void Parser::SearchForIsolatedNumbers() {
for (token_container_t::iterator token = tokens_.begin(); token != tokens_.end(); ++token) {
if (token->category != kUnknown ||
!IsNumericString(token->content) ||
!IsTokenIsolated(token))
continue;
int number = StringToInt(token->content);
// Anime year
if (number >= kAnimeYearMin && number <= kAnimeYearMax) {
if (elements_.empty(kElementAnimeYear)) {
elements_.insert(kElementAnimeYear, token->content);
token->category = kIdentifier;
continue;
}
}
// Video resolution
if (number == 480 || number == 720 || number == 1080) {
// If these numbers are isolated, it's more likely for them to be the
// video resolution rather than the episode number. Some fansub groups
// use these without the "p" suffix.
if (elements_.empty(kElementVideoResolution)) {
elements_.insert(kElementVideoResolution, token->content);
token->category = kIdentifier;
continue;
}
}
}
}
bool Parser::CheckAnimeSeasonKeyword(const token_iterator_t token)
{
auto set_anime_season = [&](token_iterator_t first, token_iterator_t second,
const string_t& content)
{
elements_.insert(kElementAnimeSeason, content);
first->category = kIdentifier;
second->category = kIdentifier;
};
auto previous_token = FindPreviousToken(tokens_, token, kFlagNotDelimiter);
if (previous_token != tokens_.end())
{
auto number = GetNumberFromOrdinal(previous_token->content);
if (!number.empty())
{
set_anime_season(previous_token, token, number);
return true;
}
}
auto next_token = FindNextToken(tokens_, token, kFlagNotDelimiter);
if (next_token != tokens_.end() &&
IsNumericString(next_token->content))
{
set_anime_season(token, next_token, next_token->content);
return true;
}
return false;
}
void Parser::SearchForKeywords() {
for (auto it = tokens_.begin(); it != tokens_.end(); ++it) {
auto& token = *it;
if (token.category != kUnknown)
continue;
auto word = token.content;
TrimString(word, L" -");
if (word.empty())
continue;
// Don't bother if the word is a number that cannot be CRC
if (word.size() != 8 && IsNumericString(word))
continue;
// Performs better than making a case-insensitive Find
auto keyword = keyword_manager.Normalize(word);
ElementCategory category = kElementUnknown;
KeywordOptions options;
if (keyword_manager.Find(keyword, category, options)) {
if (!options_.parse_release_group && category == kElementReleaseGroup)
continue;
if (!IsElementCategorySearchable(category) || !options.searchable)
continue;
if (IsElementCategorySingular(category) && !elements_.empty(category))
continue;
if (category == kElementAnimeSeasonPrefix) {
CheckAnimeSeasonKeyword(it);
continue;
} else if (category == kElementEpisodePrefix) {
if (options.valid)
CheckExtentKeyword(kElementEpisodeNumber, it);
continue;
} else if (category == kElementReleaseVersion) {
word = word.substr(1); // number without "v"
} else if (category == kElementVolumePrefix) {
CheckExtentKeyword(kElementVolumeNumber, it);
continue;
}
} else {
if (elements_.empty(kElementFileChecksum) && IsCrc32(word)) {
category = kElementFileChecksum;
} else if (elements_.empty(kElementVideoResolution) &&
IsResolution(word)) {
category = kElementVideoResolution;
}
}
if (category != kElementUnknown) {
elements_.insert(category, word);
if (options.identifiable)
token.category = kIdentifier;
}
}
}
bool CBanManager::IsValidIPPart ( const char* szIP )
{
if ( IsNumericString ( szIP ) )
{
int iIP = atoi ( szIP );
if ( iIP >= 0 && iIP < 256 )
return true;
}
else if ( strcmp ( szIP, "*" ) == 0 )
return true;
return false;
}
void Parser::SearchForEpisodeNumber() {
// List all unknown tokens that contain a number
std::vector<size_t> tokens;
for (size_t i = 0; i < tokens_.size(); ++i) {
auto& token = tokens_.at(i);
if (token.category == kUnknown)
if (FindNumberInString(token.content) != token.content.npos)
tokens.push_back(i);
}
if (tokens.empty())
return;
found_episode_keywords_ = !elements_.empty(kElementEpisodeNumber);
// If a token matches a known episode pattern, it has to be the episode number
if (SearchForEpisodePatterns(tokens))
return;
if (!elements_.empty(kElementEpisodeNumber))
return; // We have previously found an episode number via keywords
// From now on, we're only interested in numeric tokens
auto not_numeric_string = [&](size_t index) -> bool {
return !IsNumericString(tokens_.at(index).content);
};
tokens.erase(std::remove_if(tokens.begin(), tokens.end(), not_numeric_string),
tokens.end());
if (tokens.empty())
return;
// e.g. "01 (176)", "29 (04)"
if (SearchForEquivalentNumbers(tokens))
return;
// e.g. " - 08"
if (SearchForSeparatedNumbers(tokens))
return;
// e.g. "[12]", "(2006)"
if (SearchForIsolatedNumbers(tokens))
return;
// Consider using the last number as a last resort
SearchForLastNumber(tokens);
}
bool Parser::MatchEpisodePatterns(string_t word, Token& token) {
// All patterns contain at least one non-numeric character
if (IsNumericString(word))
return false;
TrimString(word, L" -");
const bool numeric_front = IsNumericChar(word.front());
const bool numeric_back = IsNumericChar(word.back());
// e.g. "01v2"
if (numeric_front && numeric_back)
if (MatchSingleEpisodePattern(word, token))
return true;
// e.g. "01-02", "03-05v2"
if (numeric_front && numeric_back)
if (MatchMultiEpisodePattern(word, token))
return true;
// e.g. "2x01", "S01E03", "S01-02xE001-150"
if (numeric_back)
if (MatchSeasonAndEpisodePattern(word, token))
return true;
// e.g. "ED1", "OP4a", "OVA2"
if (!numeric_front)
if (MatchTypeAndEpisodePattern(word, token))
return true;
// e.g. "07.5"
if (numeric_front && numeric_back)
if (MatchFractionalEpisodePattern(word, token))
return true;
// e.g. "4a", "111C"
if (numeric_front && !numeric_back)
if (MatchPartialEpisodePattern(word, token))
return true;
// e.g. "#01", "#02-03v2"
if (numeric_back)
if (MatchNumberSignPattern(word, token))
return true;
// U+8A71 is used as counter for stories, episodes of TV series, etc.
if (numeric_front)
if (MatchJapaneseCounterPattern(word, token))
return true;
return false;
}
bool Parser::NumberComesBeforeTotalNumber(const token_iterator_t token) {
auto next_token = FindNextToken(tokens_, token, kFlagNotDelimiter);
if (next_token != tokens_.end()) {
if (IsStringEqualTo(next_token->content, L"of")) {
auto other_token = FindNextToken(tokens_, next_token, kFlagNotDelimiter);
if (other_token != tokens_.end()) {
if (IsNumericString(other_token->content)) {
SetEpisodeNumber(token->content, *token, false);
next_token->category = kIdentifier;
other_token->category = kIdentifier;
return true;
}
}
}
}
return false;
}
void Parser::SearchForEpisodeNumber() {
// List all tokens that contain a number
std::vector<size_t> tokens;
for (size_t i = 0; i < tokens_.size(); ++i) {
auto& token = tokens_.at(i);
if (token.category != kUnknown)
continue; // Skip previously identified tokens
if (FindNumberInString(token.content) != token.content.npos)
tokens.push_back(i);
}
if (tokens.empty())
return;
// If a token matches a known episode pattern, it has to be the episode number
if (SearchForEpisodePatterns(tokens))
return;
// From now on, we're only interested in numeric tokens
auto not_numeric_string = [&](size_t index) -> bool {
return !IsNumericString(tokens_.at(index).content);
};
tokens.erase(std::remove_if(tokens.begin(), tokens.end(), not_numeric_string),
tokens.end());
if (tokens.empty())
return;
// e.g. "[12]", "(2006)"
if (SearchForIsolatedNumbers(tokens))
return;
// e.g. " - 08"
if (SearchForSeparatedNumbers(tokens))
return;
// Consider using the last number as a last resort
SearchForLastNumber(tokens);
}
bool Parser::SearchForEquivalentNumbers(std::vector<size_t>& tokens) {
for (auto token_index = tokens.begin();
token_index != tokens.end(); ++token_index) {
auto token = tokens_.begin() + *token_index;
if (IsTokenIsolated(token))
continue;
// Find the first enclosed, non-delimiter token
auto next_token = FindNextToken(tokens_, token, kFlagNotDelimiter);
if (next_token != tokens_.end() &&
next_token->category == kBracket) {
next_token = FindNextToken(tokens_, next_token,
kFlagEnclosed | kFlagNotDelimiter);
} else {
continue;
}
// See if it's an isolated number
if (next_token != tokens_.end() &&
next_token->category == kUnknown &&
IsTokenIsolated(next_token) &&
IsNumericString(next_token->content)) {
if (IsValidEpisodeNumber(token->content) &&
IsValidEpisodeNumber(next_token->content)) {
auto lower_token =
StringToInt(token->content) < StringToInt(next_token->content) ?
token : next_token;
SetEpisodeNumber(lower_token->content, *token, false);
next_token->category = kIdentifier;
return true;
}
}
}
return false;
}
bool Parser::not_numeric_string(size_t index) {
return !IsNumericString(tokens_.at(index).content);
};
bool CPickup::ReadSpecialData ( void )
{
unsigned short usBuffer = 0;
// Grab the "posX" data
if ( !GetCustomDataFloat ( "posX", m_vecPosition.fX, true ) )
{
CLogger::ErrorPrintf ( "Bad/missing 'posX' attribute in <pickup> (line %u)\n", m_uiLine );
return false;
}
// Grab the "posY" data
if ( !GetCustomDataFloat ( "posY", m_vecPosition.fY, true ) )
{
CLogger::ErrorPrintf ( "Bad/missing 'posY' attribute in <pickup> (line %u)\n", m_uiLine );
return false;
}
// Grab the "posZ" data
if ( !GetCustomDataFloat ( "posZ", m_vecPosition.fZ, true ) )
{
CLogger::ErrorPrintf ( "Bad/missing 'posZ' attribute in <pickup> (line %u)\n", m_uiLine );
return false;
}
// Put the collision object at the given xyz
if ( m_pCollision )
m_pCollision->SetPosition ( m_vecPosition );
// Grab the "type" data
char szBuffer [128];
if ( GetCustomDataString ( "type", szBuffer, 128, true ) )
{
// Check what type it is
m_bIsTypeRandom = false;
if ( stricmp ( szBuffer, "health" ) == 0 )
{
m_ucType = HEALTH;
m_usModel = CPickupManager::GetHealthModel ();
}
else if ( stricmp ( szBuffer, "armor" ) == 0 )
{
m_ucType = ARMOR;
m_usModel = CPickupManager::GetArmorModel ();
}
else if ( IsNumericString ( szBuffer ) )
{ // could be a weapon
usBuffer = atoi ( szBuffer );
if ( CPickupManager::IsValidWeaponID ( usBuffer ) )
{ // its a weapon
m_ucType = WEAPON;
m_usModel = CPickupManager::GetWeaponModel ( m_ucWeaponType );
}
}
else if ( stricmp ( szBuffer, "custom" ) == 0 )
{
m_ucType = CUSTOM;
m_usModel = 1700;
}
else if ( stricmp ( szBuffer, "random" ) == 0 )
{
m_ucType = HEALTH;
m_usModel = CPickupManager::GetHealthModel ();
m_bIsTypeRandom = true;
}
else
{
CLogger::LogPrintf ( "WARNING: Unknown 'type' value in <pickup>; defaulting to \"random\" (line %u)\n", m_uiLine );
m_ucType = HEALTH;
m_usModel = CPickupManager::GetHealthModel ();
m_bIsTypeRandom = true;
}
}
else
{
CLogger::ErrorPrintf ( "Bad/missing 'type' attribute in <pickup> (line %u)\n", m_uiLine );
return false;
}
// Is this a weapon pickup?
if ( m_ucType == CPickup::WEAPON || m_bIsTypeRandom )
{
// Remember the weapon type
m_ucWeaponType = static_cast < unsigned char > ( usBuffer );
m_usModel = CPickupManager::GetWeaponModel ( m_ucWeaponType );
m_bIsWeaponTypeRandom = false;
}
// Is this a health pickup?
if ( m_ucType == CPickup::HEALTH || m_ucType == CPickup::ARMOR || m_bIsTypeRandom )
{
// Grab the "health" data
if ( GetCustomDataString ( "amount", szBuffer, 128, true ) )
{
// Is it random?
if ( strcmp ( szBuffer, "random" ) == 0 )
{
m_fAmount = 100.0f;
m_bIsHealthRandom = true;
}
else
{
// Convert the health to an integer and limit it to 100
m_fAmount = static_cast < float > ( atoi ( szBuffer ) );
if ( m_fAmount > 100.0f ) m_fAmount = 100.0f;
}
}
else
{
m_fAmount = 100.0f;
m_bIsHealthRandom = false;
}
}
// Is this a weapon pickup?
int iTemp;
if ( m_ucType == CPickup::WEAPON || m_bIsTypeRandom )
{
// Grab the "ammo" data
if ( GetCustomDataInt ( "amount", iTemp, true ) )
{
// Limit it to 0-9999 if it was above
if ( iTemp > 9999 )
iTemp = 9999;
else if ( iTemp < 0 )
iTemp = 0;
// Remember it
m_usAmmo = static_cast < unsigned short > ( iTemp );
}
else
{
m_usAmmo = 100;
}
}
// Grab the "respawn" data
if ( GetCustomDataInt ( "respawn", iTemp, true ) )
{
// Make sure it's above 3 seconds
if ( iTemp < 3000 )
iTemp = 3000;
// Remember it
m_ulRespawnIntervals = static_cast < unsigned long > ( iTemp );
}
else
{
m_ulRespawnIntervals = 10000;
}
// Is this a custom pickup?
if ( m_ucType == CPickup::CUSTOM )
{
// Grab the "model" data
if ( GetCustomDataInt ( "model", iTemp, true ) )
{
// Valid id?
if ( CObjectManager::IsValidModel ( iTemp ) || iTemp == 370 ) // 370 = jetpack - sort of a hack
{
// Set the object id
m_usModel = static_cast < unsigned short > ( iTemp );
}
else
{
CLogger::ErrorPrintf ( "Bad 'model' id specified in <pickup> (line %u)\n", m_uiLine );
return false;
}
}
else
{
// Error out if custom is specified but no model id is
CLogger::ErrorPrintf ( "Pickup type set to 'custom' but no 'model' id specified (line %u)\n", m_uiLine );
return false;
}
}
if ( GetCustomDataInt ( "dimension", iTemp, true ) )
m_usDimension = static_cast < unsigned short > ( iTemp );
// Success
return true;
}
void Parser::SearchForKeywords() {
for (auto& token : tokens_) {
if (token.category != kUnknown)
continue;
auto word = token.content;
TrimString(word);
// Don't bother if the word is a number that cannot be CRC
if (word.size() != 8 && IsNumericString(word))
continue;
// Performs better than making a case-insensitive Find
auto keyword = StringToUpperCopy(word);
for (int i = kElementIterateFirst; i < kElementIterateLast; i++) {
auto category = static_cast<ElementCategory>(i);
if (!parse_options.parse_release_group)
if (category == kElementReleaseGroup)
continue;
if (!IsElementCategorySearchable(category))
continue;
if (IsElementCategorySingular(category))
if (!elements_.empty(category))
continue;
bool add_keyword = false;
KeywordOptions options;
switch (category) {
case kElementFileChecksum:
add_keyword = IsCrc32(word);
break;
case kElementVideoResolution:
add_keyword = IsResolution(word);
break;
default:
add_keyword = keyword_manager.Find(category, keyword, options);
break;
}
if (add_keyword) {
switch (category) {
case kElementReleaseVersion:
elements_.insert(category, word.substr(1)); // number without "v"
break;
default:
elements_.insert(category, word);
break;
}
if (options.safe || token.enclosed)
token.category = kIdentifier;
break;
}
}
}
}
