ИТОГОВЫЙ ПРОЕКТ 3 СПРИНТА

#include <algorithm>
#include <cmath>
#include <iostream>
#include <map>
#include <set>
#include <string>
#include <utility>
#include <vector>
#include <cassert>

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 {
    int id;
    double relevance;
    int rating;
};

enum class DocumentStatus {
    ACTUAL,
    IRRELEVANT,
    BANNED,
    REMOVED,
};

class SearchServer {
public:
    void SetStopWords(const string& text) {
        for (const string& word : SplitIntoWords(text)) {
            stop_words_.insert(word);
        }
    }

    void AddDocument(int document_id, const string& document, DocumentStatus status,
        const vector<int>& ratings) {
        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 });
    }

    vector<Document> FindTopDocuments(const string& raw_query, DocumentStatus status) const {
        return FindTopDocuments(raw_query, [status](int document_id, DocumentStatus doc_status, int rating) { return doc_status == status; });
    }


    vector<Document> FindTopDocuments(const string& raw_query) const {
        return FindTopDocuments(raw_query, DocumentStatus::ACTUAL);
    }

    template<typename Predicate>
    vector<Document> FindTopDocuments(const string& raw_query, Predicate predicate) const {
        const Query query = ParseQuery(raw_query);
        auto matched_documents = FindAllDocuments(query, 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;
    }

    int GetDocumentCount() const {
        return documents_.size();
    }

    tuple<vector<string>, DocumentStatus> MatchDocument(const string& raw_query,
        int document_id) const {
        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;
    };

    set<string> stop_words_;
    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;
    };

    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) };
    }

    struct Query {
        set<string> plus_words;
        set<string> minus_words;
    };

    Query ParseQuery(const string& text) const {
        Query query;
        for (const string& word : SplitIntoWords(text)) {
            const QueryWord query_word = ParseQueryWord(word);
            if (!query_word.is_stop) {
                if (query_word.is_minus) {
                    query.minus_words.insert(query_word.data);
                }
                else {
                    query.plus_words.insert(query_word.data);
                }
            }
        }
        return query;
    }

    // Existence required
    double ComputeWordInverseDocumentFreq(const string& word) const {
        return log(GetDocumentCount() * 1.0 / word_to_document_freqs_.at(word).size());
    }

    template<typename Docpredicate>
    vector<Document> FindAllDocuments(const Query& query, Docpredicate 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)) {
                if (predicate(document_id, documents_.at(document_id).status, documents_.at(document_id).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;
    }
};

void PrintDocument(const Document& document) {
    cout << "{ "s
        << "document_id = "s << document.id << ", "s
        << "relevance = "s << document.relevance << ", "s
        << "rating = "s << document.rating
        << " }"s << endl;
}

// Тест проверяет, что поисковая система исключает стоп-слова при добавлении документов
void TestExcludeStopWordsFromAddedDocumentContent() {
    const int doc_id = 42;
    const string content = "cat in the city"s;
    const vector<int> ratings = {1, 2, 3};
    {
        SearchServer server;
        server.AddDocument(doc_id, content, DocumentStatus::ACTUAL, ratings);
        const auto found_docs = server.FindTopDocuments("in"s);
        ASSERT_EQUAL(found_docs.size(), 1u);
        const Document& doc0 = found_docs[0];
        ASSERT_EQUAL(doc0.id, doc_id);
    }

    {
        SearchServer server;
        server.SetStopWords("in the"s);
        server.AddDocument(doc_id, content, DocumentStatus::ACTUAL, ratings);
        ASSERT_HINT(server.FindTopDocuments("in"s).empty(),
                    "Stop words must be excluded from documents"s);
    }
}

void TestAddDocument(){
    const int doc_id = 25;
    const string content = "shaman - i'm russian";
    const vector<int>ratings = {0, 1, 2};
    const vector<int>ratings1 = {1};
    {
        SearchServer server;
        server.AddDocument(doc_id, content, DocumentStatus::ACTUAL, ratings);
        server.AddDocument(1, "FUCK THE POLICE"s, DocumentStatus::ACTUAL, ratings1);
        server.AddDocument(2, "shaman forever"s, DocumentStatus::ACTUAL, {0});
        ASSERT(!server.FindTopDocuments("shaman").empty());
        ASSERT_EQUAL_HINT(server.GetDocumentCount(), 3, "Not found this doc");
    }

}

void TestMinusWords(){
    {
        SearchServer server;
        server.AddDocument(1, "shaman isn't the russian"s, DocumentStatus::ACTUAL, {2,3,1});
        server.AddDocument(2, "killreal this the superman"s, DocumentStatus::ACTUAL, {5,12,3,1,1});
        server.AddDocument(3, "the famous people of great britain", DocumentStatus::ACTUAL, {3,1,2,3,1,1});
        server.AddDocument(4, "lalipop bob dad the", DocumentStatus::ACTUAL, {5});

        assert(server.GetDocumentCount() == 4);
        auto doc_vec = server.FindTopDocuments("-shaman -killreal the"s);
        ASSERT_EQUAL(doc_vec[0].id, 4);
        ASSERT_EQUAL(doc_vec[1].id, 3);
    }

}

void TestMatchDoc(){
    const int doc_id = 1;
    const string content = "just look in my eyes and you'll see russian paradise"s;
    const string content2 = "just russian paradise"s;
    vector<int>rating {1,3,4};
    {
        SearchServer server;
        server.AddDocument(doc_id, content, DocumentStatus::ACTUAL, rating);
        server.AddDocument(2, content2, DocumentStatus::ACTUAL, {1,3,4});
        const string raw_query = "just moment russian";
        const string min_raw_query = "-see";

        auto match_doc = server.MatchDocument(min_raw_query, doc_id);
        const auto [str, doc_st] = match_doc;
        ASSERT(str.empty());

        auto pl_match_doc = server.MatchDocument(raw_query, 2);
        const auto [s, dt] = pl_match_doc;
        ASSERT(!s.empty());
    }
}

void TestRelevance(){
    {
        SearchServer server;
        server.AddDocument(3, "hello world i'm from Pesua"s, DocumentStatus::ACTUAL, {2,3,4});
        server.AddDocument(2, "hello world i'm"s, DocumentStatus::ACTUAL, {2,3,4});
        server.AddDocument(1, "hello world"s, DocumentStatus::ACTUAL, {2,3,4});

        auto find_top = server.FindTopDocuments("hello world i'm from Pesua", DocumentStatus::ACTUAL);

        ASSERT_EQUAL(find_top[0].id, 3);
        ASSERT_EQUAL(find_top[1].id, 2);
        ASSERT_EQUAL(find_top[2].id, 1);
    }
}

void TestByRating(){

    {
        SearchServer server;
        server.AddDocument(3, "hello world i'm from Pesua"s, DocumentStatus::ACTUAL, {2,3,4});
        server.AddDocument(4, "sss psss add"s, DocumentStatus::ACTUAL, {6,1,2,7});
        auto find_top = server.FindTopDocuments("hello world i'm from Pesua sss", DocumentStatus::ACTUAL);

        ASSERT_EQUAL(find_top[0].rating, 3);
        ASSERT_EQUAL(find_top[1].rating, 4);
    }

}

void TestStatus(){

    const int doc_id = 1;
    const string content = "blue cat and red dog";
    vector<int>rating = {1, 1, 2, 3, 4};

    const int doc_id2 = 2;
    const string content2 = "yellow rabbit blue cat matrix";
    vector<int>rating2 = {2, 2, 3, 3, 1};

    const int doc_id3 = 3;
    const string content3 = "slut and river blue cat minner";
    vector<int>rating3 = {2, 0, 1, 2, 3, 1};

    const int doc_id4 = 4;
    const string content4 = "upload blue image";
    vector<int>rating4 = {1, 2, 3};

    const string raw_query = "blue cat";

    {
        SearchServer server;
        server.AddDocument(doc_id, content, DocumentStatus::ACTUAL, rating);
        server.AddDocument(doc_id2, content2, DocumentStatus::BANNED, rating2);
        server.AddDocument(doc_id3, content3, DocumentStatus::IRRELEVANT, rating3);
        server.AddDocument(doc_id4, content4, DocumentStatus::REMOVED, rating4);

        auto doc_act = server.FindTopDocuments(raw_query, DocumentStatus::ACTUAL);
        auto doc_ban = server.FindTopDocuments(raw_query, DocumentStatus::BANNED);
        auto doc_irr = server.FindTopDocuments(raw_query, DocumentStatus::IRRELEVANT);
        auto doc_rem = server.FindTopDocuments(raw_query, DocumentStatus::REMOVED);

        ASSERT_EQUAL(doc_act[0].id, 1);
        ASSERT_EQUAL(doc_ban[0].id, 2);
        ASSERT_EQUAL(doc_irr[0].id, 3);
        ASSERT_EQUAL(doc_rem[0].id, 4);
    }
}

void TestPredicate(){
    {
        SearchServer server;
        server.AddDocument(1, "blue green red bad"s, DocumentStatus::ACTUAL, {0});
        server.AddDocument(2, "sick peek world word"s, DocumentStatus::BANNED, {2});

        server.FindTopDocuments("peek"s);
        server.FindTopDocuments("raw"s, DocumentStatus::BANNED);
        server.FindTopDocuments("sick", [](int document_id, DocumentStatus doc_status, int rating){ return doc_status == DocumentStatus::BANNED;

        });


    }
}

void TestRelevanceNum(){
    SearchServer search_server;
    int document_id = 1;
    vector<int> rating1 = {1, 7, 13};
    DocumentStatus status = DocumentStatus::ACTUAL;
    string document1 = "tabby cat with big eyes"s;
    search_server.AddDocument(document_id, document1, status, rating1);

    document_id = 2;
    vector<int> rating2 = {2, 4, 10};
    string document2 = "small dog and tabby bird"s;
    search_server.AddDocument(document_id, document2, status, rating2);

    vector<Document> document = search_server.FindTopDocuments("cat"s);
    ASSERT_EQUAL(round(document[0].relevance*1000000)/1000000, 0.138629);

}

/*
Разместите код остальных тестов здесь
*/

// Функция TestSearchServer является точкой входа для запуска тестов
void TestSearchServer() {
    RUN_TEST(TestExcludeStopWordsFromAddedDocumentContent);
    RUN_TEST(TestAddDocument);
    RUN_TEST(TestMinusWords);
    RUN_TEST(TestStatus);
    RUN_TEST(TestMatchDoc);
    RUN_TEST(TestRelevance);
    RUN_TEST(TestByRating);
    RUN_TEST(TestPredicate);
    RUN_TEST(TestRelevanceNum);
    // Не забудьте вызывать остальные тесты здесь
}

// --------- Окончание модульных тестов поисковой системы -----------

int main() {
    TestSearchServer();
    // Если вы видите эту строку, значит все тесты прошли успешно
    cout << "Search server testing finished"s << endl;
}