void Parser::SearchForEpisodeTitle() {
auto token_begin = tokens_.begin();
auto token_end = tokens_.begin();
do {
// Find the first non-enclosed unknown token
token_begin = FindToken(token_end, tokens_.end(),
kFlagNotEnclosed | kFlagUnknown);
if (token_begin == tokens_.end())
return;
// Continue until a bracket or identifier is found
token_end = FindToken(token_begin, tokens_.end(),
kFlagBracket | kFlagIdentifier);
// Ignore if it's only a dash
if (std::distance(token_begin, token_end) <= 2 &&
IsDashCharacter(token_begin->content)) {
continue;
}
// Build episode title
BuildElement(kElementEpisodeTitle, false, token_begin, token_end);
return;
} while (token_begin != tokens_.end());
}
void Parser::SearchForReleaseGroup() {
token_container_t::iterator token_begin = tokens_.begin();
token_container_t::iterator token_end = tokens_.begin();
do {
// Find the first enclosed unknown token
token_begin = FindToken(token_end, tokens_.end(),
kFlagEnclosed | kFlagUnknown);
if (token_begin == tokens_.end())
return;
// Continue until a bracket or identifier is found
token_end = FindToken(token_begin, tokens_.end(),
kFlagBracket | kFlagIdentifier);
if (token_end->category != kBracket)
continue;
// Ignore if it's not the first non-delimiter token in group
token_container_t::iterator previous_token = FindPreviousToken(tokens_, token_begin,
kFlagNotDelimiter);
if (previous_token != tokens_.end() &&
previous_token->category != kBracket) {
continue;
}
// Build release group
BuildElement(kElementReleaseGroup, true, token_begin, token_end);
return;
} while (token_begin != tokens_.end());
}
void Parser::SearchForAnimeTitle() {
// Find the first non-enclosed unknown token
auto token_begin = std::find_if(tokens_.begin(), tokens_.end(),
[](const Token& token) {
return !token.enclosed && token.category == kUnknown;
});
if (token_begin == tokens_.end())
return;
// Continue until an identifier is found
auto token_end = std::find_if(token_begin, tokens_.end(),
[](const Token& token) {
return token.category == kIdentifier;
});
// If within the interval there's an open bracket without its matching pair,
// move the upper endpoint back to the bracket
auto last_bracket = token_end;
bool bracket_open = false;
for (auto token = token_begin; token != token_end; ++token) {
if (token->category == kBracket) {
last_bracket = token;
bracket_open = !bracket_open;
}
}
if (bracket_open)
token_end = last_bracket;
// Build anime title
BuildElement(kElementAnimeTitle, false, token_begin, token_end);
}
void Parser::SearchForAnimeTitle() {
bool enclosed_title = false;
// Find the first non-enclosed unknown token
token_container_t::iterator token_begin = FindToken(tokens_.begin(), tokens_.end(),
kFlagNotEnclosed | kFlagUnknown);
// If that doesn't work, find the first unknown token in the second enclosed
// group, assuming that the first one is the release group
if (token_begin == tokens_.end()) {
enclosed_title = true;
token_begin = tokens_.begin();
bool skipped_previous_group = false;
do {
token_begin = FindToken(token_begin, tokens_.end(), kFlagUnknown);
if (token_begin == tokens_.end())
break;
// Ignore groups that are composed of non-Latin characters
if (IsMostlyLatinString(token_begin->content))
if (skipped_previous_group)
break; // Found it
// Get the first unknown token of the next group
token_begin = FindToken(token_begin, tokens_.end(), kFlagBracket);
token_begin = FindToken(token_begin, tokens_.end(), kFlagUnknown);
skipped_previous_group = true;
} while (token_begin != tokens_.end());
}
if (token_begin == tokens_.end())
return;
// Continue until an identifier (or a bracket, if the title is enclosed)
// is found
token_container_t::iterator token_end = FindToken(token_begin, tokens_.end(),
kFlagIdentifier | (enclosed_title ? kFlagBracket : kFlagNone));
// If within the interval there's an open bracket without its matching pair,
// move the upper endpoint back to the bracket
if (!enclosed_title) {
token_container_t::iterator last_bracket = token_end;
bool bracket_open = false;
for (token_container_t::iterator token = token_begin; token != token_end; ++token) {
if (token->category == kBracket) {
last_bracket = token;
bracket_open = !bracket_open;
}
}
if (bracket_open)
token_end = last_bracket;
}
// Build anime title
BuildElement(kElementAnimeTitle, false, token_begin, token_end);
}
void Parser::SearchForReleaseGroup() {
auto token_begin = tokens_.begin();
auto token_end = tokens_.begin();
do {
// Find the first enclosed unknown token
token_begin = std::find_if(token_end, tokens_.end(),
[](const Token& token) {
return token.enclosed && token.category == kUnknown;
});
if (token_begin == tokens_.end())
continue;
// Continue until a bracket or identifier is found
token_end = std::find_if(token_begin, tokens_.end(),
[](const Token& token) {
return token.category == kBracket || token.category == kIdentifier;
});
if (token_end->category != kBracket)
continue;
// Ignore if it's not the first token in group
auto previous_token = GetPreviousValidToken(token_begin);
if (previous_token != tokens_.end() &&
previous_token->category != kBracket) {
continue;
}
// Build release group, or anime title if it wasn't found earlier
if (elements_.empty(kElementReleaseGroup)) {
BuildElement(kElementReleaseGroup, true, token_begin, token_end);
if (elements_.empty(kElementAnimeTitle))
continue;
} else if (elements_.empty(kElementAnimeTitle)) {
BuildElement(kElementAnimeTitle, false, token_begin, token_end);
return;
}
} while (token_begin != tokens_.end());
}
void Parser::SearchForEpisodeTitle() {
// Find the first non-enclosed unknown token
token_container_t::iterator token_begin = FindToken(tokens_.begin(), tokens_.end(),
kFlagNotEnclosed | kFlagUnknown);
if (token_begin == tokens_.end())
return;
// Continue until a bracket or identifier is found
token_container_t::iterator token_end = FindToken(token_begin, tokens_.end(),
kFlagBracket | kFlagIdentifier);
// Build episode title
BuildElement(kElementEpisodeTitle, false, token_begin, token_end);
}
void Parser::SearchForEpisodeTitle() {
// Find the first non-enclosed unknown token
auto token_begin = std::find_if(tokens_.begin(), tokens_.end(),
[](const Token& token) {
return !token.enclosed && token.category == kUnknown;
});
if (token_begin == tokens_.end())
return;
// Continue until a bracket or identifier is found
auto token_end = std::find_if(token_begin, tokens_.end(),
[](const Token& token) {
return token.category == kBracket || token.category == kIdentifier;
});
// Build episode title
BuildElement(kElementEpisodeTitle, false, token_begin, token_end);
}
void Parser::SearchForAnimeTitle() {
bool enclosed_title = false;
// Find the first non-enclosed unknown token
auto token_begin = FindToken(tokens_.begin(), tokens_.end(),
kFlagNotEnclosed | kFlagUnknown);
// If that doesn't work, find the first unknown token in the second enclosed
// group, assuming that the first one is the release group
if (token_begin == tokens_.end()) {
enclosed_title = true;
token_begin = tokens_.begin();
bool skipped_previous_group = false;
do {
token_begin = FindToken(token_begin, tokens_.end(), kFlagUnknown);
if (token_begin == tokens_.end())
break;
// Ignore groups that are composed of non-Latin characters
if (IsMostlyLatinString(token_begin->content))
if (skipped_previous_group)
break; // Found it
// Get the first unknown token of the next group
token_begin = FindToken(token_begin, tokens_.end(), kFlagBracket);
token_begin = FindToken(token_begin, tokens_.end(), kFlagUnknown);
skipped_previous_group = true;
} while (token_begin != tokens_.end());
}
if (token_begin == tokens_.end())
return;
// Continue until an identifier (or a bracket, if the title is enclosed)
// is found
auto token_end = FindToken(token_begin, tokens_.end(),
kFlagIdentifier | (enclosed_title ? kFlagBracket : kFlagNone));
// If within the interval there's an open bracket without its matching pair,
// move the upper endpoint back to the bracket
if (!enclosed_title) {
auto last_bracket = token_end;
bool bracket_open = false;
for (auto token = token_begin; token != token_end; ++token) {
if (token->category == kBracket) {
last_bracket = token;
bracket_open = !bracket_open;
}
}
if (bracket_open)
token_end = last_bracket;
}
// If the interval ends with an enclosed group (e.g. "Anime Title [Fansub]"),
// move the upper endpoint back to the beginning of the group. We ignore
// parentheses in order to keep certain groups (e.g. "(TV)") intact.
if (!enclosed_title) {
auto token = FindPreviousToken(tokens_, token_end, kFlagNotDelimiter);
while (CheckTokenCategory(token, kBracket) &&
token->content.front() != ')') {
token = FindPreviousToken(tokens_, token, kFlagBracket);
if (token != tokens_.end()) {
token_end = token;
token = FindPreviousToken(tokens_, token_end, kFlagNotDelimiter);
}
}
}
// Build anime title
BuildElement(kElementAnimeTitle, false, token_begin, token_end);
}