patom/main.cpp

#include <cstdlib>
#include <stdio.h>
#include <curl/curl.h>
#include <nlohmann/json.hpp>

#define MAX_FILE_SIZE 1024 * 1024 * 1024
#define set_curl_opt_or_die(curl, option, param) do { \
    CURLcode curl_code = curl_easy_setopt(curl, option, param); \
    if (curl_code != CURLE_OK) { \
        curl_easy_cleanup(curl); \
        fprintf(stderr, "curl_easy_setopt returned %s\n", curl_easy_strerror(curl_code)); \
        curl_global_cleanup(); \
        exit(1); \
    } \
} while(0)

size_t write_memory_cb(char* contents, size_t size, size_t nmemb, std::string* str) {
    size_t realsize = size * nmemb;
    if (str->length() + realsize > MAX_FILE_SIZE) {
        fprintf(stderr, "Max file size reached\n");
        return 0;
    }
    str->append(contents, realsize);
    return realsize;
}

// behold: the why
// oh and this is stolen from https://stackoverflow.com/a/3418285
void escape_xml(std::string* str) {
    if (str->empty()) {
        return;
    }
    size_t start = 0;
    while ((start = str->find("&", start)) != std::string::npos) {
        str->replace(start, 1, "&amp;");
        start += 5;
    }
    start = 0;
    while ((start = str->find("<", start)) != std::string::npos) {
        str->replace(start, 1, "&lt;");
        start += 4;
    }
    start = 0;
    while ((start = str->find(">", start)) != std::string::npos) {
        str->replace(start, 1, "&gt;");
        start += 4;
    }
    start = 0;
    while ((start = str->find("\"", start)) != std::string::npos) {
        str->replace(start, 1, "&quot;");
        start += 6;
    }
    start = 0;
    while ((start = str->find("'", start)) != std::string::npos) {
        str->replace(start, 1, "&#x27;");
        start += 6;
    }
}

int main(int argc, char* argv[]) {
    if (argc != 2) {
        if (argc > 0) {
            fprintf(stderr, "Usage: %s <pixiv user id>\n", argv[0]);
        } else {
            fprintf(stderr, "Usage: patom <pixiv user id>\n");
        }
        return 1;
    }
    char* pixiv_user_id = argv[1];
    if (!*pixiv_user_id) {
        fprintf(stderr, "User ID is empty\n");
        return 1;
    }
    for (int i=0; pixiv_user_id[i]; i++) {
        if (!(pixiv_user_id[i] >= '0' && pixiv_user_id[i] <= '9')) {
            fprintf(stderr, "User ID is not numeric\n");
            return 1;
        }
    }
    curl_global_init(CURL_GLOBAL_ALL);
    CURL* curl = curl_easy_init();
    if (!curl) {
        fprintf(stderr, "curl_easy_init returned NULL\n");
        curl_global_cleanup();
        return 1;
    }
    set_curl_opt_or_die(curl, CURLOPT_PROTOCOLS, CURLPROTO_HTTPS);
    std::string url = "https://www.pixiv.net/touch/ajax/user/illusts?type=illust&lang=en&id=";
    url.append(argv[1]);
    set_curl_opt_or_die(curl, CURLOPT_URL, url.c_str());
    std::string response_body;
    set_curl_opt_or_die(curl, CURLOPT_WRITEFUNCTION, write_memory_cb);
    set_curl_opt_or_die(curl, CURLOPT_WRITEDATA, &response_body);
    CURLcode curl_code = curl_easy_perform(curl);
    curl_easy_cleanup(curl);
    if (curl_code != CURLE_OK) {
        fprintf(stderr, "curl_easy_perform returned %s\n", curl_easy_strerror(curl_code));
        curl_global_cleanup();
        return 1;
    }
    curl_global_cleanup();
    nlohmann::json j = nlohmann::json::parse(response_body);

    if (j["error"].get<bool>()) {
        fprintf(stderr, "Error returned by the API: %s\n", j["message"].get<std::string>().c_str());
        return 1;
    }
    std::vector<nlohmann::json> illusts = j["body"]["illusts"].get<std::vector<nlohmann::json>>();
    time_t last_updated_timestamp;
    if (illusts.size()) {
        last_updated_timestamp = illusts[0]["upload_timestamp"].get<time_t>();
    } else {
        last_updated_timestamp = time(NULL);
    }
    std::string ogp_title = j["body"]["meta"]["ogp"]["title"].get<std::string>();
    escape_xml(&ogp_title);
    char tm_str[100] = "";
    strftime(tm_str, 100, "%FT%TZ", gmtime(&last_updated_timestamp));
    // wonder if pixiv is a urn namespace
    // also cba to make another request to find the username so here we are
    printf("<?xml version=\"1.0\"?>\n<feed xmlns=\"http://www.w3.org/2005/Atom\">\n    <id>urn:pixiv:user:%s</id>\n    <author><name>%s</name></author>\n    <updated>%s</updated>\n    <title type=\"text\">%s</title>\n", argv[1], argv[1], tm_str, ogp_title.c_str());
    for (size_t i=0; i < illusts.size(); i++) {
        nlohmann::json illust = illusts[i];
        std::string title = illust["alt"].get<std::string>();
        std::string illust_id = illust["id"].get<std::string>();
        // istg if 0 isn't 1970
        time_t upload_timestamp = illust["upload_timestamp"].get<time_t>();
        strftime(tm_str, 100, "%FT%TZ", gmtime(&upload_timestamp));
        escape_xml(&title);
        printf("    <entry>\n        <id>urn:pixiv:illust:%s</id>\n        <title type=\"text\">%s</title>\n        <updated>%s</updated>\n        <link rel=\"alternate\" type=\"text/html\" href=\"https://www.pixiv.net/en/artworks/%s\"></link>\n    </entry>\n", illust_id.c_str(), title.c_str(), tm_str, illust_id.c_str());
    }
    printf("</feed>\n");
    return 0;
}