#include "parser.h"

#include <cinttypes>
#include <cstdint>
#include <cstring>
#include <curl/curl.h>
#include <libxml/parser.h>
#include <libxml/tree.h>

#include "charencoding.h"
#include "config.h"
#include "curldatareceiver.h"
#include "curlhandle.h"
#include "curlheadercontainer.h"
#include "exception.h"
#include "logger.h"
#include "remoteapi.h"
#include "rssparser.h"
#include "rssparserfactory.h"
#include "rsspp_uris.h"
#include "strprintf.h"
#include "utils.h"

using namespace newsboat;

namespace rsspp {

Parser::Parser(unsigned int timeout,
	const std::string& user_agent,
	const std::string& proxy,
	const std::string& proxy_auth,
	curl_proxytype proxy_type,
	const bool ssl_verify)
	: to(timeout)
	, ua(user_agent)
	, prx(proxy)
	, prxauth(proxy_auth)
	, prxtype(proxy_type)
	, verify_ssl(ssl_verify)
	, doc(0)
	, lm(0)
{
}

Parser::~Parser()
{
	if (doc) {
		xmlFreeDoc(doc);
	}
}

Feed Parser::parse_url(const std::string& url,
	newsboat::CurlHandle& easyhandle,
	time_t lastmodified,
	const std::string& etag,
	newsboat::RemoteApi* api,
	const std::string& cookie_cache)
{
	CURLcode ret;
	curl_slist* custom_headers{};

	if (!ua.empty()) {
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_USERAGENT, ua.c_str());
	}

	if (api) {
		api->add_custom_headers(&custom_headers);
	}
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_URL, url.c_str());
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_SSL_VERIFYPEER, verify_ssl);
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_NOSIGNAL, 1);
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_FOLLOWLOCATION, 1);
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_MAXREDIRS, 10);
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_FAILONERROR, 1);
	// Accept all of curl's built-in encodings
	curl_easy_setopt(easyhandle.ptr(), CURLOPT_ACCEPT_ENCODING, "");
	if (cookie_cache != "") {
		curl_easy_setopt(
			easyhandle.ptr(), CURLOPT_COOKIEFILE, cookie_cache.c_str());
		curl_easy_setopt(
			easyhandle.ptr(), CURLOPT_COOKIEJAR, cookie_cache.c_str());
	}
	if (to != 0) {
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_TIMEOUT, to);
	}

	if (!prx.empty()) {
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_PROXY, prx.c_str());
	}

	if (!prxauth.empty()) {
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_PROXYAUTH, CURLAUTH_ANY);
		curl_easy_setopt(
			easyhandle.ptr(), CURLOPT_PROXYUSERPWD, prxauth.c_str());
	}

	curl_easy_setopt(easyhandle.ptr(), CURLOPT_PROXYTYPE, prxtype);

	const char* curl_ca_bundle = ::getenv("CURL_CA_BUNDLE");
	if (curl_ca_bundle != nullptr) {
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_CAINFO, curl_ca_bundle);
	}

	auto curlHeaderHandler = CurlHeaderContainer::register_header_handler(easyhandle);
	auto curlDataReceiver = CurlDataReceiver::register_data_handler(easyhandle);

	if (lastmodified != 0) {
		curl_easy_setopt(easyhandle.ptr(),
			CURLOPT_TIMECONDITION,
			CURL_TIMECOND_IFMODSINCE);
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_TIMEVALUE, lastmodified);
	}

	if (etag.length() > 0) {
		auto header = strprintf::fmt("If-None-Match: %s", etag);
		custom_headers =
			curl_slist_append(custom_headers, header.c_str());
	}

	if (lastmodified != 0 || etag.length() > 0) {
		custom_headers =
			curl_slist_append(custom_headers, "A-IM: feed");
	}

	if (custom_headers) {
		curl_easy_setopt(
			easyhandle.ptr(), CURLOPT_HTTPHEADER, custom_headers);
	}

	ret = curl_easy_perform(easyhandle.ptr());

	const auto etag_headers = curlHeaderHandler->get_header_lines("ETag");
	if (etag_headers.size() >= 1) {
		std::string etag = etag_headers.back();
		utils::trim(etag);
		LOG(Level::DEBUG, "parse_url: got etag %s", etag);
		et = etag;
	}

	const auto last_modified_headers = curlHeaderHandler->get_header_lines("Last-Modified");
	if (last_modified_headers.size() >= 1) {
		const std::string header_value = last_modified_headers.back();
		time_t time = curl_getdate(header_value.c_str(), nullptr);
		if (time == -1) {
			LOG(Level::DEBUG, "parse_url: last-modified %s (curl_getdate FAILED)", header_value);
		} else {
			LOG(Level::DEBUG,
				"parse_url: got last-modified %s (%" PRId64 ")",
				header_value,
				// On GCC, `time_t` is `long int`, which is at least 32 bits.
				// On x86_64, it's 64 bits. Thus, this cast is either a no-op,
				// or an up-cast which is always safe.
				static_cast<int64_t>(time));
			lm = time;
		}
	}

	nonstd::optional<std::string> charset_content_type;
	const auto content_type_headers = curlHeaderHandler->get_header_lines("Content-Type");
	if (content_type_headers.size() >= 1) {
		std::string header_value = content_type_headers.back();
		utils::trim(header_value);
		LOG(Level::DEBUG, "parse_url: got content type %s", header_value);
		const auto input = std::vector<uint8_t>(header_value.begin(), header_value.end());
		charset_content_type = charencoding::charset_from_content_type_header(input);
	}

	if (custom_headers) {
		curl_easy_setopt(easyhandle.ptr(), CURLOPT_HTTPHEADER, 0);
		curl_slist_free_all(custom_headers);
	}

	LOG(Level::DEBUG,
		"rsspp::Parser::parse_url: ret = %d (%s)",
		ret,
		curl_easy_strerror(ret));

	long status;
	CURLcode infoOk =
		curl_easy_getinfo(easyhandle.ptr(), CURLINFO_RESPONSE_CODE, &status);

	curl_easy_reset(easyhandle.ptr());
	if (cookie_cache != "") {
		curl_easy_setopt(
			easyhandle.ptr(), CURLOPT_COOKIEJAR, cookie_cache.c_str());
	}

	if (ret != 0) {
		LOG(Level::ERROR,
			"rsspp::Parser::parse_url: curl_easy_perform returned "
			"err "
			"%d: %s",
			ret,
			curl_easy_strerror(ret));
		std::string msg;
		if (ret == CURLE_HTTP_RETURNED_ERROR && infoOk == CURLE_OK) {
			msg = strprintf::fmt(
					"%s %" PRIi64,
					curl_easy_strerror(ret),
					// `status` is `long`, which is at least 32 bits, and on x86_64
					// it's actually 64 bits. Thus casting to `int64_t` is either
					// a no-op, or an up-cast which are always safe.
					static_cast<int64_t>(status));
		} else {
			msg = curl_easy_strerror(ret);
		}
		throw Exception(msg);
	}
	if (infoOk == CURLE_OK && status == 304) {
		throw NotModifiedException();
	}

	std::string buf = curlDataReceiver->get_data();
	LOG(Level::DEBUG,
		"Parser::parse_url: retrieved data for %s: %s",
		url,
		buf);

	const std::vector<std::uint8_t> data(buf.begin(), buf.end());
	const auto charset_xml_declaration = charencoding::charset_from_xml_declaration(data);
	const auto charset_bom = charencoding::charset_from_bom(data);

	nonstd::optional<std::string> charset;
	if (charset_bom.has_value()) {
		charset = charset_bom;
	} else if (charset_xml_declaration.has_value()) {
		charset = charset_xml_declaration;
	} else if (charset_content_type.has_value()) {
		charset = charset_content_type;
	} else {
		buf = utils::string_from_utf8_lossy(data);
		charset = "utf-8";
	}

	if (buf.length() > 0) {
		LOG(Level::DEBUG,
			"Parser::parse_url: handing over data to "
			"parse_buffer()");
		return parse_buffer(buf, url, charset);
	}

	return Feed();
}

Feed Parser::parse_buffer(const std::string& buffer, const std::string& url,
	nonstd::optional<std::string> charset)
{
	if (charset.has_value()) {
		const auto buffer_utf8 = utils::convert_text(buffer, "utf-8", charset.value());
		doc = xmlReadMemory(buffer_utf8.c_str(),
				buffer_utf8.length(),
				url.c_str(),
				"UTF-8",
				XML_PARSE_RECOVER | XML_PARSE_NOERROR | XML_PARSE_NOWARNING | XML_PARSE_IGNORE_ENC);
	} else {
		doc = xmlReadMemory(buffer.c_str(),
				buffer.length(),
				url.c_str(),
				nullptr,
				XML_PARSE_RECOVER | XML_PARSE_NOERROR | XML_PARSE_NOWARNING);
	}
	if (doc == nullptr) {
		throw Exception(_("could not parse buffer"));
	}

	xmlNode* root_element = xmlDocGetRootElement(doc);

	Feed f = parse_xmlnode(root_element);

	if (doc->encoding) {
		f.encoding = (const char*)doc->encoding;
	}

	LOG(Level::INFO, "Parser::parse_buffer: encoding = %s", f.encoding);

	return f;
}

Feed Parser::parse_file(const std::string& filename)
{
	doc = xmlReadFile(filename.c_str(),
			nullptr,
			XML_PARSE_RECOVER | XML_PARSE_NOERROR | XML_PARSE_NOWARNING);
	xmlNode* root_element = xmlDocGetRootElement(doc);

	if (root_element == nullptr) {
		throw Exception(_("could not parse file"));
	}

	Feed f = parse_xmlnode(root_element);

	if (doc->encoding) {
		f.encoding = (const char*)doc->encoding;
	}

	LOG(Level::INFO, "Parser::parse_file: encoding = %s", f.encoding);

	return f;
}

Feed Parser::parse_xmlnode(xmlNode* node)
{
	Feed f;

	if (node) {
		if (node->name && node->type == XML_ELEMENT_NODE) {
			if (strcmp((const char*)node->name, "rss") == 0) {
				const char* version = (const char*)xmlGetProp(
						node, (const xmlChar*)"version");
				if (!version) {
					xmlFree((void*)version);
					throw Exception(_("no RSS version"));
				}
				if (strcmp(version, "0.91") == 0) {
					f.rss_version = Feed::RSS_0_91;
				} else if (strcmp(version, "0.92") == 0) {
					f.rss_version = Feed::RSS_0_92;
				} else if (strcmp(version, "0.94") == 0) {
					f.rss_version = Feed::RSS_0_94;
				} else if (strcmp(version, "2.0") == 0 ||
					strcmp(version, "2") == 0) {
					f.rss_version = Feed::RSS_2_0;
				} else if (strcmp(version, "1.0") == 0) {
					f.rss_version = Feed::RSS_0_91;
				} else {
					xmlFree((void*)version);
					throw Exception(
						_("invalid RSS version"));
				}
				xmlFree((void*)version);
			} else if (strcmp((const char*)node->name, "RDF") ==
				0) {
				f.rss_version = Feed::RSS_1_0;
			} else if (strcmp((const char*)node->name, "feed") ==
				0) {
				if (node->ns && node->ns->href) {
					if (strcmp((const char*)node->ns->href,
							ATOM_0_3_URI) == 0) {
						f.rss_version = Feed::ATOM_0_3;
					} else if (strcmp((const char*)node->ns
							->href,
							ATOM_1_0_URI) == 0) {
						f.rss_version = Feed::ATOM_1_0;
					} else {
						const char* version = (const char*)xmlGetProp(node, (const xmlChar*)"version");
						if (!version) {
							xmlFree((void*)version);
							throw Exception(_(
									"invalid Atom "
									"version"));
						}
						if (strcmp(version, "0.3") ==
							0) {
							xmlFree((void*)version);
							f.rss_version =
								Feed::ATOM_0_3_NONS;
						} else {
							xmlFree((void*)version);
							throw Exception(_(
									"invalid Atom "
									"version"));
						}
					}
				} else {
					throw Exception(_("no Atom version"));
				}
			}

			std::shared_ptr<RssParser> parser =
				RssParserFactory::get_object(f.rss_version, doc);

			try {
				parser->parse_feed(f, node);
			} catch (Exception& e) {
				throw;
			}
		}
	} else {
		throw Exception(_("XML root node is NULL"));
	}

	return f;
}

void Parser::global_init()
{
	LIBXML_TEST_VERSION
	curl_global_init(CURL_GLOBAL_ALL);
}

void Parser::global_cleanup()
{
	xmlCleanupParser();
	curl_global_cleanup();
}

} // namespace rsspp