introduction

libtorrent is a C++ library that aims to be a good alternative to all the other bittorrent implementations around. It is a library and not a full featured client, although it comes with a working example client.

The main goals of libtorrent are:

• to be cpu efficient
• to be memory efficient
• to be very easy to use

libtorrent is not finished. It is an ongoing project (including this documentation). The current state includes the following features:

• multitracker extension support (as described by TheShadow)
• serves multiple torrents on a single port and a single thread
• supports http proxies and proxy authentication
• gzipped tracker-responses
• piece picking on block-level (as opposed to piece-level) like in Azureus
• queues torrents for file check, instead of checking all of them in parallel.
• uses separate threads for checking files and for main downloader, with a fool-proof thread-safe library interface. (i.e. There's no way for the user to cause a deadlock).
• can limit the upload and download bandwidth usage and the maximum number of unchoked peers
• piece-wise, unordered, file allocation
• implements fair trade. User settable trade-ratio, must at least be 1:1, but one can choose to trade 1 for 2 or any other ratio that isn't unfair to the other party. (i.e. real tit for tat)
• fast resume support, a way to get rid of the costly piece check at the start of a resumed torrent. Saves the storage state, piece_picker state as well as all local peers in a separate fast-resume file.
• supports the extension protocol described by Nolar. See extensions.
• supports files > 2 gigabytes.
• supports the no_peer_id=1 extension that will ease the load off trackers.
• supports the udp-tracker protocol by Olaf van der Spek.
• possibility to limit the number of connections.
• delays have messages if there's no other outgoing traffic to the peer, and doesn't send have messages to peers that already has the piece. This saves bandwidth.
• does not have any requirements on the piece order in a torrent that it resumes. This means it can resume a torrent downloaded by any client.
• adjusts the length of the request queue depending on download rate.
• supports the compact=1 tracker parameter.

Functions that are yet to be implemented:

• large file support on linux and Mac OS X.
• better identification of peers that send bad data
• ip-filters
• file-level priority

libtorrent is portable at least among windows, macosx, and UNIX-systems. It uses boost.thread, boost.filesystem, boost.date_time and various other boost libraries as well as zlib.

libtorrent has been successfully compiled and tested on:

• Windows 2000 vc7.1
• Linux x86 (debian) GCC 3.0.4, GCC 3.2.3
• Windows 2000, msvc6 sp5 (does not support 64-bit values due to problems with operator<<(ostream&, __int64))
• MacOS X, GCC 3.3

Fails on:

• GCC 2.95.4 (std::ios_base is missing)
• Cygwin GCC 3.3.1 (builds but crashes)

libtorrent is released under the BSD-license.

downloading and building

To acquire the latest version of libtorrent, you'll have to grab it from CVS. You'll find instructions on how to do this here (see Anonymous CVS access).

The easiest way to build libtorrent is probably to use boost-build. Make sure you install it correctly by setting the environment variable BOOST_BUILD_PATH and modifying the user_config.jam to reflect the toolsets you have installed.

You also need to install boost 1.31.0.

Before you invoke bjam you have to set the environment variable BOOST_ROOT to the path where you installed boost. This will be used to build and link against the required boost libraries as well as be used as include path for boost headers.

To build you just have to run:

bjam <toolset> link=static

in the libtorrent directory.

If you're building on a platform where dlls share the same heap, you can build libtorrent as a dll too, by typing link=shared instead of link=static.

To build on MacOS X, you need the latest version of boost-build, from the boost cvs.

If you're making your own project file, note that there are two versions of the file abstraction. There's one file_win.cpp which relies on windows file API that supports files larger than 2 Gigabytes. This does not work in vc6 for some reason, possibly because it may require windows NT and above. The other file, file.cpp is the default implementation that simply relies on the standard library's fstream, and as a result does not support files larger than 2 Gigabytes.

using

The interface of libtorrent consists of a few classes. The main class is the session, it contains the main loop that serves all torrents.

The basic usage is as follows:

• conststruct a session

• parse .torrent-files and add them to the session

• main loop

  • query the torrent_handles for progress
  • query the session for information
  • add and remove torrents from the session at run-time

• destruct all torrent_handles

• destruct session object

Each class and function is described in this manual.

session

The session class has the following synopsis:

class session: public boost::noncopyable
{

        session(const fingerprint& print = libtorrent::fingerprint("LT, 0, 1, 0, 0));

        session(
                const fingerprint& print
                , std::pair<int, int> listen_port_range
                , const char* listen_interface = 0);

        torrent_handle add_torrent(
                const torrent_info& t
                , const boost::filesystem::path& save_path
                , const entry& resume_data = entry());

        void remove_torrent(const torrent_handle& h);

        void set_http_settings(const http_settings& settings);
        void set_upload_rate_limit(int bytes_per_second);
        void set_download_rate_limit(int bytes_per_second);

        bool is_listening() const;
        unsigned short listen_port() const;
        bool listen_on(
                std::pair<int, int> const& port_range
                , const char* interface = 0);


        std::auto_ptr<alert> pop_alert();
        void set_severity_level(alert::severity_t s);

};

Once it's created, it will spawn the main thread that will do all the work. The main thread will be idle as long it doesn't have any torrents to participate in. You add torrents through the add_torrent()-function where you give an object representing the information found in the torrent file and the path where you want to save the files. The save_path will be prepended to the directory structure in the torrent-file. add_torrent will throw duplicate_torrent exception if the torrent already exists in the session.

The optional last parameter, resume_data can be given if up to date fast-resume data is available. The fast-resume data can be acquired from a running torrent by calling torrent_handle::write_resume_data(). See fast resume.

remove_torrent() will close all peer connections associated with the torrent and tell the tracker that we've stopped participating in the swarm.

If the torrent you are trying to add already exists in the session (is either queued for checking, being checked or downloading) add_torrent() will throw duplicate_torrent which derives from std::exception.

The difference between the two constructors is that one of them takes a fingerprint as argument. If this is ommited, the client will get a default fingerprint stating the version of libtorrent. The fingerprint is a short string that will be used in the peer-id to identify the client and the client's version. For more details see the fingerprint class. The constructor that only takes a finger print will not open a listen port for the session, to get it running you'll have to call session::listen_on(). The other constructor, that takes a port range and an interface as well as the fingerprint will automatically try to listen on a port on the given interface. For more information about the parameters, see listen_on() function.

set_upload_rate_limit() set the maximum number of bytes allowed to be sent to peers per second. This bandwidth is distributed among all the peers. If you don't want to limit upload rate, you can set this to -1 (the default). set_download_rate_limit() works the same way but for download rate instead of upload rate.

is_listening() will tell you wether or not the session has successfully opened a listening port. If it hasn't, this function will return false, and then you can use listen_on() to make another try.

listen_port() returns the port we ended up listening on. Since you just pass a port-range to the constructor and to listen_on(), to know which port it ended up using, you have to ask the session using this function.

listen_on() will change the listen port and/or the listen interface. If the session is already listening on a port, this socket will be closed and a new socket will be opened with these new settings. The port range is the ports it will try to listen on, if the first port fails, it will continue trying the next port within the range and so on. The interface parameter can be left as 0, in that case the os will decide which interface to listen on, otherwise it should be the ip-address of the interface you want the listener socket bound to. listen_on() returns true if it managed to open the socket, and false if it failed. If it fails, it will also generate an appropriate alert (listen_failed_alert).

The interface parameter can also be a hostname that will resolve to the device you want to listen on.

The destructor of session will notify all trackers that our torrents has been shut down. If some trackers are down, they will timout. All this before the destructor of session returns. So, it's adviced that any kind of interface (such as windows) are closed before destructing the sessoin object. Because it can take a few second for it to finish. The timeout can be set with set_http_settings().

The torrent_handle returned by add_torrent can be used to retrieve information about the torrent's progress, its peers etc. It is also used to abort a torrent.

The constructor takes a range of listen ports as argument, if the first port is busy it will increase the port number by one and try again. If it still fails it will continue increasing the port number until it succeeds or has reached the end of the range. If it fails with all ports, a listen_failed_alert will be posted and the session thread will exit. The only thing you can do with your session if this alert is posted is to destruct it and possibly try again or change the port range. The listen interaface string is the name (ip address) of the interface you want to listen on. If this is left as 0, the os will decide which interface to listen on (works in most cases). All torrents will use this interface to open outgoing connections on by default. You can change which interface to use for outgoing connections on a per torrent basis. See torrent_handle.

For information about the pop_alert() function, see alerts.

parsing torrent files

The torrent files are bencoded. There are two functions in libtorrent that can encode and decode bencoded data. They are:

template<class InIt> entry bdecode(InIt start, InIt end);
template<class OutIt> void bencode(OutIt out, const entry& e);

The entry class is the internal representation of the bencoded data and it can be used to retreive information, an entry can also be build by the program and given to bencode() to encode it into the OutIt iterator.

The OutIt and InIt are iterators (InputIterator_ and OutputIterator_ respectively). They are templates and are usually instantiated as ostream_iterator_, back_insert_iterator_ or istream_iterator_. These functions will assume that the iterator refers to a character (char). So, if you want to encode entry e into a buffer in memory, you can do it like this:

std::vector<char> buffer;
bencode(std::back_insert_iterator<std::vector<char> >(buf), e);

If you want to decode a torrent file from a buffer in memory, you can do it like this:

std::vector<char> buffer;
// ...
entry e = bdecode(buf.begin(), buf.end());

Or, if you have a raw char buffer:

const char* buf;
// ...
entry e = bdecode(buf, buf + data_size);

Now we just need to know how to retrieve information from the entry.

If bdecode() encounters invalid encoded data in the range given to it it will throw invalid_encoding.

entry

The entry class represents one node in a bencoded hierarchy. It works as a variant type, it can be either a list, a dictionary (std::map), an integer or a string. This is its synopsis:

class entry
{
public:

        typedef std::list<std::pair<std::string, entry> > dictionary_type;
        typedef std::string string_type;
        typedef std::list<entry> list_type;
        typedef size_type integer_type;

        enum data_type
        {
                int_t,
                string_t,
                list_t,
                dictionary_t,
                undefined_t
        };

        data_type type() const;

        entry(const dictionary_type&);
        entry(const string_type&);
        entry(const list_type&);
        entry(const integer_type&);

        entry();
        entry(data_type t);
        entry(const entry& e);
        ~entry();

        void operator=(const entry& e);
        void operator=(const dictionary_type&);
        void operator=(const string_type&);
        void operator=(const list_type&);
        void operator=(const integer_type&);

        integer_type& integer();
        const integer_type& integer() const;
        string_type& string();
        const string_type& string() const;
        list_type& list();
        const list_type& list() const;
        dictionary_type& dict();
        const dictionary_type& dict() const;

        // these functions requires that the entry
        // is a dictionary, otherwise they will throw   
        entry& operator[](char const* key);
        entry& operator[](std::string const& key);
        const entry& operator[](char const* key) const;
        const entry& operator[](std::string const& key) const;
        entry* find_key(char const* key);
        entry const* find_key(char const* key) const;
        
        void print(std::ostream& os, int indent = 0) const;
};

The integer(), string(), list() and dict() functions are accessorts that return the respecive type. If the entry object isn't of the type you request, the accessor will throw type_error (which derives from std::runtime_error). You can ask an entry for its type through the type() function.

The print() function is there for debug purposes only.

If you want to create an entry you give it the type you want it to have in its constructor, and then use one of the non-const accessors to get a reference which you then can assign the value you want it to have.

The typical code to get info from a torrent file will then look like this:

entry torrent_file;
// ...

const entry::dictionary_type& dict = torrent_file.dict();
entry::dictionary_type::const_iterator i;
i = dict.find("announce");
if (i != dict.end())
{
        std::string tracker_url= i->second.string();
        std::cout << tracker_url << "\n";
}

To make it easier to extract information from a torren file, the class torrent_info exists.

torrent_info

The torrent_info has the following synopsis:

class torrent_info
{
public:

        torrent_info(const entry& torrent_file)
        torrent_info(int piece_size, const char* name);

        entry create_torrent() const;
        void set_comment(char const* str);
        void set_creator(char const* str);
        void set_hash(int index, const sha1_hash& h);
        void add_tracker(std::string const& url, int tier = 0);
        void add_file(boost::filesystem::path file, size_type size);

        typedef std::vector>file_entry>::const_iterator file_iterator;
        typedef std::vector<file_entry>::const_reverse_iterator reverse_file_iterator;

        file_iterator begin_files() const;
        file_iterator end_files() const;
        reverse_file_iterator rbegin_files() const;
        reverse_file_iterator rend_files() const;

        int num_files() const;
        const file_entry& file_at(int index) const;

        const std::vector<announce_entry>& trackers() const;

        int prioritize_tracker(int index);

        size_type total_size() const;
        size_type piece_length() const;
        int num_pieces() const;
        const sha1_hash& info_hash() const;
        const std::stirng& name() const;
        const std::string& comment() const;

        boost::optional<boost::posix_time::ptime>
        creation_date() const;


        void print(std::ostream& os) const;

        size_type piece_size(unsigned int index) const;
        const sha1_hash& hash_for_piece(unsigned int index) const;
};

This class will need some explanation. First of all, to get a list of all files in the torrent, you can use begin_files(), end_files(), rbegin_files() and rend_files(). These will give you standard vector iterators with the type file_entry.

struct file_entry
{
        boost::filesystem::path path;
        size_type size;
};

If you need index-access to files you can use the num_files() and file_at() to access files using indices.

The print() function is there for debug purposes only. It will print the info from the torrent file to the given outstream.

name() returns the name of the torrent.

The trackers() function will return a sorted vector of announce_entry. Each announce entry contains a string, which is the tracker url, and a tier index. The tier index is the high-level priority. No matter which trackers that works or not, the ones with lower tier will always be tried before the one with higher tier number.

struct announce_entry
{
        std::string url;
        int tier;
};

The prioritize_tracker() is used internally to move a tracker to the front of its tier group. i.e. It will never be moved pass a tracker with a different tier number. For more information about how multiple trackers are dealt with, see the specification.

total_size(), piece_length() and num_pieces() returns the total number of bytes the torrent-file represents (all the files in it), the number of byte for each piece and the total number of pieces, respectively. The difference between piece_size() and piece_length() is that piece_size() takes the piece index as argument and gives you the exact size of that piece. It will always be the same as piece_length() except in the case of the last piece, which may be smaller.

hash_for_piece() takes a piece-index and returns the 20-bytes sha1-hash for that piece and info_hash() returns the 20-bytes sha1-hash for the info-section of the torrent file. For more information on the sha1_hash, see the big_number class.

comment() returns the comment associated with the torrent. If there's no comment, it will return an empty string. creation_date() returns a boost::posix_time::ptime object, representing the time when this torrent file was created. If there's no timestamp in the torrent file, this will return a date of january 1:st 1970.

torrent_handle

You will usually have to store your torrent handles somewhere, since it's the object through which you retrieve infromation about the torrent and aborts the torrent. Its declaration looks like this:

struct torrent_handle
{
        torrent_handle();

        torrent_status status();
        void get_download_queue(std::vector<partial_piece_info>& queue);
        void get_peer_info(std::vector<peer_info>& v);
        const torrent_info& get_torrent_info();
        bool is_valid();

        entry write_resume_data();
        void force_reannounce();
        void connect_peer(const address& adr) const;

        void set_tracker_login(std::string const& username, std::string const& password);

        void set_ratio(float ratio);
        void set_max_uploads(int max_uploads);
        void set_max_connections(int max_connections);
        void set_upload_limit(int limit);
        void set_download_limit(int limit);
        void use_interface(const char* net_interface);

        void pause();
        void resume();
        bool is_paused() const;

        boost::filsystem::path save_path() const;

        sha1_hash info_hash() const;

        bool operator==(const torrent_handle&) const;
        bool operator!=(const torrent_handle&) const;
        bool operator<(const torrent_handle&) const;
};

The default constructor will initialize the handle to an invalid state. Which means you cannot perform any operation on it, unless you first assign it a valid handle. If you try to perform any operation on an uninitialized handle, it will throw invalid_handle.

save_path() returns the path that was given to add_torrent() when this torrent was started.

force_reannounce() will force this torrent to do another tracker request, to receive new peers. If the torrent is invalid, queued or in checking mode, this functions will throw invalid_handle.

connect_peer() is a way to manually connect to peers that one believe is a part of the torrent. If the peer does not respond, or is not a member of this torrent, it will simply be disconnected. No harm can be done by using this other than an unnecessary connection attempt is made. If the torrent is uninitialized or in queued or checking mode, this will throw invalid_handle.

set_ratio() sets the desired download / upload ratio. If set to 0, it is considered being infinite. i.e. the client will always upload as much as it can, no matter how much it gets back in return. With this setting it will work much like the standard clients.

Besides 0, the ratio can be set to any number greater than or equal to 1. It means how much to attempt to upload in return for each download. e.g. if set to 2, the client will try to upload 2 bytes for every byte received. The default setting for this is 0, which will make it work as a standard client.

set_upload_limit will limit the upload bandwidth used by this particular torrent to the limit you set. It is given as the number of bytes per second the torrent is allowed to upload. set_download_limit works the same way but for download bandwidth instead of upload bandwidth. Note that setting i higher limit on a torrent then the global limit (session::set_upload_rate_limit) will not override the global rate limit. The torrent can never upload more than the global rate limit.

pause(), and resume() will disconnect all peers and reconnect all peers respectively. When a torrent is paused, it will however remember all share ratios to all peers and remember all potential (not connected) peers. You can use is_paused() to determine if a torrent is currently paused. Torrents may be paused automatically if there is a file error (eg. disk full) or something similar. See file_error_alert.

set_tracker_login() sets a username and password that will be sent along in the HTTP-request of the tracker announce. Set this if the tracker requires authorization.

use_interface() sets the network interface this torrent will use when it opens outgoing connections. By default, it uses the same interface as the session uses to listen on.

info_hash() returns the info hash for the torrent.

set_max_uploads() sets the maximum number of peers that's unchoked at the same time on this torrent. If you set this to -1, there will be no limit.

set_max_connections() sets the maximum number of connection this torrent will open. If all connections are used up, incoming connections may be refused or poor connections may be closed. This must be at least 2. The default is unlimited number of connections. If -1 is given to the function, it means unlimited.

write_resume_data() generates fast-resume data and returns it as an entry. This entry is suitable for being bencoded. For more information about how fast-resume works, see fast resume. It may throw invalid_handle if the torrent handle is invalid.

struct torrent_status
{
        enum state_t
        {
                queued_for_checking,
                checking_files,
                connecting_to_tracker,
                downloading,
                seeding
        };

        state_t state;
        bool paused;
        float progress;
        boost::posix_time::time_duration next_announce;
        boost::posix_time::time_duration announce_interval;

        std::string current_tracker;

        size_type total_download;
        size_type total_upload;

        size_type total_payload_download;
        size_type total_payload_upload;

        float download_rate;
        float upload_rate;

        int num_peers;

        const std::vector<bool>* pieces;
        size_type total_done;
};

struct partial_piece_info
{
        enum { max_blocks_per_piece };
        int piece_index;
        int blocks_in_piece;
        std::bitset<max_blocks_per_piece> requested_blocks;
        std::bitset<max_blocks_per_piece> finished_blocks;
        peer_id peer[max_blocks_per_piece];
        int num_downloads[max_blocks_per_piece];
};

struct peer_info
{
        enum
        {
                interesting = 0x1,
                choked = 0x2,
                remote_interested = 0x4,
                remote_choked = 0x8,
                supports_extensions = 0x10,
                local_connection = 0x20
        };
        unsigned int flags;
        address ip;
        float up_speed;
        float down_speed;
        size_type total_download;
        size_type total_upload;
        peer_id id;
        std::vector<bool> pieces;
        int upload_limit;
        int upload_ceiling;

        size_type load_balancing;

        int download_queue_length;
        int upload_queue_length;

        int downloading_piece_index;
        int downloading_block_index;
        int downloading_progress;
        int downloading_total;
};

address

The address class represents a name of a network endpoint (usually referred to as IP-address) and a port number. This is the same thing as a sockaddr_in would contain. Its declaration looks like this:

class address
{
public:
        address();
        address(unsigned char a
                , unsigned char b
                , unsigned char c
                , unsigned char d
                , unsigned short  port);
        address(unsigned int addr, unsigned short port);
        address(const std::string& addr, unsigned short port);
        address(const address& a);
        ~address();

        std::string as_string() const;
        unsigned int ip() const;
        unsigned short port() const;

        bool operator<(const address& a) const;
        bool operator!=(const address& a) const;
        bool operator==(const address& a) const;
};

It is less-than comparable to make it possible to use it as a key in a map. as_string() may block while it does the DNS lookup, it returns a string that points to the address represented by the object.

ip() will return the 32-bit ip-address as an integer. port() returns the port number.

http_settings

You have some control over tracker requests through the http_settings object. You create it and fill it with your settings and the use session::set_http_settings() to apply them. You have control over proxy and authorization settings and also the user-agent that will be sent to the tracker. The user-agent is a good way to identify your client.

struct http_settings
{
        http_settings();
        std::string proxy_ip;
        int proxy_port;
        std::string proxy_login;
        std::string proxy_password;
        std::string user_agent;
        int tracker_timeout;
        int tracker_maximum_response_length;
};

proxy_ip may be a hostname or ip to a http proxy to use. If this is an empty string, no http proxy will be used.

proxy_port is the port on which the http proxy listens. If proxy_ip is empty, this will be ignored.

proxy_login should be the login username for the http proxy, if this empty, the http proxy will be trid to be used without authentication.

proxy_password the password string for the http proxy.

user_agent this is the client identification to the tracker. It will be followed by the string "(libtorrent)" to identify that this library is being used. This should be set to your client's name and version number.

tracker_timeout is the number of seconds the tracker connection will wait until it considers the tracker to have timed-out. Default value is 10 seconds.

tracker_maximum_response_length is the maximum number of bytes in a tracker response. If a response size passes this number it will be rejected and the connection will be closed. On gzipped responses this size is measured on the uncompressed data. So, if you get 20 bytes of gzip response that'll expand to 2 megs, it will be interrupted before the entire response has been uncompressed (given your limit is lower than 2 megs). Default limit is 1 megabyte.

big_number

Both the peer_id and sha1_hash types are typedefs of the class big_number. It represents 20 bytes of data. Its synopsis follows:

class big_number
{
public:
        bool operator==(const big_number& n) const;
        bool operator!=(const big_number& n) const;
        bool operator<(const big_number& n) const;

        const unsigned char* begin() const;
        const unsigned char* end() const;

        unsigned char* begin();
        unsigned char* end();
};

The iterators gives you access to individual bytes.

hasher

This class creates sha1-hashes. Its declaration looks like this:

class hasher
{
public:
        hasher();

        void update(const char* data, unsigned int len);
        sha1_hash final();
        void reset();
};

You use it by first instantiating it, then call update() to feed it with data. i.e. you don't have to keep the entire buffer of which you want to create the hash in memory. You can feed the hasher parts of it at a time. When You have fed the hasher with all the data, you call final() and it will return the sha1-hash of the data.

If you want to reuse the hasher object once you have created a hash, you have to call reset() to reinitialize it.

The sha1-algorithm used was implemented by Steve Reid and released as public domain. For more info, see src/sha1.c.

fingerprint

The fingerprint class represents information about a client and its version. It is used to encode this information into the client's peer id.

This is the class declaration:

struct fingerprint
{
        fingerprint(const char* id_string, int major, int minor, int revision, int tag);

        std::string to_string() const;

        char id[2];
        char major_version;
        char minor_version;
        char revision_version;
        char tag_version;

};

The constructor takes a const char* that should point to a string constant containing exactly two characters. These are the characters that should be unique for your client. Make sure not to clash with anybody else. Here are some taken id's:

The major, minor, revision and tag parameters are used to identify the version of your client. All these numbers must be within the range [0, 9].

to_string() will generate the actual string put in the peer-id, and return it.

std::string identify_client(const peer_id& id);

alerts

The pop_alert() function on session is the interface for retrieving alerts, warnings, messages and errors from libtorrent. If there hasn't occured any errors (matching your severity level) pop_alert() will return a zero pointer. If there has been some error, it will return a pointer to an alert object describing it. You can then use the alert object and query it for information about the error or message. To retrieve any alerts, you have to select a severity level using session::set_severity_level(). It defaults to alert::none, which means that you don't get any messages at all, ever. You have the following levels to select among:

When setting a severity level, you will receive messages of that severity and all messages that are more sever. If you set alert::none (the default) you will not recieve any events at all.

When you set a severuty level other than none, you have the responsibility to call pop_alert() from time to time. If you don't do that, the alert queue will just grow.

When you get an alert, you can use typeid() or dynamic_cast<> to get more detailed information on exactly which type it is. i.e. what kind of error it is. You can also use a dispatcher mechanism that's available in libtorrent.

All alert types are defined in the <libtorrent/alert_types.hpp> header file.

The alert class is the base class that specific messages are derived from. This is its synopsis:

class alert
{
public:

        enum severity_t { debug, info, warning, critital, fatal, none };

        alert(severity_t severity, const std::string& msg);
        virtual ~alert();

        const std::string& msg() const;
        severity_t severity() const;

        virtual std::auto_ptr<alert> clone() const = 0;
};

This means that all alerts have at least a string describing it. They also have a severity leve that can be used to sort them or present them to the user in different ways.

The specific alerts, that all derives from alert, are:

struct listen_failed_alert: alert
{
        listen_failed_alert(const std::string& msg);
        virtual std::auto_ptr<alert> clone() const;
};

struct file_error_alert: alert
{
        file_error_alert(
                const torrent_handle& h
                , const std::string& msg);
                
        virtual std::auto_ptr<alert> clone() const;

        torrent_handle handle;
};

struct tracker_alert: alert
{
        tracker_alert(const torrent_handle& h, const std::string& msg);
        virtual std::auto_ptr<alert> clone() const;

        torrent_handle handle;
};

struct hash_failed_alert: alert
{
        hash_failed_alert(
                const torrent_handle& h
                , int index
                , const std::string& msg);

        virtual std::auto_ptr<alert> clone() const;

        torrent_handle handle;
        int piece_index;
};

struct peer_ban_alert: alert
{
        peer_ban_alert(
                address const& pip
                , torrent_handle h
                , const std::string& msg);

        virtual std::auto_ptr<alert> clone() const;

        address ip;
        torrent_handle handle;
};

struct peer_error_alert: alert
{
        peer_error_alert(
                address const& pip
                , peer_id const& pid
                , const std::string& msg);

        virtual std::auto_ptr<alert> clone() const;

        address ip;
        peer_id id;
};

struct invalid_request_alert: alert
{
        invalid_request_alert(
                peer_request const& r
                , torrent_handle const& h
                , address const& send
                , peer_id const& pid
                , std::string const& msg);

        virtual std::auto_ptr<alert> clone() const;

        torrent_handle handle;
        address ip;
        peer_request request;
        peer_id id;
};


struct peer_request
{
        int piece;
        int start;
        int length;
        bool operator==(peer_request const& r) const;
};

struct torrent_finished_alert: alert
{
        torrent_finished_alert(
                const torrent_handle& h
                , const std::string& msg);

        virtual std::auto_ptr<alert> clone() const;

        torrent_handle handle;
};

exceptions

There are a number of exceptions that can be thrown from different places in libtorrent, here's a complete list with description.

struct invalid_handle: std::exception
{
        const char* what() const throw();
};

struct duplicate_torrent: std::exception
{
        const char* what() const throw();
};

struct invalid_encoding: std::exception
{
        const char* what() const throw();
};

struct type_error: std::runtime_error
{
        type_error(const char* error);
};

struct invalid_torrent_file: std::exception
{
        const char* what() const throw();
};

examples

#include <iostream>
#include <fstream>
#include <iterator>
#include <exception>
#include <iomanip>

#include "libtorrent/entry.hpp"
#include "libtorrent/bencode.hpp"
#include "libtorrent/torrent_info.hpp"


int main(int argc, char* argv[])
{
        using namespace libtorrent;

        if (argc != 2)
        {
                std::cerr << "usage: dump_torrent torrent-file\n";
                return 1;
        }

        try
        {
                std::ifstream in(argv[1], std::ios_base::binary);
                in.unsetf(std::ios_base::skipws);
                entry e = bdecode(std::istream_iterator<char>(in), std::istream_iterator<char>());
                torrent_info t(e);

                // print info about torrent
                std::cout << "\n\n----- torrent file info -----\n\n";
                std::cout << "trackers:\n";
                for (std::vector<announce_entry>::const_iterator i = t.trackers().begin();
                        i != t.trackers().end();
                        ++i)
                {
                        std::cout << i->tier << ": " << i->url << "\n";
                }

                std::cout << "number of pieces: " << t.num_pieces() << "\n";
                std::cout << "piece length: " << t.piece_length() << "\n";
                std::cout << "files:\n";
                for (torrent_info::file_iterator i = t.begin_files();
                        i != t.end_files();
                        ++i)
                {
                        std::cout << "  " << std::setw(11) << i->size
                        << "  " << i->path << " " << i->filename << "\n";
                }
                
        }
        catch (std::exception& e)
        {
                std::cout << e.what() << "\n";
        }

        return 0;
}

#include <iostream>
#include <fstream>
#include <iterator>
#include <exception>

#include <boost/format.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

#include "libtorrent/entry.hpp"
#include "libtorrent/bencode.hpp"
#include "libtorrent/session.hpp"
#include "libtorrent/http_settings.hpp"

int main(int argc, char* argv[])
{
        using namespace libtorrent;

        if (argc != 2)
        {
                std::cerr << "usage: ./simple_cient torrent-file\n"
                        "to stop the client, press return.\n";
                return 1;
        }

        try
        {
                session s(std::make_pair(6881, 6889));

                std::ifstream in(argv[1], std::ios_base::binary);
                in.unsetf(std::ios_base::skipws);
                entry e = bdecode(std::istream_iterator<char>(in), std::istream_iterator<char>());
                torrent_info t(e);
                s.add_torrent(t, "");
                        
                // wait for the user to end
                char a;
                std::cin.unsetf(std::ios_base::skipws);
                std::cin >> a;
        }
        catch (std::exception& e)
        {
                std::cout << e.what() << "\n";
        }
        return 0;
}

fast resume

The fast resume mechanism is a way to remember which pieces are downloaded and where they are put between sessions. You can generate fast resume data by calling torrent_handle::write_resume_data() on torrent_handle. You can then save this data to disk and use it when resuming the torrent. libtorrent will not check the piece hashes then, and rely on the information given in the fast-resume data. The fast-resume data also contains information about which blocks, in the unfinished pieces, were downloaded, so it will not have to start from scratch on the partially downloaded pieces.

To use the fast-resume data you simply give it to session::add_torrent(), and it will skip the time consuming checks. It may have to do the checking anyway, if the fast-resume data is corrupt or doesn't fit the storage for that torrent, then it will not trust the fast-resume data and just do the checking.

extensions

These extensions all operates within the extension protocol. The name of the extension is the name used in the extension-list packets, and the payload is the data in the extended message (not counting the length-prefix, message-id nor extension-id).

The extension protocol is currently disabled, since it may not be compatible with future versions of bittorrent.

filenames checks

Boost.Filesystem will by default check all its paths to make sure they conform to filename requirements on many platforms. If you don't want this check, you can set it to either only check for native filesystem requirements or turn it off alltogether. You can use: boost::filesystem::path::default_name_check(boost::filesystem::native) for example. For more information, see the Boost.Filesystem docs.

aknowledgements

Written by Arvid Norberg. Copyright (c) 2003

Contributions by Magnus Jonsson and Daniel Wallin

Thanks to Reimond Retz for bugfixes, suggestions and testing

Thanks to University of Umeå for providing development and test hardware.

Project is hosted by sourceforge.