Обработка ошибок в поисковой системе

RobertDeMilo

Nov 5th, 2023

Never

Add comment

Not a member of Pastebin yet? Sign Up, it unlocks many cool features!

C++ 9.80 KB | None | 0 0

raw download clone embed print report

#include <algorithm>
#include <cmath>
#include <iostream>
#include <map>
#include <set>
#include <string>
#include <utility>
#include <vector>
#include <exception>
using namespace std;
const int MAX_RESULT_DOCUMENT_COUNT = 5;
string ReadLine() {
string s;
getline(cin, s);
return s;
}
int ReadLineWithNumber() {
int result;
cin >> result;
ReadLine();
return result;
}
vector<string> SplitIntoWords(const string& text) {
vector<string> words;
string word;
for (const char c : text) {
if (c == ' ') {
if (!word.empty()) {
words.push_back(word);
word.clear();
}
}
else {
word += c;
}
}
if (!word.empty()) {
words.push_back(word);
}
return words;
}
struct Document {
Document() = default;
Document(int id, double relevance, int rating)
: id(id)
, relevance(relevance)
, rating(rating) {
}
int id = 0;
double relevance = 0.0;
int rating = 0;
};
enum class DocumentStatus {
ACTUAL,
IRRELEVANT,
BANNED,
REMOVED,
};
template <typename StringContainer>
set<string> MakeUniqueNonEmptyStrings(const StringContainer& strings)
{
set<string> non_empty_strings;
for (const string& str : strings)
{
if (!str.empty())
{
if (any_of(str.begin(), str.end(), [](const char& c)
{ return c >= 0 && c <= 31; }))
{
throw invalid_argument("special symbol!");
}
non_empty_strings.insert(str);
}
}
return non_empty_strings;
}
class SearchServer {
public:
template <typename StringContainer>
explicit SearchServer(const StringContainer& stop_words) : stop_words_(MakeUniqueNonEmptyStrings(stop_words))
{
}
// Invoke delegating constructor from string container
explicit SearchServer(const string& stop_words_text) : SearchServer(SplitIntoWords(stop_words_text))
{
}
void AddDocument(int document_id, const string& document, DocumentStatus status, const vector<int>& ratings)
{
// negative id
if (document_id < 0)
{
throw invalid_argument("document_id < 0!");
}
// repeated id
if (documents_.count(document_id) != 0)
{
throw invalid_argument("repeated id!");
}
// special symbol
if (any_of(document.begin(), document.end(), [](const char& c)
{ return c >= 0 && c <= 31; }))
{
throw invalid_argument("special symbol!");
}
ids_.push_back(document_id);
const vector<string> words = SplitIntoWordsNoStop(document);
const double inv_word_count = 1.0 / words.size();
for (const string& word : words) {
word_to_document_freqs_[word][document_id] += inv_word_count;
}
documents_.emplace(document_id, DocumentData{ ComputeAverageRating(ratings), status });
}
template <typename DocumentPredicate>
vector<Document> FindTopDocuments(const string& raw_query, DocumentPredicate document_predicate) const
{
if (any_of(raw_query.begin(), raw_query.end(), [](const char& c)
{ return c >= 0 && c <= 31; }))
{
throw invalid_argument("special symbol!");
}
const Query query = ParseQuery(raw_query);
auto matched_documents = FindAllDocuments(query, document_predicate);
sort(matched_documents.begin(), matched_documents.end(),
[](const Document& lhs, const Document& rhs) {
if (abs(lhs.relevance - rhs.relevance) < 1e-6) {
return lhs.rating > rhs.rating;
}
else {
return lhs.relevance > rhs.relevance;
}
});
if (matched_documents.size() > MAX_RESULT_DOCUMENT_COUNT) {
matched_documents.resize(MAX_RESULT_DOCUMENT_COUNT);
}
return matched_documents;
}
vector<Document> FindTopDocuments(const string& raw_query, DocumentStatus status) const {
return FindTopDocuments(
raw_query, [status](int document_id, DocumentStatus document_status, int rating) {
return document_status == status;
});
}
vector<Document> FindTopDocuments(const string& raw_query) const {
return FindTopDocuments(raw_query, DocumentStatus::ACTUAL);
}
int GetDocumentCount() const {
return documents_.size();
}
int GetDocumentId(int index) const
{
/*if (index < documents_.size() && index >= 0)
{
throw out_of_range("out of range");
}*/
return ids_.at(index);
}
tuple<vector<string>, DocumentStatus> MatchDocument(const string& raw_query, int document_id) const
{
if (any_of(raw_query.begin(), raw_query.end(), [](const char& c)
{ return c >= 0 && c <= 31; }))
{
throw invalid_argument("special symbol!");
}
const Query query = ParseQuery(raw_query);
vector<string> matched_words;
for (const string& word : query.plus_words) {
if (word_to_document_freqs_.count(word) == 0) {
continue;
}
if (word_to_document_freqs_.at(word).count(document_id)) {
matched_words.push_back(word);
}
}
for (const string& word : query.minus_words) {
if (word_to_document_freqs_.count(word) == 0) {
continue;
}
if (word_to_document_freqs_.at(word).count(document_id)) {
matched_words.clear();
break;
}
}
return { matched_words, documents_.at(document_id).status };
}
private:
struct DocumentData {
int rating;
DocumentStatus status;
};
const set<string> stop_words_;
vector<int> ids_;
map<string, map<int, double>> word_to_document_freqs_;
map<int, DocumentData> documents_;
bool IsStopWord(const string& word) const {
return stop_words_.count(word) > 0;
}
vector<string> SplitIntoWordsNoStop(const string& text) const {
vector<string> words;
for (const string& word : SplitIntoWords(text)) {
if (!IsStopWord(word)) {
words.push_back(word);
}
}
return words;
}
static int ComputeAverageRating(const vector<int>& ratings) {
if (ratings.empty()) {
return 0;
}
int rating_sum = 0;
for (const int rating : ratings) {
rating_sum += rating;
}
return rating_sum / static_cast<int>(ratings.size());
}
struct QueryWord {
string data;
bool is_minus;
bool is_stop;
};
struct Query {
set<string> plus_words;
set<string> minus_words;
};
//QueryWord ParseQueryWord(string text) const {
// bool is_minus = false;
// // Word shouldn't be empty
// if (text[0] == '-') {
// is_minus = true;
// text = text.substr(1);
// }
// return { text, is_minus, IsStopWord(text) };
//}
//Query ParseQuery(const string& text) const
Query ParseQuery(const string& text) const
{
Query query;
for (const string& word : SplitIntoWordsNoStop(text))
{
if (SplitIntoWordsNoStop(text).back() == "-")
{
throw invalid_argument("last symbol is - ");
}
if (int not_minus_pos = word.find_first_not_of('-'); not_minus_pos >= 2)
{
throw invalid_argument("-- and more");
}
else if (word[0] == '-')
{
query.minus_words.insert(word.substr(1));
}
else
{
query.plus_words.insert(word);
}
}
return query;
}
// Existence required
double ComputeWordInverseDocumentFreq(const string& word) const {
return log(GetDocumentCount() * 1.0 / word_to_document_freqs_.at(word).size());
}
template <typename DocumentPredicate>
vector<Document> FindAllDocuments(const Query& query,
DocumentPredicate document_predicate) const {
map<int, double> document_to_relevance;
for (const string& word : query.plus_words) {
if (word_to_document_freqs_.count(word) == 0) {
continue;
}
const double inverse_document_freq = ComputeWordInverseDocumentFreq(word);
for (const auto [document_id, term_freq] : word_to_document_freqs_.at(word)) {
const auto& document_data = documents_.at(document_id);
if (document_predicate(document_id, document_data.status, document_data.rating)) {
document_to_relevance[document_id] += term_freq * inverse_document_freq;
}
}
}
for (const string& word : query.minus_words) {
if (word_to_document_freqs_.count(word) == 0) {
continue;
}
for (const auto [document_id, _] : word_to_document_freqs_.at(word)) {
document_to_relevance.erase(document_id);
}
}
vector<Document> matched_documents;
for (const auto [document_id, relevance] : document_to_relevance) {
matched_documents.push_back(
{ document_id, relevance, documents_.at(document_id).rating });
}
return matched_documents;
}
};
// ==================== для примера =========================
void PrintDocument(const Document& document) {
cout << "{ "s
<< "document_id = "s << document.id << ", "s
<< "relevance = "s << document.relevance << ", "s
<< "rating = "s << document.rating << " }"s << endl;
}
int main() {
SearchServer search_server("и в на"s);
search_server.AddDocument(0, "белый кот и модный ошейник"s, DocumentStatus::ACTUAL, { 8, -3 });
search_server.AddDocument(1, "пушистый кот пушистый хвост"s, DocumentStatus::ACTUAL, { 7, 2, 7 });
search_server.AddDocument(2, "ухоженный пёс выразительные глаза"s, DocumentStatus::ACTUAL,
{ 5, -12, 2, 1 });
search_server.AddDocument(3, "ухоженный скворец евгений"s, DocumentStatus::BANNED, { 9 });
search_server.GetDocumentId(123);
cout << "ACTUAL by default:"s << endl;
for (const Document& document : search_server.FindTopDocuments("пушистый ухоженный кот"s)) {
PrintDocument(document);
}
cout << "BANNED:"s << endl;
for (const Document& document :
search_server.FindTopDocuments("пушистый ухоженный кот"s, DocumentStatus::BANNED)) {
PrintDocument(document);
}
cout << "Even ids:"s << endl;
for (const Document& document :
search_server.FindTopDocuments("пушистый ухоженный кот"s,
[](int document_id, DocumentStatus status, int rating) {
return document_id % 2 == 0;
})) //
{
PrintDocument(document);
}
return 0;
}

Add Comment

Please, Sign In to add comment