СПРИНТ № 5 | Распределение кода по файлам | Делим проект на файлы

document.cpp

#include "document.h"
using namespace std;

std::ostream& operator<<(std::ostream& out, const Document& document) {
    out << "{ "s
        << "document_id = "s << document.id << ", "s
        << "relevance = "s << document.relevance << ", "s
        << "rating = "s << document.rating << " }"s;
    return out;
}
***************************************************************************************************************************************

main.cpp

#include <iostream>
#include "paginator.h"
#include "read_input_functions.h"
#include "request_queue.h"
#include "search_server.h"



int main() {
    setlocale(0, "");
    SearchServer search_server("and in at"s);
    RequestQueue request_queue(search_server);
    search_server.AddDocument(1, "curly cat curly tail"s, DocumentStatus::ACTUAL, { 7, 2, 7 });
    search_server.AddDocument(2, "curly dog and fancy collar"s, DocumentStatus::ACTUAL, { 1, 2, 3 });
    search_server.AddDocument(3, "big cat fancy collar "s, DocumentStatus::ACTUAL, { 1, 2, 8 });
    search_server.AddDocument(4, "big dog sparrow Eugene"s, DocumentStatus::ACTUAL, { 1, 3, 2 });
    search_server.AddDocument(5, "big dog sparrow Vasiliy"s, DocumentStatus::ACTUAL, { 1, 1, 1 });
    for (int i = 0; i < 1439; ++i) {
        request_queue.AddFindRequest("empty request"s);
    }

    request_queue.AddFindRequest("curly dog"s);

    request_queue.AddFindRequest("big collar"s);

    request_queue.AddFindRequest("sparrow"s);
    cout << "Total empty requests: "s << request_queue.GetNoResultRequests() << endl;
    return 0;
}
***************************************************************************************************************************************

document.h

#pragma once
#include <iostream>

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,
};

std::ostream& operator<<(std::ostream& out, const Document& document);
***************************************************************************************************************************************

paginator.h

#pragma once
#include <algorithm>
#include <vector>
#include <iostream>

template <typename Iterator>
class IteratorRange {
public:
    IteratorRange(Iterator begin, Iterator end)
        : first_(begin)
        , last_(end)
        , size_(distance(first_, last_)) {
    }

    Iterator begin() const {
        return first_;
    }

    Iterator end() const {
        return last_;
    }

    size_t size() const {
        return size_;
    }

private:
    Iterator first_, last_;
    size_t size_;
};

template <typename Iterator>
std::ostream& operator<<(std::ostream& out, const IteratorRange<Iterator>& range) {
    for (Iterator it = range.begin(); it != range.end(); ++it) {
        out << *it;
    }
    return out;
}

template <typename Iterator>
class Paginator {
public:
    Paginator(Iterator begin, Iterator end, size_t page_size) {
        for (size_t left = distance(begin, end); left > 0;) {
            const size_t current_page_size = std::min(page_size, left);
            const Iterator current_page_end = next(begin, current_page_size);
            pages_.push_back({begin, current_page_end});

            left -= current_page_size;
            begin = current_page_end;
        }
    }

    auto begin() const {
        return pages_.begin();
    }

    auto end() const {
        return pages_.end();
    }

    size_t size() const {
        return pages_.size();
    }

private:
    std::vector<IteratorRange<Iterator>> pages_;
};

template <typename Container>
auto Paginate(const Container& c, size_t page_size) {
    return Paginator(begin(c), end(c), page_size);
}
***************************************************************************************************************************************

read_input_functions.cpp

#include "read_input_functions.h"


std::string ReadLine()
{
    std::string s;
    getline(std::cin, s);
    return s;
}

int ReadLineWithNumber()
{
    int result;
    std::cin >> result;
    ReadLine();
    return result;
}

***************************************************************************************************************************************

read_input_functions.h

#pragma once
#include <string>
#include <iostream>

std::string ReadLine();

int ReadLineWithNumber();
***************************************************************************************************************************************

request_queue.cpp

#include "request_queue.h"


RequestQueue::RequestQueue(const SearchServer& search_server) : search_server_(search_server), count_request_(0)
{
}

std::vector<Document> RequestQueue::AddFindRequest(const std::string& raw_query, DocumentStatus status)
{
    const auto result = search_server_.FindTopDocuments(raw_query, status);
    AddRequest(result.size());
    return result;
}

std::vector<Document> RequestQueue::AddFindRequest(const std::string& raw_query)
{
    const auto result = search_server_.FindTopDocuments(raw_query);
    AddRequest(result.size());
    return result;
}

int RequestQueue::GetNoResultRequests() const
{
    return min_in_day_ - count_request_;
}

void RequestQueue::AddRequest(int num_request)
{
    if (num_request != 0)
    {
        ++current_time_;
        ++count_request_;
        requests_.push_back({ num_request, current_time_ });
    }
}
***************************************************************************************************************************************

request_queue.h

#pragma once

#include <string>
#include <vector>
#include <deque>
#include "search_server.h"
#include "document.h"

class RequestQueue {
public:
    explicit RequestQueue(const SearchServer& search_server);

    template <typename DocumentPredicate>
    std::vector<Document> AddFindRequest(const std::string& raw_query, DocumentPredicate document_predicate);
    std::vector<Document> AddFindRequest(const std::string& raw_query, DocumentStatus status);
    std::vector<Document> AddFindRequest(const std::string& raw_query);
    int GetNoResultRequests() const;

private:
    struct QueryResult {
        int req;
        uint64_t timestamp;
    };
    std::deque<QueryResult> requests_;
    const SearchServer& search_server_;
    int count_request_ = 0;
    uint64_t current_time_ = 0;
    const static int min_in_day_ = 1440;


    void AddRequest(int num_request);

};

template<typename DocumentPredicate>
inline std::vector<Document> RequestQueue::AddFindRequest(const std::string& raw_query, DocumentPredicate document_predicate)
{
    const auto result = search_server_.FindTopDocuments(raw_query, document_predicate);
    AddRequest(result.size());
    return result;
}

***************************************************************************************************************************************

search_server.cpp

#include "search_server.h"

SearchServer::SearchServer(const string& stop_words_text) : SearchServer(SplitIntoWords(stop_words_text))  // Invoke delegating constructor
// from string container
{
}

void SearchServer::AddDocument(int document_id, const string& document, DocumentStatus status, const vector<int>& ratings)
{
    if ((document_id < 0) || (documents_.count(document_id) > 0)) {
        throw invalid_argument("Invalid document_id"s);
    }
    const auto 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 });
    document_ids_.push_back(document_id);
}

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

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

int SearchServer::GetDocumentId(int index) const {
    return document_ids_.at(index);
}

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

vector<string> SearchServer::SplitIntoWordsNoStop(const string& text) const {
    vector<string> words;
    for (const string& word : SplitIntoWords(text)) {
        if (!IsValidWord(word)) {
            throw invalid_argument("Word "s + word + " is invalid"s);
        }
        if (!IsStopWord(word)) {
            words.push_back(word);
        }
    }
    return words;
}

int SearchServer::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());
}

SearchServer::QueryWord SearchServer::ParseQueryWord(const string& text) const
{
    if (text.empty()) {
        throw invalid_argument("Query word is empty"s);
    }
    string word = text;
    bool is_minus = false;
    if (word[0] == '-') {
        is_minus = true;
        word = word.substr(1);
    }
    if (word.empty() || word[0] == '-' || !IsValidWord(word)) {
        throw invalid_argument("Query word "s + text + " is invalid");
    }

    return { word, is_minus, IsStopWord(word) };
}


bool SearchServer::IsValidWord(const string& word) {
    // A valid word must not contain special characters
    return none_of(word.begin(), word.end(), [](char c) {
        return c >= '\0' && c < ' ';
        });
}



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

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

search_server.h

#pragma once
#include <algorithm>
#include <map>
#include <vector>
#include <cmath>
#include "read_input_functions.h"
#include "string_processing.h"
#include "document.h"

using namespace std;

const int MAX_RESULT_DOCUMENT_COUNT = 5;

class SearchServer {
public:
    template <typename StringContainer>
    explicit SearchServer(const StringContainer& stop_words);

    explicit SearchServer(const string& stop_words_text);


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

    template <typename DocumentPredicate>
    vector<Document> FindTopDocuments(const string& raw_query, DocumentPredicate document_predicate) const;

    vector<Document> FindTopDocuments(const string& raw_query, DocumentStatus status) const;

    vector<Document> FindTopDocuments(const string& raw_query) const;

    int GetDocumentCount() const;

    int GetDocumentId(int index) const;

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

private:
    struct DocumentData {
        int rating;
        DocumentStatus status;
    };
    const set<string> stop_words_;
    map<string, map<int, double>> word_to_document_freqs_;
    map<int, DocumentData> documents_;
    vector<int> document_ids_;

    bool IsStopWord(const string& word) const {
        return stop_words_.count(word) > 0;
    }

    static bool IsValidWord(const string& word);

    vector<string> SplitIntoWordsNoStop(const string& text) const;

    static int ComputeAverageRating(const vector<int>& ratings);

    struct QueryWord {
        string data;
        bool is_minus;
        bool is_stop;
    };

    QueryWord ParseQueryWord(const string& text) const;

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

    Query ParseQuery(const string& text) const;

    // Existence required
    double ComputeWordInverseDocumentFreq(const string& word) const;

    template <typename DocumentPredicate>
    vector<Document> FindAllDocuments(const Query& query, DocumentPredicate document_predicate) const;
};





template<typename StringContainer>
inline SearchServer::SearchServer(const StringContainer& stop_words)
    : stop_words_(MakeUniqueNonEmptyStrings(stop_words))  // Extract non-empty stop words
{
    if (!all_of(stop_words_.begin(), stop_words_.end(), IsValidWord)) {
        throw invalid_argument("Some of stop words are invalid"s);
    }
}

template<typename DocumentPredicate>
inline vector<Document> SearchServer::FindTopDocuments(const string& raw_query, DocumentPredicate document_predicate) const
{
    const auto 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;
}

template<typename DocumentPredicate>
inline vector<Document> SearchServer::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;
}
***************************************************************************************************************************************

string_processing.cpp

#include "string_processing.h"


std::vector<std::string> SplitIntoWords(const std::string& text)
{
    std::vector<std::string> words;
    std::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;
}
***************************************************************************************************************************************

string_processing.h

#pragma once
#include <string>
#include <vector>
#include <set>

std::vector<std::string> SplitIntoWords(const std::string& text);

template <typename StringContainer>
std::set<std::string> MakeUniqueNonEmptyStrings(const StringContainer& strings) {
    std::set<std::string> non_empty_strings;
    for (const std::string& str : strings) {
        if (!str.empty()) {
            non_empty_strings.insert(str);
        }
    }
    return non_empty_strings;
}