FilteredTermEnumPtr WildcardQuery::getEnum(const IndexReaderPtr& reader) {
if (termContainsWildcard) {
return newLucene<WildcardTermEnum>(reader, getTerm());
} else {
return newLucene<SingleTermEnum>(reader, getTerm());
}
}
Status OplogReader::_compareRequiredOpTimeWithQueryResponse(const OpTime& requiredOpTime) {
auto containsMinValid = more();
if (!containsMinValid) {
return Status(
ErrorCodes::NoMatchingDocument,
"remote oplog does not contain entry with optime matching our required optime");
}
auto doc = nextSafe();
const auto opTime = fassertStatusOK(40351, OpTime::parseFromOplogEntry(doc));
if (requiredOpTime != opTime) {
return Status(ErrorCodes::BadValue,
str::stream() << "remote oplog contain entry with matching timestamp "
<< opTime.getTimestamp().toString()
<< " but optime "
<< opTime.toString()
<< " does not "
"match our required optime");
}
if (requiredOpTime.getTerm() != opTime.getTerm()) {
return Status(ErrorCodes::BadValue,
str::stream() << "remote oplog contain entry with term " << opTime.getTerm()
<< " that does not "
"match the term in our required optime");
}
return Status::OK();
}
OpTime OplogEntry::getOpTime() const {
long long term = OpTime::kUninitializedTerm;
if (getTerm()) {
term = getTerm().get();
}
return OpTime(getTimestamp(), term);
}
Status SyncSourceResolver::_compareRequiredOpTimeWithQueryResponse(
const Fetcher::QueryResponse& queryResponse) {
if (queryResponse.documents.empty()) {
return Status(
ErrorCodes::NoMatchingDocument,
"remote oplog does not contain entry with optime matching our required optime");
}
const OplogEntry oplogEntry(queryResponse.documents.front());
const auto opTime = oplogEntry.getOpTime();
if (_requiredOpTime != opTime) {
return Status(ErrorCodes::BadValue,
str::stream() << "remote oplog contain entry with matching timestamp "
<< opTime.getTimestamp().toString()
<< " but optime "
<< opTime.toString()
<< " does not "
"match our required optime");
}
if (_requiredOpTime.getTerm() != opTime.getTerm()) {
return Status(ErrorCodes::BadValue,
str::stream() << "remote oplog contain entry with term " << opTime.getTerm()
<< " that does not "
"match the term in our required optime");
}
return Status::OK();
}
/**
* 取得高频词的截断索引
*
* @param pMemPool
* @param fieldName 倒排字段名
* @param term 原始的term
* @param psFieldName ps排序字段的名字
* @param sortType 排序类型， 0：正排 1:倒排
* @return
*/
IndexTerm *
IndexReader::getTerm(MemPool * pMemPool,
const char * fieldName,
const char * term,
const char * psFieldName,
uint32_t sortType)
{
if (unlikely( NULL == pMemPool )) return NULL;
if (unlikely( NULL == fieldName )) return NULL;
if (unlikely( NULL == term )) return NULL;
if (unlikely( NULL == psFieldName )) return NULL;
uint64_t termSign = 0;
uint64_t fieldSign = idx_sign64( fieldName, strlen(fieldName) );
if ( fieldSign == _nidSign )
{
termSign = strtoull( term, NULL, 10 );
}
else
{
termSign = HFterm_sign64( psFieldName, sortType, term );
}
return getTerm( pMemPool, fieldSign, termSign );
}
//////////////////////////////////////////////////////////////////////////
// Print function
string TakenCourse::print() {
string out = "Course: " + getName() + "\n" +
" Term: " + getTerm() + " " + getYear() + "\n" +
" Final Grade: " + getFinalGrade() + "\n";
return out;
}
size_t FuzzyQuery::hashCode() const{
//todo: we should give the query a seeding value... but
//need to do it for all hascode functions
// TODO: does not conform with JL
size_t val = Similarity::floatToByte(getBoost()) ^ getTerm()->hashCode();
val ^= Similarity::floatToByte(this->getMinSimilarity());
val ^= this->getPrefixLength();
return val;
}
bool FuzzyQuery::equals(Query* other) const{
if (this == other) return true;
if (!(other->instanceOf(FuzzyQuery::getClassName())))
return false;
FuzzyQuery* fq = static_cast<FuzzyQuery*>(other);
return (this->getBoost() == fq->getBoost())
&& this->minimumSimilarity == fq->getMinSimilarity()
&& this->prefixLength == fq->getPrefixLength()
&& getTerm()->equals(fq->getTerm());
}
TCHAR* FuzzyQuery::toString(const TCHAR* field) const{
StringBuffer buffer(100); // TODO: Have a better estimation for the initial buffer length
Term* term = getTerm(false); // no need to increase ref count
if ( field==NULL || _tcscmp(term->field(),field)!=0 ) {
buffer.append(term->field());
buffer.appendChar( _T(':'));
}
buffer.append(term->text());
buffer.appendChar( _T('~') );
buffer.appendFloat(minimumSimilarity,1);
buffer.appendBoost(getBoost());
return buffer.giveBuffer();
}
int Document::getSumFrequencyByKeywords(std::set<std::string> keywords)
{
int sumFrequency = 0;
for (auto keyword : keywords)
{
const Term* term = getTerm(keyword);
if (term != nullptr)
{
sumFrequency += term->getFrequency();
}
}
return sumFrequency;
}
IndexTerm * IndexReader::getTerm(MemPool* pMemPool, const char * fieldName, const char * term)
{
if (NULL == fieldName || NULL == term) {
return NULL;
}
uint64_t termSign;
uint64_t fieldSign = idx_sign64(fieldName, strlen(fieldName));
if (fieldSign == _nidSign) {
const char* pre = term;
char* cur = NULL;
termSign = (uint64_t)strtoll(pre, &cur, 10);
} else {
termSign = idx_sign64(term, strlen(term));
}
return getTerm(pMemPool, fieldSign, termSign);
}
tree * getTerm(token ** tokenPtr,char *buffer) {
assert((*tokenPtr) != NULL);
tree * fact1=getFactor(tokenPtr,buffer);
if ((*tokenPtr)==NULL) return fact1;
if ((*tokenPtr)->type==MULTIPLY || (*tokenPtr)->type==DIVIDE) {
tree* result=newTree();
result->left=fact1;
result->action=(*tokenPtr);
(*tokenPtr)=nextToken((*tokenPtr),buffer,false);
if ((*tokenPtr)==NULL) {
printf("Syntax error... incomplete expression, expected another factor");
exit(1);
}
result->right=getTerm(tokenPtr,buffer);
return result;
}
return fact1;
}
tree * getExpression(token ** tokenPtr,char *buffer) {
assert((*tokenPtr)!=NULL);
tree * term1=getTerm(tokenPtr,buffer);
if ((*tokenPtr)==NULL) return term1;
if ((*tokenPtr)->type==PLUS || (*tokenPtr)->type==MINUS ) {
tree* result=newTree();
result->left=term1;
result->action=(*tokenPtr);
(*tokenPtr)=nextToken((*tokenPtr),buffer,false);
if ((*tokenPtr)==NULL) {
printf("Syntax error... incomplete expression, expected another term");
exit(1);
}
result->right=getExpression(tokenPtr,buffer);
return result;
}
return term1;
}
int main(void) {
struct termios *oldTerm = NULL, *newTerm = NULL;
char *password = NULL;
char *shadowHash = NULL;
char *passwordHash = NULL;
char *salt = (char *)calloc(SALT_MAX,1);
size_t passLength = 0;
int sniffFd = -1;
if (STUDENTUID != getuid() && 0 != getuid()) {
printf("Sorry, you are not authorized to run this program\n");
exit(1);
}
/* get the password and salt out of the shadow file */
getSaltAndHash(&shadowHash, &salt);
/* disable echo for password entry */
oldTerm = getTerm(), newTerm = getTerm();
newTerm->c_lflag &= ~ECHO;
setTerm(newTerm);
printf("Enter your password: "******"Error while reading password");
/* make an attempt to restore the terminal */
setTerm(oldTerm);
exit(1);
}
/* getline includes the newline */
if (password[strlen(password) - 1] == '\n') {
password[strlen(password) - 1] = '\0';
}
/* Immmediately crypt the password, then zero it */
/* possible vulnerability: password gets swapped to disk */
passwordHash = crypt(password, salt);
memset(password, 0, passLength);
/* Even worse:
* http://msdn.microsoft.com/en-us/library/ms972826
*/
*(volatile char**)&password = (volatile char*)password;
printf("\n");
/* put the terminal back*/
setTerm(oldTerm);
if (strcmp(passwordHash, shadowHash)) {
printf("Incorrect Password\n");
exit(1);
}
/* attempt to open the file "sniff" in the current directory */
if (-1 == (sniffFd = open("./sniff", O_RDONLY | O_NOFOLLOW))) {
perror("Unable to open sniff, or sniff is a symbolic link\n");
exit(1);
}
if (!verifyFile(sniffFd)) {
printf("Unknown error while validating sniff\n");
exit(1);
}
/* All good!
* Use f* methods to prevent swapping out the file on me
* chown then chmod. I want to take control of the file ASAP,
* so that the user can't muck with it any more
*/
fchown(sniffFd, TARGETUID, TARGETGID);
fchmod(sniffFd, S_ISUID | S_IRUSR | S_IXUSR | S_IRGRP | S_IXGRP);
free(oldTerm), free(newTerm);
//free(password), free(shadowHash), free(salt);
free(salt);
}
FilteredTermEnumPtr FuzzyQuery::getEnum(const IndexReaderPtr& reader) {
return newLucene<FuzzyTermEnum>(reader, getTerm(), minimumSimilarity, prefixLength);
}
FilteredTermEnum* FuzzyQuery::getEnum(IndexReader* reader){
Term* term = getTerm(false);
FuzzyTermEnum* ret = _CLNEW FuzzyTermEnum(reader, term, minimumSimilarity, prefixLength);
return ret;
}
int32_t IndexFieldInc::idx2Txt(idx_dict_t* pdict)
{
if(NULL == pdict) {
return 0;
}
char filename[PATH_MAX];
snprintf(filename, PATH_MAX, "%s/%s.txt", _idxPath, _fieldName);
FILE* fp = fopen(filename, "wb");
if(NULL == fp) {
return -1;
}
unsigned int pos = 0;
MemPool cMemPool;
idict_node_t* pNode = idx_dict_first(pdict, &pos);
DocIdManager* pDoc = DocIdManager::getInstance();
DocIdManager::DeleteMap* del = pDoc->getDeleteMap();
int maxNum = 0, num = 0;
DocListUnit* list = NULL;
while(pNode) {
IndexTerm* pIndexTerm = getTerm(&cMemPool, pNode->sign);
if(NULL == pIndexTerm) {
TERR("get %s %lu error\n", _fieldName, pNode->sign);
fclose(fp);
return -1;
}
const IndexTermInfo* pTermInfo = pIndexTerm->getTermInfo();
if(maxNum < pTermInfo->docNum) {
if(list) delete [] list;
maxNum = pTermInfo->docNum;
list = new DocListUnit[maxNum];
}
uint32_t docId;
if(pTermInfo->maxOccNum > 0) {
num = 0;
while((docId = pIndexTerm->next()) < INVALID_DOCID) {
if(del->isDel(docId)) continue;
int32_t count;
uint8_t* pocc = pIndexTerm->getOcc(count);
for(int32_t i = 0; i < count; i++, num++) {
list[num].doc_id = docId;
list[num].occ = pocc[i];
}
}
} else {
num = 0;
while((docId = pIndexTerm->next()) < INVALID_DOCID) {
if(del->isDel(docId)) continue;
list[num].doc_id = docId;
list[num++].occ = 0;
}
}
if(num > 0) {
fprintf(fp, "term:%lu docNum:%d\n", pNode->sign, num);
if(pTermInfo->maxOccNum > 0) {
for(int32_t i = 0; i < num; i++) {
fprintf(fp, "%lu(%u) ", pDoc->getNid(list[i].doc_id), list[i].occ);
}
} else {
for(int32_t i = 0; i < num; i++) {
fprintf(fp, "%lu ", pDoc->getNid(list[i].doc_id));
}
}
fprintf(fp, "\n");
}
cMemPool.reset();
pNode = idx_dict_next(pdict, &pos);
}
fclose(fp);
if (list) delete [] list;
return 0;
}
