reader.hpp

Go to the documentation of this file.

#ifndef OSMIUM_IO_READER_HPP
#define OSMIUM_IO_READER_HPP

/*

This file is part of Osmium (https://osmcode.org/libosmium).

Copyright 2013-2021 Jochen Topf <jochen@topf.org> and others (see README).

Boost Software License - Version 1.0 - August 17th, 2003

Permission is hereby granted, free of charge, to any person or organization
obtaining a copy of the software and accompanying documentation covered by
this license (the "Software") to use, reproduce, display, distribute,
execute, and transmit the Software, and to prepare derivative works of the
Software, and to permit third-parties to whom the Software is furnished to
do so, all subject to the following:

The copyright notices in the Software and this entire statement, including
the above license grant, this restriction and the following disclaimer,
must be included in all copies of the Software, in whole or in part, and
all derivative works of the Software, unless such copies or derivative
works are solely in the form of machine-executable object code generated by
a source language processor.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.

*/

#include <osmium/io/compression.hpp>
#include <osmium/io/detail/input_format.hpp>
#include <osmium/io/detail/queue_util.hpp>
#include <osmium/io/detail/read_thread.hpp>
#include <osmium/io/detail/read_write.hpp>
#include <osmium/io/error.hpp>
#include <osmium/io/file.hpp>
#include <osmium/io/header.hpp>
#include <osmium/memory/buffer.hpp>
#include <osmium/osm/entity_bits.hpp>
#include <osmium/thread/pool.hpp>
#include <osmium/thread/util.hpp>
#include <osmium/util/config.hpp>

#include <cerrno>
#include <cstdlib>
#include <fcntl.h>
#include <future>
#include <memory>
#include <string>
#include <system_error>
#include <thread>
#include <utility>

#ifndef _WIN32
# include <sys/wait.h>
#endif

#ifndef _MSC_VER
# include <unistd.h>
#endif

namespace osmium {

    namespace io {

        namespace detail {

            inline std::size_t get_input_queue_size() noexcept {
                return osmium::config::get_max_queue_size("INPUT", 20);
            }

            inline std::size_t get_osmdata_queue_size() noexcept {
                return osmium::config::get_max_queue_size("OSMDATA", 20);
            }

        } // namespace detail

        class Reader {

            // The Reader::read() function reads from a queue of buffers which
            // can contain nested buffers. These nested buffers will be in
            // here, because read() can only return a single unnested buffer.
            osmium::memory::Buffer m_back_buffers{};

            osmium::io::File m_file;

            osmium::thread::Pool* m_pool = nullptr;

            detail::ParserFactory::create_parser_type m_creator;

            enum class status {
                okay   = 0, // normal reading
                error  = 1, // some error occurred while reading
                closed = 2, // close() called
                eof    = 3  // eof of file was reached without error
            } m_status = status::okay;

            int m_childpid = 0;

            detail::future_string_queue_type m_input_queue;

            std::unique_ptr<osmium::io::Decompressor> m_decompressor;

            osmium::io::detail::ReadThreadManager m_read_thread_manager;

            detail::future_buffer_queue_type m_osmdata_queue;
            detail::queue_wrapper<osmium::memory::Buffer> m_osmdata_queue_wrapper;

            std::future<osmium::io::Header> m_header_future{};
            osmium::io::Header m_header{};

            osmium::thread::thread_handler m_thread{};

            std::size_t m_file_size = 0;

            osmium::osm_entity_bits::type m_read_which_entities = osmium::osm_entity_bits::all;
            osmium::io::read_meta m_read_metadata = osmium::io::read_meta::yes;

            void set_option(osmium::thread::Pool& pool) noexcept {
                m_pool = &pool;
            }

            void set_option(osmium::osm_entity_bits::type value) noexcept {
                m_read_which_entities = value;
            }

            void set_option(osmium::io::read_meta value) noexcept {
                m_read_metadata = value;
            }

            // This function will run in a separate thread.
            static void parser_thread(osmium::thread::Pool& pool,
                                      const detail::ParserFactory::create_parser_type& creator,
                                      detail::future_string_queue_type& input_queue,
                                      detail::future_buffer_queue_type& osmdata_queue,
                                      std::promise<osmium::io::Header>&& header_promise,
                                      osmium::osm_entity_bits::type read_which_entities,
                                      osmium::io::read_meta read_metadata) {
                std::promise<osmium::io::Header> promise{std::move(header_promise)};
                osmium::io::detail::parser_arguments args = {
                    pool,
                    input_queue,
                    osmdata_queue,
                    promise,
                    read_which_entities,
                    read_metadata
                };
                creator(args)->parse();
            }

#ifndef _WIN32

            static int execute(const std::string& command, const std::string& filename, int* childpid) {
                int pipefd[2];
                if (pipe(pipefd) < 0) {
                    throw std::system_error{errno, std::system_category(), "opening pipe failed"};
                }
                const pid_t pid = fork();
                if (pid < 0) {
                    throw std::system_error{errno, std::system_category(), "fork failed"};
                }
                if (pid == 0) { // child
                    // close all file descriptors except one end of the pipe
                    for (int i = 0; i < 32; ++i) {
                        if (i != pipefd[1]) {
                            ::close(i);
                        }
                    }
                    if (dup2(pipefd[1], 1) < 0) { // put end of pipe as stdout/stdin
                        exit(1);
                    }

                    ::open("/dev/null", O_RDONLY); // stdin
                    ::open("/dev/null", O_WRONLY); // stderr
                    // hack: -g switches off globbing in curl which allows [] to be used in file names
                    // this is important for XAPI URLs
                    // in theory this execute() function could be used for other commands, but it is
                    // only used for curl at the moment, so this is okay.
                    if (::execlp(command.c_str(), command.c_str(), "-g", filename.c_str(), nullptr) < 0) {
                        exit(1);
                    }
                }
                // parent
                *childpid = pid;
                ::close(pipefd[1]);
                return pipefd[0];
            }
#endif

            static int open_input_file_or_url(const std::string& filename, int* childpid) {
                const std::string protocol{filename.substr(0, filename.find_first_of(':'))};
                if (protocol == "http" || protocol == "https" || protocol == "ftp" || protocol == "file") {
#ifndef _WIN32
                    return execute("curl", filename, childpid);
#else
                    throw io_error{"Reading OSM files from the network currently not supported on Windows."};
#endif
                }
                return osmium::io::detail::open_for_reading(filename);
            }

        public:

            template <typename... TArgs>
            explicit Reader(const osmium::io::File& file, TArgs&&... args) :
                m_file(file.check()),
                m_creator(detail::ParserFactory::instance().get_creator_function(m_file)),
                m_input_queue(detail::get_input_queue_size(), "raw_input"),
                m_decompressor(m_file.buffer() ?
                    osmium::io::CompressionFactory::instance().create_decompressor(file.compression(), m_file.buffer(), m_file.buffer_size()) :
                    osmium::io::CompressionFactory::instance().create_decompressor(file.compression(), open_input_file_or_url(m_file.filename(), &m_childpid))),
                m_read_thread_manager(*m_decompressor, m_input_queue),
                m_osmdata_queue(detail::get_osmdata_queue_size(), "parser_results"),
                m_osmdata_queue_wrapper(m_osmdata_queue),
                m_file_size(m_decompressor->file_size()) {

                (void)std::initializer_list<int>{
                    (set_option(args), 0)...
                };

                if (!m_pool) {
                    m_pool = &thread::Pool::default_instance();
                }

                std::promise<osmium::io::Header> header_promise;
                m_header_future = header_promise.get_future();
                m_thread = osmium::thread::thread_handler{parser_thread, std::ref(*m_pool), std::ref(m_creator), std::ref(m_input_queue), std::ref(m_osmdata_queue), std::move(header_promise), m_read_which_entities, m_read_metadata};
            }

            template <typename... TArgs>
            explicit Reader(const std::string& filename, TArgs&&... args) :
                Reader(osmium::io::File(filename), std::forward<TArgs>(args)...) {
            }

            template <typename... TArgs>
            explicit Reader(const char* filename, TArgs&&... args) :
                Reader(osmium::io::File(filename), std::forward<TArgs>(args)...) {
            }

            Reader(const Reader&) = delete;
            Reader& operator=(const Reader&) = delete;

            Reader(Reader&&) = delete;
            Reader& operator=(Reader&&) = delete;

            ~Reader() noexcept {
                try {
                    close();
                } catch (...) {
                    // Ignore any exceptions because destructor must not throw.
                }
            }

            void close() {
                m_status = status::closed;

                m_read_thread_manager.stop();

                m_osmdata_queue_wrapper.drain();

                try {
                    m_read_thread_manager.close();
                } catch (...) {
                    // Ignore any exceptions.
                }

#ifndef _WIN32
                if (m_childpid) {
                    int status = 0;
                    const pid_t pid = ::waitpid(m_childpid, &status, 0);
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wold-style-cast"
                    if (pid < 0 || !WIFEXITED(status) || WEXITSTATUS(status) != 0) { // NOLINT(hicpp-signed-bitwise)
                        throw std::system_error{errno, std::system_category(), "subprocess returned error"};
                    }
#pragma GCC diagnostic pop
                    m_childpid = 0;
                }
#endif
            }

            osmium::io::Header header() {
                if (m_status == status::error) {
                    throw io_error{"Can not get header from reader when in status 'error'"};
                }

                try {
                    if (m_header_future.valid()) {
                        m_header = m_header_future.get();
                    }
                } catch (...) {
                    close();
                    m_status = status::error;
                    throw;
                }

                return m_header;
            }

            osmium::memory::Buffer read() {
                osmium::memory::Buffer buffer;

                // If there are buffers on the stack, return those first.
                if (m_back_buffers) {
                    if (m_back_buffers.has_nested_buffers()) {
                        buffer = std::move(*m_back_buffers.get_last_nested());
                    } else {
                        buffer = std::move(m_back_buffers);
                        m_back_buffers = osmium::memory::Buffer{};
                    }
                    return buffer;
                }

                if (m_status != status::okay) {
                    throw io_error{"Can not read from reader when in status 'closed', 'eof', or 'error'"};
                }

                if (m_read_which_entities == osmium::osm_entity_bits::nothing) {
                    m_status = status::eof;
                    return buffer;
                }

                try {
                    // m_input_format.read() can return an invalid buffer to signal EOF,
                    // or a valid buffer with or without data. A valid buffer
                    // without data is not an error, it just means we have to
                    // keep getting the next buffer until there is one with data.
                    while (true) {
                        buffer = m_osmdata_queue_wrapper.pop();
                        if (detail::at_end_of_data(buffer)) {
                            m_status = status::eof;
                            m_read_thread_manager.close();
                            return buffer;
                        }
                        if (buffer.has_nested_buffers()) {
                            m_back_buffers = std::move(buffer);
                            buffer = std::move(*m_back_buffers.get_last_nested());
                        }
                        if (buffer.committed() > 0) {
                            return buffer;
                        }
                    }
                } catch (...) {
                    close();
                    m_status = status::error;
                    throw;
                }
            }

            bool eof() const {
                return m_status == status::eof || m_status == status::closed;
            }

            std::size_t file_size() const noexcept {
                return m_file_size;
            }

            std::size_t offset() const noexcept {
                return m_decompressor->offset();
            }

        }; // class Reader

        template <typename... TArgs>
        osmium::memory::Buffer read_file(TArgs&&... args) {
            osmium::memory::Buffer buffer{1024 * 1024, osmium::memory::Buffer::auto_grow::yes};

            Reader reader{std::forward<TArgs>(args)...};
            while (auto read_buffer = reader.read()) {
                buffer.add_buffer(read_buffer);
                buffer.commit();
            }

            return buffer;
        }

    } // namespace io

} // namespace osmium

#endif // OSMIUM_IO_READER_HPP