << i->first << " -> "<< "Percent: "<< percentage << " (hash = " << hashfunc( i->first ) << ")" << "\r\n";
}
bOb.append(m_col, bArr.arr());
m_schema = bOb.obj();
}
// sets current position to the start of a specific field number
// field numbers are zero based
void CSVFile::setToField (int fieldNum)
{
if (fieldNum == END)
{
curPos = lineLength;
return;
}
curPos = 0;
skipField (fieldNum);
}
int MongoSchema::extractBSON(mongo::BSONObj bo, int& depth, std::string parent){
if(depth >= m_depth){
return 0;
}
depth++;
for( BSONObj::iterator i = bo.begin(); i.more(); ) {
BSONElement e = i.next();
if(skipField(e.fieldName())){
continue;
}
std::string fieldname = parent ;
fieldname.append(".");
fieldname.append(e.fieldName());
if(e.isSimpleType()){
hashmap::const_iterator keyexsit = m_map.find(fieldname);
SchemaModel* sm = new SchemaModel();
if(keyexsit != m_map.end()){
sm = &m_map[fieldname];
sm->count ++;
}else{
sm->count = 1;
sm->datatype = getType(e);
m_map[fieldname] = *sm;
}
}else if(e.isABSONObj()){
extractBSON(e.Obj(), depth, fieldname);
}
}
return 0;
}
int handleKey(TDA_Task* task, char** key, char** content) {
int out = -1;
if (strcmp(*key, "data") == 0) {
match(content, "{");
openBrackets++;
out = parse(content, key);
while (out > -1) {
out = handleKey(task, key, content);
if (!out) { // out == 0, Stop condition
return out;
} else if (out < 0) {
return out;
} else {
out = parse(content, key);
}
}
} else if (strcmp(*key, "assignee") == 0) {
NamedObject n;
out = readNamedObject(content, key, &n);
setTaskAssignee(task, &n);
} else if (strcmp(*key, "id") == 0) {
out = getInt(content, key);
setTaskId(task, *key);
} else if (strcmp(*key, "name") == 0) {
out = getString(content, key);
setTaskName(task, *key);
} else if (strcmp(*key, "notes") == 0) {
out = getString(content, key);
setTaskNotes(task, *key);
} else if (strcmp(*key, "created_at") == 0) {
out = getDate(content, key);
setTaskCreationDate(task, *key);
} else if (strcmp(*key, "modified_at") == 0) {
out = getDate(content, key);
setTaskModificationDate(task, *key);
} else if (strcmp(*key, "completed_at") == 0) {
out = getDate(content, key);
setTaskCompletionDate(task, *key);
} else if (strcmp(*key, "due_on") == 0) {
out = getDate(content, key);
setTaskDueDate(task, *key);
} else if (strcmp(*key, "completed") == 0) {
out = getBool(content, key);
if (out > -1) setTaskCompleted(task, (bool) out);
out = 2; // Can't return out, because it's 0 on false and 0 is parsing complete
} else if (strcmp(*key, "tags") == 0) {
out = match(content, "[");
while(out > -1 && out != ']') {
NamedObject n;
out = readNamedObject(content, key, &n);
addTaskTag(task, &n);
}
out = match(content, ",");
} else {
out = skipField(content);
if (out == 0) {
return 0;
}
}
return out;
}
