~ubuntu-branches/ubuntu/maverick/libtorrent-rasterbar/maverick

<li><a class="reference internal" href="#set-upload-rate-limit-set-download-rate-limit-upload-rate-limit-download-rate-limit" id="id27">set_upload_rate_limit() set_download_rate_limit() upload_rate_limit() download_rate_limit()</a></li>

<li><a class="reference internal" href="#set-max-uploads-set-max-connections" id="id28">set_max_uploads() set_max_connections()</a></li>

<li><a class="reference internal" href="#num-uploads-num-connections" id="id29">num_uploads() num_connections()</a></li>

<li><a class="reference internal" href="#set-max-half-open-connections-max-half-open-connections" id="id30">set_max_half_open_connections() max_half_open_connections()</a></li>

<li><a class="reference internal" href="#set-ip-filter" id="id31">set_ip_filter()</a></li>

<li><a class="reference internal" href="#status" id="id32">status()</a></li>

<li><a class="reference internal" href="#is-listening-listen-port-listen-on" id="id33">is_listening() listen_port() listen_on()</a></li>

<li><a class="reference internal" href="#pop-alert-set-severity-level-wait-for-alert" id="id34">pop_alert() set_severity_level() wait_for_alert()</a></li>

<li><a class="reference internal" href="#add-extension" id="id35">add_extension()</a></li>

<li><a class="reference internal" href="#set-settings-set-pe-settings" id="id36">set_settings() set_pe_settings()</a></li>

<li><a class="reference internal" href="#set-peer-proxy-set-web-seed-proxy-set-tracker-proxy-set-dht-proxy" id="id37">set_peer_proxy() set_web_seed_proxy() set_tracker_proxy() set_dht_proxy()</a></li>

<li><a class="reference internal" href="#peer-proxy-web-seed-proxy-tracker-proxy-dht-proxy" id="id38">peer_proxy() web_seed_proxy() tracker_proxy() dht_proxy()</a></li>

<li><a class="reference internal" href="#start-dht-stop-dht-set-dht-settings-dht-state" id="id39">start_dht() stop_dht() set_dht_settings() dht_state()</a></li>

<li><a class="reference internal" href="#add-dht-node-add-dht-router" id="id40">add_dht_node() add_dht_router()</a></li>

<li><a class="reference internal" href="#start-lsd-stop-lsd" id="id41">start_lsd() stop_lsd()</a></li>

<li><a class="reference internal" href="#start-upnp-stop-upnp" id="id42">start_upnp() stop_upnp()</a></li>

<li><a class="reference internal" href="#start-natpmp-stop-natpmp" id="id43">start_natpmp() stop_natpmp()</a></li>

</ul>

</li>

<li><a class="reference internal" href="#entry" id="id44">entry</a><ul>

<li><a class="reference internal" href="#integer-string-list-dict-type" id="id45">integer() string() list() dict() type()</a></li>

<li><a class="reference internal" href="#operator" id="id46">operator[]</a></li>

</ul>

</li>

<li><a class="reference internal" href="#torrent-info" id="id48">torrent_info</a><ul>

<li><a class="reference internal" href="#id3" id="id49">torrent_info()</a></li>

<li><a class="reference internal" href="#set-comment-set-piece-size-set-creator-set-hash-add-tracker-add-file" id="id50">set_comment() set_piece_size() set_creator() set_hash() add_tracker() add_file()</a></li>

<li><a class="reference internal" href="#create-torrent" id="id51">create_torrent()</a></li>

<li><a class="reference internal" href="#remap-files" id="id52">remap_files()</a></li>

<li><a class="reference internal" href="#begin-files-end-files-rbegin-files-rend-files" id="id53">begin_files() end_files() rbegin_files() rend_files()</a></li>

<li><a class="reference internal" href="#num-files-file-at" id="id54">num_files() file_at()</a></li>

<li><a class="reference internal" href="#map-block" id="id55">map_block()</a></li>

<li><a class="reference internal" href="#url-seeds-add-url-seed" id="id57">url_seeds() add_url_seed()</a></li>

<li><a class="reference internal" href="#print" id="id58">print()</a></li>

<li><a class="reference internal" href="#trackers" id="id59">trackers()</a></li>

<li><a class="reference internal" href="#total-size-piece-length-piece-size-num-pieces" id="id60">total_size() piece_length() piece_size() num_pieces()</a></li>

<li><a class="reference internal" href="#hash-for-piece-info-hash" id="id61">hash_for_piece() info_hash()</a></li>

<li><a class="reference internal" href="#name-comment-creation-date-creator" id="id62">name() comment() creation_date() creator()</a></li>

<li><a class="reference internal" href="#nodes" id="id64">nodes()</a></li>

</ul>

</li>

<li><a class="reference internal" href="#torrent-handle" id="id66">torrent_handle</a><ul>

<li><a class="reference internal" href="#piece-priority-prioritize-pieces-piece-priorities-prioritize-files" id="id67">piece_priority() prioritize_pieces() piece_priorities() prioritize_files()</a></li>

<li><a class="reference internal" href="#file-progress" id="id68">file_progress()</a></li>

<li><a class="reference internal" href="#move-storage" id="id70">move_storage()</a></li>

<li><a class="reference internal" href="#force-reannounce" id="id71">force_reannounce()</a></li>

<li><a class="reference internal" href="#scrape-tracker" id="id72">scrape_tracker()</a></li>

<li><a class="reference internal" href="#connect-peer" id="id73">connect_peer()</a></li>

<li><a class="reference internal" href="#set-ratio" id="id75">set_ratio()</a></li>

<li><a class="reference internal" href="#set-upload-limit-set-download-limit-upload-limit-download-limit" id="id76">set_upload_limit() set_download_limit() upload_limit() download_limit()</a></li>

<li><a class="reference internal" href="#set-sequenced-download-threshold" id="id77">set_sequenced_download_threshold()</a></li>

<li><a class="reference internal" href="#set-peer-upload-limit-set-peer-download-limit" id="id78">set_peer_upload_limit() set_peer_download_limit()</a></li>

<li><a class="reference internal" href="#pause-resume-is-paused" id="id79">pause() resume() is_paused()</a></li>

<li><a class="reference internal" href="#resolve-countries" id="id80">resolve_countries()</a></li>

<li><a class="reference internal" href="#has-metadata" id="id82">has_metadata()</a></li>

<li><a class="reference internal" href="#set-tracker-login" id="id83">set_tracker_login()</a></li>

<li><a class="reference internal" href="#trackers-replace-trackers" id="id84">trackers() replace_trackers()</a></li>

100

<li><a class="reference internal" href="#add-url-seed-remove-url-seed-url-seeds" id="id85">add_url_seed() remove_url_seed() url_seeds()</a></li>

101

<li><a class="reference internal" href="#use-interface" id="id86">use_interface()</a></li>

102

103

<li><a class="reference internal" href="#id5" id="id88">set_max_uploads() set_max_connections()</a></li>

104

<li><a class="reference internal" href="#write-resume-data" id="id89">write_resume_data()</a></li>

105

<li><a class="reference internal" href="#id6" id="id90">status()</a></li>

106

<li><a class="reference internal" href="#get-download-queue" id="id91">get_download_queue()</a></li>

107

108

<li><a class="reference internal" href="#get-torrent-info" id="id93">get_torrent_info()</a></li>

109

<li><a class="reference internal" href="#is-valid" id="id94">is_valid()</a></li>

110

</ul>

111

</li>

112

<li><a class="reference internal" href="#torrent-status" id="id95">torrent_status</a></li>

113

114

<li><a class="reference internal" href="#session-settings" id="id97">session_settings</a></li>

115

<li><a class="reference internal" href="#pe-settings" id="id98">pe_settings</a></li>

116

<li><a class="reference internal" href="#proxy-settings" id="id99">proxy_settings</a></li>

117

<li><a class="reference internal" href="#ip-filter" id="id100">ip_filter</a><ul>

118

<li><a class="reference internal" href="#id9" id="id101">ip_filter()</a></li>

119

120

<li><a class="reference internal" href="#access" id="id103">access()</a></li>

121

<li><a class="reference internal" href="#export-filter" id="id104">export_filter()</a></li>

122

</ul>

123

</li>

124

<li><a class="reference internal" href="#big-number" id="id105">big_number</a></li>

125

<li><a class="reference internal" href="#hasher" id="id106">hasher</a></li>

126

<li><a class="reference internal" href="#fingerprint" id="id107">fingerprint</a></li>

127

<li><a class="reference internal" href="#free-functions" id="id108">free functions</a><ul>

128

<li><a class="reference internal" href="#identify-client" id="id109">identify_client()</a></li>

129

<li><a class="reference internal" href="#client-fingerprint" id="id110">client_fingerprint()</a></li>

130

<li><a class="reference internal" href="#bdecode-bencode" id="id111">bdecode() bencode()</a></li>

131

</ul>

132

</li>

133

<li><a class="reference internal" href="#alerts" id="id112">alerts</a><ul>

134

<li><a class="reference internal" href="#listen-failed-alert" id="id113">listen_failed_alert</a></li>

135

<li><a class="reference internal" href="#portmap-error-alert" id="id114">portmap_error_alert</a></li>

136

<li><a class="reference internal" href="#portmap-alert" id="id115">portmap_alert</a></li>

137

<li><a class="reference internal" href="#file-error-alert" id="id116">file_error_alert</a></li>

138

<li><a class="reference internal" href="#tracker-announce-alert" id="id117">tracker_announce_alert</a></li>

139

<li><a class="reference internal" href="#tracker-alert" id="id118">tracker_alert</a></li>

140

<li><a class="reference internal" href="#tracker-reply-alert" id="id119">tracker_reply_alert</a></li>

141

<li><a class="reference internal" href="#tracker-warning-alert" id="id120">tracker_warning_alert</a></li>

142

<li><a class="reference internal" href="#scrape-reply-alert" id="id121">scrape_reply_alert</a></li>

143

<li><a class="reference internal" href="#scrape-failed-alert" id="id122">scrape_failed_alert</a></li>

144

<li><a class="reference internal" href="#url-seed-alert" id="id123">url_seed_alert</a></li>

145

<li><a class="reference internal" href="#hash-failed-alert" id="id124">hash_failed_alert</a></li>

146

<li><a class="reference internal" href="#peer-ban-alert" id="id125">peer_ban_alert</a></li>

147

<li><a class="reference internal" href="#peer-error-alert" id="id126">peer_error_alert</a></li>

148

<li><a class="reference internal" href="#invalid-request-alert" id="id127">invalid_request_alert</a></li>

149

<li><a class="reference internal" href="#torrent-finished-alert" id="id128">torrent_finished_alert</a></li>

150

<li><a class="reference internal" href="#metadata-failed-alert" id="id129">metadata_failed_alert</a></li>

151

<li><a class="reference internal" href="#metadata-received-alert" id="id130">metadata_received_alert</a></li>

152

<li><a class="reference internal" href="#fastresume-rejected-alert" id="id131">fastresume_rejected_alert</a></li>

153

<li><a class="reference internal" href="#peer-blocked-alert" id="id132">peer_blocked_alert</a></li>

154

<li><a class="reference internal" href="#storage-moved-alert" id="id133">storage_moved_alert</a></li>

155

<li><a class="reference internal" href="#torrent-paused-alert" id="id134">torrent_paused_alert</a></li>

156

<li><a class="reference internal" href="#dispatcher" id="id135">dispatcher</a></li>

157

</ul>

158

</li>

159

<li><a class="reference internal" href="#exceptions" id="id136">exceptions</a><ul>

160

<li><a class="reference internal" href="#invalid-handle" id="id137">invalid_handle</a></li>

161

<li><a class="reference internal" href="#duplicate-torrent" id="id138">duplicate_torrent</a></li>

162

<li><a class="reference internal" href="#invalid-encoding" id="id139">invalid_encoding</a></li>

163

<li><a class="reference internal" href="#type-error" id="id140">type_error</a></li>

164

<li><a class="reference internal" href="#invalid-torrent-file" id="id141">invalid_torrent_file</a></li>

165

</ul>

166

</li>

167

<li><a class="reference internal" href="#storage-interface" id="id142">storage_interface</a><ul>

168

<li><a class="reference internal" href="#initialize" id="id143">initialize()</a></li>

169

170

<li><a class="reference internal" href="#write" id="id145">write()</a></li>

171

<li><a class="reference internal" href="#id11" id="id146">move_storage()</a></li>

172

<li><a class="reference internal" href="#verify-resume-data" id="id147">verify_resume_data()</a></li>

173

<li><a class="reference internal" href="#id12" id="id148">write_resume_data( )</a></li>

174

175

<li><a class="reference internal" href="#swap-slots" id="id150">swap_slots()</a></li>

176

<li><a class="reference internal" href="#swap-slots3" id="id151">swap_slots3()</a></li>

177

178

<li><a class="reference internal" href="#release-files" id="id153">release_files()</a></li>

179

<li><a class="reference internal" href="#delete-files" id="id154">delete_files()</a></li>

180

</ul>

181

</li>

182

<li><a class="reference internal" href="#fast-resume" id="id155">fast resume</a><ul>

183

<li><a class="reference internal" href="#file-format" id="id156">file format</a></li>

184

</ul>

185

</li>

186

<li><a class="reference internal" href="#threads" id="id157">threads</a></li>

187

<li><a class="reference internal" href="#storage-allocation" id="id158">storage allocation</a><ul>

188

<li><a class="reference internal" href="#sparse-allocation" id="id159">sparse allocation</a></li>

189

<li><a class="reference internal" href="#full-allocation" id="id160">full allocation</a></li>

190

<li><a class="reference internal" href="#compact-allocation" id="id161">compact allocation</a></li>

191

</ul>

192

</li>

193

<li><a class="reference internal" href="#extensions" id="id162">extensions</a><ul>

194

<li><a class="reference internal" href="#metadata-from-peers" id="id163">metadata from peers</a></li>

195

<li><a class="reference internal" href="#http-seeding" id="id164">HTTP seeding</a></li>

196

</ul>

197

</li>

198

<li><a class="reference internal" href="#filename-checks" id="id165">filename checks</a></li>

199

<li><a class="reference internal" href="#acknowledgments" id="id166">acknowledgments</a></li>

200

</ul>

201

</div>

202

203

<h1>overview</h1>

204

<p>The interface of libtorrent consists of a few classes. The main class is

205

the <tt class="docutils literal"><span class="pre">session</span></tt>, it contains the main loop that serves all torrents.</p>

206

<p>The basic usage is as follows:</p>

207

<ul>

208

<li><p class="first">construct a session</p>

209

</li>

210

<li><p class="first">parse .torrent-files and add them to the session (see <a class="reference internal" href="#bdecode-bencode">bdecode() bencode()</a> and <a class="reference internal" href="#add-torrent">add_torrent()</a>)</p>

211

</li>

212

<li><p class="first">main loop (see <a class="reference internal" href="#session">session</a>)</p>

213

214

215

<li>query the torrent_handles for progress (see <a class="reference internal" href="#torrent-handle">torrent_handle</a>)</li>

216

<li>query the session for information</li>

217

<li>add and remove torrents from the session at run-time</li>

218

</ul>

219

</blockquote>

220

</li>

221

<li><p class="first">save resume data for all torrent_handles (optional, see

222

<a class="reference internal" href="#write-resume-data">write_resume_data()</a>)</p>

223

</li>

224

<li><p class="first">destruct session object</p>

225

</li>

226

</ul>

227

<p>Each class and function is described in this manual.</p>

228

</div>

229

230

<h1>network primitives</h1>

231

<p>There are a few typedefs in the <tt class="docutils literal"><span class="pre">libtorrent</span></tt> namespace which pulls

232

in network types from the <tt class="docutils literal"><span class="pre">asio</span></tt> namespace. These are:</p>

233

234

typedef asio::ip::address address;

235

typedef asio::ip::address_v4 address_v4;

236

typedef asio::ip::address_v6 address_v6;

237

using asio::ip::tcp;

238

using asio::ip::udp;

239

</pre>

240

<p>These are declared in the <tt class="docutils literal"><span class="pre"><libtorrent/socket.hpp></span></tt> header.</p>

241

<p>The <tt class="docutils literal"><span class="pre">using</span></tt> statements will give easy access to:</p>

242

243

tcp::endpoint

244

udp::endpoint

245

</pre>

246

<p>Which are the endpoint types used in libtorrent. An endpoint is an address

247

with an associated port.</p>

248

<p>For documentation on these types, please refer to the <a class="reference external" href="http://asio.sourceforge.net/asio-0.3.8/doc/asio/reference.html">asio documentation</a>.</p>

249

</div>

250

251

<h1>session</h1>

252

<p>The <tt class="docutils literal"><span class="pre">session</span></tt> class has the following synopsis:</p>

253

254

class session: public boost::noncopyable

255

{

256

257

session(fingerprint const& print

258

= libtorrent::fingerprint(

259

"LT", 0, 1, 0, 0));

260

261

session(

262

fingerprint const& print

263

, std::pair<int, int> listen_port_range

264

, char const* listen_interface = 0);

265

266

torrent_handle add_torrent(

267

boost::intrusive_ptr<torrent_info> const& ti

268

, boost::filesystem::path const& save_path

269

, entry const& resume_data = entry()

270

, storage_mode_t storage_mode = storage_mode_sparse

271

, bool paused = false

272

, storage_constructor_type sc = default_storage_constructor

273

, void* userdata = 0);

274

275

torrent_handle add_torrent(

276

char const* tracker_url

277

, sha1_hash const& info_hash

278

, char const* name

279

, boost::filesystem::path const& save_path

280

, entry const& resume_data = entry()

281

, storage_mode_t storage_mode = storage_mode_sparse

282

, bool paused = false

283

, storage_constructor_type sc = default_storage_constructor

284

, void* userdata = 0);

285

286

session_proxy abort();

287

288

enum options_t

289

{

290

none = 0,

291

delete_files = 1

292

};

293

294

void remove_torrent(torrent_handle const& h, int options = none);

295

torrent_handle find_torrent(sha_hash const& ih);

296

std::vector<torrent_handle> get_torrents() const;

297

298

void set_settings(session_settings const& settings);

299

void set_pe_settings(pe_settings const& settings);

300

301

void set_upload_rate_limit(int bytes_per_second);

302

int upload_rate_limit() const;

303

void set_download_rate_limit(int bytes_per_second);

304

int download_rate_limit() const;

305

void set_max_uploads(int limit);

306

void set_max_connections(int limit);

307

void set_max_half_open_connections(int limit);

308

int max_half_open_connections() const;

309

310

void set_peer_proxy(proxy_settings const& s);

311

void set_web_seed_proxy(proxy_settings const& s);

312

void set_tracker_proxy(proxy_settings const& s);

313

314

proxy_settings const& peer_proxy() const;

315

proxy_settings const& web_seed_proxy() const;

316

proxy_settings const& tracker_proxy() const;

317

318

int num_uploads() const;

319

int num_connections() const;

320

321

void set_ip_filter(ip_filter const& f);

322

323

session_status status() const;

324

325

bool is_listening() const;

326

unsigned short listen_port() const;

327

bool listen_on(

328

std::pair<int, int> const& port_range

329

, char const* interface = 0);

330

331

std::auto_ptr<alert> pop_alert();

332

alert const* wait_for_alert(time_duration max_wait);

333

void set_severity_level(alert::severity_t s);

334

335

void add_extension(boost::function<

336

boost::shared_ptr<torrent_plugin>(torrent*)> ext);

337

338

void start_dht();

339

void stop_dht();

340

void set_dht_settings(

341

dht_settings const& settings);

342

entry dht_state() const;

343

void add_dht_node(std::pair<std::string

344

, int> const& node);

345

void add_dht_router(std::pair<std::string

346

, int> const& node);

347

348

void start_lsd();

349

void stop_lsd();

350

351

void start_upnp();

352

void stop_upnp();

353

354

void start_natpmp();

355

void stop_natpmp();

356

};

357

</pre>

358

<p>Once it's created, the session object will spawn the main thread that will do all the work.

359

The main thread will be idle as long it doesn't have any torrents to participate in.</p>

360

361

<h2>session()</h2>

362

363

364

session(fingerprint const& print

365

= libtorrent::fingerprint("LT", 0, 1, 0, 0));

366

session(fingerprint const& print

367

, std::pair<int, int> listen_port_range

368

, char const* listen_interface = 0);

369

</pre>

370

</blockquote>

371

<p>If the fingerprint in the first overload is omited, the client will get a default

372

fingerprint stating the version of libtorrent. The fingerprint is a short string that will be

373

used in the peer-id to identify the client and the client's version. For more details see the

374

<a class="reference internal" href="#fingerprint">fingerprint</a> class. The constructor that only takes a fingerprint will not open a

375

listen port for the session, to get it running you'll have to call <tt class="docutils literal"><span class="pre">session::listen_on()</span></tt>.

376

The other constructor, that takes a port range and an interface as well as the fingerprint

377

will automatically try to listen on a port on the given interface. For more information about

378

the parameters, see <tt class="docutils literal"><span class="pre">listen_on()</span></tt> function.</p>

379

</div>

380

381

<h2>~session()</h2>

382

<p>The destructor of session will notify all trackers that our torrents have been shut down.

383

If some trackers are down, they will time out. All this before the destructor of session

384

returns. So, it's advised that any kind of interface (such as windows) are closed before

385

destructing the session object. Because it can take a few second for it to finish. The

386

timeout can be set with <tt class="docutils literal"><span class="pre">set_settings()</span></tt>.</p>

387

</div>

388

389

<h2>abort()</h2>

390

391

session_proxy abort();

392

</pre>

393

<p>In case you want to destruct the session asynchrounously, you can request a session

394

destruction proxy. If you don't do this, the destructor of the session object will

395

block while the trackers are contacted. If you keep one <tt class="docutils literal"><span class="pre">session_proxy</span></tt> to the

396

session when destructing it, the destructor will not block, but start to close down

397

the session, the destructor of the proxy will then synchronize the threads. So, the

398

destruction of the session is performed from the <tt class="docutils literal"><span class="pre">session</span></tt> destructor call until the

399

<tt class="docutils literal"><span class="pre">session_proxy</span></tt> destructor call. The <tt class="docutils literal"><span class="pre">session_proxy</span></tt> does not have any operations

400

on it (since the session is being closed down, no operations are allowed on it). The

401

only valid operation is calling the destructor:</p>

402

403

class session_proxy

404

{

405

public:

406

session_proxy();

407

~session_proxy()

408

};

409

</pre>

410

</div>

411

412

<h2>add_torrent()</h2>

413

414

415

typedef storage_interface* (&storage_constructor_type)(

416

boost::intrusive_ptr<torrent_info const>, fs::path const&

417

, file_pool&);

418

419

torrent_handle add_torrent(

420

boost::intrusive_ptr<torrent_info> const& ti

421

, boost::filesystem::path const& save_path

422

, entry const& resume_data = entry()

423

, storage_mode_t storage_mode = storage_mode_sparse

424

, bool paused = false

425

, storage_constructor_type sc = default_storage_constructor

426

, void* userdata = 0);

427

428

torrent_handle add_torrent(

429

char const* tracker_url

430

, sha1_hash const& info_hash

431

, char const* name

432

, boost::filesystem::path const& save_path

433

, entry const& resume_data = entry()

434

, storage_mode_t storage_mode = storage_mode_sparse

435

, bool paused = false

436

, storage_constructor_type sc = default_storage_constructor

437

, void* userdata = 0);

438

</pre>

439

</blockquote>

440

<p>You add torrents through the <tt class="docutils literal"><span class="pre">add_torrent()</span></tt> function where you give an

441

object representing the information found in the torrent file and the path where you

442

want to save the files. The <tt class="docutils literal"><span class="pre">save_path</span></tt> will be prepended to the directory

443

structure in the torrent-file.</p>

444

<p>If the torrent you are trying to add already exists in the session (is either queued

445

for checking, being checked or downloading) <tt class="docutils literal"><span class="pre">add_torrent()</span></tt> will throw

446

<a class="reference internal" href="#duplicate-torrent">duplicate_torrent</a> which derives from <tt class="docutils literal"><span class="pre">std::exception</span></tt>.</p>

447

<p>The optional parameter, <tt class="docutils literal"><span class="pre">resume_data</span></tt> can be given if up to date fast-resume data

448

is available. The fast-resume data can be acquired from a running torrent by calling

449

<tt class="docutils literal"><span class="pre">torrent_handle::write_resume_data()</span></tt>. See <a class="reference internal" href="#fast-resume">fast resume</a>.</p>

450

<p>The <tt class="docutils literal"><span class="pre">storage_mode</span></tt> parameter refers to the layout of the storage for this torrent.

451

There are 3 different modes:</p>

452

453

<dt>storage_mode_sparse</dt>

454

<dd>All pieces will be written to the place where they belong and sparse files

455

will be used. This is the recommended, and default mode.</dd>

456

<dt>storage_mode_allocate</dt>

457

<dd>All pieces will be allocated, zeroes will be written to the files, before

458

the data is downloaded and written to the file. This might be useful for

459

filesystems that don't support sparse files.</dd>

460

<dt>storage_mode_compact</dt>

461

<dd>The storage will grow as more pieces are downloaded, and pieces

462

are rearranged to finally be in their correct places once the entire torrent has been

463

downloaded.</dd>

464

</dl>

465

<p>For more information, see <a class="reference internal" href="#storage-allocation">storage allocation</a>.</p>

466

<p><tt class="docutils literal"><span class="pre">paused</span></tt> is a boolean that specifies whether or not the torrent is to be started in

467

a paused state. I.e. it won't connect to the tracker or any of the peers until it's

468

resumed. This is typically a good way of avoiding race conditions when setting

469

configuration options on torrents before starting them.</p>

470

<p><tt class="docutils literal"><span class="pre">storage_constructor</span></tt> can be used to customize how the data is stored. The default

471

storage will simply write the data to the files it belongs to, but it could be

472

overridden to save everything to a single file at a specific location or encrypt the

473

content on disk for instance. For more information about the <tt class="docutils literal"><span class="pre">storage_interface</span></tt>

474

that needs to be implemented for a custom storage, see <a class="reference internal" href="#storage-interface">storage_interface</a>.</p>

475

<p>The <a class="reference internal" href="#torrent-handle">torrent_handle</a> returned by <tt class="docutils literal"><span class="pre">add_torrent()</span></tt> can be used to retrieve information

476

about the torrent's progress, its peers etc. It is also used to abort a torrent.</p>

477

<p>The <tt class="docutils literal"><span class="pre">userdata</span></tt> parameter is optional and will be passed on to the extension

478

constructor functions, if any (see <a class="reference internal" href="#add-extension">add_extension()</a>).</p>

479

<p>The second overload that takes a tracker url and an info-hash instead of metadata

480

(<tt class="docutils literal"><span class="pre">torrent_info</span></tt>) can be used with torrents where (at least some) peers support

481

the metadata extension. For the overload to be available, libtorrent must be built

482

with extensions enabled (<tt class="docutils literal"><span class="pre">TORRENT_DISABLE_EXTENSIONS</span></tt> must not be defined). It also

483

takes an optional <tt class="docutils literal"><span class="pre">name</span></tt> argument. This may be 0 in case no name should be assigned

484

to the torrent. In case it's not 0, the name is used for the torrent as long as it doesn't

485

have metadata. See <tt class="docutils literal"><span class="pre">torrent_handle::name</span></tt>.</p>

486

<p>If the torrent doesn't have a tracker, but relies on the DHT to find peers, the

487

<tt class="docutils literal"><span class="pre">tracker_url</span></tt> can be 0.</p>

488

</div>

489

490

<h2>remove_torrent()</h2>

491

492

493

void remove_torrent(torrent_handle const& h, int options = none);

494

</pre>

495

</blockquote>

496

<p><tt class="docutils literal"><span class="pre">remove_torrent()</span></tt> will close all peer connections associated with the torrent and tell

497

the tracker that we've stopped participating in the swarm. The optional second argument

498

<tt class="docutils literal"><span class="pre">options</span></tt> can be used to delete all the files downloaded by this torrent. To do this, pass

499

in the value <tt class="docutils literal"><span class="pre">session::delete_files</span></tt>. The removal of the torrent is asyncronous, there is

500

no guarantee that adding the same torrent immediately after it was removed will not throw

501

a <a class="reference internal" href="#duplicate-torrent">duplicate_torrent</a> exception.</p>

502

</div>

503

504

<h2>find_torrent() get_torrents()</h2>

505

506

507

torrent_handle find_torrent(sha_hash const& ih);

508

std::vector<torrent_handle> get_torrents() const;

509

</pre>

510

</blockquote>

511

<p><tt class="docutils literal"><span class="pre">find_torrent()</span></tt> looks for a torrent with the given info-hash. In case there

512

is such a torrent in the session, a torrent_handle to that torrent is returned.

513

In case the torrent cannot be found, an invalid torrent_handle is returned.</p>

514

<p>See <tt class="docutils literal"><span class="pre">torrent_handle::is_valid()</span></tt> to know if the torrent was found or not.</p>

515

<p><tt class="docutils literal"><span class="pre">get_torrents()</span></tt> returns a vector of torrent_handles to all the torrents

516

currently in the session.</p>

517

</div>

518

519

<h2>set_upload_rate_limit() set_download_rate_limit() upload_rate_limit() download_rate_limit()</h2>

520

521

522

void set_upload_rate_limit(int bytes_per_second);

523

void set_download_rate_limit(int bytes_per_second);

524

int upload_rate_limit() const;

525

int download_rate_limit() const;

526

</pre>

527

</blockquote>

528

<p><tt class="docutils literal"><span class="pre">set_upload_rate_limit()</span></tt> set the maximum number of bytes allowed to be

529

sent to peers per second. This bandwidth is distributed among all the peers. If

530

you don't want to limit upload rate, you can set this to -1 (the default).

531

<tt class="docutils literal"><span class="pre">set_download_rate_limit()</span></tt> works the same way but for download rate instead

532

of upload rate.

533

<tt class="docutils literal"><span class="pre">download_rate_limit()</span></tt> and <tt class="docutils literal"><span class="pre">upload_rate_limit()</span></tt> returns the previously

534

set limits.</p>

535

</div>

536

537

<h2>set_max_uploads() set_max_connections()</h2>

538

539

540

void set_max_uploads(int limit);

541

void set_max_connections(int limit);

542

</pre>

543

</blockquote>

544

<p>These functions will set a global limit on the number of unchoked peers (uploads)

545

and the number of connections opened. The number of connections is set to a hard

546

minimum of at least two connections per torrent, so if you set a too low

547

connections limit, and open too many torrents, the limit will not be met. The

548

number of uploads is at least one per torrent.</p>

549

</div>

550

551

<h2>num_uploads() num_connections()</h2>

552

553

554

int num_uploads() const;

555

int num_connections() const;

556

</pre>

557

</blockquote>

558

<p>Returns the number of currently unchoked peers and the number of connections

559

(including half-open ones) respectively.</p>

560

</div>

561

562

<h2>set_max_half_open_connections() max_half_open_connections()</h2>

563

564

565

void set_max_half_open_connections(int limit);

566

int max_half_open_connections() const;

567

</pre>

568

</blockquote>

569

<p>Sets the maximum number of half-open connections libtorrent will have when

570

connecting to peers. A half-open connection is one where connect() has been

571

called, but the connection still hasn't been established (nor failed). Windows

572

XP Service Pack 2 sets a default, system wide, limit of the number of half-open

573

connections to 10. So, this limit can be used to work nicer together with

574

other network applications on that system. The default is to have no limit,

575

and passing -1 as the limit, means to have no limit. When limiting the number

576

of simultaneous connection attempts, peers will be put in a queue waiting for

577

their turn to get connected.</p>

578

<p><tt class="docutils literal"><span class="pre">max_half_open_connections()</span></tt> returns the set limit. This limit defaults

579

to 8 on windows.</p>

580

</div>

581

582

<h2>set_ip_filter()</h2>

583

584

585

void set_ip_filter(ip_filter const& filter);

586

</pre>

587

</blockquote>

588

<p>Sets a filter that will be used to reject and accept incoming as well as outgoing

589

connections based on their originating ip address. The default filter will allow

590

connections to any ip address. To build a set of rules for which addresses are

591

accepted and not, see <a class="reference internal" href="#ip-filter">ip_filter</a>.</p>

592

<p>Each time a peer is blocked because of the IP filter, a <a class="reference internal" href="#peer-blocked-alert">peer_blocked_alert</a> is

593

generated.</p>

594

</div>

595

596

<h2>status()</h2>

597

598

599

session_status status() const;

600

</pre>

601

</blockquote>

602

<p><tt class="docutils literal"><span class="pre">status()</span></tt> returns session wide-statistics and status. The <tt class="docutils literal"><span class="pre">session_status</span></tt>

603

struct has the following members:</p>

604

605

struct session_status

606

{

607

bool has_incoming_connections;

608

609

float upload_rate;

610

float download_rate;

611

612

float payload_upload_rate;

613

float payload_download_rate;

614

615

size_type total_download;

616

size_type total_upload;

617

618

size_type total_payload_download;

619

size_type total_payload_upload;

620

621

int num_peers;

622

623

int dht_nodes;

624

int dht_cache_nodes;

625

int dht_torrents;

626

int dht_global_nodes;

627

};

628

</pre>

629

<p><tt class="docutils literal"><span class="pre">has_incoming_connections</span></tt> is false as long as no incoming connections have been

630

established on the listening socket. Every time you change the listen port, this will

631

be reset to false.</p>

632

<p><tt class="docutils literal"><span class="pre">upload_rate</span></tt>, <tt class="docutils literal"><span class="pre">download_rate</span></tt>, <tt class="docutils literal"><span class="pre">payload_download_rate</span></tt> and <tt class="docutils literal"><span class="pre">payload_upload_rate</span></tt>

633

are the total download and upload rates accumulated from all torrents. The payload

634

versions is the payload download only.</p>

635

<p><tt class="docutils literal"><span class="pre">total_download</span></tt> and <tt class="docutils literal"><span class="pre">total_upload</span></tt> are the total number of bytes downloaded and

636

uploaded to and from all torrents. <tt class="docutils literal"><span class="pre">total_payload_download</span></tt> and <tt class="docutils literal"><span class="pre">total_payload_upload</span></tt>

637

are the same thing but where only the payload is considered.</p>

638

<p><tt class="docutils literal"><span class="pre">num_peers</span></tt> is the total number of peer connections this session has. This includes

639

incoming connections that still hasn't sent their handshake or outgoing connections

640

that still hasn't completed the TCP connection. This number may be slightly higher

641

than the sum of all peers of all torrents because the incoming connections may not

642

be assigned a torrent yet.</p>

643

<p><tt class="docutils literal"><span class="pre">dht_nodes</span></tt>, <tt class="docutils literal"><span class="pre">dht_cache_nodes</span></tt> and <tt class="docutils literal"><span class="pre">dht_torrents</span></tt> are only available when

644

built with DHT support. They are all set to 0 if the DHT isn't running. When

645

the DHT is running, <tt class="docutils literal"><span class="pre">dht_nodes</span></tt> is set to the number of nodes in the routing

646

table. This number only includes <em>active</em> nodes, not cache nodes. The

647

<tt class="docutils literal"><span class="pre">dht_cache_nodes</span></tt> is set to the number of nodes in the node cache. These nodes

648

are used to replace the regular nodes in the routing table in case any of them

649

becomes unresponsive.</p>

650

<p><tt class="docutils literal"><span class="pre">dht_torrents</span></tt> are the number of torrents tracked by the DHT at the moment.</p>

651

<p><tt class="docutils literal"><span class="pre">dht_global_nodes</span></tt> is an estimation of the total number of nodes in the DHT

652

network.</p>

653

</div>

654

655

<h2>is_listening() listen_port() listen_on()</h2>

656

657

658

bool is_listening() const;

659

unsigned short listen_port() const;

660

bool listen_on(

661

std::pair<int, int> const& port_range

662

, char const* interface = 0);

663

</pre>

664

</blockquote>

665

<p><tt class="docutils literal"><span class="pre">is_listening()</span></tt> will tell you whether or not the session has successfully

666

opened a listening port. If it hasn't, this function will return false, and

667

then you can use <tt class="docutils literal"><span class="pre">listen_on()</span></tt> to make another try.</p>

668

<p><tt class="docutils literal"><span class="pre">listen_port()</span></tt> returns the port we ended up listening on. Since you just pass

669

a port-range to the constructor and to <tt class="docutils literal"><span class="pre">listen_on()</span></tt>, to know which port it

670

ended up using, you have to ask the session using this function.</p>

671

<p><tt class="docutils literal"><span class="pre">listen_on()</span></tt> will change the listen port and/or the listen interface. If the

672

session is already listening on a port, this socket will be closed and a new socket

673

will be opened with these new settings. The port range is the ports it will try

674

to listen on, if the first port fails, it will continue trying the next port within

675

the range and so on. The interface parameter can be left as 0, in that case the

676

os will decide which interface to listen on, otherwise it should be the ip-address

677

of the interface you want the listener socket bound to. <tt class="docutils literal"><span class="pre">listen_on()</span></tt> returns true

678

if it managed to open the socket, and false if it failed. If it fails, it will also

679

generate an appropriate alert (<a class="reference internal" href="#listen-failed-alert">listen_failed_alert</a>).</p>

680

<p>The interface parameter can also be a hostname that will resolve to the device you

681

want to listen on.</p>

682

<p>If you're also starting the DHT, it is a good idea to do that after you've called

683

<tt class="docutils literal"><span class="pre">listen_on()</span></tt>, since the default listen port for the DHT is the same as the tcp

684

listen socket. If you start the DHT first, it will assume the tcp port is free and

685

open the udp socket on that port, then later, when <tt class="docutils literal"><span class="pre">listen_on()</span></tt> is called, it

686

may turn out that the tcp port is in use. That results in the DHT and the bittorrent

687

socket listening on different ports. If the DHT is active when <tt class="docutils literal"><span class="pre">listen_on</span></tt> is

688

called, the udp port will be rebound to the new port, if it was configured to use

689

the same port as the tcp socket, and if the listen_on call failed to bind to the

690

same port that the udp uses.</p>

691

<p>The reason why it's a good idea to run the DHT and the bittorrent socket on the same

692

port is because that is an assumption that may be used to increase performance. One

693

way to accelerate the connecting of peers on windows may be to first ping all peers

694

with a DHT ping packet, and connect to those that responds first. On windows one

695

can only connect to a few peers at a time because of a built in limitation (in XP

696

Service pack 2).</p>

697

</div>

698

699

<h2>pop_alert() set_severity_level() wait_for_alert()</h2>

700

701

702

std::auto_ptr<alert> pop_alert();

703

alert const* wait_for_alert(time_duration max_wait);

704

void set_severity_level(alert::severity_t s);

705

</pre>

706

</blockquote>

707

<p><tt class="docutils literal"><span class="pre">pop_alert()</span></tt> is used to ask the session if any errors or events has occurred. With

708

<tt class="docutils literal"><span class="pre">set_severity_level()</span></tt> you can filter how serious the event has to be for you to

709

receive it through <tt class="docutils literal"><span class="pre">pop_alert()</span></tt>. For information, see <a class="reference internal" href="#alerts">alerts</a>.</p>

710

<p><tt class="docutils literal"><span class="pre">wait_for_alert</span></tt> blocks until an alert is available, or for no more than <tt class="docutils literal"><span class="pre">max_wait</span></tt>

711

time. If <tt class="docutils literal"><span class="pre">wait_for_alert</span></tt> returns because of the time-out, and no alerts are available,

712

it returns 0. If at least one alert was generated, a pointer to that alert is returned.

713

The alert is not popped, any subsequent calls to <tt class="docutils literal"><span class="pre">wait_for_alert</span></tt> will return the

714

same pointer until the alert is popped by calling <tt class="docutils literal"><span class="pre">pop_alert</span></tt>. This is useful for

715

leaving any alert dispatching mechanism independent of this blocking call, the dispatcher

716

can be called and it can pop the alert independently.</p>

717

</div>

718

719

<h2>add_extension()</h2>

720

721

722

void add_extension(boost::function<

723

boost::shared_ptr<torrent_plugin>(torrent*, void*)> ext);

724

</pre>

725

</blockquote>

726

<p>This function adds an extension to this session. The argument is a function

727

object that is called with a <tt class="docutils literal"><span class="pre">torrent*</span></tt> and which should return a

728

<tt class="docutils literal"><span class="pre">boost::shared_ptr<torrent_plugin></span></tt>. To write custom plugins, see

729

<a class="reference external" href="libtorrent_plugins.html">libtorrent plugins</a>. The main plugins implemented in libtorrent are:</p>

730

731

<dt>metadata extension</dt>

732

<dd>Allows peers to download the metadata (.torren files) from the swarm

733

directly. Makes it possible to join a swarm with just a tracker and

734

info-hash.</dd>

735

<dt>uTorrent peer exchange</dt>

736

<dd>Exchanges peers between clients.</dd>

737

</dl>

738

<p>To use these, imclude <tt class="docutils literal"><span class="pre"><libtorrent/extensions/metadata_transfer.hpp></span></tt>

739

or <tt class="docutils literal"><span class="pre"><libtorrent/extensions/ut_pex.hpp></span></tt>. The functions to pass in to

740

<tt class="docutils literal"><span class="pre">add_extension()</span></tt> are <tt class="docutils literal"><span class="pre">libtorrent::create_metadata_plugin</span></tt> and

741

<tt class="docutils literal"><span class="pre">libtorrent::create_ut_pex_plugin</span></tt> respectively.</p>

742

743

744

ses.add_extension(&libtorrent::create_metadata_plugin);

745

ses.add_extension(&libtorrent::create_ut_pex_plugin);

746

</pre>

747

</div>

748

749

<h2>set_settings() set_pe_settings()</h2>

750

751

752

void set_settings(session_settings const& settings);

753

void set_pe_settings(pe_settings const& settings);

754

</pre>

755

</blockquote>

756

<p>Sets the session settings and the packet encryption settings respectively.

757

See <a class="reference internal" href="#session-settings">session_settings</a> and <a class="reference internal" href="#pe-settings">pe_settings</a> for more information on available

758

options.</p>

759

</div>

760

761

<h2>set_peer_proxy() set_web_seed_proxy() set_tracker_proxy() set_dht_proxy()</h2>

762

763

764

void set_peer_proxy(proxy_settings const& s);

765

void set_web_seed_proxy(proxy_settings const& s);

766

void set_tracker_proxy(proxy_settings const& s);

767

void set_dht_proxy(proxy_settings const& s);

768

</pre>

769

</blockquote>

770

<p>The <tt class="docutils literal"><span class="pre">set_dht_proxy</span></tt> is not available when DHT is disabled. These functions

771

sets the proxy settings for different kinds of connections, bittorrent peers,

772

web seeds, trackers and the DHT traffic.</p>

773

<p><tt class="docutils literal"><span class="pre">set_peer_proxy</span></tt> affects regular bittorrent peers. <tt class="docutils literal"><span class="pre">set_web_seed_proxy</span></tt>

774

affects only web seeds. see <a class="reference internal" href="#http-seeding">HTTP seeding</a>.</p>

775

<p><tt class="docutils literal"><span class="pre">set_tracker_proxy</span></tt> only affects HTTP tracker connections (UDP tracker

776

connections are affected if the given proxy supports UDP, e.g. SOCKS5).</p>

777

<p><tt class="docutils literal"><span class="pre">set_dht_proxy</span></tt> affects the DHT messages. Since they are sent over UDP,

778

it only has any effect if the proxy supports UDP.</p>

779

<p>For more information on what settings are available for proxies, see

780

<a class="reference internal" href="#proxy-settings">proxy_settings</a>.</p>

781

</div>

782

783

<h2>peer_proxy() web_seed_proxy() tracker_proxy() dht_proxy()</h2>

784

785

786

proxy_settings const& peer_proxy() const;

787

proxy_settings const& web_seed_proxy() const;

788

proxy_settings const& tracker_proxy() const;

789

proxy_settings const& dht_proxy() const;

790

</pre>

791

</blockquote>

792

<p>These functions returns references to their respective current settings.</p>

793

<p>The <tt class="docutils literal"><span class="pre">dht_proxy</span></tt> is not available when DHT is disabled.</p>

794

</div>

795

796

<h2>start_dht() stop_dht() set_dht_settings() dht_state()</h2>

797

798

799

void start_dht(entry const& startup_state);

800

void stop_dht();

801

void set_dht_settings(dht_settings const& settings);

802

entry dht_state() const;

803

</pre>

804

</blockquote>

805

<p>These functions are not available in case <tt class="docutils literal"><span class="pre">TORRENT_DISABLE_DHT</span></tt> is

806

defined. <tt class="docutils literal"><span class="pre">start_dht</span></tt> starts the dht node and makes the trackerless service

807

available to torrents. The startup state is optional and can contain nodes

808

and the node id from the previous session. The dht node state is a bencoded

809

dictionary with the following entries:</p>

810

811

<dt><tt class="docutils literal"><span class="pre">nodes</span></tt></dt>

812

<dd>A list of strings, where each string is a node endpoint encoded in binary. If

813

the string is 6 bytes long, it is an IPv4 address of 4 bytes, encoded in

814

network byte order (big endian), followed by a 2 byte port number (also

815

network byte order). If the string is 18 bytes long, it is 16 bytes of IPv6

816

address followed by a 2 bytes port number (also network byte order).</dd>

817

818

<dd>The node id written as a readable string as a hexadecimal number.</dd>

819

</dl>

820

<p><tt class="docutils literal"><span class="pre">dht_state</span></tt> will return the current state of the dht node, this can be used

821

to start up the node again, passing this entry to <tt class="docutils literal"><span class="pre">start_dht</span></tt>. It is a good

822

idea to save this to disk when the session is closed, and read it up again

823

when starting.</p>

824

<p>If the port the DHT is supposed to listen on is already in use, and exception

825

is thrown, <tt class="docutils literal"><span class="pre">asio::error</span></tt>.</p>

826

<p><tt class="docutils literal"><span class="pre">stop_dht</span></tt> stops the dht node.</p>

827

<p><tt class="docutils literal"><span class="pre">add_dht_node</span></tt> adds a node to the routing table. This can be used if your

828

client has its own source of bootstrapping nodes.</p>

829

<p><tt class="docutils literal"><span class="pre">set_dht_settings</span></tt> sets some parameters availavle to the dht node. The

830

struct has the following members:</p>

831

832

struct dht_settings

833

{

834

int max_peers_reply;

835

int search_branching;

836

int service_port;

837

int max_fail_count;

838

};

839

</pre>

840

<p><tt class="docutils literal"><span class="pre">max_peers_reply</span></tt> is the maximum number of peers the node will send in

841

response to a <tt class="docutils literal"><span class="pre">get_peers</span></tt> message from another node.</p>

842

<p><tt class="docutils literal"><span class="pre">search_branching</span></tt> is the number of concurrent search request the node will

843

send when announcing and refreshing the routing table. This parameter is

844

called alpha in the kademlia paper.</p>

845

<p><tt class="docutils literal"><span class="pre">service_port</span></tt> is the udp port the node will listen to. This will default

846

to 0, which means the udp listen port will be the same as the tcp listen

847

port. This is in general a good idea, since some NAT implementations

848

reserves the udp port for any mapped tcp port, and vice versa. NAT-PMP

849

guarantees this for example.</p>

850

<p><tt class="docutils literal"><span class="pre">max_fail_count</span></tt> is the maximum number of failed tries to contact a node

851

before it is removed from the routing table. If there are known working nodes

852

that are ready to replace a failing node, it will be replaced immediately,

853

this limit is only used to clear out nodes that don't have any node that can

854

replace them.</p>

855

</div>

856

857

<h2>add_dht_node() add_dht_router()</h2>

858

859

860

void add_dht_node(std::pair<std::string, int> const& node);

861

void add_dht_router(std::pair<std::string, int> const& node);

862

</pre>

863

</blockquote>

864

<p><tt class="docutils literal"><span class="pre">add_dht_node</span></tt> takes a host name and port pair. That endpoint will be

865

pinged, and if a valid DHT reply is received, the node will be added to

866

the routing table.</p>

867

<p><tt class="docutils literal"><span class="pre">add_dht_router</span></tt> adds the given endpoint to a list of DHT router nodes.

868

If a search is ever made while the routing table is empty, those nodes will

869

be used as backups. Nodes in the router node list will also never be added

870

to the regular routing table, which effectively means they are only used

871

for bootstrapping, to keep the load off them.</p>

872

<p>An example routing node that you could typically add is

873

<tt class="docutils literal"><span class="pre">router.bittorrent.com</span></tt>.</p>

874

</div>

875

876

<h2>start_lsd() stop_lsd()</h2>

877

878

879

void start_lsd();

880

void stop_lsd();

881

</pre>

882

</blockquote>

883

<p>Starts and stops Local Service Discovery. This service will broadcast

884

the infohashes of all the non-private torrents on the local network to

885

look for peers on the same swarm within multicast reach.</p>

886

<p>It is turned off by default.</p>

887

</div>

888

889

<h2>start_upnp() stop_upnp()</h2>

890

891

892

void start_upnp();

893

void stop_upnp();

894

</pre>

895

</blockquote>

896

<p>Starts and stops the UPnP service. When started, the listen port and the DHT

897

port are attempted to be forwarded on local UPnP router devices.</p>

898

<p>It is off by default.</p>

899

</div>

900

901

<h2>start_natpmp() stop_natpmp()</h2>

902

903

904

void start_natpmp();

905

void stop_natpmp();

906

</pre>

907

</blockquote>

908

<p>Starts and stops the NAT-PMP service. When started, the listen port and the DHT

909

port are attempted to be forwarded on the router through NAT-PMP.</p>

910

<p>It is off by default.</p>

911

</div>

912

</div>

913

914

<h1>entry</h1>

915

<p>The <tt class="docutils literal"><span class="pre">entry</span></tt> class represents one node in a bencoded hierarchy. It works as a

916

variant type, it can be either a list, a dictionary (<tt class="docutils literal"><span class="pre">std::map</span></tt>), an integer

917

or a string. This is its synopsis:</p>

918

919

class entry

920

{

921

public:

922

923

typedef std::map<std::string, entry> dictionary_type;

924

typedef std::string string_type;

925

typedef std::list<entry> list_type;

926

typedef size_type integer_type;

927

928

enum data_type

929

{

930

int_t,

931

string_t,

932

list_t,

933

dictionary_t,

934

undefined_t

935

};

936

937

data_type type() const;

938

939

entry(dictionary_type const&);

940

entry(string_type const&);

941

entry(list_type const&);

942

entry(integer_type const&);

943

944

entry();

945

entry(data_type t);

946

entry(entry const& e);

947

~entry();

948

949

void operator=(entry const& e);

950

void operator=(dictionary_type const&);

951

void operator=(string_type const&);

952

void operator=(list_type const&);

953

void operator=(integer_type const&);

954

955

integer_type& integer();

956

integer_type const& integer() const;

957

string_type& string();

958

string_type const& string() const;

959

list_type& list();

960

list_type const& list() const;

961

dictionary_type& dict();

962

dictionary_type const& dict() const;

963

964

// these functions requires that the entry

965

// is a dictionary, otherwise they will throw

966

entry& operator[](char const* key);

967

entry& operator[](std::string const& key);

968

entry const& operator[](char const* key) const;

969

entry const& operator[](std::string const& key) const;

970

entry* find_key(char const* key);

971

entry const* find_key(char const* key) const;

972

973

void print(std::ostream& os, int indent = 0) const;

974

};

975

</pre>

976

<p><em>TODO: finish documentation of entry.</em></p>

977

978

<h2>integer() string() list() dict() type()</h2>

979

980

981

integer_type& integer();

982

integer_type const& integer() const;

983

string_type& string();

984

string_type const& string() const;

985

list_type& list();

986

list_type const& list() const;

987

dictionary_type& dict();

988

dictionary_type const& dict() const;

989

</pre>

990

</blockquote>

991

<p>The <tt class="docutils literal"><span class="pre">integer()</span></tt>, <tt class="docutils literal"><span class="pre">string()</span></tt>, <tt class="docutils literal"><span class="pre">list()</span></tt> and <tt class="docutils literal"><span class="pre">dict()</span></tt> functions

992

are accessors that return the respective type. If the <tt class="docutils literal"><span class="pre">entry</span></tt> object isn't of the

993

type you request, the accessor will throw <a class="reference internal" href="#type-error">type_error</a> (which derives from

994

<tt class="docutils literal"><span class="pre">std::runtime_error</span></tt>). You can ask an <tt class="docutils literal"><span class="pre">entry</span></tt> for its type through the

995

<tt class="docutils literal"><span class="pre">type()</span></tt> function.</p>

996

<p>The <tt class="docutils literal"><span class="pre">print()</span></tt> function is there for debug purposes only.</p>

997

<p>If you want to create an <tt class="docutils literal"><span class="pre">entry</span></tt> you give it the type you want it to have in its

998

constructor, and then use one of the non-const accessors to get a reference which you then

999

can assign the value you want it to have.</p>

1000

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

1001

1002

entry torrent_file;

1003

// ...

1004

1005

// throws if this is not a dictionary

1006

entry::dictionary_type const& dict = torrent_file.dict();

1007

entry::dictionary_type::const_iterator i;

1008

i = dict.find("announce");

1009

if (i != dict.end())

1010

{

1011

std::string tracker_url = i->second.string();

1012

std::cout << tracker_url << "\n";

1013

}

1014

</pre>

1015

<p>The following code is equivalent, but a little bit shorter:</p>

1016

1017

entry torrent_file;

1018

// ...

1019

1020

// throws if this is not a dictionary

1021

if (entry* i = torrent_file.find_key("announce"))

1022

{

1023

std::string tracker_url = i->string();

1024

std::cout << tracker_url << "\n";

1025

}

1026

</pre>

1027

<p>To make it easier to extract information from a torrent file, the class <a class="reference internal" href="#torrent-info">torrent_info</a>

1028

exists.</p>

1029

</div>

1030

1031

<h2>operator[]</h2>

1032

1033

1034

entry& operator[](char const* key);

1035

entry& operator[](std::string const& key);

1036

entry const& operator[](char const* key) const;

1037

entry const& operator[](std::string const& key) const;

1038

</pre>

1039

</blockquote>

1040

<p>All of these functions requires the entry to be a dictionary, if it isn't they

1041

will throw <tt class="docutils literal"><span class="pre">libtorrent::type_error</span></tt>.</p>

1042

<p>The non-const versions of the <tt class="docutils literal"><span class="pre">operator[]</span></tt> will return a reference to either

1043

the existing element at the given key or, if there is no element with the

1044

given key, a reference to a newly inserted element at that key.</p>

1045

<p>The const version of <tt class="docutils literal"><span class="pre">operator[]</span></tt> will only return a reference to an

1046

existing element at the given key. If the key is not found, it will throw

1047

<tt class="docutils literal"><span class="pre">libtorrent::type_error</span></tt>.</p>

1048

</div>

1049

1050

1051

1052

1053

entry* find_key(char const* key);

1054

entry const* find_key(char const* key) const;

1055

</pre>

1056

</blockquote>

1057

<p>These functions requires the entry to be a dictionary, if it isn't they

1058

will throw <tt class="docutils literal"><span class="pre">libtorrent::type_error</span></tt>.</p>

1059

<p>They will look for an element at the given key in the dictionary, if the

1060

element cannot be found, they will return 0. If an element with the given

1061

key is found, the return a pointer to it.</p>

1062

</div>

1063

</div>

1064

1065

<h1>torrent_info</h1>

1066

<p>The <tt class="docutils literal"><span class="pre">torrent_info</span></tt> has the following synopsis:</p>

1067

1068

class torrent_info

1069

{

1070

public:

1071

1072

torrent_info();

1073

torrent_info(sha1_hash const& info_hash);

1074

torrent_info(entry const& torrent_file);

1075

1076

entry create_torrent() const;

1077

void set_comment(char const* str);

1078

void set_piece_size(int size);

1079

void set_creator(char const* str);

1080

void set_hash(int index, sha1_hash const& h);

1081

void add_tracker(std::string const& url, int tier = 0);

1082

void add_file(boost::filesystem::path file, size_type size);

1083

void add_url_seed(std::string const& url);

1084

1085

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

1086

typedef std::vector<file_entry>::const_reverse_iterator

1087

reverse_file_iterator;

1088

1089

bool remap_files(std::vector<file_entry> const& map);

1090

1091

file_iterator begin_files(bool storage = false) const;

1092

file_iterator end_files(bool storage = false) const;

1093

reverse_file_iterator rbegin_files(bool storage = false) const;

1094

reverse_file_iterator rend_files(bool storage = false) const;

1095

1096

int num_files(bool storage = false) const;

1097

file_entry const& file_at(int index, bool storage = false) const;

1098

1099

std::vector<file_slice> map_block(int piece, size_type offset

1100

, int size, bool storage = false) const;

1101

peer_request map_file(int file_index, size_type file_offset

1102

, int size, bool storage = false) const;

1103

1104

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

1105

1106

bool priv() const;

1107

void set_priv(bool v);

1108

1109

std::vector<std::string> const& url_seeds() const;

1110

1111

size_type total_size() const;

1112

size_type piece_length() const;

1113

int num_pieces() const;

1114

sha1_hash const& info_hash() const;

1115

std::string const& name() const;

1116

std::string const& comment() const;

1117

std::string const& creator() const;

1118

1119

std::vector<std::pair<std::string, int> > const& nodes() const;

1120

void add_node(std::pair<std::string, int> const& node);

1121

1122

boost::optional<boost::posix_time::ptime>

1123

creation_date() const;

1124

1125

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

1126

1127

size_type piece_size(unsigned int index) const;

1128

sha1_hash const& hash_for_piece(unsigned int index) const;

1129

};

1130

</pre>

1131

1132

<h2>torrent_info()</h2>

1133

1134

1135

torrent_info();

1136

torrent_info(sha1_hash const& info_hash);

1137

torrent_info(entry const& torrent_file);

1138

</pre>

1139

</blockquote>

1140

<p>The default constructor of <tt class="docutils literal"><span class="pre">torrent_info</span></tt> is used when creating torrent files. It will

1141

initialize the object to an empty torrent, containing no files. The info hash will be set

1142

to 0 when this constructor is used. To use the empty <tt class="docutils literal"><span class="pre">torrent_info</span></tt> object, add files

1143

and piece hashes, announce URLs and optionally a creator tag and comment. To do this you

1144

use the members <tt class="docutils literal"><span class="pre">set_comment()</span></tt>, <tt class="docutils literal"><span class="pre">set_piece_size()</span></tt>, <tt class="docutils literal"><span class="pre">set_creator()</span></tt>, <tt class="docutils literal"><span class="pre">set_hash()</span></tt>

1145

etc.</p>

1146

<p>The constructor that takes an info-hash is identical to the default constructor with the

1147

exception that it will initialize the info-hash to the given value. This is used internally

1148

when downloading torrents without the metadata. The metadata will be created by libtorrent

1149

as soon as it has been downloaded from the swarm.</p>

1150

<p>The last constructor is the one that is used in most cases. It will create a <tt class="docutils literal"><span class="pre">torrent_info</span></tt>

1151

object from the information found in the given torrent_file. The <tt class="docutils literal"><span class="pre">entry</span></tt> represents a tree

1152

node in an bencoded file. To load an ordinary .torrent file into an <tt class="docutils literal"><span class="pre">entry</span></tt>, use bdecode(),

1153

see <a class="reference internal" href="#bdecode-bencode">bdecode() bencode()</a>.</p>

1154

</div>

1155

1156

<h2>set_comment() set_piece_size() set_creator() set_hash() add_tracker() add_file()</h2>

1157

1158

1159

void set_comment(char const* str);

1160

void set_piece_size(int size);

1161

void set_creator(char const* str);

1162

void set_hash(int index, sha1_hash const& h);

1163

void add_tracker(std::string const& url, int tier = 0);

1164

void add_file(boost::filesystem::path file, size_type size);

1165

</pre>

1166

</blockquote>

1167

<p>These files are used when creating a torrent file. <tt class="docutils literal"><span class="pre">set_comment()</span></tt> will simply set

1168

the comment that belongs to this torrent. The comment can be retrieved with the

1169

<tt class="docutils literal"><span class="pre">comment()</span></tt> member. The string should be UTF-8 encoded.</p>

1170

<p><tt class="docutils literal"><span class="pre">set_piece_size()</span></tt> will set the size of each piece in this torrent. The piece size must

1171

be an even multiple of 2. i.e. usually something like 256 kiB, 512 kiB, 1024 kiB etc. The

1172

size is given in number of bytes.</p>

1173

<p><tt class="docutils literal"><span class="pre">set_creator()</span></tt> is an optional attribute that can be used to identify your application

1174

that was used to create the torrent file. The string should be UTF-8 encoded.</p>

1175

<p><tt class="docutils literal"><span class="pre">set_hash()</span></tt> writes the hash for the piece with the given piece-index. You have to call

1176

this function for every piece in the torrent. Usually the <a class="reference internal" href="#hasher">hasher</a> is used to calculate

1177

the sha1-hash for a piece.</p>

1178

<p><tt class="docutils literal"><span class="pre">add_tracker()</span></tt> adds a tracker to the announce-list. The <tt class="docutils literal"><span class="pre">tier</span></tt> determines the order in

1179

which the trackers are to be tried. For more information see <a class="reference internal" href="#trackers">trackers()</a>.</p>

1180

<p><tt class="docutils literal"><span class="pre">add_file()</span></tt> adds a file to the torrent. The order in which you add files will determine

1181

the order in which they are placed in the torrent file. You have to add at least one file

1182

to the torrent. The <tt class="docutils literal"><span class="pre">path</span></tt> you give has to be a relative path from the root directory

1183

of the torrent. The <tt class="docutils literal"><span class="pre">size</span></tt> is given in bytes.</p>

1184

<p>When you have added all the files and hashes to your torrent, you can generate an <tt class="docutils literal"><span class="pre">entry</span></tt>

1185

which then can be encoded as a .torrent file. You do this by calling <a class="reference internal" href="#create-torrent">create_torrent()</a>.</p>

1186

<p>For a complete example of how to create a torrent from a file structure, see <a class="reference external" href="examples.html#make_torrent">make_torrent</a>.</p>

1187

</div>

1188

1189

<h2>create_torrent()</h2>

1190

1191

1192

entry create_torrent();

1193

</pre>

1194

</blockquote>

1195

<p>Returns an <tt class="docutils literal"><span class="pre">entry</span></tt> representing the bencoded tree of data that makes up a .torrent file.

1196

You can save this data as a torrent file with bencode() (see <a class="reference internal" href="#bdecode-bencode">bdecode() bencode()</a>), for a

1197

complete example, see <a class="reference external" href="examples.html#make_torrent">make_torrent</a>.</p>

1198

<p>This function is not const because it will also set the info-hash of the <tt class="docutils literal"><span class="pre">torrent_info</span></tt>

1199

object.</p>

1200

<p>Note that a torrent file must include at least one file, and it must have at

1201

least one tracker url or at least one DHT node.</p>

1202

</div>

1203

1204

<h2>remap_files()</h2>

1205

1206

1207

bool remap_files(std::vector<file_entry> const& map);

1208

</pre>

1209

</blockquote>

1210

<p>This call will create a new mapping of the data in this torrent to other files. The

1211

<tt class="docutils literal"><span class="pre">torrent_info</span></tt> maintains 2 views of the file storage. One that is true to the torrent

1212

file, and one that represents what is actually saved on disk. This call will change

1213

what the files on disk are called.</p>

1214

<p>The each entry in the vector <tt class="docutils literal"><span class="pre">map</span></tt> is a <tt class="docutils literal"><span class="pre">file_entry</span></tt>. The only fields in this struct

1215

that are used in this case are <tt class="docutils literal"><span class="pre">path</span></tt>, <tt class="docutils literal"><span class="pre">size</span></tt> and <tt class="docutils literal"><span class="pre">file_base</span></tt>.</p>

1216

<p>The return value indicates if the remap was successful or not. True means success and

1217

false means failure. The sum of all the files passed in through <tt class="docutils literal"><span class="pre">map</span></tt> has to be exactly

1218

the same as the total_size of the torrent. If the number of bytes that are mapped do not

1219

match, false will be returned (this is the only case this function may fail).</p>

1220

<p>Changing this mapping for an existing torrent will not move or rename files. If some files

1221

should be renamed, this can be done before the torrent is added.</p>

1222

</div>

1223

1224

<h2>begin_files() end_files() rbegin_files() rend_files()</h2>

1225

1226

1227

file_iterator begin_files(bool storage = false) const;

1228

file_iterator end_files(bool storage = false) const;

1229

reverse_file_iterator rbegin_files(bool storage = false) const;

1230

reverse_file_iterator rend_files(bool storage = false) const;

1231

</pre>

1232

</blockquote>

1233

<p>This class will need some explanation. First of all, to get a list of all files

1234

in the torrent, you can use <tt class="docutils literal"><span class="pre">begin_files()</span></tt>, <tt class="docutils literal"><span class="pre">end_files()</span></tt>,

1235

<tt class="docutils literal"><span class="pre">rbegin_files()</span></tt> and <tt class="docutils literal"><span class="pre">rend_files()</span></tt>. These will give you standard vector

1236

iterators with the type <tt class="docutils literal"><span class="pre">file_entry</span></tt>.</p>

1237

<p>The <tt class="docutils literal"><span class="pre">storage</span></tt> parameter specifies which view of the files you want. The default

1238

is false, which means you will see the content of the torrent file. If set to

1239

true, you will see the file that the storage class uses to save the files to

1240

disk. Typically these views are the same, but in case the files have been

1241

remapped, they may differ. For more info, see <a class="reference internal" href="#remap-files">remap_files()</a>.</p>

1242

1243

struct file_entry

1244

{

1245

boost::filesystem::path path;

1246

size_type offset;

1247

size_type size;

1248

size_type file_base;

1249

boost::shared_ptr<const boost::filesystem::path> orig_path;

1250

};

1251

</pre>

1252

<p>The <tt class="docutils literal"><span class="pre">path</span></tt> is the full (relative) path of each file. i.e. if it is a multi-file

1253

torrent, all the files starts with a directory with the same name as <tt class="docutils literal"><span class="pre">torrent_info::name()</span></tt>.

1254

The filenames are encoded with UTF-8.</p>

1255

<p><tt class="docutils literal"><span class="pre">size</span></tt> is the size of the file (in bytes) and <tt class="docutils literal"><span class="pre">offset</span></tt> is the byte offset

1256

of the file within the torrent. i.e. the sum of all the sizes of the files

1257

before it in the list.</p>

1258

<p><tt class="docutils literal"><span class="pre">file_base</span></tt> is the offset in the file where the storage should start. The normal

1259

case is to have this set to 0, so that the storage starts saving data at the start

1260

if the file. In cases where multiple files are mapped into the same file though,

1261

the <tt class="docutils literal"><span class="pre">file_base</span></tt> should be set to an offset so that the different regions do

1262

not overlap. This is used when mapping "unselected" files into a so-called part

1263

file.</p>

1264

<p><tt class="docutils literal"><span class="pre">orig_path</span></tt> is set to 0 in case the path element is an exact copy of that

1265

found in the metadata. In case the path in the original metadata was

1266

incorrectly encoded, and had to be fixed in order to be acceptable utf-8,

1267

the original string is preserved in <tt class="docutils literal"><span class="pre">orig_path</span></tt>. The reason to keep it

1268

is to be able to reproduce the info-section exactly, with the correct

1269

info-hash.</p>

1270

</div>

1271

1272

<h2>num_files() file_at()</h2>

1273

1274

1275

int num_files(bool storage = false) const;

1276

file_entry const& file_at(int index, bool storage = false) const;

1277

</pre>

1278

</blockquote>

1279

<p>If you need index-access to files you can use the <tt class="docutils literal"><span class="pre">num_files()</span></tt> and <tt class="docutils literal"><span class="pre">file_at()</span></tt>

1280

to access files using indices.</p>

1281

<p>The <tt class="docutils literal"><span class="pre">storage</span></tt> parameter specifies which view of the files you want. The default

1282

is false, which means you will see the content of the torrent file. If set to

1283

true, you will see the file that the storage class uses to save the files to

1284

disk. Typically these views are the same, but in case the files have been

1285

remapped, they may differ. For more info, see <a class="reference internal" href="#remap-files">remap_files()</a>.</p>

1286

</div>

1287

1288

<h2>map_block()</h2>

1289

1290

1291

std::vector<file_slice> map_block(int piece, size_type offset

1292

, int size, bool storage = false) const;

1293

</pre>

1294

</blockquote>

1295

<p>This function will map a piece index, a byte offset within that piece and

1296

a size (in bytes) into the corresponding files with offsets where that data

1297

for that piece is supposed to be stored.</p>

1298

<p>The file slice struct looks like this:</p>

1299

1300

struct file_slice

1301

{

1302

int file_index;

1303

size_type offset;

1304

size_type size;

1305

};

1306

</pre>

1307

<p>The <tt class="docutils literal"><span class="pre">file_index</span></tt> refers to the index of the file (in the torrent_info).

1308

To get the path and filename, use <tt class="docutils literal"><span class="pre">file_at()</span></tt> and give the <tt class="docutils literal"><span class="pre">file_index</span></tt>

1309

as argument. The <tt class="docutils literal"><span class="pre">offset</span></tt> is the byte offset in the file where the range

1310

starts, and <tt class="docutils literal"><span class="pre">size</span></tt> is the number of bytes this range is. The size + offset

1311

will never be greater than the file size.</p>

1312

<p>The <tt class="docutils literal"><span class="pre">storage</span></tt> parameter specifies which view of the files you want. The default

1313

is false, which means you will see the content of the torrent file. If set to

1314

true, you will see the file that the storage class uses to save the files to

1315

disk. Typically these views are the same, but in case the files have been

1316

remapped, they may differ. For more info, see <a class="reference internal" href="#remap-files">remap_files()</a>.</p>

1317

</div>

1318

1319

1320

1321

1322

peer_request map_file(int file_index, size_type file_offset

1323

, int size, bool storage = false) const;

1324

</pre>

1325

</blockquote>

1326

<p>This function will map a range in a specific file into a range in the torrent.

1327

The <tt class="docutils literal"><span class="pre">file_offset</span></tt> parameter is the offset in the file, given in bytes, where

1328

0 is the start of the file.

1329

The <tt class="docutils literal"><span class="pre">peer_request</span></tt> structure looks like this:</p>

1330

1331

struct peer_request

1332

{

1333

int piece;

1334

int start;

1335

int length;

1336

bool operator==(peer_request const& r) const;

1337

};

1338

</pre>

1339

<p><tt class="docutils literal"><span class="pre">piece</span></tt> is the index of the piece in which the range starts.

1340

<tt class="docutils literal"><span class="pre">start</span></tt> is the offset within that piece where the range starts.

1341

<tt class="docutils literal"><span class="pre">length</span></tt> is the size of the range, in bytes.</p>

1342

<p>The input range is assumed to be valid within the torrent. <tt class="docutils literal"><span class="pre">file_offset</span></tt>

1343

+ <tt class="docutils literal"><span class="pre">size</span></tt> is not allowed to be greater than the file size. <tt class="docutils literal"><span class="pre">file_index</span></tt>

1344

must refer to a valid file, i.e. it cannot be >= <tt class="docutils literal"><span class="pre">num_files()</span></tt>.</p>

1345

</div>

1346

1347

<h2>url_seeds() add_url_seed()</h2>

1348

1349

1350

std::vector<std::string> const& url_seeds() const;

1351

void add_url_seed(std::string const& url);

1352

</pre>

1353

</blockquote>

1354

<p>If there are any url-seeds in this torrent, <tt class="docutils literal"><span class="pre">url_seeds()</span></tt> will return a

1355

vector of those urls. If you're creating a torrent file, <tt class="docutils literal"><span class="pre">add_url_seed()</span></tt>

1356

adds one url to the list of url-seeds. Currently, the only transport protocol

1357

supported for the url is http.</p>

1358

<p>The <tt class="docutils literal"><span class="pre">storage</span></tt> parameter specifies which view of the files you want. The default

1359

is false, which means you will see the content of the torrent file. If set to

1360

true, you will see the file that the storage class uses to save the files to

1361

disk. Typically these views are the same, but in case the files have been

1362

remapped, they may differ. For more info, see <a class="reference internal" href="#remap-files">remap_files()</a>.</p>

1363

<p>See <a class="reference internal" href="#http-seeding">HTTP seeding</a> for more information.</p>

1364

</div>

1365

1366

<h2>print()</h2>

1367

1368

1369

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

1370

</pre>

1371

</blockquote>

1372

<p>The <tt class="docutils literal"><span class="pre">print()</span></tt> function is there for debug purposes only. It will print the info from

1373

the torrent file to the given outstream. This function has been deprecated and will

1374

be removed from future releases.</p>

1375

</div>

1376

1377

<h2>trackers()</h2>

1378

1379

1380

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

1381

</pre>

1382

</blockquote>

1383

<p>The <tt class="docutils literal"><span class="pre">trackers()</span></tt> function will return a sorted vector of <tt class="docutils literal"><span class="pre">announce_entry</span></tt>.

1384

Each announce entry contains a string, which is the tracker url, and a tier index. The

1385

tier index is the high-level priority. No matter which trackers that works or not, the

1386

ones with lower tier will always be tried before the one with higher tier number.</p>

1387

1388

struct announce_entry

1389

{

1390

announce_entry(std::string const& url);

1391

std::string url;

1392

int tier;

1393

};

1394

</pre>

1395

</div>

1396

1397

<h2>total_size() piece_length() piece_size() num_pieces()</h2>

1398

1399

1400

size_type total_size() const;

1401

size_type piece_length() const;

1402

size_type piece_size(unsigned int index) const;

1403

int num_pieces() const;

1404

</pre>

1405

</blockquote>

1406

<p><tt class="docutils literal"><span class="pre">total_size()</span></tt>, <tt class="docutils literal"><span class="pre">piece_length()</span></tt> and <tt class="docutils literal"><span class="pre">num_pieces()</span></tt> returns the total

1407

number of bytes the torrent-file represents (all the files in it), the number of byte for

1408

each piece and the total number of pieces, respectively. The difference between

1409

<tt class="docutils literal"><span class="pre">piece_size()</span></tt> and <tt class="docutils literal"><span class="pre">piece_length()</span></tt> is that <tt class="docutils literal"><span class="pre">piece_size()</span></tt> takes

1410

the piece index as argument and gives you the exact size of that piece. It will always

1411

be the same as <tt class="docutils literal"><span class="pre">piece_length()</span></tt> except in the case of the last piece, which may

1412

be smaller.</p>

1413

</div>

1414

1415

<h2>hash_for_piece() info_hash()</h2>

1416

1417

1418

size_type piece_size(unsigned int index) const;

1419

sha1_hash const& hash_for_piece(unsigned int index) const;

1420

</pre>

1421

</blockquote>

1422

<p><tt class="docutils literal"><span class="pre">hash_for_piece()</span></tt> takes a piece-index and returns the 20-bytes sha1-hash for that

1423

piece and <tt class="docutils literal"><span class="pre">info_hash()</span></tt> returns the 20-bytes sha1-hash for the info-section of the

1424

torrent file. For more information on the <tt class="docutils literal"><span class="pre">sha1_hash</span></tt>, see the <a class="reference internal" href="#big-number">big_number</a> class.

1425

<tt class="docutils literal"><span class="pre">info_hash()</span></tt> will only return a valid hash if the torrent_info was read from a

1426

<tt class="docutils literal"><span class="pre">.torrent</span></tt> file or if an <tt class="docutils literal"><span class="pre">entry</span></tt> was created from it (through <tt class="docutils literal"><span class="pre">create_torrent</span></tt>).</p>

1427

</div>

1428

1429

<h2>name() comment() creation_date() creator()</h2>

1430

1431

1432

std::string const& name() const;

1433

std::string const& comment() const;

1434

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

1435

</pre>

1436

</blockquote>

1437

<p><tt class="docutils literal"><span class="pre">name()</span></tt> returns the name of the torrent.</p>

1438

<p><tt class="docutils literal"><span class="pre">comment()</span></tt> returns the comment associated with the torrent. If there's no comment,

1439

it will return an empty string. <tt class="docutils literal"><span class="pre">creation_date()</span></tt> returns a <a class="reference external" href="http://www.boost.org/doc/html/date_time/posix_time.html#date_time.posix_time.ptime_class">boost::posix_time::ptime</a>

1440

object, representing the time when this torrent file was created. If there's no time stamp

1441

in the torrent file, this will return a date of January 1:st 1970.</p>

1442

<p>Both the name and the comment is UTF-8 encoded strings.</p>

1443

<p><tt class="docutils literal"><span class="pre">creator()</span></tt> returns the creator string in the torrent. If there is no creator string

1444

it will return an empty string.</p>

1445

</div>

1446

1447

1448

1449

1450

bool priv() const;

1451

void set_priv(bool v);

1452

</pre>

1453

</blockquote>

1454

<p><tt class="docutils literal"><span class="pre">priv()</span></tt> returns true if this torrent is private. i.e., it should not be

1455

distributed on the trackerless network (the kademlia DHT).</p>

1456

<p><tt class="docutils literal"><span class="pre">set_priv()</span></tt> sets or clears the private flag on this torrent.</p>

1457

</div>

1458

1459

<h2>nodes()</h2>

1460

1461

1462

std::vector<std::pair<std::string, int> > const& nodes() const;

1463

</pre>

1464

</blockquote>

1465

<p>If this torrent contains any DHT nodes, they are put in this vector in their original

1466

form (host name and port number).</p>

1467

</div>

1468

1469

1470

1471

1472

void add_node(std::pair<std::string, int> const& node);

1473

</pre>

1474

</blockquote>

1475

<p>This is used when creating torrent. Use this to add a known DHT node. It may

1476

be used, by the client, to bootstrap into the DHT network.</p>

1477

</div>

1478

</div>

1479

1480

<h1>torrent_handle</h1>

1481

<p>You will usually have to store your torrent handles somewhere, since it's the

1482

object through which you retrieve information about the torrent and aborts the torrent.

1483

Its declaration looks like this:</p>

1484

1485

struct torrent_handle

1486

{

1487

torrent_handle();

1488

1489

torrent_status status();

1490

void file_progress(std::vector<float>& fp);

1491

void get_download_queue(std::vector<partial_piece_info>& queue) const;

1492

void get_peer_info(std::vector<peer_info>& v) const;

1493

torrent_info const& get_torrent_info() const;

1494

bool is_valid() const;

1495

1496

std::string name() const;

1497

1498

entry write_resume_data() const;

1499

void force_reannounce() const;

1500

void force_reannounce(boost::posix_time::time_duration) const;

1501

void scrape_tracker() const;

1502

void connect_peer(asio::ip::tcp::endpoint const& adr, int source = 0) const;

1503

1504

void set_tracker_login(std::string const& username

1505

, std::string const& password) const;

1506

1507

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

1508

void replace_trackers(std::vector<announce_entry> const&);

1509

1510

void add_url_seed(std::string const& url);

1511

void remove_url_seed(std::string const& url);

1512

std::set<std::string> url_seeds() const;

1513

1514

void set_ratio(float ratio) const;

1515

void set_max_uploads(int max_uploads) const;

1516

void set_max_connections(int max_connections) const;

1517

void set_upload_limit(int limit) const;

1518

int upload_limit() const;

1519

void set_download_limit(int limit) const;

1520

int download_limit() const;

1521

void set_sequenced_download_threshold(int threshold) const;

1522

1523

void set_peer_upload_limit(asio::ip::tcp::endpoint ip, int limit) const;

1524

void set_peer_download_limit(asio::ip::tcp::endpoint ip, int limit) const;

1525

1526

void use_interface(char const* net_interface) const;

1527

1528

void pause() const;

1529

void resume() const;

1530

bool is_paused() const;

1531

bool is_seed() const;

1532

1533

void resolve_countries(bool r);

1534

bool resolve_countries() const;

1535

1536

void piece_priority(int index, int priority) const;

1537

int piece_priority(int index) const;

1538

1539

void prioritize_pieces(std::vector<int> const& pieces) const;

1540

std::vector<int> piece_priorities() const;

1541

1542

void prioritize_files(std::vector<int> const& files) const;

1543

1544

// these functions are deprecated

1545

void filter_piece(int index, bool filter) const;

1546

void filter_pieces(std::vector<bool> const& bitmask) const;

1547

bool is_piece_filtered(int index) const;

1548

std::vector<bool> filtered_pieces() const;

1549

void filter_files(std::vector<bool> const& files) const;

1550

1551

bool has_metadata() const;

1552

1553

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

1554

void move_storage(boost::filesystem::path const& save_path) const;

1555

1556

sha1_hash info_hash() const;

1557

1558

bool operator==(torrent_handle const&) const;

1559

bool operator!=(torrent_handle const&) const;

1560

bool operator<(torrent_handle const&) const;

1561

};

1562

</pre>

1563

<p>The default constructor will initialize the handle to an invalid state. Which

1564

means you cannot perform any operation on it, unless you first assign it a

1565

valid handle. If you try to perform any operation on an uninitialized handle,

1566

it will throw <tt class="docutils literal"><span class="pre">invalid_handle</span></tt>.</p>

1567

1568

<p class="first admonition-title">Warning</p>

1569

<p class="last">All operations on a <tt class="docutils literal"><span class="pre">torrent_handle</span></tt> may throw <a class="reference internal" href="#invalid-handle">invalid_handle</a>

1570

exception, in case the handle is no longer refering to a torrent. There are

1571

two exceptions, <tt class="docutils literal"><span class="pre">info_hash()</span></tt> and <tt class="docutils literal"><span class="pre">is_valid()</span></tt> will never throw.

1572

Since the torrents are processed by a background thread, there is no

1573

guarantee that a handle will remain valid between two calls.</p>

1574

</div>

1575

1576

<h2>piece_priority() prioritize_pieces() piece_priorities() prioritize_files()</h2>

1577

1578

1579

void piece_priority(int index, int priority) const;

1580

int piece_priority(int index) const;

1581

void prioritize_pieces(std::vector<int> const& pieces) const;

1582

std::vector<int> piece_priorities() const;

1583

void prioritize_files(std::vector<int> const& files) const;

1584

</pre>

1585

</blockquote>

1586

<p>These functions are used to set and get the prioritiy of individual pieces.

1587

By default all pieces have priority 1. That means that the random rarest

1588

first algorithm is effectively active for all pieces. You may however

1589

change the priority of individual pieces. There are 8 different priority

1590

levels:</p>

1591

1592

1593

<li>piece is not downloaded at all</li>

1594

<li>normal priority. Download order is dependent on availability</li>

1595

<li>higher than normal priority. Pieces are preferred over pieces with

1596

the same availability, but not over pieces with lower availability</li>

1597

<li>pieces are as likely to be picked as partial pieces.</li>

1598

<li>pieces are preferred over partial pieces, but not over pieces with

1599

lower availability</li>

1600

<li><em>currently the same as 4</em></li>

1601

<li>piece is as likely to be picked as any piece with availability 1</li>

1602

<li>maximum priority, availability is disregarded, the piece is preferred

1603

over any other piece with lower priority</li>

1604

</ol>

1605

</blockquote>

1606

<p>The exact definitions of these priorities are implementation details, and

1607

subject to change. The interface guarantees that higher number means higher

1608

priority, and that 0 means do not download.</p>

1609

<p><tt class="docutils literal"><span class="pre">piece_priority</span></tt> sets or gets the priority for an individual piece,

1610

specified by <tt class="docutils literal"><span class="pre">index</span></tt>.</p>

1611

<p><tt class="docutils literal"><span class="pre">prioritize_pieces</span></tt> takes a vector of integers, one integer per piece in

1612

the torrent. All the piece priorities will be updated with the priorities

1613

in the vector.</p>

1614

<p><tt class="docutils literal"><span class="pre">piece_priorities</span></tt> returns a vector with one element for each piece in the

1615

torrent. Each element is the current priority of that piece.</p>

1616

<p><tt class="docutils literal"><span class="pre">prioritize_files</span></tt> takes a vector that has at as many elements as there are

1617

files in the torrent. Each entry is the priority of that file. The function

1618

sets the priorities of all the pieces in the torrent based on the vector.</p>

1619

</div>

1620

1621

<h2>file_progress()</h2>

1622

1623

1624

void file_progress(std::vector<float>& fp);

1625

</pre>

1626

</blockquote>

1627

<p>This function fills in the supplied vector with the progress (a value in the

1628

range [0, 1]) describing the download progress of each file in this torrent.

1629

The progress values are ordered the same as the files in the <a class="reference internal" href="#torrent-info">torrent_info</a>.

1630

This operation is not very cheap.</p>

1631

</div>

1632

1633

1634

1635

1636

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

1637

</pre>

1638

</blockquote>

1639

<p><tt class="docutils literal"><span class="pre">save_path()</span></tt> returns the path that was given to <a class="reference internal" href="#add-torrent">add_torrent()</a> when this torrent

1640

was started.</p>

1641

</div>

1642

1643

<h2>move_storage()</h2>

1644

1645

1646

void move_storage(boost::filesystem::path const& save_path) const;

1647

</pre>

1648

</blockquote>

1649

<p>Moves the file(s) that this torrent are currently seeding from or downloading to. This

1650

operation will only have the desired effect if the given <tt class="docutils literal"><span class="pre">save_path</span></tt> is located on

1651

the same drive as the original save path. Since disk IO is performed in a separate

1652

thread, this operation is also asynchronous. Once the operation completes, the

1653

<tt class="docutils literal"><span class="pre">storage_moved_alert</span></tt> is generated, with the new path as the message.</p>

1654

</div>

1655

1656

<h2>force_reannounce()</h2>

1657

1658

1659

void force_reannounce() const;

1660

void force_reannounce(boost::posix_time::time_duration) const;

1661

</pre>

1662

</blockquote>

1663

<p><tt class="docutils literal"><span class="pre">force_reannounce()</span></tt> will force this torrent to do another tracker request, to receive new

1664

peers. The second overload of <tt class="docutils literal"><span class="pre">force_reannounce</span></tt> that takes a <tt class="docutils literal"><span class="pre">time_duration</span></tt> as

1665

argument will schedule a reannounce in that amount of time from now.</p>

1666

</div>

1667

1668

<h2>scrape_tracker()</h2>

1669

1670

1671

void scrape_tracker() const;

1672

</pre>

1673

</blockquote>

1674

<p><tt class="docutils literal"><span class="pre">scrape_tracker()</span></tt> will send a scrape request to the tracker. A scrape request queries the

1675

tracker for statistics such as total number of incomplete peers, complete peers, number of

1676

downloads etc.</p>

1677

<p>This request will specifically update the <tt class="docutils literal"><span class="pre">num_complete</span></tt> and <tt class="docutils literal"><span class="pre">num_incomplete</span></tt> fields in

1678

the <a class="reference internal" href="#torrent-status">torrent_status</a> struct once it completes. When it completes, it will generate a

1679

<a class="reference internal" href="#scrape-reply-alert">scrape_reply_alert</a>. If it fails, it will generate a <a class="reference internal" href="#scrape-failed-alert">scrape_failed_alert</a>.</p>

1680

</div>

1681

1682

<h2>connect_peer()</h2>

1683

1684

1685

void connect_peer(asio::ip::tcp::endpoint const& adr, int source = 0) const;

1686

</pre>

1687

</blockquote>

1688

<p><tt class="docutils literal"><span class="pre">connect_peer()</span></tt> is a way to manually connect to peers that one believe is a part of the

1689

torrent. If the peer does not respond, or is not a member of this torrent, it will simply

1690

be disconnected. No harm can be done by using this other than an unnecessary connection

1691

attempt is made. If the torrent is uninitialized or in queued or checking mode, this

1692

will throw <a class="reference internal" href="#invalid-handle">invalid_handle</a>. The second (optional) argument will be bitwised ORed into

1693

the source mask of this peer. Typically this is one of the source flags in <a class="reference internal" href="#peer-info">peer_info</a>.

1694

i.e. <tt class="docutils literal"><span class="pre">tracker</span></tt>, <tt class="docutils literal"><span class="pre">pex</span></tt>, <tt class="docutils literal"><span class="pre">dht</span></tt> etc.</p>

1695

</div>

1696

1697

1698

1699

1700

std::string name() const;

1701

</pre>

1702

</blockquote>

1703

<p>Returns the name of the torrent. i.e. the name from the metadata associated with it. In

1704

case the torrent was started without metadata, and hasn't completely received it yet,

1705

it returns the name given to it when added to the session. See <tt class="docutils literal"><span class="pre">session::add_torrent</span></tt>.</p>

1706

</div>

1707

1708

<h2>set_ratio()</h2>

1709

1710

1711

void set_ratio(float ratio) const;

1712

</pre>

1713

</blockquote>

1714

<p><tt class="docutils literal"><span class="pre">set_ratio()</span></tt> sets the desired download / upload ratio. If set to 0, it is considered being

1715

infinite. i.e. the client will always upload as much as it can, no matter how much it gets back

1716

in return. With this setting it will work much like the standard clients.</p>

1717

<p>Besides 0, the ratio can be set to any number greater than or equal to 1. It means how much to

1718

attempt to upload in return for each download. e.g. if set to 2, the client will try to upload

1719

2 bytes for every byte received. The default setting for this is 0, which will make it work

1720

as a standard client.</p>

1721

</div>

1722

1723

<h2>set_upload_limit() set_download_limit() upload_limit() download_limit()</h2>

1724

1725

1726

void set_upload_limit(int limit) const;

1727

void set_download_limit(int limit) const;

1728

int upload_limit() const;

1729

int download_limit() const;

1730

</pre>

1731

</blockquote>

1732

<p><tt class="docutils literal"><span class="pre">set_upload_limit</span></tt> will limit the upload bandwidth used by this particular torrent to the

1733

limit you set. It is given as the number of bytes per second the torrent is allowed to upload.

1734

<tt class="docutils literal"><span class="pre">set_download_limit</span></tt> works the same way but for download bandwidth instead of upload bandwidth.

1735

Note that setting a higher limit on a torrent then the global limit (<tt class="docutils literal"><span class="pre">session::set_upload_rate_limit</span></tt>)

1736

will not override the global rate limit. The torrent can never upload more than the global rate

1737

limit.</p>

1738

<p><tt class="docutils literal"><span class="pre">upload_limit</span></tt> and <tt class="docutils literal"><span class="pre">download_limit</span></tt> will return the current limit setting, for upload and

1739

download, respectively.</p>

1740

</div>

1741

1742

<h2>set_sequenced_download_threshold()</h2>

1743

1744

1745

void set_sequenced_download_threshold(int threshold);

1746

</pre>

1747

</blockquote>

1748

<p>sequenced-download threshold is the limit on how popular a piece has to be

1749

(popular == inverse of rarity) to be downloaded in sequence instead of in

1750

random (rarest first) order. It can be used to tweak disk performance in

1751

settings where the random download property is less necessary. For example, if

1752

the threshold is 10, all pieces which 10 or more peers have, will be downloaded

1753

in index order. This setting defaults to 100, which means that it is disabled

1754

in practice.</p>

1755

<p>Setting this threshold to a very small value will affect the piece distribution

1756

negatively in the swarm. It should basically only be used in situations where

1757

the random seeks on the disk is the download bottleneck.</p>

1758

</div>

1759

1760

<h2>set_peer_upload_limit() set_peer_download_limit()</h2>

1761

1762

1763

void set_peer_upload_limit(asio::ip::tcp::endpoint ip, int limit) const;

1764

void set_peer_download_limit(asio::ip::tcp::endpoint ip, int limit) const;

1765

</pre>

1766

</blockquote>

1767

<p>Works like <tt class="docutils literal"><span class="pre">set_upload_limit</span></tt> and <tt class="docutils literal"><span class="pre">set_download_limit</span></tt> respectively, but controls individual

1768

peer instead of the whole torrent.</p>

1769

</div>

1770

1771

<h2>pause() resume() is_paused()</h2>

1772

1773

1774

void pause() const;

1775

void resume() const;

1776

bool is_paused() const;

1777

</pre>

1778

</blockquote>

1779

<p><tt class="docutils literal"><span class="pre">pause()</span></tt>, and <tt class="docutils literal"><span class="pre">resume()</span></tt> will disconnect all peers and reconnect all peers respectively.

1780

When a torrent is paused, it will however remember all share ratios to all peers and remember

1781

all potential (not connected) peers. You can use <tt class="docutils literal"><span class="pre">is_paused()</span></tt> to determine if a torrent

1782

is currently paused. Torrents may be paused automatically if there is a file error (e.g. disk full)

1783

or something similar. See <a class="reference internal" href="#file-error-alert">file_error_alert</a>.</p>

1784

</div>

1785

1786

<h2>resolve_countries()</h2>

1787

1788

1789

void resolve_countries(bool r);

1790

bool resolve_countries() const;

1791

</pre>

1792

</blockquote>

1793

<p>Sets or gets the flag that derermines if countries should be resolved for the peers of this

1794

torrent. It defaults to false. If it is set to true, the <a class="reference internal" href="#peer-info">peer_info</a> structure for the peers

1795

in this torrent will have their <tt class="docutils literal"><span class="pre">country</span></tt> member set. See <a class="reference internal" href="#peer-info">peer_info</a> for more information

1796

on how to interpret this field.</p>

1797

</div>

1798

1799

1800

1801

1802

bool is_seed() const;

1803

</pre>

1804

</blockquote>

1805

<p>Returns true if the torrent is in seed mode (i.e. if it has finished downloading).</p>

1806

</div>

1807

1808

<h2>has_metadata()</h2>

1809

1810

1811

bool has_metadata() const;

1812

</pre>

1813

</blockquote>

1814

<p>Returns true if this torrent has metadata (either it was started from a .torrent file or the

1815

metadata has been downloaded). The only scenario where this can return false is when the torrent

1816

was started torrent-less (i.e. with just an info-hash and tracker ip). Note that if the torrent

1817

doesn't have metadata, the member <a class="reference internal" href="#get-torrent-info">get_torrent_info()</a> will throw.</p>

1818

</div>

1819

1820

<h2>set_tracker_login()</h2>

1821

1822

1823

void set_tracker_login(std::string const& username

1824

, std::string const& password) const;

1825

</pre>

1826

</blockquote>

1827

<p><tt class="docutils literal"><span class="pre">set_tracker_login()</span></tt> sets a username and password that will be sent along in the HTTP-request

1828

of the tracker announce. Set this if the tracker requires authorization.</p>

1829

</div>

1830

1831

<h2>trackers() replace_trackers()</h2>

1832

1833

1834

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

1835

void replace_trackers(std::vector<announce_entry> const&) const;

1836

</pre>

1837

</blockquote>

1838

<p><tt class="docutils literal"><span class="pre">trackers()</span></tt> will return the list of trackers for this torrent. The

1839

announce entry contains both a string <tt class="docutils literal"><span class="pre">url</span></tt> which specify the announce url

1840

for the tracker as well as an int <tt class="docutils literal"><span class="pre">tier</span></tt>, which is specifies the order in

1841

which this tracker is tried. If you want libtorrent to use another list of

1842

trackers for this torrent, you can use <tt class="docutils literal"><span class="pre">replace_trackers()</span></tt> which takes

1843

a list of the same form as the one returned from <tt class="docutils literal"><span class="pre">trackers()</span></tt> and will

1844

replace it. If you want an immediate effect, you have to call

1845

<a class="reference internal" href="#force-reannounce">force_reannounce()</a>.</p>

1846

</div>

1847

1848

<h2>add_url_seed() remove_url_seed() url_seeds()</h2>

1849

1850

1851

void add_url_seed(std::string const& url);

1852

void remove_url_seed(std::string const& url);

1853

std::set<std::string> url_seeds() const;

1854

</pre>

1855

</blockquote>

1856

<p><tt class="docutils literal"><span class="pre">add_url_seed()</span></tt> adds another url to the torrent's list of url seeds. If the

1857

given url already exists in that list, the call has no effect. The torrent

1858

will connect to the server and try to download pieces from it, unless it's

1859

paused, queued, checking or seeding. <tt class="docutils literal"><span class="pre">remove_url_seed()</span></tt> removes the given

1860

url if it exists already. <tt class="docutils literal"><span class="pre">url_seeds()</span></tt> return a set of the url seeds

1861

currently in this torrent. Note that urls that fails may be removed

1862

automatically from the list.</p>

1863

<p>See <a class="reference internal" href="#http-seeding">HTTP seeding</a> for more information.</p>

1864

</div>

1865

1866

<h2>use_interface()</h2>

1867

1868

1869

void use_interface(char const* net_interface) const;

1870

</pre>

1871

</blockquote>

1872

<p><tt class="docutils literal"><span class="pre">use_interface()</span></tt> sets the network interface this torrent will use when it opens outgoing

1873

connections. By default, it uses the same interface as the <a class="reference internal" href="#session">session</a> uses to listen on. The

1874

parameter must be a string containing an ip-address (either an IPv4 or IPv6 address). If

1875

the string does not conform to this format and exception is thrown.</p>

1876

</div>

1877

1878

1879

1880

1881

sha1_hash info_hash() const;

1882

</pre>

1883

</blockquote>

1884

<p><tt class="docutils literal"><span class="pre">info_hash()</span></tt> returns the info-hash for the torrent.</p>

1885

</div>

1886

1887

<h2>set_max_uploads() set_max_connections()</h2>

1888

1889

1890

void set_max_uploads(int max_uploads) const;

1891

void set_max_connections(int max_connections) const;

1892

</pre>

1893

</blockquote>

1894

<p><tt class="docutils literal"><span class="pre">set_max_uploads()</span></tt> sets the maximum number of peers that's unchoked at the same time on this

1895

torrent. If you set this to -1, there will be no limit.</p>

1896

<p><tt class="docutils literal"><span class="pre">set_max_connections()</span></tt> sets the maximum number of connection this torrent will open. If all

1897

connections are used up, incoming connections may be refused or poor connections may be closed.

1898

This must be at least 2. The default is unlimited number of connections. If -1 is given to the

1899

function, it means unlimited.</p>

1900

</div>

1901

1902

<h2>write_resume_data()</h2>

1903

1904

1905

entry write_resume_data() const;

1906

</pre>

1907

</blockquote>

1908

<p><tt class="docutils literal"><span class="pre">write_resume_data()</span></tt> generates fast-resume data and returns it as an <a class="reference internal" href="#entry">entry</a>. This <a class="reference internal" href="#entry">entry</a>

1909

is suitable for being bencoded. For more information about how fast-resume works, see <a class="reference internal" href="#fast-resume">fast resume</a>.</p>

1910

<p>There are three cases where this function will just return an empty <tt class="docutils literal"><span class="pre">entry</span></tt>:</p>

1911

1912

1913

<li>The torrent handle is invalid.</li>

1914

<li>The torrent is checking (or is queued for checking) its storage, it will obviously

1915

not be ready to write resume data.</li>

1916

<li>The torrent hasn't received valid metadata and was started without metadata

1917

(see libtorrent's <a class="reference internal" href="#metadata-from-peers">metadata from peers</a> extension)</li>

1918

</ol>

1919

</blockquote>

1920

<p>Note that by the time this function returns, the resume data may already be invalid if the torrent

1921

is still downloading! The recommended practice is to first pause the torrent, then generate the

1922

fast resume data, and then close it down. Since the disk IO is done in a separate thread, in order

1923

to synchronize, you shoule to wait for the <tt class="docutils literal"><span class="pre">torrent_paused_alert</span></tt> before you write the resume

1924

data.</p>

1925

<p>In full allocation mode the reume data is never invalidated by subsequent

1926

writes to the files, since pieces won't move around. This means that you don't need to

1927

pause before writing resume data in full or sparse mode. If you don't, however, any data written to

1928

disk after you saved resume data and before the session closed is lost.</p>

1929

<p>It also means that if the resume data is out dated, libtorrent will not re-check the files, but assume

1930

that it is fairly recent. The assumption is that it's better to loose a little bit than to re-check

1931

the entire file.</p>

1932

<p>It is still a good idea to save resume data periodically during download as well as when

1933

closing down.</p>

1934

</div>

1935

1936

<h2>status()</h2>

1937

1938

1939

torrent_status status() const;

1940

</pre>

1941

</blockquote>

1942

<p><tt class="docutils literal"><span class="pre">status()</span></tt> will return a structure with information about the status of this

1943

torrent. If the <a class="reference internal" href="#torrent-handle">torrent_handle</a> is invalid, it will throw <a class="reference internal" href="#invalid-handle">invalid_handle</a> exception.

1944

See <a class="reference internal" href="#torrent-status">torrent_status</a>.</p>

1945

</div>

1946

1947

<h2>get_download_queue()</h2>

1948

1949

1950

void get_download_queue(std::vector<partial_piece_info>& queue) const;

1951

</pre>

1952

</blockquote>

1953

<p><tt class="docutils literal"><span class="pre">get_download_queue()</span></tt> takes a non-const reference to a vector which it will fill with

1954

information about pieces that are partially downloaded or not downloaded at all but partially

1955

requested. The entry in the vector (<tt class="docutils literal"><span class="pre">partial_piece_info</span></tt>) looks like this:</p>

1956

1957

struct partial_piece_info

1958

{

1959

int piece_index;

1960

int blocks_in_piece;

1961

block_info blocks[256];

1962

enum state_t { none, slow, medium, fast };

1963

state_t piece_state;

1964

};

1965

</pre>

1966

<p><tt class="docutils literal"><span class="pre">piece_index</span></tt> is the index of the piece in question. <tt class="docutils literal"><span class="pre">blocks_in_piece</span></tt> is the

1967

number of blocks in this particular piece. This number will be the same for most pieces, but

1968

the last piece may have fewer blocks than the standard pieces.</p>

1969

<p><tt class="docutils literal"><span class="pre">piece_state</span></tt> is set to either <tt class="docutils literal"><span class="pre">fast</span></tt>, <tt class="docutils literal"><span class="pre">medium</span></tt>, <tt class="docutils literal"><span class="pre">slow</span></tt> or <tt class="docutils literal"><span class="pre">none</span></tt>. It tells which

1970

download rate category the peers downloading this piece falls into. <tt class="docutils literal"><span class="pre">none</span></tt> means that no

1971

peer is currently downloading any part of the piece. Peers prefer picking pieces from

1972

the same category as themselves. The reason for this is to keep the number of partially

1973

downloaded pieces down. Pieces set to <tt class="docutils literal"><span class="pre">none</span></tt> can be converted into any of <tt class="docutils literal"><span class="pre">fast</span></tt>,

1974

<tt class="docutils literal"><span class="pre">medium</span></tt> or <tt class="docutils literal"><span class="pre">slow</span></tt> as soon as a peer want to download from it.</p>

1975

1976

struct block_info

1977

{

1978

enum block_state_t

1979

{ none, requested, writing, finished };

1980

1981

tcp::endpoint peer;

1982

unsigned state:2;

1983

unsigned num_peers:14;

1984

};

1985

</pre>

1986

<p>The <tt class="docutils literal"><span class="pre">block_info</span></tt> array contains data for each individual block in the piece. Each block has

1987

a state (<tt class="docutils literal"><span class="pre">state</span></tt>) which is any of:</p>

1988

1989

<li><tt class="docutils literal"><span class="pre">none</span></tt> - This block has not been downloaded or requested form any peer.</li>

1990

<li><tt class="docutils literal"><span class="pre">requested</span></tt> - The block has been requested, but not completely downloaded yet.</li>

1991

<li><tt class="docutils literal"><span class="pre">writing</span></tt> - The block has been downloaded and is currently queued for being written to disk.</li>

1992

<li><tt class="docutils literal"><span class="pre">finished</span></tt> - The block has been written to disk.</li>

1993

</ul>

1994

<p>The <tt class="docutils literal"><span class="pre">peer</span></tt> field is the ip address of the peer this block was downloaded from.

1995

<tt class="docutils literal"><span class="pre">num_peers</span></tt> is the number of peers that is currently requesting this block. Typically this

1996

is 0 or 1, but at the end of the torrent blocks may be requested by more peers in parallel to

1997

speed things up.</p>

1998

</div>

1999

2000

2001

2002

2003

void get_peer_info(std::vector<peer_info>&) const;

2004

</pre>

2005

</blockquote>

2006

<p><tt class="docutils literal"><span class="pre">get_peer_info()</span></tt> takes a reference to a vector that will be cleared and filled

2007

with one entry for each peer connected to this torrent, given the handle is valid. If the

2008

<a class="reference internal" href="#torrent-handle">torrent_handle</a> is invalid, it will throw <a class="reference internal" href="#invalid-handle">invalid_handle</a> exception. Each entry in

2009

the vector contains information about that particular peer. See <a class="reference internal" href="#peer-info">peer_info</a>.</p>

2010

</div>

2011

2012

<h2>get_torrent_info()</h2>

2013

2014

2015

torrent_info const& get_torrent_info() const;

2016

</pre>

2017

</blockquote>

2018

<p>Returns a const reference to the <a class="reference internal" href="#torrent-info">torrent_info</a> object associated with this torrent.

2019

This reference is valid as long as the <a class="reference internal" href="#torrent-handle">torrent_handle</a> is valid, no longer. If the

2020

<a class="reference internal" href="#torrent-handle">torrent_handle</a> is invalid or if it doesn't have any metadata, <a class="reference internal" href="#invalid-handle">invalid_handle</a>

2021

exception will be thrown. The torrent may be in a state without metadata only if

2022

it was started without a .torrent file, i.e. by using the libtorrent extension of

2023

just supplying a tracker and info-hash.</p>

2024

</div>

2025

2026

<h2>is_valid()</h2>

2027

2028

2029

bool is_valid() const;

2030

</pre>

2031

</blockquote>

2032

<p>Returns true if this handle refers to a valid torrent and false if it hasn't been initialized

2033

or if the torrent it refers to has been aborted. Note that a handle may become invalid after

2034

it has been added to the session. Usually this is because the storage for the torrent is

2035

somehow invalid or if the filenames are not allowed (and hence cannot be opened/created) on

2036

your filesystem. If such an error occurs, a <a class="reference internal" href="#file-error-alert">file_error_alert</a> is generated and all handles

2037

that refers to that torrent will become invalid.</p>

2038

<p><em>TODO: document storage</em></p>

2039

</div>

2040

</div>

2041

2042

<h1>torrent_status</h1>

2043

<p>It contains the following fields:</p>

2044

2045

struct torrent_status

2046

{

2047

enum state_t

2048

{

2049

queued_for_checking,

2050

checking_files,

2051

connecting_to_tracker,

2052

downloading_metadata,

2053

downloading,

2054

finished,

2055

seeding,

2056

allocating

2057

};

2058

2059

state_t state;

2060

bool paused;

2061

float progress;

2062

boost::posix_time::time_duration next_announce;

2063

boost::posix_time::time_duration announce_interval;

2064

2065

std::string current_tracker;

2066

2067

size_type total_download;

2068

size_type total_upload;

2069

2070

size_type total_payload_download;

2071

size_type total_payload_upload;

2072

2073

size_type total_failed_bytes;

2074

size_type total_redundant_bytes;

2075

2076

float download_rate;

2077

float upload_rate;

2078

2079

float download_payload_rate;

2080

float upload_payload_rate;

2081

2082

int num_peers;

2083

2084

int num_complete;

2085

int num_incomplete;

2086

2087

int list_seeds;

2088

int list_peers;

2089

2090

const std::vector<bool>* pieces;

2091

int num_pieces;

2092

2093

size_type total_done;

2094

size_type total_wanted_done;

2095

size_type total_wanted;

2096

2097

int num_seeds;

2098

float distributed_copies;

2099

2100

int block_size;

2101

2102

int num_uploads;

2103

int num_connections;

2104

int uploads_limit;

2105

int connections_limit;

2106

2107

storage_mode_t storage_mode;

2108

2109

int up_bandwidth_queue;

2110

int down_bandwidth_queue;

2111

};

2112

</pre>

2113

<p><tt class="docutils literal"><span class="pre">progress</span></tt> is a value in the range [0, 1], that represents the progress of the

2114

torrent's current task. It may be checking files or downloading. The torrent's

2115

current task is in the <tt class="docutils literal"><span class="pre">state</span></tt> member, it will be one of the following:</p>

2116

2117

2118

2119

2120

</colgroup>

2121

2122

<tr><td><tt class="docutils literal"><span class="pre">queued_for_checking</span></tt></td>

2123

<td>The torrent is in the queue for being checked. But there

2124

currently is another torrent that are being checked.

2125

This torrent will wait for its turn.</td>

2126

</tr>

2127

<tr><td><tt class="docutils literal"><span class="pre">checking_files</span></tt></td>

2128

<td>The torrent has not started its download yet, and is

2129

currently checking existing files.</td>

2130

</tr>

2131

<tr><td><tt class="docutils literal"><span class="pre">connecting_to_tracker</span></tt></td>

2132

<td>The torrent has sent a request to the tracker and is

2133

currently waiting for a response</td>

2134

</tr>

2135

<tr><td><tt class="docutils literal"><span class="pre">downloading_metadata</span></tt></td>

2136

<td>The torrent is trying to download metadata from peers.

2137

This assumes the metadata_transfer extension is in use.</td>

2138

</tr>

2139

<tr><td><tt class="docutils literal"><span class="pre">downloading</span></tt></td>

2140

<td>The torrent is being downloaded. This is the state

2141

most torrents will be in most of the time. The progress

2142

meter will tell how much of the files that has been

2143

downloaded.</td>

2144

</tr>

2145

<tr><td><tt class="docutils literal"><span class="pre">finished</span></tt></td>

2146

<td>In this state the torrent has finished downloading but

2147

still doesn't have the entire torrent. i.e. some pieces

2148

are filtered and won't get downloaded.</td>

2149

</tr>

2150

<tr><td><tt class="docutils literal"><span class="pre">seeding</span></tt></td>

2151

<td>In this state the torrent has finished downloading and

2152

is a pure seeder.</td>

2153

</tr>

2154

<tr><td><tt class="docutils literal"><span class="pre">allocating</span></tt></td>

2155

<td>If the torrent was started in full allocation mode, this

2156

indicates that the (disk) storage for the torrent is

2157

allocated.</td>

2158

</tr>

2159

</tbody>

2160

</table>

2161

<p>When downloading, the progress is <tt class="docutils literal"><span class="pre">total_wanted_done</span></tt> / <tt class="docutils literal"><span class="pre">total_wanted</span></tt>.</p>

2162

<p><tt class="docutils literal"><span class="pre">paused</span></tt> is set to true if the torrent is paused and false otherwise.</p>

2163

<p><tt class="docutils literal"><span class="pre">next_announce</span></tt> is the time until the torrent will announce itself to the tracker. And

2164

<tt class="docutils literal"><span class="pre">announce_interval</span></tt> is the time the tracker want us to wait until we announce ourself

2165

again the next time.</p>

2166

<p><tt class="docutils literal"><span class="pre">current_tracker</span></tt> is the URL of the last working tracker. If no tracker request has

2167

been successful yet, it's set to an empty string.</p>

2168

<p><tt class="docutils literal"><span class="pre">total_download</span></tt> and <tt class="docutils literal"><span class="pre">total_upload</span></tt> is the number of bytes downloaded and

2169

uploaded to all peers, accumulated, <em>this session</em> only.</p>

2170

<p><tt class="docutils literal"><span class="pre">total_payload_download</span></tt> and <tt class="docutils literal"><span class="pre">total_payload_upload</span></tt> counts the amount of bytes

2171

send and received this session, but only the actual payload data (i.e the interesting

2172

data), these counters ignore any protocol overhead.</p>

2173

<p><tt class="docutils literal"><span class="pre">total_failed_bytes</span></tt> is the number of bytes that has been downloaded and that

2174

has failed the piece hash test. In other words, this is just how much crap that

2175

has been downloaded.</p>

2176

<p><tt class="docutils literal"><span class="pre">total_redundant_bytes</span></tt> is the number of bytes that has been downloaded even

2177

though that data already was downloaded. The reason for this is that in some

2178

situations the same data can be downloaded by mistake. When libtorrent sends

2179

requests to a peer, and the peer doesn't send a response within a certain

2180

timeout, libtorrent will re-request that block. Another situation when

2181

libtorrent may re-request blocks is when the requests it sends out are not

2182

replied in FIFO-order (it will re-request blocks that are skipped by an out of

2183

order block). This is supposed to be as low as possible.</p>

2184

<p><tt class="docutils literal"><span class="pre">pieces</span></tt> is the bitmask that represents which pieces we have (set to true) and

2185

the pieces we don't have. It's a pointer and may be set to 0 if the torrent isn't

2186

downloading or seeding.</p>

2187

<p><tt class="docutils literal"><span class="pre">num_pieces</span></tt> is the number of pieces that has been downloaded. It is equivalent

2188

to: <tt class="docutils literal"><span class="pre">std::accumulate(pieces->begin(),</span> <span class="pre">pieces->end())</span></tt>. So you don't have to

2189

count yourself. This can be used to see if anything has updated since last time

2190

if you want to keep a graph of the pieces up to date.</p>

2191

<p><tt class="docutils literal"><span class="pre">download_rate</span></tt> and <tt class="docutils literal"><span class="pre">upload_rate</span></tt> are the total rates for all peers for this

2192

torrent. These will usually have better precision than summing the rates from

2193

all peers. The rates are given as the number of bytes per second. The

2194

<tt class="docutils literal"><span class="pre">download_payload_rate</span></tt> and <tt class="docutils literal"><span class="pre">upload_payload_rate</span></tt> respectively is the

2195

total transfer rate of payload only, not counting protocol chatter. This might

2196

be slightly smaller than the other rates, but if projected over a long time

2197

(e.g. when calculating ETA:s) the difference may be noticeable.</p>

2198

<p><tt class="docutils literal"><span class="pre">num_peers</span></tt> is the number of peers this torrent currently is connected to.

2199

Peer connections that are in the half-open state (is attempting to connect)

2200

or are queued for later connection attempt do not count. Although they are

2201

visible in the peer list when you call <a class="reference internal" href="#get-peer-info">get_peer_info()</a>.</p>

2202

<p><tt class="docutils literal"><span class="pre">num_complete</span></tt> and <tt class="docutils literal"><span class="pre">num_incomplete</span></tt> are set to -1 if the tracker did not

2203

send any scrape data in its announce reply. This data is optional and may

2204

not be available from all trackers. If these are not -1, they are the total

2205

number of peers that are seeding (complete) and the total number of peers

2206

that are still downloading (incomplete) this torrent.</p>

2207

<p><tt class="docutils literal"><span class="pre">list_seeds</span></tt> and <tt class="docutils literal"><span class="pre">list_peers</span></tt> are the number of seeds in our peer list

2208

and the total number of peers (including seeds) respectively. We are not

2209

necessarily connected to all the peers in our peer list. This is the number

2210

of peers we know of in total, including banned peers and peers that we have

2211

failed to connect to.</p>

2212

<p><tt class="docutils literal"><span class="pre">total_done</span></tt> is the total number of bytes of the file(s) that we have. All

2213

this does not necessarily has to be downloaded during this session (that's

2214

<tt class="docutils literal"><span class="pre">total_payload_download</span></tt>).</p>

2215

<p><tt class="docutils literal"><span class="pre">total_wanted_done</span></tt> is the number of bytes we have downloaded, only counting the

2216

pieces that we actually want to download. i.e. excluding any pieces that we have but

2217

are filtered as not wanted.</p>

2218

<p><tt class="docutils literal"><span class="pre">total_wanted</span></tt> is the total number of bytes we want to download. This is also

2219

excluding pieces that have been filtered.</p>

2220

<p><tt class="docutils literal"><span class="pre">num_seeds</span></tt> is the number of peers that are seeding that this client is

2221

currently connected to.</p>

2222

<p><tt class="docutils literal"><span class="pre">distributed_copies</span></tt> is the number of distributed copies of the torrent.

2223

Note that one copy may be spread out among many peers. The integer part

2224

tells how many copies there are currently of the rarest piece(s) among the

2225

peers this client is connected to. The fractional part tells the share of

2226

pieces that have more copies than the rarest piece(s). For example: 2.5 would

2227

mean that the rarest pieces have only 2 copies among the peers this torrent is

2228

connected to, and that 50% of all the pieces have more than two copies.</p>

2229

<p>If we are a seed, the piece picker is deallocated as an optimization, and

2230

piece availability is no longer tracked. In this case the distributed

2231

copies is set to -1.</p>

2232

<p><tt class="docutils literal"><span class="pre">block_size</span></tt> is the size of a block, in bytes. A block is a sub piece, it

2233

is the number of bytes that each piece request asks for and the number of

2234

bytes that each bit in the <tt class="docutils literal"><span class="pre">partial_piece_info</span></tt>'s bitset represents

2235

(see <a class="reference internal" href="#get-download-queue">get_download_queue()</a>). This is typically 16 kB, but it may be

2236

larger if the pieces are larger.</p>

2237

<p><tt class="docutils literal"><span class="pre">num_uploads</span></tt> is the number of unchoked peers in this torrent.</p>

2238

<p><tt class="docutils literal"><span class="pre">num_connections</span></tt> is the number of peer connections this torrent has, including

2239

half-open connections that hasn't completed the bittorrent handshake yet. This is

2240

always <= <tt class="docutils literal"><span class="pre">num_peers</span></tt>.</p>

2241

<p><tt class="docutils literal"><span class="pre">uploads_limit</span></tt> is the set limit of upload slots (unchoked peers) for this torrent.</p>

2242

<p><tt class="docutils literal"><span class="pre">connections_limit</span></tt> is the set limit of number of connections for this torrent.</p>

2243

<p><tt class="docutils literal"><span class="pre">storage_mode</span></tt> is one of <tt class="docutils literal"><span class="pre">storage_mode_allocate</span></tt>, <tt class="docutils literal"><span class="pre">storage_mode_sparse</span></tt> or

2244

<tt class="docutils literal"><span class="pre">storage_mode_compact</span></tt>. Identifies which storage mode this torrent is being saved

2245

with. See <a class="reference internal" href="#storage-allocation">Storage allocation</a>.</p>

2246

<p><tt class="docutils literal"><span class="pre">up_bandwidth_queue</span></tt> and <tt class="docutils literal"><span class="pre">down_bandwidth_queue</span></tt> are the number of peers in this

2247

torrent that are waiting for more bandwidth quota from the torrent rate limiter.

2248

This can determine if the rate you get from this torrent is bound by the torrents

2249

limit or not. If there is no limit set on this torrent, the peers might still be

2250

waiting for bandwidth quota from the global limiter, but then they are counted in

2251

the <tt class="docutils literal"><span class="pre">session_status</span></tt> object.</p>

2252

</div>

2253

2254

2255

<p>It contains the following fields:</p>

2256

2257

struct peer_info

2258

{

2259

enum

2260

{

2261

interesting = 0x1,

2262

choked = 0x2,

2263

remote_interested = 0x4,

2264

remote_choked = 0x8,

2265

supports_extensions = 0x10,

2266

local_connection = 0x20,

2267

handshake = 0x40,

2268

connecting = 0x80,

2269

queued = 0x100,

2270

on_parole = 0x200,

2271

seed = 0x400,

2272

optimistic_unchoke = 0x800,

2273

rc4_encrypted = 0x100000,

2274

plaintext_encrypted = 0x200000

2275

};

2276

2277

unsigned int flags;

2278

2279

enum peer_source_flags

2280

{

2281

tracker = 0x1,

2282

dht = 0x2,

2283

pex = 0x4,

2284

lsd = 0x8

2285

};

2286

2287

int source;

2288

2289

asio::ip::tcp::endpoint ip;

2290

float up_speed;

2291

float down_speed;

2292

float payload_up_speed;

2293

float payload_down_speed;

2294

size_type total_download;

2295

size_type total_upload;

2296

peer_id pid;

2297

std::vector<bool> pieces;

2298

int upload_limit;

2299

int download_limit;

2300

2301

time_duration last_request;

2302

time_duration last_active;

2303

2304

int send_buffer_size;

2305

int used_send_buffer;

2306

2307

int num_hashfails;

2308

2309

char country[2];

2310

2311

size_type load_balancing;

2312

2313

int download_queue_length;

2314

int upload_queue_length;

2315

2316

int failcount;

2317

2318

int downloading_piece_index;

2319

int downloading_block_index;

2320

int downloading_progress;

2321

int downloading_total;

2322

2323

std::string client;

2324

2325

enum

2326

{

2327

standard_bittorrent = 0,

2328

web_seed = 1

2329

};

2330

int connection_type;

2331

};

2332

</pre>

2333

<p>The <tt class="docutils literal"><span class="pre">flags</span></tt> attribute tells you in which state the peer is. It is set to

2334

any combination of the enums above. The following table describes each flag:</p>

2335

2336

2337

2338

2339

</colgroup>

2340

2341

<tr><td><tt class="docutils literal"><span class="pre">interesting</span></tt></td>

2342

<td><strong>we</strong> are interested in pieces from this peer.</td>

2343

</tr>

2344

<tr><td><tt class="docutils literal"><span class="pre">choked</span></tt></td>

2345

<td><strong>we</strong> have choked this peer.</td>

2346

</tr>

2347

<tr><td><tt class="docutils literal"><span class="pre">remote_interested</span></tt></td>

2348

<td>the peer is interested in <strong>us</strong></td>

2349

</tr>

2350

<tr><td><tt class="docutils literal"><span class="pre">remote_choked</span></tt></td>

2351

<td>the peer has choked <strong>us</strong>.</td>

2352

</tr>

2353

<tr><td><tt class="docutils literal"><span class="pre">support_extensions</span></tt></td>

2354

<td>means that this peer supports the

2355

<a class="reference external" href="extension_protocol.html">extension protocol</a>.</td>

2356

</tr>

2357

<tr><td><tt class="docutils literal"><span class="pre">local_connection</span></tt></td>

2358

<td>The connection was initiated by us, the peer has a

2359

listen port open, and that port is the same as in the

2360

address of this peer. If this flag is not set, this

2361

peer connection was opened by this peer connecting to

2362

us.</td>

2363

</tr>

2364

<tr><td><tt class="docutils literal"><span class="pre">handshake</span></tt></td>

2365

<td>The connection is opened, and waiting for the

2366

handshake. Until the handshake is done, the peer

2367

cannot be identified.</td>

2368

</tr>

2369

<tr><td><tt class="docutils literal"><span class="pre">connecting</span></tt></td>

2370

<td>The connection is in a half-open state (i.e. it is

2371

being connected).</td>

2372

</tr>

2373

<tr><td><tt class="docutils literal"><span class="pre">queued</span></tt></td>

2374

<td>The connection is currently queued for a connection

2375

attempt. This may happen if there is a limit set on

2376

the number of half-open TCP connections.</td>

2377

</tr>

2378

<tr><td><tt class="docutils literal"><span class="pre">on_parole</span></tt></td>

2379

<td>The peer has participated in a piece that failed the

2380

hash check, and is now "on parole", which means we're

2381

only requesting whole pieces from this peer until

2382

it either fails that piece or proves that it doesn't

2383

send bad data.</td>

2384

</tr>

2385

2386

<td>This peer is a seed (it has all the pieces).</td>

2387

</tr>

2388

<tr><td><tt class="docutils literal"><span class="pre">optimistic_unchoke</span></tt></td>

2389

<td>This peer is subject to an optimistic unchoke. It has

2390

been unchoked for a while to see if it might unchoke

2391

us in return an earn an upload/unchoke slot. If it

2392

doesn't within some period of time, it will be choked

2393

and another peer will be optimistically unchoked.</td>

2394

</tr>

2395

</tbody>

2396

</table>

2397

<p><tt class="docutils literal"><span class="pre">source</span></tt> is a combination of flags describing from which sources this peer

2398

was received. The flags are:</p>

2399

2400

2401

2402

2403

</colgroup>

2404

2405

<tr><td><tt class="docutils literal"><span class="pre">tracker</span></tt></td>

2406

<td>The peer was received from the tracker.</td>

2407

</tr>

2408

2409

<td>The peer was received from the kademlia DHT.</td>

2410

</tr>

2411

2412

<td>The peer was received from the peer exchange

2413

extension.</td>

2414

</tr>

2415

2416

<td>The peer was received from the local service

2417

discovery (The peer is on the local network).</td>

2418

</tr>

2419

<tr><td><tt class="docutils literal"><span class="pre">resume_data</span></tt></td>

2420

<td>The peer was added from the fast resume data.</td>

2421

</tr>

2422

</tbody>

2423

</table>

2424

<p>The <tt class="docutils literal"><span class="pre">ip</span></tt> field is the IP-address to this peer. The type is an asio endpoint. For

2425

more info, see the <a class="reference external" href="http://asio.sf.net">asio</a> documentation.</p>

2426

<p><tt class="docutils literal"><span class="pre">up_speed</span></tt> and <tt class="docutils literal"><span class="pre">down_speed</span></tt> contains the current upload and download speed

2427

we have to and from this peer (including any protocol messages). The transfer rates

2428

of payload data only are found in <tt class="docutils literal"><span class="pre">payload_up_speed</span></tt> and <tt class="docutils literal"><span class="pre">payload_down_speed</span></tt>.

2429

These figures are updated approximately once every second.</p>

2430

<p><tt class="docutils literal"><span class="pre">total_download</span></tt> and <tt class="docutils literal"><span class="pre">total_upload</span></tt> are the total number of bytes downloaded

2431

from and uploaded to this peer. These numbers do not include the protocol chatter, but only

2432

the payload data.</p>

2433

<p><tt class="docutils literal"><span class="pre">pid</span></tt> is the peer's id as used in the bit torrent protocol. This id can be used to

2434

extract 'fingerprints' from the peer. Sometimes it can tell you which client the peer

2435

is using. See identify_client()_</p>

2436

<p><tt class="docutils literal"><span class="pre">pieces</span></tt> is a vector of booleans that has as many entries as there are pieces

2437

in the torrent. Each boolean tells you if the peer has that piece (if it's set to true)

2438

or if the peer miss that piece (set to false).</p>

2439

2440

<p><tt class="docutils literal"><span class="pre">upload_limit</span></tt> is the number of bytes per second we are allowed to send to this

2441

peer every second. It may be -1 if there's no local limit on the peer. The global

2442

limit and the torrent limit is always enforced anyway.</p>

2443

<p><tt class="docutils literal"><span class="pre">download_limit</span></tt> is the number of bytes per second this peer is allowed to

2444

receive. -1 means it's unlimited.</p>

2445

<p><tt class="docutils literal"><span class="pre">last_request</span></tt> and <tt class="docutils literal"><span class="pre">last_active</span></tt> is the time since we last sent a request

2446

to this peer and since any transfer occurred with this peer, respectively.</p>

2447

<p><tt class="docutils literal"><span class="pre">send_buffer_size</span></tt> and <tt class="docutils literal"><span class="pre">used_send_buffer</span></tt> is the number of bytes allocated

2448

and used for the peer's send buffer, respectively.</p>

2449

<p><tt class="docutils literal"><span class="pre">num_hashfails</span></tt> is the number of pieces this peer has participated in

2450

sending us that turned out to fail the hash check.</p>

2451

<p><tt class="docutils literal"><span class="pre">country</span></tt> is the two letter <a class="reference external" href="http://www.iso.org/iso/en/prods-services/iso3166ma/02iso-3166-code-lists/list-en1.html">ISO 3166 country code</a> for the country the peer

2452

is connected from. If the country hasn't been resolved yet, both chars are set

2453

to 0. If the resolution failed for some reason, the field is set to "--". If the

2454

resolution service returns an invalid country code, it is set to "!!".

2455

The <tt class="docutils literal"><span class="pre">countries.nerd.dk</span></tt> service is used to look up countries. This field will

2456

remain set to 0 unless the torrent is set to resolve countries, see <a class="reference internal" href="#resolve-countries">resolve_countries()</a>.</p>

2457

<p><tt class="docutils literal"><span class="pre">load_balancing</span></tt> is a measurement of the balancing of free download (that we get)

2458

and free upload that we give. Every peer gets a certain amount of free upload, but

2459

this member says how much <em>extra</em> free upload this peer has got. If it is a negative

2460

number it means that this was a peer from which we have got this amount of free

2461

download.</p>

2462

<p><tt class="docutils literal"><span class="pre">download_queue_length</span></tt> is the number of piece-requests we have sent to this peer

2463

that hasn't been answered with a piece yet.</p>

2464

<p><tt class="docutils literal"><span class="pre">upload_queue_length</span></tt> is the number of piece-requests we have received from this peer

2465

that we haven't answered with a piece yet.</p>

2466

<p><tt class="docutils literal"><span class="pre">failcount</span></tt> is the number of times this peer has "failed". i.e. failed to connect

2467

or disconnected us. The failcount is decremented when we see this peer in a tracker

2468

response or peer exchange message.</p>

2469

<p>You can know which piece, and which part of that piece, that is currently being

2470

downloaded from a specific peer by looking at the next four members.

2471

<tt class="docutils literal"><span class="pre">downloading_piece_index</span></tt> is the index of the piece that is currently being downloaded.

2472

This may be set to -1 if there's currently no piece downloading from this peer. If it is

2473

>= 0, the other three members are valid. <tt class="docutils literal"><span class="pre">downloading_block_index</span></tt> is the index of the

2474

block (or sub-piece) that is being downloaded. <tt class="docutils literal"><span class="pre">downloading_progress</span></tt> is the number

2475

of bytes of this block we have received from the peer, and <tt class="docutils literal"><span class="pre">downloading_total</span></tt> is

2476

the total number of bytes in this block.</p>

2477

<p><tt class="docutils literal"><span class="pre">client</span></tt> is a string describing the software at the other end of the connection.

2478

In some cases this information is not available, then it will contain a string

2479

that may give away something about which software is running in the other end.

2480

In the case of a web seed, the server type and version will be a part of this

2481

string.</p>

2482

<p><tt class="docutils literal"><span class="pre">connection_type</span></tt> can currently be one of <tt class="docutils literal"><span class="pre">standard_bittorrent</span></tt> or

2483

<tt class="docutils literal"><span class="pre">web_seed</span></tt>. These are currently the only implemented protocols.</p>

2484

</div>

2485

2486

<h1>session_settings</h1>

2487

<p>You have some control over tracker requests through the <tt class="docutils literal"><span class="pre">session_settings</span></tt> object. You

2488

create it and fill it with your settings and then use <tt class="docutils literal"><span class="pre">session::set_settings()</span></tt>

2489

to apply them. You have control over proxy and authorization settings and also the user-agent

2490

that will be sent to the tracker. The user-agent is a good way to identify your client.</p>

2491

2492

struct session_settings

2493

{

2494

session_settings();

2495

std::string user_agent;

2496

int tracker_completion_timeout;

2497

int tracker_receive_timeout;

2498

int stop_tracker_timeout;

2499

int tracker_maximum_response_length;

2500

2501

int piece_timeout;

2502

float request_queue_time;

2503

int max_allowed_in_request_queue;

2504

int max_out_request_queue;

2505

int whole_pieces_threshold;

2506

int peer_timeout;

2507

int urlseed_timeout;

2508

int urlseed_pipeline_size;

2509

int file_pool_size;

2510

bool allow_multiple_connections_per_ip;

2511

int max_failcount;

2512

int min_reconnect_time;

2513

int peer_connect_timeout;

2514

bool ignore_limits_on_local_network;

2515

int connection_speed;

2516

bool send_redundant_have;

2517

bool lazy_bitfields;

2518

int inactivity_timeout;

2519

int unchoke_interval;

2520

int optimistic_unchoke_multiplier;

2521

address announce_ip;

2522

int num_want;

2523

int initial_picker_threshold;

2524

int allowed_fast_set_size;

2525

int max_outstanding_disk_bytes_per_connection;

2526

int handshake_timeout;

2527

bool use_dht_as_fallback;

2528

bool free_torrent_hashes;

2529

bool upnp_ignore_nonrouters;

2530

};

2531

</pre>

2532

<p><tt class="docutils literal"><span class="pre">user_agent</span></tt> this is the client identification to the tracker.

2533

The recommended format of this string is:

2534

"ClientName/ClientVersion libtorrent/libtorrentVersion".

2535

This name will not only be used when making HTTP requests, but also when

2536

sending extended headers to peers that support that extension.</p>

2537

<p><tt class="docutils literal"><span class="pre">tracker_completion_timeout</span></tt> is the number of seconds the tracker

2538

connection will wait from when it sent the request until it considers the

2539

tracker to have timed-out. Default value is 60 seconds.</p>

2540

<p><tt class="docutils literal"><span class="pre">tracker_receive_timeout</span></tt> is the number of seconds to wait to receive

2541

any data from the tracker. If no data is received for this number of

2542

seconds, the tracker will be considered as having timed out. If a tracker

2543

is down, this is the kind of timeout that will occur. The default value

2544

is 20 seconds.</p>

2545

<p><tt class="docutils literal"><span class="pre">stop_tracker_timeout</span></tt> is the time to wait for tracker responses when

2546

shutting down the session object. This is given in seconds. Default is

2547

10 seconds.</p>

2548

<p><tt class="docutils literal"><span class="pre">tracker_maximum_response_length</span></tt> is the maximum number of bytes in a

2549

tracker response. If a response size passes this number it will be rejected

2550

and the connection will be closed. On gzipped responses this size is measured

2551

on the uncompressed data. So, if you get 20 bytes of gzip response that'll

2552

expand to 2 megs, it will be interrupted before the entire response has been

2553

uncompressed (given your limit is lower than 2 megs). Default limit is

2554

1 megabyte.</p>

2555

<p><tt class="docutils literal"><span class="pre">piece_timeout</span></tt> controls the number of seconds from a request is sent until

2556

it times out if no piece response is returned.</p>

2557

<p><tt class="docutils literal"><span class="pre">request_queue_time</span></tt> is the length of the request queue given in the number

2558

of seconds it should take for the other end to send all the pieces. i.e. the

2559

actual number of requests depends on the download rate and this number.</p>

2560

<p><tt class="docutils literal"><span class="pre">max_allowed_in_request_queue</span></tt> is the number of outstanding block requests

2561

a peer is allowed to queue up in the client. If a peer sends more requests

2562

than this (before the first one has been handled) the last request will be

2563

dropped. The higher this is, the faster upload speeds the client can get to a

2564

single peer.</p>

2565

<p><tt class="docutils literal"><span class="pre">max_out_request_queue</span></tt> is the maximum number of outstanding requests to

2566

send to a peer. This limit takes precedence over <tt class="docutils literal"><span class="pre">request_queue_time</span></tt>. i.e.

2567

no matter the download speed, the number of outstanding requests will never

2568

exceed this limit.</p>

2569

<p><tt class="docutils literal"><span class="pre">whole_pieces_threshold</span></tt> is a limit in seconds. if a whole piece can be

2570

downloaded in at least this number of seconds from a specific peer, the

2571

peer_connection will prefer requesting whole pieces at a time from this peer.

2572

The benefit of this is to better utilize disk caches by doing localized

2573

accesses and also to make it easier to identify bad peers if a piece fails

2574

the hash check.</p>

2575

<p><tt class="docutils literal"><span class="pre">peer_timeout</span></tt> is the number of seconds the peer connection should

2576

wait (for any activity on the peer connection) before closing it due

2577

to time out. This defaults to 120 seconds, since that's what's specified

2578

in the protocol specification. After half the time out, a keep alive message

2579

is sent.</p>

2580

<p><tt class="docutils literal"><span class="pre">urlseed_timeout</span></tt> is the same as <tt class="docutils literal"><span class="pre">peer_timeout</span></tt> but applies only to

2581

url seeds. This value defaults to 20 seconds.</p>

2582

<p><tt class="docutils literal"><span class="pre">urlseed_pipeline_size</span></tt> controls the pipelining with the web server. When

2583

using persistent connections to HTTP 1.1 servers, the client is allowed to

2584

send more requests before the first response is received. This number controls

2585

the number of outstanding requests to use with url-seeds. Default is 5.</p>

2586

<p><tt class="docutils literal"><span class="pre">file_pool_size</span></tt> is the the upper limit on the total number of files this

2587

session will keep open. The reason why files are left open at all is that

2588

some anti virus software hooks on every file close, and scans the file for

2589

viruses. deferring the closing of the files will be the difference between

2590

a usable system and a completely hogged down system. Most operating systems

2591

also has a limit on the total number of file descriptors a process may have

2592

open. It is usually a good idea to find this limit and set the number of

2593

connections and the number of files limits so their sum is slightly below it.</p>

2594

<p><tt class="docutils literal"><span class="pre">allow_multiple_connections_per_ip</span></tt> determines if connections from the

2595

same IP address as existing connections should be rejected or not. Multiple

2596

connections from the same IP address is not allowed by default, to prevent

2597

abusive behavior by peers. It may be useful to allow such connections in

2598

cases where simulations are run on the same machie, and all peers in a

2599

swarm has the same IP address.</p>

2600

<p><tt class="docutils literal"><span class="pre">max_failcount</span></tt> is the maximum times we try to connect to a peer before

2601

stop connecting again. If a peer succeeds, the failcounter is reset. If

2602

a peer is retrieved from a peer source (other than DHT) the failcount is

2603

decremented by one, allowing another try.</p>

2604

<p><tt class="docutils literal"><span class="pre">min_reconnect_time</span></tt> is the time to wait between connection attempts. If

2605

the peer fails, the time is multiplied by fail counter.</p>

2606

<p><tt class="docutils literal"><span class="pre">peer_connect_timeout</span></tt> the number of seconds to wait after a connection

2607

attempt is initiated to a peer until it is considered as having timed out.

2608

The default is 10 seconds. This setting is especially important in case

2609

the number of half-open connections are limited, since stale half-open

2610

connection may delay the connection of other peers considerably.</p>

2611

<p><tt class="docutils literal"><span class="pre">ignore_limits_on_local_network</span></tt>, if set to true, upload, download and

2612

unchoke limits are ignored for peers on the local network.</p>

2613

<p><tt class="docutils literal"><span class="pre">connection_speed</span></tt> is the number of connection attempts that

2614

are made per second. If a number <= 0 is specified, it will default to

2615

200 connections per second.</p>

2616

<p><tt class="docutils literal"><span class="pre">send_redundant_have</span></tt> controls if have messages will be sent

2617

to peers that already have the piece. This is typically not necessary,

2618

but it might be necessary for collecting statistics in some cases.

2619

Default is false.</p>

2620

<p><tt class="docutils literal"><span class="pre">lazy_bitfields</span></tt> prevents outgoing bitfields from being full. If the

2621

client is seed, a few bits will be set to 0, and later filled in with

2622

have-messages. This is to prevent certain ISPs from stopping people

2623

from seeding.</p>

2624

<p><tt class="docutils literal"><span class="pre">inactivity_timeout</span></tt>, if a peer is uninteresting and uninterested

2625

for longer than this number of seconds, it will be disconnected.

2626

Default is 10 minutes</p>

2627

<p><tt class="docutils literal"><span class="pre">unchoke_interval</span></tt> is the number of seconds between chokes/unchokes.

2628

On this interval, peers are re-evaluated for being choked/unchoked. This

2629

is defined as 30 seconds in the protocol, and it should be significantly

2630

longer than what it takes for TCP to ramp up to it's max rate.</p>

2631

<p><tt class="docutils literal"><span class="pre">optimistic_unchoke_multiplier</span></tt> is the number of unchoke intervals between

2632

each <em>optimistic</em> unchoke interval. On this timer, the currently optimistically

2633

unchoked peer will change.</p>

2634

<p><tt class="docutils literal"><span class="pre">announce_ip</span></tt> is the ip address passed along to trackers as the <tt class="docutils literal"><span class="pre">&ip=</span></tt> parameter.

2635

If left as the default (default constructed), that parameter is ommited.</p>

2636

<p><tt class="docutils literal"><span class="pre">num_want</span></tt> is the number of peers we want from each tracker request. It defines

2637

what is sent as the <tt class="docutils literal"><span class="pre">&num_want=</span></tt> parameter to the tracker.</p>

2638

<p><tt class="docutils literal"><span class="pre">initial_picker_threshold</span></tt> specifies the number of pieces we need before we

2639

switch to rarest first picking. This defaults to 4, which means the 4 first

2640

pieces in any torrent are picked at random, the following pieces are picked

2641

in rarest first order.</p>

2642

<p><tt class="docutils literal"><span class="pre">allowed_fast_set_size</span></tt> is the number of pieces we allow peers to download

2643

from us without being unchoked.</p>

2644

<p><tt class="docutils literal"><span class="pre">max_outstanding_disk_bytes_per_connection</span></tt> is the number of bytes each

2645

connection is allowed to have waiting in the disk I/O queue before it is

2646

throttled back. This limit is meant to stop fast internet connections to

2647

queue up bufferes indefinitely on slow hard-drives or storage.</p>

2648

<p><tt class="docutils literal"><span class="pre">handshake_timeout</span></tt> specifies the number of seconds we allow a peer to

2649

delay responding to a protocol handshake. If no response is received within

2650

this time, the connection is closed.</p>

2651

<p><tt class="docutils literal"><span class="pre">use_dht_as_fallback</span></tt> determines how the DHT is used. If this is true

2652

(which it is by default), the DHT will only be used for torrents where

2653

all trackers in its tracker list has failed. Either by an explicit error

2654

message or a time out.</p>

2655

<p><tt class="docutils literal"><span class="pre">free_torrent_hashes</span></tt> determines whether or not the torrent's piece hashes

2656

are kept in memory after the torrent becomes a seed or not. If it is set to

2657

<tt class="docutils literal"><span class="pre">true</span></tt> the hashes are freed once the torrent is a seed (they're not

2658

needed anymore since the torrent won't download anything more). If it's set

2659

to false they are not freed. If they are freed, the <a class="reference internal" href="#torrent-info">torrent_info</a> returned

2660

by get_torrent_info() will return an object that may be incomplete, that

2661

cannot be passed back to <a class="reference internal" href="#add-torrent">add_torrent()</a> for instance.</p>

2662

<p><tt class="docutils literal"><span class="pre">upnp_ignore_nonrouters</span></tt> indicates whether or not the UPnP implementation

2663

should ignore any broadcast response from a device whose address is not the

2664

configured router for this machine. i.e. it's a way to not talk to other

2665

people's routers by mistake.</p>

2666

</div>

2667

2668

<h1>pe_settings</h1>

2669

<p>The <tt class="docutils literal"><span class="pre">pe_settings</span></tt> structure is used to control the settings related

2670

to peer protocol encryption:</p>

2671

2672

struct pe_settings

2673

{

2674

pe_settings();

2675

2676

enum enc_policy

2677

{

2678

forced,

2679

enabled,

2680

disabled

2681

};

2682

2683

enum enc_level

2684

{

2685

plaintext,

2686

rc4,

2687

both

2688

};

2689

2690

enc_policy out_enc_policy;

2691

enc_policy in_enc_policy;

2692

enc_level allowed_enc_level;

2693

bool prefer_rc4;

2694

};

2695

</pre>

2696

<p><tt class="docutils literal"><span class="pre">in_enc_policy</span></tt> and <tt class="docutils literal"><span class="pre">out_enc_policy</span></tt> control the settings for incoming

2697

and outgoing connections respectively. The settings for these are:</p>

2698

2699

2700

<li><tt class="docutils literal"><span class="pre">forced</span></tt> - Only encrypted connections are allowed. Incoming connections

2701

that are not encrypted are closed and if the encrypted outgoing connection

2702

fails, a non-encrypted retry will not be made.</li>

2703

<li><tt class="docutils literal"><span class="pre">enabled</span></tt> - encrypted connections are enabled, but non-encrypted

2704

connections are allowed. An incoming non-encrypted connection will

2705

be accepted, and if an outgoing encrypted connection fails, a non-

2706

encrypted connection will be tried.</li>

2707

<li><tt class="docutils literal"><span class="pre">disabled</span></tt> - only non-encrypted connections are allowed.</li>

2708

</ul>

2709

</blockquote>

2710

<p><tt class="docutils literal"><span class="pre">allowed_enc_level</span></tt> determines the encryption level of the

2711

connections. This setting will adjust which encryption scheme is

2712

offered to the other peer, as well as which encryption scheme is

2713

selected by the client. The settings are:</p>

2714

2715

2716

<li><tt class="docutils literal"><span class="pre">plaintext</span></tt> - only the handshake is encrypted, the bulk of the traffic

2717

remains unchanged.</li>

2718

<li><tt class="docutils literal"><span class="pre">rc4</span></tt> - the entire stream is encrypted with RC4</li>

2719

<li><tt class="docutils literal"><span class="pre">both</span></tt> - both RC4 and plaintext connections are allowed.</li>

2720

</ul>

2721

</blockquote>

2722

<p><tt class="docutils literal"><span class="pre">prefer_rc4</span></tt> can be set to true if you want to prefer the RC4 encrypted stream.</p>

2723

</div>

2724

2725

<h1>proxy_settings</h1>

2726

<p>The <tt class="docutils literal"><span class="pre">proxy_settings</span></tt> structs contains the information needed to

2727

direct certain traffic to a proxy.</p>

2728

2729

2730

struct proxy_settings

2731

{

2732

proxy_settings();

2733

2734

std::string hostname;

2735

int port;

2736

2737

std::string username;

2738

std::string password;

2739

2740

enum proxy_type

2741

{

2742

none,

2743

socks4,

2744

socks5,

2745

socks5_pw,

2746

http,

2747

http_pw

2748

};

2749

2750

proxy_type type;

2751

};

2752

</pre>

2753

</blockquote>

2754

<p><tt class="docutils literal"><span class="pre">hostname</span></tt> is the name or IP of the proxy server. <tt class="docutils literal"><span class="pre">port</span></tt> is the

2755

port number the proxy listens to. If required, <tt class="docutils literal"><span class="pre">username</span></tt> and <tt class="docutils literal"><span class="pre">password</span></tt>

2756

can be set to authenticate with the proxy.</p>

2757

<p>The <tt class="docutils literal"><span class="pre">type</span></tt> tells libtorrent what kind of proxy server it is. The following

2758

options are available:</p>

2759

2760

2761

<li><tt class="docutils literal"><span class="pre">none</span></tt> - This is the default, no proxy server is used, all other fields

2762

are ignored.</li>

2763

<li><tt class="docutils literal"><span class="pre">socks4</span></tt> - The server is assumed to be a <a class="reference external" href="http://www.ufasoft.com/doc/socks4_protocol.htm">SOCKS4 server</a> that

2764

requires a username.</li>

2765

<li><tt class="docutils literal"><span class="pre">socks5</span></tt> - The server is assumed to be a SOCKS5 server (<a class="reference external" href="http://www.faqs.org/rfcs/rfc1928.html">RFC 1928</a>) that

2766

does not require any authentication. The username and password are ignored.</li>

2767

<li><tt class="docutils literal"><span class="pre">socks5_pw</span></tt> - The server is assumed to be a SOCKS5 server that supports

2768

plain text username and password authentication (<a class="reference external" href="http://www.faqs.org/rfcs/rfc1929.html">RFC 1929</a>). The username

2769

and password specified may be sent to the proxy if it requires.</li>

2770

<li><tt class="docutils literal"><span class="pre">http</span></tt> - The server is assumed to be an HTTP proxy. If the transport used

2771

for the connection is non-HTTP, the server is assumed to support the

2772

<a class="reference external" href="draft-luotonen-web-proxy-tunneling-01.txt">CONNECT</a> method. i.e. for web seeds and HTTP trackers, a plain proxy will

2773

suffice. The proxy is assumed to not require authorization. The username

2774

and password will not be used.</li>

2775

<li><tt class="docutils literal"><span class="pre">http_pw</span></tt> - The server is assumed to be an HTTP proxy that requires

2776

user authorization. The username and password will be sent to the proxy.</li>

2777

</ul>

2778

</blockquote>

2779

</div>

2780

2781

<h1>ip_filter</h1>

2782

<p>The <tt class="docutils literal"><span class="pre">ip_filter</span></tt> class is a set of rules that uniquely categorizes all

2783

ip addresses as allowed or disallowed. The default constructor creates

2784

a single rule that allows all addresses (0.0.0.0 - 255.255.255.255 for

2785

the IPv4 range, and the equivalent range covering all addresses for the

2786

IPv6 range).</p>

2787

2788

2789

template <class Addr>

2790

struct ip_range

2791

{

2792

Addr first;

2793

Addr last;

2794

int flags;

2795

};

2796

2797

class ip_filter

2798

{

2799

public:

2800

enum access_flags { blocked = 1 };

2801

2802

ip_filter();

2803

void add_rule(address first, address last, int flags);

2804

int access(address const& addr) const;

2805

2806

typedef boost::tuple<std::vector<ip_range<address_v4> >

2807

, std::vector<ip_range<address_v6> > > filter_tuple_t;

2808

2809

filter_tuple_t export_filter() const;

2810

};

2811

</pre>

2812

</blockquote>

2813

2814

<h2>ip_filter()</h2>

2815

2816

2817

ip_filter()

2818

</pre>

2819

</blockquote>

2820

<p>Creates a default filter that doesn't filter any address.</p>

2821

<p>postcondition:

2822

<tt class="docutils literal"><span class="pre">access(x)</span> <span class="pre">==</span> <span class="pre">0</span></tt> for every <tt class="docutils literal"><span class="pre">x</span></tt></p>

2823

</div>

2824

2825

2826

2827

2828

void add_rule(address first, address last, int flags);

2829

</pre>

2830

</blockquote>

2831

<p>Adds a rule to the filter. <tt class="docutils literal"><span class="pre">first</span></tt> and <tt class="docutils literal"><span class="pre">last</span></tt> defines a range of

2832

ip addresses that will be marked with the given flags. The <tt class="docutils literal"><span class="pre">flags</span></tt>

2833

can currently be 0, which means allowed, or <tt class="docutils literal"><span class="pre">ip_filter::blocked</span></tt>, which

2834

means disallowed.</p>

2835

<p>precondition:

2836

<tt class="docutils literal"><span class="pre">first.is_v4()</span> <span class="pre">==</span> <span class="pre">last.is_v4()</span> <span class="pre">&&</span> <span class="pre">first.is_v6()</span> <span class="pre">==</span> <span class="pre">last.is_v6()</span></tt></p>

2837

<p>postcondition:

2838

<tt class="docutils literal"><span class="pre">access(x)</span> <span class="pre">==</span> <span class="pre">flags</span></tt> for every <tt class="docutils literal"><span class="pre">x</span></tt> in the range [<tt class="docutils literal"><span class="pre">first</span></tt>, <tt class="docutils literal"><span class="pre">last</span></tt>]</p>

2839

<p>This means that in a case of overlapping ranges, the last one applied takes

2840

precedence.</p>

2841

</div>

2842

2843

<h2>access()</h2>

2844

2845

2846

int access(address const& addr) const;

2847

</pre>

2848

</blockquote>

2849

<p>Returns the access permissions for the given address (<tt class="docutils literal"><span class="pre">addr</span></tt>). The permission

2850

can currently be 0 or <tt class="docutils literal"><span class="pre">ip_filter::blocked</span></tt>. The complexity of this operation

2851

is O(<tt class="docutils literal"><span class="pre">log</span></tt> n), where n is the minimum number of non-overlapping ranges to describe

2852

the current filter.</p>

2853

</div>

2854

2855

<h2>export_filter()</h2>

2856

2857

2858

boost::tuple<std::vector<ip_range<address_v4> >

2859

, std::vector<ip_range<address_v6> > > export_filter() const;

2860

</pre>

2861

</blockquote>

2862

<p>This function will return the current state of the filter in the minimum number of

2863

ranges possible. They are sorted from ranges in low addresses to high addresses. Each

2864

entry in the returned vector is a range with the access control specified in its

2865

<tt class="docutils literal"><span class="pre">flags</span></tt> field.</p>

2866

<p>The return value is a tuple containing two range-lists. One for IPv4 addresses

2867

and one for IPv6 addresses.</p>

2868

</div>

2869

</div>

2870

2871

<h1>big_number</h1>

2872

<p>Both the <tt class="docutils literal"><span class="pre">peer_id</span></tt> and <tt class="docutils literal"><span class="pre">sha1_hash</span></tt> types are typedefs of the class

2873

<tt class="docutils literal"><span class="pre">big_number</span></tt>. It represents 20 bytes of data. Its synopsis follows:</p>

2874

2875

class big_number

2876

{

2877

public:

2878

bool operator==(const big_number& n) const;

2879

bool operator!=(const big_number& n) const;

2880

bool operator<(const big_number& n) const;

2881

2882

const unsigned char* begin() const;

2883

const unsigned char* end() const;

2884

2885

unsigned char* begin();

2886

unsigned char* end();

2887

};

2888

</pre>

2889

<p>The iterators gives you access to individual bytes.</p>

2890

</div>

2891

2892

<h1>hasher</h1>

2893

<p>This class creates sha1-hashes. Its declaration looks like this:</p>

2894

2895

class hasher

2896

{

2897

public:

2898

hasher();

2899

hasher(char const* data, unsigned int len);

2900

2901

void update(char const* data, unsigned int len);

2902

sha1_hash final();

2903

void reset();

2904

};

2905

</pre>

2906

<p>You use it by first instantiating it, then call <tt class="docutils literal"><span class="pre">update()</span></tt> to feed it

2907

with data. i.e. you don't have to keep the entire buffer of which you want to

2908

create the hash in memory. You can feed the hasher parts of it at a time. When

2909

You have fed the hasher with all the data, you call <tt class="docutils literal"><span class="pre">final()</span></tt> and it

2910

will return the sha1-hash of the data.</p>

2911

<p>The constructor that takes a <tt class="docutils literal"><span class="pre">char</span> <span class="pre">const*</span></tt> and an integer will construct the

2912

sha1 context and feed it the data passed in.</p>

2913

<p>If you want to reuse the hasher object once you have created a hash, you have to

2914

call <tt class="docutils literal"><span class="pre">reset()</span></tt> to reinitialize it.</p>

2915

<p>The sha1-algorithm used was implemented by Steve Reid and released as public domain.

2916

For more info, see <tt class="docutils literal"><span class="pre">src/sha1.cpp</span></tt>.</p>

2917

</div>

2918

2919

<h1>fingerprint</h1>

2920

<p>The fingerprint class represents information about a client and its version. It is used

2921

to encode this information into the client's peer id.</p>

2922

<p>This is the class declaration:</p>

2923

2924

struct fingerprint

2925

{

2926

fingerprint(const char* id_string, int major, int minor

2927

, int revision, int tag);

2928

2929

std::string to_string() const;

2930

2931

char name[2];

2932

char major_version;

2933

char minor_version;

2934

char revision_version;

2935

char tag_version;

2936

2937

};

2938

</pre>

2939

<p>The constructor takes a <tt class="docutils literal"><span class="pre">char</span> <span class="pre">const*</span></tt> that should point to a string constant containing

2940

exactly two characters. These are the characters that should be unique for your client. Make

2941

sure not to clash with anybody else. Here are some taken id's:</p>

2942

2943

2944

2945

2946

</colgroup>

2947

2948

<tr><th class="head">id chars</th>

2949

<th class="head">client</th>

2950

</tr>

2951

</thead>

2952

2953

2954

<td>Azureus</td>

2955

</tr>

2956

2957

<td>libtorrent (default)</td>

2958

</tr>

2959

2960

<td>BittorrentX</td>

2961

</tr>

2962

2963

<td>Moonlight Torrent</td>

2964

</tr>

2965

2966

<td>Torrent Storm</td>

2967

</tr>

2968

2969

<td>Swarm Scope</td>

2970

</tr>

2971

2972

<td>Xan Torrent</td>

2973

</tr>

2974

</tbody>

2975

</table>

2976

<p>There's currently an informal directory of client id's <a class="reference external" href="http://wiki.theory.org/BitTorrentSpecification#peer_id">here</a>.</p>

2977

<p>The <tt class="docutils literal"><span class="pre">major</span></tt>, <tt class="docutils literal"><span class="pre">minor</span></tt>, <tt class="docutils literal"><span class="pre">revision</span></tt> and <tt class="docutils literal"><span class="pre">tag</span></tt> parameters are used to identify the

2978

version of your client. All these numbers must be within the range [0, 9].</p>

2979

<p><tt class="docutils literal"><span class="pre">to_string()</span></tt> will generate the actual string put in the peer-id, and return it.</p>

2980

</div>

2981

2982

<h1>free functions</h1>

2983

2984

<h2>identify_client()</h2>

2985

2986

2987

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

2988

</pre>

2989

</blockquote>

2990

<p>This function is declared in the header <tt class="docutils literal"><span class="pre"><libtorrent/identify_client.hpp></span></tt>. It can can be used

2991

to extract a string describing a client version from its peer-id. It will recognize most clients

2992

that have this kind of identification in the peer-id.</p>

2993

</div>

2994

2995

<h2>client_fingerprint()</h2>

2996

2997

2998

boost::optional<fingerprint> client_fingerprint(peer_id const& p);

2999

</pre>

3000

</blockquote>

3001

<p>Returns an optional fingerprint if any can be identified from the peer id. This can be used

3002

to automate the identification of clients. It will not be able to identify peers with non-

3003

standard encodings. Only Azureus style, Shadow's style and Mainline style. This function is

3004

declared in the header <tt class="docutils literal"><span class="pre"><libtorrent/identify_client.hpp></span></tt>.</p>

3005

</div>

3006

3007

<h2>bdecode() bencode()</h2>

3008

3009

3010

template<class InIt> entry bdecode(InIt start, InIt end);

3011

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

3012

</pre>

3013

</blockquote>

3014

<p>These functions will encode data to <a class="reference external" href="http://wiki.theory.org/index.php/BitTorrentSpecification">bencoded</a> or decode <a class="reference external" href="http://wiki.theory.org/index.php/BitTorrentSpecification">bencoded</a> data.</p>

3015

<p>The <a class="reference internal" href="#entry">entry</a> class is the internal representation of the bencoded data

3016

and it can be used to retrieve information, an <a class="reference internal" href="#entry">entry</a> can also be build by

3017

the program and given to <tt class="docutils literal"><span class="pre">bencode()</span></tt> to encode it into the <tt class="docutils literal"><span class="pre">OutIt</span></tt>

3018

iterator.</p>

3019

<p>The <tt class="docutils literal"><span class="pre">OutIt</span></tt> and <tt class="docutils literal"><span class="pre">InIt</span></tt> are iterators

3020

(<a class="reference external" href="http://www.sgi.com/tech/stl/InputIterator.html">InputIterator</a> and <a class="reference external" href="http://www.sgi.com/tech/stl/OutputIterator.html">OutputIterator</a> respectively). They

3021

are templates and are usually instantiated as <a class="reference external" href="http://www.sgi.com/tech/stl/ostream_iterator.html">ostream_iterator</a>,

3022

<a class="reference external" href="http://www.sgi.com/tech/stl/back_insert_iterator.html">back_insert_iterator</a> or <a class="reference external" href="http://www.sgi.com/tech/stl/istream_iterator.html">istream_iterator</a>. These

3023

functions will assume that the iterator refers to a character

3024

(<tt class="docutils literal"><span class="pre">char</span></tt>). So, if you want to encode entry <tt class="docutils literal"><span class="pre">e</span></tt> into a buffer

3025

in memory, you can do it like this:</p>

3026

3027

std::vector<char> buffer;

3028

bencode(std::back_inserter(buf), e);

3029

</pre>

3030

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

3031

3032

std::vector<char> buffer;

3033

// ...

3034

entry e = bdecode(buf.begin(), buf.end());

3035

</pre>

3036

<p>Or, if you have a raw char buffer:</p>

3037

3038

const char* buf;

3039

// ...

3040

entry e = bdecode(buf, buf + data_size);

3041

</pre>

3042

<p>Now we just need to know how to retrieve information from the <a class="reference internal" href="#entry">entry</a>.</p>

3043

<p>If <tt class="docutils literal"><span class="pre">bdecode()</span></tt> encounters invalid encoded data in the range given to it

3044

it will throw <a class="reference internal" href="#invalid-encoding">invalid_encoding</a>.</p>

3045

</div>

3046

</div>

3047

3048

<h1>alerts</h1>

3049

<p>The <tt class="docutils literal"><span class="pre">pop_alert()</span></tt> function on session is the interface for retrieving

3050

alerts, warnings, messages and errors from libtorrent. If there hasn't

3051

occurred any errors (matching your severity level) <tt class="docutils literal"><span class="pre">pop_alert()</span></tt> will

3052

return a zero pointer. If there has been some error, it will return a pointer

3053

to an alert object describing it. You can then use the alert object and query

3054

it for information about the error or message. To retrieve any alerts, you have

3055

to select a severity level using <tt class="docutils literal"><span class="pre">session::set_severity_level()</span></tt>. It defaults to

3056

<tt class="docutils literal"><span class="pre">alert::none</span></tt>, which means that you don't get any messages at all, ever.

3057

You have the following levels to select among:</p>

3058

3059

3060

3061

3062

</colgroup>

3063

3064

3065

<td>No alert will ever have this severity level, which

3066

effectively filters all messages.</td>

3067

</tr>

3068

<tr><td><tt class="docutils literal"><span class="pre">fatal</span></tt></td>

3069

<td>Fatal errors will have this severity level. Examples can

3070

be disk full or something else that will make it

3071

impossible to continue normal execution.</td>

3072

</tr>

3073

<tr><td><tt class="docutils literal"><span class="pre">critical</span></tt></td>

3074

<td>Signals errors that requires user interaction or

3075

messages that almost never should be ignored. For

3076

example, a chat message received from another peer is

3077

announced as severity <tt class="docutils literal"><span class="pre">critical</span></tt>.</td>

3078

</tr>

3079

<tr><td><tt class="docutils literal"><span class="pre">warning</span></tt></td>

3080

<td>Messages with the warning severity can be a tracker that

3081

times out or responds with invalid data. It will be

3082

retried automatically, and the possible next tracker in

3083

a multitracker sequence will be tried. It does not

3084

require any user interaction.</td>

3085

</tr>

3086

3087

<td>Events that can be considered normal, but still deserves

3088

an event. This could be a piece hash that fails.</td>

3089

</tr>

3090

<tr><td><tt class="docutils literal"><span class="pre">debug</span></tt></td>

3091

<td>This will include a lot of debug events that can be used

3092

both for debugging libtorrent but also when debugging

3093

other clients that are connected to libtorrent. It will

3094

report strange behaviors among the connected peers.</td>

3095

</tr>

3096

</tbody>

3097

</table>

3098

<p>When setting a severity level, you will receive messages of that severity and all

3099

messages that are more sever. If you set <tt class="docutils literal"><span class="pre">alert::none</span></tt> (the default) you will not receive

3100

any events at all.</p>

3101

<p>When you set a severity level other than <tt class="docutils literal"><span class="pre">none</span></tt>, you have the responsibility to call

3102

<tt class="docutils literal"><span class="pre">pop_alert()</span></tt> from time to time. If you don't do that, the alert queue will just grow.</p>

3103

<p>When you get an alert, you can use <tt class="docutils literal"><span class="pre">typeid()</span></tt> or <tt class="docutils literal"><span class="pre">dynamic_cast<></span></tt> to get more detailed

3104

information on exactly which type it is. i.e. what kind of error it is. You can also use a

3105

<a class="reference internal" href="#dispatcher">dispatcher</a> mechanism that's available in libtorrent.</p>

3106

<p>All alert types are defined in the <tt class="docutils literal"><span class="pre"><libtorrent/alert_types.hpp></span></tt> header file.</p>

3107

<p>The <tt class="docutils literal"><span class="pre">alert</span></tt> class is the base class that specific messages are derived from. This

3108

is its synopsis:</p>

3109

3110

class alert

3111

{

3112

public:

3113

3114

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

3115

3116

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

3117

virtual ~alert();

3118

3119

std::string const& msg() const;

3120

severity_t severity() const;

3121

3122

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

3123

};

3124

</pre>

3125

<p>This means that all alerts have at least a string describing it. They also

3126

have a severity level that can be used to sort them or present them to the

3127

user in different ways.</p>

3128

<p>There's another alert base class that all most alerts derives from, all the

3129

alerts that are generated for a specific torrent are derived from:</p>

3130

3131

struct torrent_alert: alert

3132

{

3133

torrent_alert(torrent_handle const& h, severity_t s, std::string const& msg);

3134

3135

torrent_handle handle;

3136

};

3137

</pre>

3138

<p>The specific alerts, that all derives from <tt class="docutils literal"><span class="pre">alert</span></tt>, are:</p>

3139

3140

<h2>listen_failed_alert</h2>

3141

<p>This alert is generated when none of the ports, given in the port range, to

3142

<a class="reference internal" href="#session">session</a> can be opened for listening. This alert is generated as severity

3143

level <tt class="docutils literal"><span class="pre">fatal</span></tt>.</p>

3144

3145

struct listen_failed_alert: alert

3146

{

3147

listen_failed_alert(const std::string& msg);

3148

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

3149

};

3150

</pre>

3151

</div>

3152

3153

<h2>portmap_error_alert</h2>

3154

<p>This alert is generated when a NAT router was successfully found but some

3155

part of the port mapping request failed. It contains a text message that

3156

may help the user figure out what is wrong. This alert is not generated in

3157

case it appears the client is not running on a NAT:ed network or if it

3158

appears there is no NAT router that can be remote controlled to add port

3159

mappings.</p>

3160

<p>The alert is generated as severity <tt class="docutils literal"><span class="pre">warning</span></tt>, since it should be displayed

3161

to the user somehow, and could mean reduced preformance.</p>

3162

3163

struct portmap_error_alert: alert

3164

{

3165

portmap_error_alert(const std::string& msg);

3166

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

3167

};

3168

</pre>

3169

</div>

3170

3171

<h2>portmap_alert</h2>

3172

<p>This alert is generated when a NAT router was successfully found and

3173

a port was successfully mapped on it. On a NAT:ed network with a NAT-PMP

3174

capable router, this is typically generated once when mapping the TCP

3175

port and, if DHT is enabled, when the UDP port is mapped. This is merely

3176

an informational alert, and is generated at severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3177

3178

struct portmap_alert: alert

3179

{

3180

portmap_alert(const std::string& msg);

3181

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

3182

};

3183

</pre>

3184

</div>

3185

3186

<h2>file_error_alert</h2>

3187

<p>If the storage fails to read or write files that it needs access to, this alert is

3188

generated and the torrent is paused. It is generated as severity level <tt class="docutils literal"><span class="pre">fatal</span></tt>.</p>

3189

3190

struct file_error_alert: torrent_alert

3191

{

3192

file_error_alert(

3193

const torrent_handle& h

3194

, const std::string& msg);

3195

3196

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

3197

};

3198

</pre>

3199

</div>

3200

3201

<h2>tracker_announce_alert</h2>

3202

<p>This alert is generated each time a tracker announce is sent (or attempted to be sent).

3203

It is generated at severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3204

3205

struct tracker_announce_alert: torrent_alert

3206

{

3207

tracker_announce_alert(

3208

const torrent_handle& h

3209

, const std::string& msg);

3210

3211

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

3212

};

3213

</pre>

3214

</div>

3215

3216

<h2>tracker_alert</h2>

3217

<p>This alert is generated on tracker time outs, premature disconnects, invalid response or

3218

a HTTP response other than "200 OK". From the alert you can get the handle to the torrent

3219

the tracker belongs to. This alert is generated as severity level <tt class="docutils literal"><span class="pre">warning</span></tt>.</p>

3220

<p>The <tt class="docutils literal"><span class="pre">times_in_row</span></tt> member says how many times in a row this tracker has failed.

3221

<tt class="docutils literal"><span class="pre">status_code</span></tt> is the code returned from the HTTP server. 401 means the tracker needs

3222

authentication, 404 means not found etc. If the tracker timed out, the code will be set

3223

to 0.</p>

3224

3225

struct tracker_alert: torrent_alert

3226

{

3227

tracker_alert(torrent_handle const& h, int times, int status

3228

, const std::string& msg);

3229

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

3230

3231

int times_in_row;

3232

int status_code;

3233

};

3234

</pre>

3235

</div>

3236

3237

<h2>tracker_reply_alert</h2>

3238

<p>This alert is only for informational purpose. It is generated when a tracker announce

3239

succeeds. It is generated regardless what kind of tracker was used, be it UDP, HTTP or

3240

the DHT. It is generated with severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3241

3242

struct tracker_reply_alert: torrent_alert

3243

{

3244

tracker_reply_alert(const torrent_handle& h

3245

, int num_peers

3246

, const std::string& msg);

3247

3248

int num_peers;

3249

3250

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

3251

};

3252

</pre>

3253

<p>The <tt class="docutils literal"><span class="pre">num_peers</span></tt> tells how many peers were returned from the tracker. This is

3254

not necessarily all new peers, some of them may already be connected.</p>

3255

</div>

3256

3257

<h2>tracker_warning_alert</h2>

3258

<p>This alert is triggered if the tracker reply contains a warning field. Usually this

3259

means that the tracker announce was successful, but the tracker has a message to

3260

the client. The message string in the alert will contain the warning message from

3261

the tracker. It is generated with severity level <tt class="docutils literal"><span class="pre">warning</span></tt>.</p>

3262

3263

struct tracker_warning_alert: torrent_alert

3264

{

3265

tracker_warning_alert(torrent_handle const& h

3266

, std::string const& msg);

3267

3268

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

3269

};

3270

</pre>

3271

</div>

3272

3273

<h2>scrape_reply_alert</h2>

3274

3275

struct scrape_reply_alert: torrent_alert

3276

{

3277

scrape_reply_alert(torrent_handle const& h

3278

, int incomplete_

3279

, int complete_

3280

, std::string const& msg);

3281

3282

int incomplete;

3283

int complete;

3284

3285

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

3286

};

3287

</pre>

3288

<p>This alert is generated when a scrape request succeeds. <tt class="docutils literal"><span class="pre">incomplete</span></tt>

3289

and <tt class="docutils literal"><span class="pre">complete</span></tt> is the data returned in the scrape response. These numbers

3290

may be -1 if the reponse was malformed.</p>

3291

</div>

3292

3293

<h2>scrape_failed_alert</h2>

3294

3295

struct scrape_failed_alert: torrent_alert

3296

{

3297

scrape_failed_alert(torrent_handle const& h

3298

, std::string const& msg);

3299

3300

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

3301

};

3302

</pre>

3303

<p>If a scrape request fails, this alert is generated. This might be due

3304

to the tracker timing out, refusing connection or returning an http response

3305

code indicating an error.</p>

3306

</div>

3307

3308

<h2>url_seed_alert</h2>

3309

<p>This alert is generated when a HTTP seed name lookup fails. This alert is

3310

generated as severity level <tt class="docutils literal"><span class="pre">warning</span></tt>.</p>

3311

<p>It contains <tt class="docutils literal"><span class="pre">url</span></tt> to the HTTP seed that failed along with an error message.</p>

3312

3313

struct url_seed_alert: torrent_alert

3314

{

3315

url_seed_alert(torrent_handle const& h, std::string const& url

3316

, const std::string& msg);

3317

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

3318

3319

std::string url;

3320

};

3321

</pre>

3322

</div>

3323

3324

<h2>hash_failed_alert</h2>

3325

<p>This alert is generated when a finished piece fails its hash check. You can get the handle

3326

to the torrent which got the failed piece and the index of the piece itself from the alert.

3327

This alert is generated as severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3328

3329

struct hash_failed_alert: torrent_alert

3330

{

3331

hash_failed_alert(

3332

torrent_handle const& h

3333

, int index

3334

, const std::string& msg);

3335

3336

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

3337

3338

int piece_index;

3339

};

3340

</pre>

3341

</div>

3342

3343

<h2>peer_ban_alert</h2>

3344

<p>This alert is generated when a peer is banned because it has sent too many corrupt pieces

3345

to us. It is generated at severity level <tt class="docutils literal"><span class="pre">info</span></tt>. The <tt class="docutils literal"><span class="pre">handle</span></tt> member is a <a class="reference internal" href="#torrent-handle">torrent_handle</a>

3346

to the torrent that this peer was a member of.</p>

3347

3348

struct peer_ban_alert: torrent_alert

3349

{

3350

peer_ban_alert(

3351

asio::ip::tcp::endpoint const& pip

3352

, torrent_handle h

3353

, const std::string& msg);

3354

3355

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

3356

3357

asio::ip::tcp::endpoint ip;

3358

};

3359

</pre>

3360

</div>

3361

3362

<h2>peer_error_alert</h2>

3363

<p>This alert is generated when a peer sends invalid data over the peer-peer protocol. The peer

3364

will be disconnected, but you get its ip address from the alert, to identify it. This alert

3365

is generated as severity level <tt class="docutils literal"><span class="pre">debug</span></tt>.</p>

3366

3367

struct peer_error_alert: alert

3368

{

3369

peer_error_alert(

3370

asio::ip::tcp::endpoint const& pip

3371

, peer_id const& pid

3372

, const std::string& msg);

3373

3374

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

3375

asio::ip::tcp::endpoint ip;

3376

peer_id id;

3377

};

3378

</pre>

3379

</div>

3380

3381

<h2>invalid_request_alert</h2>

3382

<p>This is a debug alert that is generated by an incoming invalid piece request. The <tt class="docutils literal"><span class="pre">handle</span></tt>

3383

is a handle to the torrent the peer is a member of. <tt class="docutils literal"><span class="pre">ìp</span></tt> is the address of the peer and the

3384

<tt class="docutils literal"><span class="pre">request</span></tt> is the actual incoming request from the peer. The alert is generated as severity level

3385

<tt class="docutils literal"><span class="pre">debug</span></tt>.</p>

3386

3387

struct invalid_request_alert: torrent_alert

3388

{

3389

invalid_request_alert(

3390

peer_request const& r

3391

, torrent_handle const& h

3392

, asio::ip::tcp::endpoint const& send

3393

, peer_id const& pid

3394

, std::string const& msg);

3395

3396

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

3397

3398

asio::ip::tcp::endpoint ip;

3399

peer_request request;

3400

peer_id id;

3401

};

3402

3403

3404

struct peer_request

3405

{

3406

int piece;

3407

int start;

3408

int length;

3409

bool operator==(peer_request const& r) const;

3410

};

3411

</pre>

3412

<p>The <tt class="docutils literal"><span class="pre">peer_request</span></tt> contains the values the client sent in its <tt class="docutils literal"><span class="pre">request</span></tt> message. <tt class="docutils literal"><span class="pre">piece</span></tt> is

3413

the index of the piece it want data from, <tt class="docutils literal"><span class="pre">start</span></tt> is the offset within the piece where the data

3414

should be read, and <tt class="docutils literal"><span class="pre">length</span></tt> is the amount of data it wants.</p>

3415

</div>

3416

3417

<h2>torrent_finished_alert</h2>

3418

<p>This alert is generated when a torrent switches from being a downloader to a seed.

3419

It will only be generated once per torrent. It contains a torrent_handle to the

3420

torrent in question. This alert is generated as severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3421

3422

struct torrent_finished_alert: torrent_alert

3423

{

3424

torrent_finished_alert(

3425

const torrent_handle& h

3426

, const std::string& msg);

3427

3428

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

3429

};

3430

</pre>

3431

</div>

3432

3433

<h2>metadata_failed_alert</h2>

3434

<p>This alert is generated when the metadata has been completely received and the info-hash

3435

failed to match it. i.e. the metadata that was received was corrupt. libtorrent will

3436

automatically retry to fetch it in this case. This is only relevant when running a

3437

torrent-less download, with the metadata extension provided by libtorrent.

3438

It is generated at severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3439

3440

struct metadata_failed_alert: torrent_alert

3441

{

3442

metadata_failed_alert(

3443

torrent_handle const& h

3444

, std::string const& msg);

3445

3446

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

3447

};

3448

</pre>

3449

</div>

3450

3451

<h2>metadata_received_alert</h2>

3452

<p>This alert is generated when the metadata has been completely received and the torrent

3453

can start downloading. It is not generated on torrents that are started with metadata, but

3454

only those that needs to download it from peers (when utilizing the libtorrent extension).

3455

It is generated at severity level <tt class="docutils literal"><span class="pre">info</span></tt>.</p>

3456

3457

struct metadata_received_alert: torrent_alert

3458

{

3459

metadata_received_alert(

3460

torrent_handle const_& h

3461

, std::string const& msg);

3462

3463

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

3464

};

3465

</pre>

3466

</div>

3467

3468

<h2>fastresume_rejected_alert</h2>

3469

<p>This alert is generated when a fastresume file has been passed to <tt class="docutils literal"><span class="pre">add_torrent</span></tt> but the

3470

files on disk did not match the fastresume file. The string explains the reason why the

3471

resume file was rejected. It is generated at severity level <tt class="docutils literal"><span class="pre">warning</span></tt>.</p>

3472

3473

struct fastresume_rejected_alert: torrent_alert

3474

{

3475

fastresume_rejected_alert(torrent_handle const& h

3476

, std::string const& msg);

3477

3478

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

3479

};

3480

</pre>

3481

</div>

3482

3483

<h2>peer_blocked_alert</h2>

3484

<p>This alert is generated when a peer is blocked by the IP filter. It has the severity leve

3485

<tt class="docutils literal"><span class="pre">info</span></tt>. The <tt class="docutils literal"><span class="pre">ip</span></tt> member is the address that was blocked.</p>

3486

3487

struct peer_blocked_alert: alert

3488

{

3489

peer_blocked_alert(address const& ip_

3490

, std::string const& msg);

3491

3492

address ip;

3493

3494

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

3495

};

3496

</pre>

3497

</div>

3498

3499

<h2>storage_moved_alert</h2>

3500

<p>The <tt class="docutils literal"><span class="pre">storage_moved_alert</span></tt> is generated when all the disk IO has completed and the

3501

files have been moved, as an effect of a call to <tt class="docutils literal"><span class="pre">torrent_handle::move_storage</span></tt>. This

3502

is useful to synchronize with the actual disk.</p>

3503

3504

struct storage_moved_alert: torrent_alert

3505

{

3506

storage_moved_alert(torrent_handle const& h, std::string const& path);

3507

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

3508

};

3509

</pre>

3510

</div>

3511

3512

<h2>torrent_paused_alert</h2>

3513

<p>This alert is generated as a response to a <tt class="docutils literal"><span class="pre">torrent_handle::pause</span></tt> request. It is

3514

generated once all disk IO is complete and the files in the torrent have been closed.

3515

This is useful for synchronizing with the disk.</p>

3516

3517

struct torrent_paused_alert: torrent_alert

3518

{

3519

torrent_paused_alert(torrent_handle const& h, std::string const& msg);

3520

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

3521

};

3522

</pre>

3523

</div>

3524

3525

<h2>dispatcher</h2>

3526

<p>The <tt class="docutils literal"><span class="pre">handle_alert</span></tt> class is defined in <tt class="docutils literal"><span class="pre"><libtorrent/alert.hpp></span></tt>.</p>

3527

<p>Examples usage:</p>

3528

3529

struct my_handler

3530

{

3531

void operator()(portmap_error_alert const& a)

3532

{

3533

std::cout << "Portmapper: " << a.msg << std::endl;

3534

}

3535

3536

void operator()(tracker_warning_alert const& a)

3537

{

3538

std::cout << "Tracker warning: " << a.msg << std::endl;

3539

}

3540

3541

void operator()(torrent_finished_alert const& a)

3542

{

3543

// write fast resume data

3544

// ...

3545

3546

std::cout << a.handle.get_torrent_info().name() << "completed"

3547

<< std::endl;

3548

}

3549

};

3550

</pre>

3551

3552

std::auto_ptr<alert> a;

3553

a = ses.pop_alert();

3554

my_handler h;

3555

while (a.get())

3556

{

3557

handle_alert<portmap_error_alert

3558

, tracker_warning_alert

3559

, torrent_finished_alert

3560

>::handle_alert(h, a);

3561

a = ses.pop_alert();

3562

}

3563

</pre>

3564

<p>In this example 3 alert types are used. You can use any number of template

3565

parameters to select between more types. If the number of types are more than

3566

15, you can define <tt class="docutils literal"><span class="pre">TORRENT_MAX_ALERT_TYPES</span></tt> to a greater number before

3567

including <tt class="docutils literal"><span class="pre"><libtorrent/alert.hpp></span></tt>.</p>

3568

</div>

3569

</div>

3570

3571

<h1>exceptions</h1>

3572

<p>There are a number of exceptions that can be thrown from different places in libtorrent,

3573

here's a complete list with description.</p>

3574

3575

<h2>invalid_handle</h2>

3576

<p>This exception is thrown when querying information from a <a class="reference internal" href="#torrent-handle">torrent_handle</a> that hasn't

3577

been initialized or that has become invalid.</p>

3578

3579

struct invalid_handle: std::exception

3580

{

3581

const char* what() const throw();

3582

};

3583

</pre>

3584

</div>

3585

3586

<h2>duplicate_torrent</h2>

3587

<p>This is thrown by <a class="reference internal" href="#add-torrent">add_torrent()</a> if the torrent already has been added to

3588

the session. Since <a class="reference internal" href="#remove-torrent">remove_torrent()</a> is asynchronous, this exception may

3589

be thrown if the torrent is removed and then immediately added again.</p>

3590

3591

struct duplicate_torrent: std::exception

3592

{

3593

const char* what() const throw();

3594

};

3595

</pre>

3596

</div>

3597

3598

<h2>invalid_encoding</h2>

3599

<p>This is thrown by <tt class="docutils literal"><span class="pre">bdecode()</span></tt> if the input data is not a valid bencoding.</p>

3600

3601

struct invalid_encoding: std::exception

3602

{

3603

const char* what() const throw();

3604

};

3605

</pre>

3606

</div>

3607

3608

<h2>type_error</h2>

3609

<p>This is thrown from the accessors of <tt class="docutils literal"><span class="pre">entry</span></tt> if the data type of the <tt class="docutils literal"><span class="pre">entry</span></tt> doesn't

3610

match the type you want to extract from it.</p>

3611

3612

struct type_error: std::runtime_error

3613

{

3614

type_error(const char* error);

3615

};

3616

</pre>

3617

</div>

3618

3619

<h2>invalid_torrent_file</h2>

3620

<p>This exception is thrown from the constructor of <tt class="docutils literal"><span class="pre">torrent_info</span></tt> if the given bencoded information

3621

doesn't meet the requirements on what information has to be present in a torrent file.</p>

3622

3623

struct invalid_torrent_file: std::exception

3624

{

3625

const char* what() const throw();

3626

};

3627

</pre>

3628

</div>

3629

</div>

3630

3631

<h1>storage_interface</h1>

3632

<p>The storage interface is a pure virtual class that can be implemented to

3633

change the behavior of the actual file storage. The interface looks like

3634

this:</p>

3635

3636

struct storage_interface

3637

{

3638

virtual void initialize(bool allocate_files) = 0;

3639

virtual size_type read(char* buf, int slot, int offset, int size) = 0;

3640

virtual void write(const char* buf, int slot, int offset, int size) = 0;

3641

virtual bool move_storage(fs::path save_path) = 0;

3642

virtual bool verify_resume_data(entry& rd, std::string& error) = 0;

3643

virtual void write_resume_data(entry& rd) const = 0;

3644

virtual void move_slot(int src_slot, int dst_slot) = 0;

3645

virtual void swap_slots(int slot1, int slot2) = 0;

3646

virtual void swap_slots3(int slot1, int slot2, int slot3) = 0;

3647

virtual sha1_hash hash_for_slot(int slot, partial_hash& h, int piece_size) = 0;

3648

virtual void release_files() = 0;

3649

virtual void delete_files() = 0;

3650

virtual ~storage_interface() {}

3651

};

3652

</pre>

3653

3654

<h2>initialize()</h2>

3655

3656

3657

void initialize(bool allocate_files) = 0;

3658

</pre>

3659

</blockquote>

3660

<p>This function is called when the storage is to be initialized. The default storage

3661

will create directories and empty files at this point. If <tt class="docutils literal"><span class="pre">allocate_files</span></tt> is true,

3662

it will also <tt class="docutils literal"><span class="pre">ftruncate</span></tt> all files to their target size.</p>

3663

</div>

3664

3665

3666

3667

3668

size_type read(char* buf, int slot, int offset, int size) = 0;

3669

</pre>

3670

</blockquote>

3671

<p>This function should read the data in the given slot and at the given offset

3672

and <tt class="docutils literal"><span class="pre">size</span></tt> number of bytes. The data is to be copied to <tt class="docutils literal"><span class="pre">buf</span></tt>.</p>

3673

<p>The return value is the number of bytes actually read.</p>

3674

</div>

3675

3676

<h2>write()</h2>

3677

3678

3679

void write(const char* buf, int slot, int offset, int size) = 0;

3680

</pre>

3681

</blockquote>

3682

<p>This function should write the data in <tt class="docutils literal"><span class="pre">buf</span></tt> to the given slot (<tt class="docutils literal"><span class="pre">slot</span></tt>) at offset

3683

<tt class="docutils literal"><span class="pre">offset</span></tt> in that slot. The buffer size is <tt class="docutils literal"><span class="pre">size</span></tt>.</p>

3684

</div>

3685

3686

<h2>move_storage()</h2>

3687

3688

3689

bool move_storage(fs::path save_path) = 0;

3690

</pre>

3691

</blockquote>

3692

<p>This function should move all the files belonging to the storage to the new save_path.

3693

The default storage moves the single file or the directory of the torrent.</p>

3694

<p>Before moving the files, any open file handles may have to be closed, like

3695

<tt class="docutils literal"><span class="pre">release_files()</span></tt>.</p>

3696

</div>

3697

3698

<h2>verify_resume_data()</h2>

3699

3700

3701

bool verify_resume_data(entry& rd, std::string& error) = 0;

3702

</pre>

3703

</blockquote>

3704

<p>This function should verify the resume data <tt class="docutils literal"><span class="pre">rd</span></tt> with the files

3705

on disk. If the resume data seems to be up-to-date, return true. If

3706

not, set <tt class="docutils literal"><span class="pre">error</span></tt> to a description of what mismatched and return false.</p>

3707

<p>The default storage may compare file sizes and time stamps of the files.</p>

3708

</div>

3709

3710

<h2>write_resume_data( )</h2>

3711

3712

3713

void write_resume_data(entry& rd) const = 0;

3714

</pre>

3715

</blockquote>

3716

<p>This function should fill in resume data, the current state of the

3717

storage, in <tt class="docutils literal"><span class="pre">rd</span></tt>. The default storage adds file timestamps and

3718

sizes.</p>

3719

</div>

3720

3721

3722

3723

3724

void move_slot(int src_slot, int dst_slot) = 0;

3725

</pre>

3726

</blockquote>

3727

<p>This function should copy or move the data in slot <tt class="docutils literal"><span class="pre">src_slot</span></tt> to

3728

the slot <tt class="docutils literal"><span class="pre">dst_slot</span></tt>. This is only used in compact mode.</p>

3729

<p>If the storage caches slots, this could be implemented more

3730

efficient than reading and writing the data.</p>

3731

</div>

3732

3733

<h2>swap_slots()</h2>

3734

3735

3736

void swap_slots(int slot1, int slot2) = 0;

3737

</pre>

3738

</blockquote>

3739

<p>This function should swap the data in <tt class="docutils literal"><span class="pre">slot1</span></tt> and <tt class="docutils literal"><span class="pre">slot2</span></tt>. The default

3740

storage uses a scratch buffer to read the data into, then moving the other

3741

slot and finally writing back the temporary slot's data</p>

3742

<p>This is only used in compact mode.</p>

3743

</div>

3744

3745

<h2>swap_slots3()</h2>

3746

3747

3748

void swap_slots3(int slot1, int slot2, int slot3) = 0;

3749

</pre>

3750

</blockquote>

3751

<p>This function should do a 3-way swap, or shift of the slots. <tt class="docutils literal"><span class="pre">slot1</span></tt>

3752

should move to <tt class="docutils literal"><span class="pre">slot2</span></tt>, which should be moved to <tt class="docutils literal"><span class="pre">slot3</span></tt> which in turn

3753

should be moved to <tt class="docutils literal"><span class="pre">slot1</span></tt>.</p>

3754

<p>This is only used in compact mode.</p>

3755

</div>

3756

3757

3758

3759

3760

sha1_hash hash_for_slot(int slot, partial_hash& h, int piece_size) = 0;

3761

</pre>

3762

</blockquote>

3763

<p>The function should read the remaining bytes of the slot and hash it with the

3764

sha-1 state in <tt class="docutils literal"><span class="pre">partion_hash</span></tt>. The <tt class="docutils literal"><span class="pre">partial_hash</span></tt> struct looks like this:</p>

3765

3766

struct partial_hash

3767

{

3768

partial_hash();

3769

int offset;

3770

hasher h;

3771

};

3772

</pre>

3773

<p><tt class="docutils literal"><span class="pre">offset</span></tt> is the number of bytes in the slot that has already been hashed, and

3774

<tt class="docutils literal"><span class="pre">h</span></tt> is the sha-1 state of that hash. <tt class="docutils literal"><span class="pre">piece_size</span></tt> is the size of the piece

3775

that is stored in the given slot.</p>

3776

<p>The function should return the hash of the piece stored in the slot.</p>

3777

</div>

3778

3779

<h2>release_files()</h2>

3780

3781

3782

void release_files() = 0;

3783

</pre>

3784

</blockquote>

3785

<p>This function should release all the file handles that it keeps open to files

3786

belonging to this storage. The default implementation just calls

3787

<tt class="docutils literal"><span class="pre">file_pool::release_files(this)</span></tt>.</p>

3788

</div>

3789

3790

<h2>delete_files()</h2>

3791

3792

3793

void delete_files() = 0;

3794

</pre>

3795

</blockquote>

3796

<p>This function should delete all files and directories belonging to this storage.</p>

3797

</div>

3798

</div>

3799

3800

<h1>fast resume</h1>

3801

<p>The fast resume mechanism is a way to remember which pieces are downloaded

3802

and where they are put between sessions. You can generate fast resume data by

3803

calling <tt class="docutils literal"><span class="pre">torrent_handle::write_resume_data()</span></tt> on <a class="reference internal" href="#torrent-handle">torrent_handle</a>. You can

3804

then save this data to disk and use it when resuming the torrent. libtorrent

3805

will not check the piece hashes then, and rely on the information given in the

3806

fast-resume data. The fast-resume data also contains information about which

3807

blocks, in the unfinished pieces, were downloaded, so it will not have to

3808

start from scratch on the partially downloaded pieces.</p>

3809

<p>To use the fast-resume data you simply give it to <a class="reference internal" href="#add-torrent">add_torrent()</a>, and it

3810

will skip the time consuming checks. It may have to do the checking anyway, if

3811

the fast-resume data is corrupt or doesn't fit the storage for that torrent,

3812

then it will not trust the fast-resume data and just do the checking.</p>

3813

3814

<h2>file format</h2>

3815

<p>The file format is a bencoded dictionary containing the following fields:</p>

3816

3817

3818

3819

3820

</colgroup>

3821

3822

<tr><td><tt class="docutils literal"><span class="pre">file-format</span></tt></td>

3823

<td>string: "libtorrent resume file"</td>

3824

</tr>

3825

<tr><td><tt class="docutils literal"><span class="pre">file-version</span></tt></td>

3826

<td>integer: 1</td>

3827

</tr>

3828

3829

<td>string, the info hash of the torrent this data is saved for.</td>

3830

</tr>

3831

<tr><td><tt class="docutils literal"><span class="pre">blocks</span> <span class="pre">per</span> <span class="pre">piece</span></tt></td>

3832

<td>integer, the number of blocks per piece. Must be: piece_size

3833

/ (16 * 1024). Clamped to be within the range [1, 256]. It

3834

is the number of blocks per (normal sized) piece. Usually

3835

each block is 16 * 1024 bytes in size. But if piece size is

3836

greater than 4 megabytes, the block size will increase.</td>

3837

</tr>

3838

<tr><td><tt class="docutils literal"><span class="pre">slots</span></tt></td>

3839

<td><p class="first">list of integers. The list maps slots to piece indices. It

3840

tells which piece is on which slot. If piece index is -2 it

3841

means it is free, that there's no piece there. If it is -1,

3842

means the slot isn't allocated on disk yet. The pieces have

3843

to meet the following requirement:</p>

3844

<p class="last">If there's a slot at the position of the piece index,

3845

the piece must be located in that slot.</p>

3846

</td>

3847

</tr>

3848

<tr><td><tt class="docutils literal"><span class="pre">peers</span></tt></td>

3849

<td><p class="first">list of dictionaries. Each dictionary has the following

3850

layout:</p>

3851

3852

3853

3854

3855

</colgroup>

3856

3857

3858

<td>string, the ip address of the peer. This is

3859

not a binary representation of the ip

3860

address, but the string representation. It

3861

may be an IPv6 string or an IPv4 string.</td>

3862

</tr>

3863

3864

<td>integer, the listen port of the peer</td>

3865

</tr>

3866

</tbody>

3867

</table>

3868

<p class="last">These are the local peers we were connected to when this

3869

fast-resume data was saved.</p>

3870

</td>

3871

</tr>

3872

<tr><td><tt class="docutils literal"><span class="pre">unfinished</span></tt></td>

3873

<td><p class="first">list of dictionaries. Each dictionary represents an

3874

piece, and has the following layout:</p>

3875

3876

3877

3878

3879

</colgroup>

3880

3881

<tr><td><tt class="docutils literal"><span class="pre">piece</span></tt></td>

3882

<td>integer, the index of the piece this entry

3883

refers to.</td>

3884

</tr>

3885

<tr><td><tt class="docutils literal"><span class="pre">bitmask</span></tt></td>

3886

<td>string, a binary bitmask representing the

3887

blocks that have been downloaded in this

3888

piece.</td>

3889

</tr>

3890

<tr><td><tt class="docutils literal"><span class="pre">adler32</span></tt></td>

3891

<td>The adler32 checksum of the data in the

3892

blocks specified by <tt class="docutils literal"><span class="pre">bitmask</span></tt>.</td>

3893

</tr>

3894

</tbody>

3895

</table>

3896

</td>

3897

</tr>

3898

<tr><td><tt class="docutils literal"><span class="pre">file</span> <span class="pre">sizes</span></tt></td>

3899

<td>list where each entry corresponds to a file in the file list

3900

in the metadata. Each entry has a list of two values, the

3901

first value is the size of the file in bytes, the second

3902

is the time stamp when the last time someone wrote to it.

3903

This information is used to compare with the files on disk.

3904

All the files must match exactly this information in order

3905

to consider the resume data as current. Otherwise a full

3906

re-check is issued.</td>

3907

</tr>

3908

<tr><td><tt class="docutils literal"><span class="pre">allocation</span></tt></td>

3909

<td>The allocation mode for the storage. Can be either <tt class="docutils literal"><span class="pre">full</span></tt>

3910

or <tt class="docutils literal"><span class="pre">compact</span></tt>. If this is full, the file sizes and

3911

timestamps are disregarded. Pieces are assumed not to have

3912

moved around even if the files have been modified after the

3913

last resume data checkpoint.</td>

3914

</tr>

3915

</tbody>

3916

</table>

3917

</div>

3918

</div>

3919

3920

<h1>threads</h1>

3921

<p>libtorrent starts 2 or 3 threads.</p>

3922

3923

3924

<li>The first thread is the main thread that will sit

3925

idle in a <tt class="docutils literal"><span class="pre">select()</span></tt> call most of the time. This thread runs the main loop

3926

that will send and receive data on all connections.</li>

3927

<li>The second thread is a hash-check thread. Whenever a torrent is added it will

3928

first be passed to this thread for checking the files that may already have been

3929

downloaded. If there is any resume data this thread will make sure it is valid

3930

and matches the files. Once the torrent has been checked, it is passed on to the

3931

main thread that will start it. The hash-check thread has a queue of torrents,

3932

it will only check one torrent at a time.</li>

3933

<li>The third thread is spawned by asio on systems that don't support

3934

non-blocking host name resolution to simulate non-blocking behavior.</li>

3935

</ul>

3936

</blockquote>

3937

</div>

3938

3939

<h1>storage allocation</h1>

3940

<p>There are three modes in which storage (files on disk) are allocated in libtorrent.</p>

3941

3942

<li>The traditional <em>full allocation</em> mode, where the entire files are filled up with

3943

zeros before anything is downloaded. libtorrent will look for sparse files support

3944

in the filesystem that is used for storage, and use sparse files or file system

3945

zero fill support if present. This means that on NTFS, full allocation mode will

3946

only allocate storage for the downloaded pieces.</li>

3947

<li>The <em>compact allocation</em> mode, where only files are allocated for actual

3948

pieces that have been downloaded. This is the default allocation mode in libtorrent.</li>

3949

<li>The <em>sparce allocation</em>, sparse files are used, and pieces are downloaded directly

3950

to where they belong. This is the recommended (and default) mode.</li>

3951

</ol>

3952

<p>The allocation mode is selected when a torrent is started. It is passed as an

3953

argument to <tt class="docutils literal"><span class="pre">session::add_torrent()</span></tt> (see <a class="reference internal" href="#add-torrent">add_torrent()</a>).</p>

3954

<p>The decision to use full allocation or compact allocation typically depends on whether

3955

any files are filtered and if the filesystem supports sparse files.</p>

3956

3957

<h2>sparse allocation</h2>

3958

<p>On filesystems that supports sparse files, this allocation mode will only use

3959

as much space as has been downloaded.</p>

3960

3961

3962

<li>It does not require an allocation pass on startup.</li>

3963

<li>It supports skipping files (setting prioirty to 0 to not download).</li>

3964

<li>Fast resume data will remain valid even when file time stamps are out of date.</li>

3965

</ul>

3966

</blockquote>

3967

</div>

3968

3969

<h2>full allocation</h2>

3970

<p>When a torrent is started in full allocation mode, the checker thread (see <a class="reference internal" href="#threads">threads</a>)

3971

will make sure that the entire storage is allocated, and fill any gaps with zeros.

3972

This will be skipped if the filesystem supports sparse files or automatic zero filling.

3973

It will of course still check for existing pieces and fast resume data. The main

3974

drawbacks of this mode are:</p>

3975

3976

3977

<li>It may take longer to start the torrent, since it will need to fill the files

3978

with zeros on some systems. This delay is linearly dependent on the size of

3979

the download.</li>

3980

<li>The download may occupy unnecessary disk space between download sessions. In case

3981

sparse files are not supported.</li>

3982

<li>Disk caches usually perform extremely poorly with random access to large files

3983

and may slow down a download considerably.</li>

3984

</ul>

3985

</blockquote>

3986

<p>The benefits of this mode are:</p>

3987

3988

3989

<li>Downloaded pieces are written directly to their final place in the files and the

3990

total number of disk operations will be fewer and may also play nicer to

3991

filesystems' file allocation, and reduce fragmentation.</li>

3992

<li>No risk of a download failing because of a full disk during download. Unless

3993

sparse files are being used.</li>

3994

<li>The fast resume data will be more likely to be usable, regardless of crashes or

3995

out of date data, since pieces won't move around.</li>

3996

<li>Can be used with the filter files feature.</li>

3997

</ul>

3998

</blockquote>

3999

</div>

4000

4001

<h2>compact allocation</h2>

4002

<p>The compact allocation will only allocate as much storage as it needs to keep the

4003

pieces downloaded so far. This means that pieces will be moved around to be placed

4004

at their final position in the files while downloading (to make sure the completed

4005

download has all its pieces in the correct place). So, the main drawbacks are:</p>

4006

4007

4008

<li>More disk operations while downloading since pieces are moved around.</li>

4009

<li>Potentially more fragmentation in the filesystem.</li>

4010

<li>Cannot be used while filtering files.</li>

4011

</ul>

4012

</blockquote>

4013

<p>The benefits though, are:</p>

4014

4015

4016

<li>No startup delay, since the files doesn't need allocating.</li>

4017

<li>The download will not use unnecessary disk space.</li>

4018

<li>Disk caches perform much better than in full allocation and raises the download

4019

speed limit imposed by the disk.</li>

4020

<li>Works well on filesystems that doesn't support sparse files.</li>

4021

</ul>

4022

</blockquote>

4023

<p>The algorithm that is used when allocating pieces and slots isn't very complicated.

4024

For the interested, a description follows.</p>

4025

<p>storing a piece:</p>

4026

4027

<li>let <strong>A</strong> be a newly downloaded piece, with index <strong>n</strong>.</li>

4028

<li>let <strong>s</strong> be the number of slots allocated in the file we're

4029

downloading to. (the number of pieces it has room for).</li>

4030

<li>if <strong>n</strong> >= <strong>s</strong> then allocate a new slot and put the piece there.</li>

4031

<li>if <strong>n</strong> < <strong>s</strong> then allocate a new slot, move the data at

4032

slot <strong>n</strong> to the new slot and put <strong>A</strong> in slot <strong>n</strong>.</li>

4033

</ol>

4034

<p>allocating a new slot:</p>

4035

4036

<li>if there's an unassigned slot (a slot that doesn't

4037

contain any piece), return that slot index.</li>

4038

<li>append the new slot at the end of the file (or find an unused slot).</li>

4039

<li>let <strong>i</strong> be the index of newly allocated slot</li>

4040

<li>if we have downloaded piece index <strong>i</strong> already (to slot <strong>j</strong>) then<ol class="arabic">

4041

4042

<li>return slot index <strong>j</strong> as the newly allocated free slot.</li>

4043

</ol>

4044

</li>

4045

<li>return <strong>i</strong> as the newly allocated slot.</li>

4046

</ol>

4047

</div>

4048

</div>

4049

4050

<h1>extensions</h1>

4051

<p>These extensions all operates within the <a class="reference external" href="extension_protocol.html">extension protocol</a>. The

4052

name of the extension is the name used in the extension-list packets,

4053

and the payload is the data in the extended message (not counting the

4054

length-prefix, message-id nor extension-id).</p>

4055

<p>Note that since this protocol relies on one of the reserved bits in the

4056

handshake, it may be incompatible with future versions of the mainline

4057

bittorrent client.</p>

4058

<p>These are the extensions that are currently implemented.</p>

4059

4060

<h2>metadata from peers</h2>

4061

<p>Extension name: "LT_metadata"</p>

4062

<p>The point with this extension is that you don't have to distribute the

4063

metadata (.torrent-file) separately. The metadata can be distributed

4064

through the bittorrent swarm. The only thing you need to download such

4065

a torrent is the tracker url and the info-hash of the torrent.</p>

4066

<p>It works by assuming that the initial seeder has the metadata and that

4067

the metadata will propagate through the network as more peers join.</p>

4068

<p>There are three kinds of messages in the metadata extension. These packets

4069

are put as payload to the extension message. The three packets are:</p>

4070

4071

4072

<li>request metadata</li>

4073

<li>metadata</li>

4074

<li>don't have metadata</li>

4075

</ul>

4076

</blockquote>

4077

<p>request metadata:</p>

4078

4079

4080

4081

4082

4083

</colgroup>

4084

4085

4086

4087

<th class="head">description</th>

4088

</tr>

4089

</thead>

4090

4091

4092

4093

<td>Determines the kind of message this is

4094

0 means 'request metadata'</td>

4095

</tr>

4096

4097

<td>start</td>

4098

<td>The start of the metadata block that

4099

is requested. It is given in 256:ths

4100

of the total size of the metadata,

4101

since the requesting client don't know

4102

the size of the metadata.</td>

4103

</tr>

4104

4105

4106

<td>The size of the metadata block that is

4107

requested. This is also given in

4108

256:ths of the total size of the

4109

metadata. The size is given as size-1.

4110

That means that if this field is set

4111

0, the request wants one 256:th of the

4112

metadata.</td>

4113

</tr>

4114

</tbody>

4115

</table>

4116

<p>metadata:</p>

4117

4118

4119

4120

4121

4122

</colgroup>

4123

4124

4125

4126

<th class="head">description</th>

4127

</tr>

4128

</thead>

4129

4130

4131

4132

<td>1 means 'metadata'</td>

4133

</tr>

4134

4135

<td>total_size</td>

4136

<td>The total size of the metadata, given

4137

in number of bytes.</td>

4138

</tr>

4139

4140

<td>offset</td>

4141

<td>The offset of where the metadata block

4142

in this message belongs in the final

4143

metadata. This is given in bytes.</td>

4144

</tr>

4145

4146

<td>metadata</td>

4147

<td>The actual metadata block. The size of

4148

this part is given implicit by the

4149

length prefix in the bittorrent

4150

protocol packet.</td>

4151

</tr>

4152

</tbody>

4153

</table>

4154

<p>Don't have metadata:</p>

4155

4156

4157

4158

4159

4160

</colgroup>

4161

4162

4163

4164

<th class="head">description</th>

4165

</tr>

4166

</thead>

4167

4168

4169

4170

<td>2 means 'I don't have metadata'.

4171

This message is sent as a reply to a

4172

metadata request if the the client

4173

doesn't have any metadata.</td>

4174

</tr>

4175

</tbody>

4176

</table>

4177

</div>

4178

4179

<h2>HTTP seeding</h2>

4180

<p>The HTTP seed extension implements <a class="reference external" href="http://www.getright.com/seedtorrent.html">this specification</a>.</p>

4181

<p>The libtorrent implementation assumes that, if the URL ends with a slash

4182

('/'), the filename should be appended to it in order to request pieces from

4183

that file. The way this works is that if the torrent is a single-file torrent,

4184

only that filename is appended. If the torrent is a multi-file torrent, the

4185

torrent's name '/' the file name is appended. This is the same directory

4186

structure that libtorrent will download torrents into.</p>

4187

</div>

4188

</div>

4189

4190

<h1>filename checks</h1>

4191

<p>Boost.Filesystem will by default check all its paths to make sure they conform

4192

to filename requirements on many platforms. If you don't want this check, you can

4193

set it to either only check for native filesystem requirements or turn it off

4194

altogether. You can use:</p>

4195

4196

boost::filesystem::path::default_name_check(boost::filesystem::native);

4197

</pre>

4198

<p>for example. For more information, see the <a class="reference external" href="http://www.boost.org/libs/filesystem/doc/index.htm">Boost.Filesystem docs</a>.</p>

4199

</div>

4200

4201

<h1>acknowledgments</h1>

4202

4203

<p>Contributions by Magnus Jonsson, Daniel Wallin and Cory Nelson</p>

4204

<p>Lots of testing, suggestions and contributions by Massaroddel and Tianhao Qiu.</p>

4205

<p>Big thanks to Michael Wojciechowski and Peter Koeleman for making the autotools

4206

scripts.</p>

4207

<p>Thanks to Reimond Retz for bugfixes, suggestions and testing</p>

4208

<p>Thanks to <a class="reference external" href="http://www.cs.umu.se">University of Umeå</a> for providing development and test hardware.</p>

4209

<p>Project is hosted by sourceforge.</p>

4210

4211

</div>

4212

</div>

4213

</body>

4214

</html>

Older »