/// 문자열 치환
void replace(const char* niddle, const char* changeTo)
{
int pos = find(niddle);
if(pos < 0)
return;
FastString<MaxSize> destStr;
int begin = 0;
int end = pos;
int lenNiddle = strlen(niddle);
while(end >= 0)
{
destStr.append(&buf[begin], end-begin);
destStr.append(changeTo);
begin = end+lenNiddle;
end = find(niddle, begin);
}
// 뒤에 남은 부분 붙여 넣기
if(begin < static_cast<int>(size()))
{
destStr.append(&buf[begin], size()-begin);
}
assign(destStr);
}
// Convert UTF8 FastString to a wide char string
ReadOnlyUnicodeString convert(const FastString & string)
{
// Create some space for the UCS-2 string
ReadOnlyUnicodeString ret(0, string.getLength() + 1);
ReadOnlyUnicodeString::CharType * data = const_cast<ReadOnlyUnicodeString::CharType *>(ret.getData());
if (!data) return ret;
ReadOnlyUnicodeString::CharType * dest = data;
// Direct access to the buffer (don't bound check for every access)
const uint8 * buffer = (const uint8*)string;
for (int i = 0; i < string.getLength();)
{
// Get the current char
const uint8 c = buffer [i++];
// Check for UTF8 code
if ((c & 0x80) != 0)
{
// The data is in the 7 low bits
uint32 dataMask = 0x7f;
// The count bit mask
uint32 bitCountMask = 0x40;
// The consumption count
int charCount = 0;
while ((c & bitCountMask) != 0 && bitCountMask)
{
++charCount;
dataMask >>= 1; bitCountMask >>= 1;
}
// Get the few bits remaining here
int n = (c & dataMask);
// Then extract the remaining bits
while (--charCount >= 0 && i < string.getLength())
{
const uint8 extra = buffer[i];
// Make sure it's a valid UTF8 encoding
if ((extra & 0xc0) != 0x80) break;
// Store the new bits too
n <<= 6; n |= (extra & 0x3f);
++i;
}
*dest++ = (ReadOnlyUnicodeString::CharType)n;
}
else // Append the char as-is
int main(int argc,char* argv[])
{
if (argc != 3)
{
cerr << "Usage: " << argv[0] << " <-s|-n|-S|-N> <email>" << endl;
exit(-1);
}
int feed_type = 0;
if (string(argv[1]) == "-s") feed_type = FEED_SPAM;
else if (string(argv[1]) == "-n") feed_type = FEED_HAM;
else if (string(argv[1]) == "-S") feed_type = UN_FEED_SPAM;
else if (string(argv[1]) == "-N") feed_type = UN_FEED_HAM;
else { cerr << "Usage: " << argv[0] << " <-s|-n|-S|-N> <email>" << endl; exit(-1); }
FastString email_data = get_file_content(argv[2]);
MimeMessage msg(email_data);
FastString charset="";
FastString plain = "";
FastString html = "";
FastString subject="";
set<string> result;
CFenci myFenci;
myFenci.setDict("/usr/local/etc/dict_chs.utf8.xdb");
myFenci.setRule("/usr/local/etc/rules.utf8.ini");
myFenci.setCharset("utf-8");
myFenci.setIgnoreSign();
/* parse subject */
msg.getSubject(subject,charset);
myFenci.getFenciResult(format_to_check(subject.c_str(),charset.c_str()),result);
/* parse plain */
msg.getTextPlain(plain,charset);
myFenci.getFenciResult(format_to_check(subject.c_str(),charset.c_str()),result);
/* parse html */
msg.getTextHtml(html,charset);
string html_to_check = get_text_from_html(html.c_str());
myFenci.getFenciResult(format_to_check(html_to_check,charset.c_str()),result);
/* connect redis */
CRedis myRedis("127.0.0.1",6379);
const int CONNECT_TIMEOUT = 1;
myRedis.setTimeout(CONNECT_TIMEOUT);
if (false == myRedis.Connect())
{
cerr << myRedis.getError() << endl;
return -1;
}
/* feed spam */
if (feed_type == FEED_SPAM)
{
for (set<string>::iterator it = result.begin(); it != result.end(); ++it)
{
string buffer = myRedis.Get(*it);
if (buffer != "")
{
CDataParse myData(buffer," ");
vector<string> ret = myData.getVector();
ret[0] = my_int2str((my_str2int(ret[0]) + 1));
myData.setVector(ret);
if (false == myRedis.Set(*it,myData.getString()))
{
cerr << myRedis.getError() << endl;
return -1;
}
}
else
{
if (false == myRedis.Set(*it,"1 0"))
{
cerr << myRedis.getError() << endl;
return -1;
}
}
}
}
/* feed ham */
if (feed_type == FEED_HAM)
{
for (set<string>::iterator it = result.begin(); it != result.end(); ++it)
{
string buffer = myRedis.Get(*it);
if (buffer != "")
{
CDataParse myData(buffer," ");
vector<string> ret = myData.getVector();
ret[1] = my_int2str((my_str2int(ret[1]) + 1));
myData.setVector(ret);
if (false == myRedis.Set(*it,myData.getString()))
{
cerr << myRedis.getError() << endl;
return -1;
}
}
else
{
if (false == myRedis.Set(*it,"0 1"))
{
cerr << myRedis.getError() << endl;
return -1;
}
}
}
}
/* unfeed spam */
if (feed_type == UN_FEED_SPAM)
{
for (set<string>::iterator it = result.begin(); it != result.end(); ++it)
{
string buffer = myRedis.Get(*it);
if (buffer != "")
{
CDataParse myData(buffer," ");
vector<string> ret = myData.getVector();
if (my_str2int(ret[0]) > 0) ret[0] = my_int2str(my_str2int(ret[0]) - 1);
myData.setVector(ret);
if (false == myRedis.Set(*it,myData.getString()))
{
cerr << myRedis.getError() << endl;
return -1;
}
}
}
}
/* unfeed ham */
if (feed_type == UN_FEED_HAM)
{
for (set<string>::iterator it = result.begin(); it != result.end(); ++it)
{
string buffer = myRedis.Get(*it);
if (buffer != "")
{
CDataParse myData(buffer," ");
vector<string> ret = myData.getVector();
if (my_str2int(ret[1]) > 0) ret[1] = my_int2str(my_str2int(ret[1]) - 1);
myData.setVector(ret);
if (false == myRedis.Set(*it,myData.getString()))
{
cerr << myRedis.getError() << endl;
return -1;
}
}
}
}
/* close redis */
myRedis.Close();
exit(0);
}
int append(const FastString<SrcMaxSize>& value)
{
return append(value.c_str(), value.size());
}
