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);
}
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) {
Token& token = tokens_.at(i);
if (token.category == kUnknown)
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
tokens.erase(std::remove_if(tokens.begin(), tokens.end(), std::bind1st(std::mem_fun(&Parser::not_numeric_string), this)),
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);
}
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);
}