Use CURLU* for URLs

client.cpp

@@ -4,6 +4,7 @@
 
 #include "client.h"
 #include "models.h"
+#include "curlu_wrapper.h"
 #include "hiredis_wrapper.h"
 
 MastodonClient mastodon_client;
@@ -85,8 +86,13 @@ std::optional<Account> MastodonClient::get_account_by_username(std::string host,
         username.erase(username.size() - host.size() - 1);
     }
 
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/accounts/lookup");
+    url.set(CURLUPART_QUERY, "acct="s + url_encode(username));
     try {
-        Account account = this->_send_request("coyote:"s + host + ":@" + username, "https://"s + host + "/api/v1/accounts/lookup?acct=" + url_encode(username));
+        Account account = this->_send_request("coyote:"s + host + ":@" + username, url);
         account.same_server = host == account.server;
         return account;
     } catch (const MastodonException& e) {
@@ -102,7 +108,12 @@ std::vector<Post> MastodonClient::get_pinned_posts(std::string host, const std::
     using namespace std::string_literals;
     lowercase(host);
 
-    std::vector<Post> posts = this->_send_request("coyote:"s + host + ':' + account_id + ":pinned", "https://"s + host + "/api/v1/accounts/" + account_id + "/statuses?pinned=true");
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/accounts/"s + url_encode(account_id) + "/statuses");
+    url.set(CURLUPART_QUERY, "pinned=true");
+    std::vector<Post> posts = this->_send_request("coyote:"s + host + ':' + account_id + ":pinned", url);
 
     for (Post& post : posts) {
         handle_post_server(post, host);
@@ -113,22 +124,17 @@ std::vector<Post> MastodonClient::get_pinned_posts(std::string host, const std::
 
 std::vector<Post> MastodonClient::get_posts(const std::string& host, const std::string& account_id, PostSortingMethod sorting_method, std::optional<std::string> max_id) {
     using namespace std::string_literals;
-
     const char* sorting_parameters[3] = {"exclude_replies=true", "", "only_media=true"};
-    std::string query = sorting_parameters[sorting_method];
+
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/accounts/"s + url_encode(account_id) + "/statuses");
+    url.set(CURLUPART_QUERY, sorting_parameters[sorting_method]);
     if (max_id) {
-        if (!query.empty()) {
+        url.set(CURLUPART_QUERY, "max_id="s + std::move(*max_id), CURLU_URLENCODE | CURLU_APPENDQUERY);
-            query += '&';
-        }
-        query += "max_id=";
-        query += url_encode(std::move(*max_id));
     }
 
-    std::string url = "https://"s + host + "/api/v1/accounts/" + account_id + "/statuses";
-    if (!query.empty()) {
-        url += '?';
-        url += query;
-    }
     std::vector<Post> posts = this->_send_request(std::nullopt, url);
 
     for (Post& post : posts) {
@@ -141,8 +147,12 @@ std::vector<Post> MastodonClient::get_posts(const std::string& host, const std::
 std::optional<Post> MastodonClient::get_post(const std::string& host, std::string id) {
     using namespace std::string_literals;
 
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/statuses/"s + url_encode(std::move(id)));
     try {
-        Post post = this->_send_request(std::nullopt, "https://"s + host + "/api/v1/statuses/" + url_encode(std::move(id)));
+        Post post = this->_send_request(std::nullopt, url);
         handle_post_server(post, host);
         return post;
     } catch (const MastodonException& e) {
@@ -157,7 +167,11 @@ std::optional<Post> MastodonClient::get_post(const std::string& host, std::strin
 PostContext MastodonClient::get_post_context(const std::string& host, std::string id) {
     using namespace std::string_literals;
 
-    PostContext context = this->_send_request(std::nullopt, "https://"s + host + "/api/v1/statuses/" + url_encode(std::move(id)) + "/context");
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/statuses/"s + url_encode(std::move(id)) + "/context");
+    PostContext context = this->_send_request(std::nullopt, url);
 
     for (Post& post : context.ancestors) {
         handle_post_server(post, host);
@@ -172,10 +186,12 @@ PostContext MastodonClient::get_post_context(const std::string& host, std::strin
 std::vector<Post> MastodonClient::get_tag_timeline(const std::string& host, const std::string& tag, std::optional<std::string> max_id) {
     using namespace std::string_literals;
 
-    std::string url = "https://"s + host + "/api/v1/timelines/tag/" + url_encode(tag);
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/timelines/tag/"s + url_encode(tag));
     if (max_id) {
-        url += "?max_id=";
+        url.set(CURLUPART_QUERY, "max_id="s + std::move(*max_id), CURLU_URLENCODE | CURLU_APPENDQUERY);
-        url += url_encode(std::move(*max_id));
     }
 
     std::vector<Post> posts = this->_send_request(std::nullopt, url);
@@ -189,7 +205,11 @@ Instance MastodonClient::get_instance(std::string host) {
     using namespace std::string_literals;
     lowercase(host);
 
-    Instance instance = this->_send_request("coyote:"s + host + ":instance", "https://"s + host + "/api/v2/instance");
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v2/instance");
+    Instance instance = this->_send_request("coyote:"s + host + ":instance", url);
     instance.contact_account.same_server = instance.contact_account.server == host;
     return instance;
 }
@@ -198,7 +218,11 @@ blankie::html::HTMLString MastodonClient::get_extended_description(std::string h
     using namespace std::string_literals;
     lowercase(host);
 
-    nlohmann::json j = this->_send_request("coyote:"s + host + ":desc", "https://"s + host + "/api/v1/instance/extended_description");
+    CurlUrl url;
+    url.set(CURLUPART_SCHEME, "https");
+    url.set(CURLUPART_HOST, host);
+    url.set(CURLUPART_PATH, "/api/v1/instance/extended_description");
+    nlohmann::json j = this->_send_request("coyote:"s + host + ":desc", url);
     return blankie::html::HTMLString(j.at("content").get<std::string>());
 }
 
@@ -234,7 +258,7 @@ CURL* MastodonClient::_get_easy() {
     return curl;
 }
 
-nlohmann::json MastodonClient::_send_request(std::optional<std::string> cache_key, const std::string& url) {
+nlohmann::json MastodonClient::_send_request(std::optional<std::string> cache_key, const CurlUrl& url) {
     std::optional<std::string> cached;
     if (redis && cache_key && (cached = redis->get(*cache_key))) {
         return nlohmann::json::parse(std::move(*cached));
@@ -243,10 +267,11 @@ nlohmann::json MastodonClient::_send_request(std::optional<std::string> cache_ke
     std::string res;
     CURL* curl = this->_get_easy();
 
-    setopt(curl, CURLOPT_URL, url.c_str());
+    setopt(curl, CURLOPT_CURLU, url.get());
     setopt(curl, CURLOPT_WRITEFUNCTION, curl_write_cb);
     setopt(curl, CURLOPT_WRITEDATA, &res);
     CURLcode code = curl_easy_perform(curl);
+    setopt(curl, CURLOPT_CURLU, nullptr);
     if (code) {
         throw CurlException(code);
     }

client.h

@@ -8,6 +8,7 @@
 #include <nlohmann/json.hpp>
 
 #include "models.h"
+class CurlUrl; // forward declaration from curlu_wrapper.h
 
 class CurlException : public std::exception {
 public:
@@ -78,7 +79,7 @@ public:
 
 private:
     CURL* _get_easy();
-    nlohmann::json _send_request(std::optional<std::string> cache_key, const std::string& url);
+    nlohmann::json _send_request(std::optional<std::string> cache_key, const CurlUrl& url);
     long _response_status_code();
 
     std::mutex _share_locks[CURL_LOCK_DATA_LAST];

curlu_wrapper.h

@@ -0,0 +1,77 @@
+#pragma once
+
+#include <cstdio>
+#include <string>
+#include <memory>
+#include <exception>
+#include <stdexcept>
+#include <curl/curl.h>
+
+class CurlUrlException : public std::exception {
+public:
+    CurlUrlException(CURLUcode code_) : code(code_) {
+#if !CURL_AT_LEAST_VERSION(7, 80, 0)
+        snprintf(this->_id_buf, 64, "curl url error %d", this->code);
+#endif
+    }
+
+    const char* what() const noexcept {
+#if CURL_AT_LEAST_VERSION(7, 80, 0)
+        return curl_url_strerror(this->code);
+#else
+        return this->_id_buf;
+#endif
+    }
+
+    CURLUcode code;
+
+private:
+#if !CURL_AT_LEAST_VERSION(7, 80, 0)
+    char _id_buf[64];
+#endif
+};
+
+using CurlStr = std::unique_ptr<char, decltype(&curl_free)>;
+class CurlUrl {
+public:
+    CurlUrl(const CurlUrl&) = delete;
+    CurlUrl& operator=(const CurlUrl&) = delete;
+
+    CurlUrl() {
+        this->_ptr = curl_url();
+        if (!this->_ptr) {
+            throw std::bad_alloc();
+        }
+    }
+    ~CurlUrl() {
+        curl_url_cleanup(this->_ptr);
+    }
+
+    constexpr CURLU* get() const noexcept {
+        return this->_ptr;
+    }
+
+    CurlStr get(CURLUPart part, unsigned int flags = 0) const {
+        char* content;
+        CURLUcode code = curl_url_get(this->_ptr, part, &content, flags);
+        if (code) {
+            throw CurlUrlException(code);
+        }
+
+        return CurlStr(content, curl_free);
+    }
+
+    void set(CURLUPart part, const char* content, unsigned int flags = 0) {
+        CURLUcode code = curl_url_set(this->_ptr, part, content, flags);
+        if (code) {
+            throw CurlUrlException(code);
+        }
+    }
+
+    void set(CURLUPart part, const std::string& content, unsigned int flags = 0) {
+        this->set(part, content.c_str(), flags);
+    }
+
+private:
+    CURLU* _ptr;
+};

servehelper.cpp

@@ -10,6 +10,7 @@
 #include "timeutils.h"
 #include "servehelper.h"
 #include "lxb_wrapper.h"
+#include "curlu_wrapper.h"
 #include "routes/routes.h"
 #include "blankie/escape.h"
 
@@ -27,30 +28,6 @@ static Element serialize_post(const httplib::Request& req, const std::string& se
 static inline Element serialize_media(const Media& media);
 static inline Element serialize_poll(const httplib::Request& req, const Poll& poll);
 
-class CurlUrlException : public std::exception {
-public:
-    CurlUrlException(CURLUcode code_) : code(code_) {
-#if !CURL_AT_LEAST_VERSION(7, 80, 0)
-        snprintf(this->_id_buf, 64, "curl url error %d", this->code);
-#endif
-    }
-
-    const char* what() const noexcept {
-#if CURL_AT_LEAST_VERSION(7, 80, 0)
-        return curl_url_strerror(this->code);
-#else
-        return this->_id_buf;
-#endif
-    }
-
-    CURLUcode code;
-
-private:
-#if !CURL_AT_LEAST_VERSION(7, 80, 0)
-    char _id_buf[64];
-#endif
-};
-
 
 void serve(const httplib::Request& req, httplib::Response& res, std::string title, Element element, Nodes extra_head) {
     using namespace std::string_literals;
@@ -118,6 +95,21 @@ void serve_redirect(const httplib::Request& req, httplib::Response& res, std::st
 
 
 
+bool starts_with(const CurlUrl& url, const CurlUrl& base) {
+    if (strcmp(url.get(CURLUPART_SCHEME).get(), base.get(CURLUPART_SCHEME).get()) != 0) {
+        return false;
+    }
+    if (strcmp(url.get(CURLUPART_HOST).get(), base.get(CURLUPART_HOST).get()) != 0) {
+        return false;
+    }
+
+    CurlStr url_path = url.get(CURLUPART_PATH);
+    CurlStr base_path = base.get(CURLUPART_PATH);
+    size_t base_path_len = strlen(base_path.get());
+    return memcpy(url_path.get(), base_path.get(), base_path_len) == 0
+        && (url_path.get()[base_path_len] == '/' || url_path.get()[base_path_len] == '\0');
+}
+
 std::string get_origin(const httplib::Request& req) {
     if (req.has_header("X-Canonical-Origin")) {
         return req.get_header_value("X-Canonical-Origin");
@@ -139,43 +131,28 @@ std::string get_origin(const httplib::Request& req) {
 }
 
 std::string proxy_mastodon_url(const httplib::Request& req, const std::string& url_str) {
-    using CurlStr = std::unique_ptr<char, decltype(&curl_free)>;
+    CurlUrl url;
+    url.set(CURLUPART_URL, url_str.c_str());
 
-    std::unique_ptr<CURLU, decltype(&curl_url_cleanup)> url(curl_url(), curl_url_cleanup);
+    std::string new_url = get_origin(req) + '/' + url.get(CURLUPART_HOST).get() + url.get(CURLUPART_PATH).get();
-    if (!url) {
-        throw std::bad_alloc();
-    }
 
-    // in a block to avoid a (potential) gcc bug where it thinks the lambda below
+    try {
-    // shadows `code`, even if i do `[&url](...) { ... }`
+        CurlStr query = url.get(CURLUPART_QUERY);
-    {
-        CURLUcode code = curl_url_set(url.get(), CURLUPART_URL, url_str.c_str(), 0);
-        if (code) {
-            throw CurlUrlException(code);
-        }
-    }
-
-    auto get_part = [&](CURLUPart part, CURLUcode ignore = CURLUE_OK) {
-        char* content = nullptr;
-        CURLUcode code = curl_url_get(url.get(), part, &content, 0);
-        if (code && code != ignore) {
-            throw CurlUrlException(code);
-        }
-        return CurlStr(content, curl_free);
-    };
-    CurlStr host = get_part(CURLUPART_HOST);
-    CurlStr path = get_part(CURLUPART_PATH);
-    CurlStr query = get_part(CURLUPART_QUERY, CURLUE_NO_QUERY);
-    CurlStr fragment = get_part(CURLUPART_FRAGMENT, CURLUE_NO_FRAGMENT);
-
-    std::string new_url = get_origin(req) + '/' + host.get() + path.get();
-    if (query) {
         new_url += '?';
         new_url += query.get();
+    } catch (const CurlUrlException& e) {
+        if (e.code != CURLUE_NO_QUERY) {
+            throw;
         }
-    if (fragment) {
+    }
+    try {
+        CurlStr fragment = url.get(CURLUPART_FRAGMENT);
         new_url += '#';
         new_url += fragment.get();
+    } catch (const CurlUrlException& e) {
+        if (e.code != CURLUE_NO_FRAGMENT) {
+            throw;
+        }
     }
     return new_url;
 }
@@ -263,10 +240,14 @@ static inline void preprocess_link(const httplib::Request& req, const std::strin
     const lxb_char_t* cls_c = lxb_dom_element_class(element, &cls_c_len);
     std::string cls = cls_c ? std::string(reinterpret_cast<const char*>(cls_c), cls_c_len) : "";
 
-    std::string instance_url_base = "https://"s + domain_name;
+    CurlUrl href_url;
+    href_url.set(CURLUPART_URL, href);
+    CurlUrl instance_url_base;
+    instance_url_base.set(CURLUPART_SCHEME, "https");
+    instance_url_base.set(CURLUPART_HOST, domain_name);
     // .mention is used in note and posts
     // Instance base is used for link fields
-    if (std::regex_search(cls, mention_class_re) || href.starts_with(instance_url_base + '/') || href == instance_url_base) {
+    if (std::regex_search(cls, mention_class_re) || starts_with(href_url, instance_url_base)) {
         // Proxy this instance's URLs to Coyote
         href = proxy_mastodon_url(req, std::move(href));
 

servehelper.h

@@ -6,6 +6,7 @@
 #include "blankie/serializer.h"
 struct Post; // forward declaration from models.h
 struct Emoji; // forward declaration from models.h
+class CurlUrl; // forward declaration from curlu_wrapper.h
 
 using Element = blankie::html::Element;
 using Node = blankie::html::Node;
@@ -16,6 +17,7 @@ void serve_error(const httplib::Request& req, httplib::Response& res,
         std::string title, std::optional<std::string> subtitle = std::nullopt, std::optional<std::string> info = std::nullopt);
 void serve_redirect(const httplib::Request& req, httplib::Response& res, std::string url, bool permanent = false);
 
+bool starts_with(const CurlUrl& url, const CurlUrl& base);
 std::string get_origin(const httplib::Request& req);
 std::string proxy_mastodon_url(const httplib::Request& req, const std::string& url_str);
 bool should_send_304(const httplib::Request& req, uint64_t hash);