Robert Ly | 35f2fda | 2013-01-29 16:27:05 -0800 | [diff] [blame] | 1 | page.title=Dalvik Executable Format |
| 2 | @jd:body |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 3 | |
Robert Ly | 35f2fda | 2013-01-29 16:27:05 -0800 | [diff] [blame] | 4 | <!-- |
Clay Murphy | 768b82a | 2013-11-12 11:32:41 -0800 | [diff] [blame^] | 5 | Copyright 2013 The Android Open Source Project |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 6 | |
Robert Ly | 35f2fda | 2013-01-29 16:27:05 -0800 | [diff] [blame] | 7 | Licensed under the Apache License, Version 2.0 (the "License"); |
| 8 | you may not use this file except in compliance with the License. |
| 9 | You may obtain a copy of the License at |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 10 | |
Robert Ly | 35f2fda | 2013-01-29 16:27:05 -0800 | [diff] [blame] | 11 | http://www.apache.org/licenses/LICENSE-2.0 |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 12 | |
Robert Ly | 35f2fda | 2013-01-29 16:27:05 -0800 | [diff] [blame] | 13 | Unless required by applicable law or agreed to in writing, software |
| 14 | distributed under the License is distributed on an "AS IS" BASIS, |
| 15 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 16 | See the License for the specific language governing permissions and |
| 17 | limitations under the License. |
| 18 | --> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 19 | <p>Copyright © 2007 The Android Open Source Project |
| 20 | |
| 21 | <p>This document describes the layout and contents of <code>.dex</code> |
| 22 | files, which are used to hold a set of class definitions and their associated |
| 23 | adjunct data.</p> |
| 24 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 25 | <h2>Guide To Types</h2> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 26 | |
| 27 | <table class="guide"> |
| 28 | <thead> |
| 29 | <tr> |
| 30 | <th>Name</th> |
| 31 | <th>Description</th> |
| 32 | </tr> |
| 33 | </thead> |
| 34 | <tbody> |
| 35 | <tr> |
| 36 | <td>byte</td> |
| 37 | <td>8-bit signed int</td> |
| 38 | </tr> |
| 39 | <tr> |
| 40 | <td>ubyte</td> |
| 41 | <td>8-bit unsigned int</td> |
| 42 | </tr> |
| 43 | <tr> |
| 44 | <td>short</td> |
| 45 | <td>16-bit signed int, little-endian</td> |
| 46 | </tr> |
| 47 | <tr> |
| 48 | <td>ushort</td> |
| 49 | <td>16-bit unsigned int, little-endian</td> |
| 50 | </tr> |
| 51 | <tr> |
| 52 | <td>int</td> |
| 53 | <td>32-bit signed int, little-endian</td> |
| 54 | </tr> |
| 55 | <tr> |
| 56 | <td>uint</td> |
| 57 | <td>32-bit unsigned int, little-endian</td> |
| 58 | </tr> |
| 59 | <tr> |
| 60 | <td>long</td> |
| 61 | <td>64-bit signed int, little-endian</td> |
| 62 | </tr> |
| 63 | <tr> |
| 64 | <td>ulong</td> |
| 65 | <td>64-bit unsigned int, little-endian</td> |
| 66 | </tr> |
| 67 | <tr> |
| 68 | <td>sleb128</td> |
| 69 | <td>signed LEB128, variable-length (see below)</td> |
| 70 | </tr> |
| 71 | <tr> |
| 72 | <td>uleb128</td> |
| 73 | <td>unsigned LEB128, variable-length (see below)</td> |
| 74 | </tr> |
| 75 | <tr> |
| 76 | <td>uleb128p1</td> |
| 77 | <td>unsigned LEB128 plus <code>1</code>, variable-length (see below)</td> |
| 78 | </tr> |
| 79 | </tbody> |
| 80 | </table> |
| 81 | |
| 82 | <h3>LEB128</h3> |
| 83 | |
| 84 | <p>LEB128 ("<b>L</b>ittle-<b>E</b>ndian <b>B</b>ase <b>128</b>") is a |
| 85 | variable-length encoding for |
| 86 | arbitrary signed or unsigned integer quantities. The format was |
| 87 | borrowed from the <a href="http://dwarfstd.org/Dwarf3Std.php">DWARF3</a> |
| 88 | specification. In a <code>.dex</code> file, LEB128 is only ever used to |
| 89 | encode 32-bit quantities.</p> |
| 90 | |
| 91 | <p>Each LEB128 encoded value consists of one to five |
| 92 | bytes, which together represent a single 32-bit value. Each |
| 93 | byte has its most significant bit set except for the final byte in the |
| 94 | sequence, which has its most significant bit clear. The remaining |
| 95 | seven bits of each byte are payload, with the least significant seven |
| 96 | bits of the quantity in the first byte, the next seven in the second |
| 97 | byte and so on. In the case of a signed LEB128 (<code>sleb128</code>), |
| 98 | the most significant payload bit of the final byte in the sequence is |
| 99 | sign-extended to produce the final value. In the unsigned case |
| 100 | (<code>uleb128</code>), any bits not explicitly represented are |
| 101 | interpreted as <code>0</code>. |
| 102 | |
| 103 | <table class="leb128Bits"> |
| 104 | <thead> |
| 105 | <tr><th colspan="16">Bitwise diagram of a two-byte LEB128 value</th></tr> |
| 106 | <tr> |
| 107 | <th colspan="8">First byte</td> |
| 108 | <th colspan="8">Second byte</td> |
| 109 | </tr> |
| 110 | </thead> |
| 111 | <tbody> |
| 112 | <tr> |
| 113 | <td class="start1"><code>1</code></td> |
| 114 | <td>bit<sub>6</sub></td> |
| 115 | <td>bit<sub>5</sub></td> |
| 116 | <td>bit<sub>4</sub></td> |
| 117 | <td>bit<sub>3</sub></td> |
| 118 | <td>bit<sub>2</sub></td> |
| 119 | <td>bit<sub>1</sub></td> |
| 120 | <td>bit<sub>0</sub></td> |
| 121 | <td class="start2"><code>0</code></td> |
| 122 | <td>bit<sub>13</sub></td> |
| 123 | <td>bit<sub>12</sub></td> |
| 124 | <td>bit<sub>11</sub></td> |
| 125 | <td>bit<sub>10</sub></td> |
| 126 | <td>bit<sub>9</sub></td> |
| 127 | <td>bit<sub>8</sub></td> |
| 128 | <td class="end2">bit<sub>7</sub></td> |
| 129 | </tr> |
| 130 | </tbody> |
| 131 | </table> |
| 132 | |
| 133 | <p>The variant <code>uleb128p1</code> is used to represent a signed |
| 134 | value, where the representation is of the value <i>plus one</i> encoded |
| 135 | as a <code>uleb128</code>. This makes the encoding of <code>-1</code> |
| 136 | (alternatively thought of as the unsigned value <code>0xffffffff</code>) |
| 137 | — but no other negative number — a single byte, and is |
| 138 | useful in exactly those cases where the represented number must either |
| 139 | be non-negative or <code>-1</code> (or <code>0xffffffff</code>), |
| 140 | and where no other negative values are allowed (or where large unsigned |
| 141 | values are unlikely to be needed).</p> |
| 142 | |
| 143 | <p>Here are some examples of the formats:</p> |
| 144 | |
| 145 | <table class="leb128"> |
| 146 | <thead> |
| 147 | <tr> |
| 148 | <th>Encoded Sequence</th> |
| 149 | <th>As <code>sleb128</code></th> |
| 150 | <th>As <code>uleb128</code></th> |
| 151 | <th>As <code>uleb128p1</code></th> |
| 152 | </tr> |
| 153 | </thead> |
| 154 | <tbody> |
| 155 | <tr><td>00</td><td>0</td><td>0</td><td>-1</td></tr> |
| 156 | <tr><td>01</td><td>1</td><td>1</td><td>0</td></tr> |
| 157 | <tr><td>7f</td><td>-1</td><td>127</td><td>126</td></tr> |
| 158 | <tr><td>80 7f</td><td>-128</td><td>16256</td><td>16255</td></tr> |
| 159 | </tbody> |
| 160 | </table> |
| 161 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 162 | <h2>Overall File Layout</h2> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 163 | |
| 164 | <table class="format"> |
| 165 | <thead> |
| 166 | <tr> |
| 167 | <th>Name</th> |
| 168 | <th>Format</th> |
| 169 | <th>Description</th> |
| 170 | </tr> |
| 171 | </thead> |
| 172 | <tbody> |
| 173 | <tr> |
| 174 | <td>header</td> |
| 175 | <td>header_item</td> |
| 176 | <td>the header</td> |
| 177 | </tr> |
| 178 | <tr> |
| 179 | <td>string_ids</td> |
| 180 | <td>string_id_item[]</td> |
| 181 | <td>string identifiers list. These are identifiers for all the strings |
| 182 | used by this file, either for internal naming (e.g., type descriptors) |
| 183 | or as constant objects referred to by code. This list must be sorted |
| 184 | by string contents, using UTF-16 code point values (not in a |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 185 | locale-sensitive manner), and it must not contain any duplicate entries. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 186 | </td> |
| 187 | </tr> |
| 188 | <tr> |
| 189 | <td>type_ids</td> |
| 190 | <td>type_id_item[]</td> |
| 191 | <td>type identifiers list. These are identifiers for all types (classes, |
| 192 | arrays, or primitive types) referred to by this file, whether defined |
| 193 | in the file or not. This list must be sorted by <code>string_id</code> |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 194 | index, and it must not contain any duplicate entries. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 195 | </td> |
| 196 | </tr> |
| 197 | <tr> |
| 198 | <td>proto_ids</td> |
| 199 | <td>proto_id_item[]</td> |
| 200 | <td>method prototype identifiers list. These are identifiers for all |
| 201 | prototypes referred to by this file. This list must be sorted in |
| 202 | return-type (by <code>type_id</code> index) major order, and then |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 203 | by arguments (also by <code>type_id</code> index). The list must not |
| 204 | contain any duplicate entries. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 205 | </td> |
| 206 | </tr> |
| 207 | <tr> |
| 208 | <td>field_ids</td> |
| 209 | <td>field_id_item[]</td> |
| 210 | <td>field identifiers list. These are identifiers for all fields |
| 211 | referred to by this file, whether defined in the file or not. This |
| 212 | list must be sorted, where the defining type (by <code>type_id</code> |
| 213 | index) is the major order, field name (by <code>string_id</code> index) |
| 214 | is the intermediate order, and type (by <code>type_id</code> index) |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 215 | is the minor order. The list must not contain any duplicate entries. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 216 | </td> |
| 217 | </tr> |
| 218 | <tr> |
| 219 | <td>method_ids</td> |
| 220 | <td>method_id_item[]</td> |
| 221 | <td>method identifiers list. These are identifiers for all methods |
| 222 | referred to by this file, whether defined in the file or not. This |
| 223 | list must be sorted, where the defining type (by <code>type_id</code> |
| 224 | index) is the major order, method name (by <code>string_id</code> |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 225 | index) is the intermediate order, and method prototype (by |
| 226 | <code>proto_id</code> index) is the minor order. The list must not |
| 227 | contain any duplicate entries. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 228 | </td> |
| 229 | </tr> |
| 230 | <tr> |
| 231 | <td>class_defs</td> |
| 232 | <td>class_def_item[]</td> |
| 233 | <td>class definitions list. The classes must be ordered such that a given |
| 234 | class's superclass and implemented interfaces appear in the |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 235 | list earlier than the referring class. Furthermore, it is invalid for |
| 236 | a definition for the same-named class to appear more than once in |
| 237 | the list. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 238 | </td> |
| 239 | </tr> |
| 240 | <tr> |
| 241 | <td>data</td> |
| 242 | <td>ubyte[]</td> |
| 243 | <td>data area, containing all the support data for the tables listed above. |
| 244 | Different items have different alignment requirements, and |
| 245 | padding bytes are inserted before each item if necessary to achieve |
| 246 | proper alignment. |
| 247 | </td> |
| 248 | </tr> |
| 249 | <tr> |
| 250 | <td>link_data</td> |
| 251 | <td>ubyte[]</td> |
| 252 | <td>data used in statically linked files. The format of the data in |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 253 | this section is left unspecified by this document. |
| 254 | This section is empty in unlinked files, and runtime implementations |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 255 | may use it as they see fit. |
| 256 | </td> |
| 257 | </tr> |
| 258 | </tbody> |
| 259 | </table> |
| 260 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 261 | <h2>Bitfield, String, and Constant Definitions</h2> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 262 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 263 | <h3>DEX_FILE_MAGIC</h3> |
| 264 | <h4>embedded in header_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 265 | |
| 266 | <p>The constant array/string <code>DEX_FILE_MAGIC</code> is the list of |
| 267 | bytes that must appear at the beginning of a <code>.dex</code> file |
| 268 | in order for it to be recognized as such. The value intentionally |
| 269 | contains a newline (<code>"\n"</code> or <code>0x0a</code>) and a |
| 270 | null byte (<code>"\0"</code> or <code>0x00</code>) in order to help |
| 271 | in the detection of certain forms of corruption. The value also |
| 272 | encodes a format version number as three decimal digits, which is |
| 273 | expected to increase monotonically over time as the format evolves.</p> |
| 274 | |
| 275 | <pre> |
| 276 | ubyte[8] DEX_FILE_MAGIC = { 0x64 0x65 0x78 0x0a 0x30 0x33 0x35 0x00 } |
| 277 | = "dex\n035\0" |
| 278 | </pre> |
| 279 | |
| 280 | <p><b>Note:</b> At least a couple earlier versions of the format have |
| 281 | been used in widely-available public software releases. For example, |
| 282 | version <code>009</code> was used for the M3 releases of the |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 283 | Android platform (November–December 2007), |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 284 | and version <code>013</code> was used for the M5 releases of the Android |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 285 | platform (February–March 2008). In several respects, these earlier |
| 286 | versions of the format differ significantly from the version described in this |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 287 | document.</p> |
| 288 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 289 | <h3>ENDIAN_CONSTANT and REVERSE_ENDIAN_CONSTANT</h3> |
| 290 | <h4>embedded in header_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 291 | |
| 292 | <p>The constant <code>ENDIAN_CONSTANT</code> is used to indicate the |
| 293 | endianness of the file in which it is found. Although the standard |
| 294 | <code>.dex</code> format is little-endian, implementations may choose |
| 295 | to perform byte-swapping. Should an implementation come across a |
| 296 | header whose <code>endian_tag</code> is <code>REVERSE_ENDIAN_CONSTANT</code> |
| 297 | instead of <code>ENDIAN_CONSTANT</code>, it would know that the file |
| 298 | has been byte-swapped from the expected form.</p> |
| 299 | |
| 300 | <pre> |
| 301 | uint ENDIAN_CONSTANT = 0x12345678; |
| 302 | uint REVERSE_ENDIAN_CONSTANT = 0x78563412; |
| 303 | </pre> |
| 304 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 305 | <h3>NO_INDEX</h3> |
| 306 | <h4>embedded in class_def_item and debug_info_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 307 | |
| 308 | <p>The constant <code>NO_INDEX</code> is used to indicate that |
| 309 | an index value is absent.</p> |
| 310 | |
| 311 | <p><b>Note:</b> This value isn't defined to be |
| 312 | <code>0</code>, because that is in fact typically a valid index.</p> |
| 313 | |
| 314 | <p><b>Also Note:</b> The chosen value for <code>NO_INDEX</code> is |
| 315 | representable as a single byte in the <code>uleb128p1</code> encoding.</p> |
| 316 | |
| 317 | <pre> |
| 318 | uint NO_INDEX = 0xffffffff; // == -1 if treated as a signed int |
| 319 | </pre> |
| 320 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 321 | <h3>access_flags Definitions</h3> |
| 322 | <h4>embedded in class_def_item, encoded_field, encoded_method, and |
| 323 | InnerClass</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 324 | |
| 325 | <p>Bitfields of these flags are used to indicate the accessibility and |
| 326 | overall properties of classes and class members.</p> |
| 327 | |
| 328 | <table class="accessFlags"> |
| 329 | <thead> |
| 330 | <tr> |
| 331 | <th>Name</th> |
| 332 | <th>Value</th> |
| 333 | <th>For Classes (and <code>InnerClass</code> annotations)</th> |
| 334 | <th>For Fields</th> |
| 335 | <th>For Methods</th> |
| 336 | </tr> |
| 337 | </thead> |
| 338 | <tbody> |
| 339 | <tr> |
| 340 | <td>ACC_PUBLIC</td> |
| 341 | <td>0x1</td> |
| 342 | <td><code>public</code>: visible everywhere</td> |
| 343 | <td><code>public</code>: visible everywhere</td> |
| 344 | <td><code>public</code>: visible everywhere</td> |
| 345 | </tr> |
| 346 | <tr> |
| 347 | <td>ACC_PRIVATE</td> |
| 348 | <td>0x2</td> |
| 349 | <td><super>*</super> |
| 350 | <code>private</code>: only visible to defining class |
| 351 | </td> |
| 352 | <td><code>private</code>: only visible to defining class</td> |
| 353 | <td><code>private</code>: only visible to defining class</td> |
| 354 | </tr> |
| 355 | <tr> |
| 356 | <td>ACC_PROTECTED</td> |
| 357 | <td>0x4</td> |
| 358 | <td><super>*</super> |
| 359 | <code>protected</code>: visible to package and subclasses |
| 360 | </td> |
| 361 | <td><code>protected</code>: visible to package and subclasses</td> |
| 362 | <td><code>protected</code>: visible to package and subclasses</td> |
| 363 | </tr> |
| 364 | <tr> |
| 365 | <td>ACC_STATIC</td> |
| 366 | <td>0x8</td> |
| 367 | <td><super>*</super> |
| 368 | <code>static</code>: is not constructed with an outer |
| 369 | <code>this</code> reference</td> |
| 370 | <td><code>static</code>: global to defining class</td> |
| 371 | <td><code>static</code>: does not take a <code>this</code> argument</td> |
| 372 | </tr> |
| 373 | <tr> |
| 374 | <td>ACC_FINAL</td> |
| 375 | <td>0x10</td> |
| 376 | <td><code>final</code>: not subclassable</td> |
| 377 | <td><code>final</code>: immutable after construction</td> |
| 378 | <td><code>final</code>: not overridable</td> |
| 379 | </tr> |
| 380 | <tr> |
| 381 | <td>ACC_SYNCHRONIZED</td> |
| 382 | <td>0x20</td> |
| 383 | <td> </td> |
| 384 | <td> </td> |
| 385 | <td><code>synchronized</code>: associated lock automatically acquired |
| 386 | around call to this method. <b>Note:</b> This is only valid to set when |
| 387 | <code>ACC_NATIVE</code> is also set.</td> |
| 388 | </tr> |
| 389 | <tr> |
| 390 | <td>ACC_VOLATILE</td> |
| 391 | <td>0x40</td> |
| 392 | <td> </td> |
| 393 | <td><code>volatile</code>: special access rules to help with thread |
| 394 | safety</td> |
| 395 | <td> </td> |
| 396 | </tr> |
| 397 | <tr> |
| 398 | <td>ACC_BRIDGE</td> |
| 399 | <td>0x40</td> |
| 400 | <td> </td> |
| 401 | <td> </td> |
| 402 | <td>bridge method, added automatically by compiler as a type-safe |
| 403 | bridge</td> |
| 404 | </tr> |
| 405 | <tr> |
| 406 | <td>ACC_TRANSIENT</td> |
| 407 | <td>0x80</td> |
| 408 | <td> </td> |
| 409 | <td><code>transient</code>: not to be saved by default serialization</td> |
| 410 | <td> </td> |
| 411 | </tr> |
| 412 | <tr> |
| 413 | <td>ACC_VARARGS</td> |
| 414 | <td>0x80</td> |
| 415 | <td> </td> |
| 416 | <td> </td> |
| 417 | <td>last argument should be treated as a "rest" argument by compiler</td> |
| 418 | </tr> |
| 419 | <tr> |
| 420 | <td>ACC_NATIVE</td> |
| 421 | <td>0x100</td> |
| 422 | <td> </td> |
| 423 | <td> </td> |
| 424 | <td><code>native</code>: implemented in native code</td> |
| 425 | </tr> |
| 426 | <tr> |
| 427 | <td>ACC_INTERFACE</td> |
| 428 | <td>0x200</td> |
| 429 | <td><code>interface</code>: multiply-implementable abstract class</td> |
| 430 | <td> </td> |
| 431 | <td> </td> |
| 432 | </tr> |
| 433 | <tr> |
| 434 | <td>ACC_ABSTRACT</td> |
| 435 | <td>0x400</td> |
| 436 | <td><code>abstract</code>: not directly instantiable</td> |
| 437 | <td> </td> |
| 438 | <td><code>abstract</code>: unimplemented by this class</td> |
| 439 | </tr> |
| 440 | <tr> |
| 441 | <td>ACC_STRICT</td> |
| 442 | <td>0x800</td> |
| 443 | <td> </td> |
| 444 | <td> </td> |
| 445 | <td><code>strictfp</code>: strict rules for floating-point arithmetic</td> |
| 446 | </tr> |
| 447 | <tr> |
| 448 | <td>ACC_SYNTHETIC</td> |
| 449 | <td>0x1000</td> |
| 450 | <td>not directly defined in source code</td> |
| 451 | <td>not directly defined in source code</td> |
| 452 | <td>not directly defined in source code</td> |
| 453 | </tr> |
| 454 | <tr> |
| 455 | <td>ACC_ANNOTATION</td> |
| 456 | <td>0x2000</td> |
| 457 | <td>declared as an annotation class</td> |
| 458 | <td> </td> |
| 459 | <td> </td> |
| 460 | </tr> |
| 461 | <tr> |
| 462 | <td>ACC_ENUM</td> |
| 463 | <td>0x4000</td> |
| 464 | <td>declared as an enumerated type</td> |
| 465 | <td>declared as an enumerated value</td> |
| 466 | <td> </td> |
| 467 | </tr> |
| 468 | <tr> |
| 469 | <td><i>(unused)</i></td> |
| 470 | <td>0x8000</td> |
| 471 | <td> </td> |
| 472 | <td> </td> |
| 473 | <td> </td> |
| 474 | </tr> |
| 475 | <tr> |
| 476 | <td>ACC_CONSTRUCTOR</td> |
| 477 | <td>0x10000</td> |
| 478 | <td> </td> |
| 479 | <td> </td> |
| 480 | <td>constructor method (class or instance initializer)</td> |
| 481 | </tr> |
| 482 | <tr> |
| 483 | <td>ACC_DECLARED_<br/>SYNCHRONIZED</td> |
| 484 | <td>0x20000</td> |
| 485 | <td> </td> |
| 486 | <td> </td> |
| 487 | <td>declared <code>synchronized</code>. <b>Note:</b> This has no effect on |
| 488 | execution (other than in reflection of this flag, per se). |
| 489 | </td> |
| 490 | </tr> |
| 491 | </tbody> |
| 492 | </table> |
| 493 | |
| 494 | <p><super>*</super> Only allowed on for <code>InnerClass</code> annotations, |
| 495 | and must not ever be on in a <code>class_def_item</code>.</p> |
| 496 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 497 | <h3>MUTF-8 (Modified UTF-8) Encoding</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 498 | |
| 499 | <p>As a concession to easier legacy support, the <code>.dex</code> format |
| 500 | encodes its string data in a de facto standard modified UTF-8 form, hereafter |
| 501 | referred to as MUTF-8. This form is identical to standard UTF-8, except:</p> |
| 502 | |
| 503 | <ul> |
| 504 | <li>Only the one-, two-, and three-byte encodings are used.</li> |
| 505 | <li>Code points in the range <code>U+10000</code> … |
| 506 | <code>U+10ffff</code> are encoded as a surrogate pair, each of |
| 507 | which is represented as a three-byte encoded value.</li> |
| 508 | <li>The code point <code>U+0000</code> is encoded in two-byte form.</li> |
| 509 | <li>A plain null byte (value <code>0</code>) indicates the end of |
| 510 | a string, as is the standard C language interpretation.</li> |
| 511 | </ul> |
| 512 | |
| 513 | <p>The first two items above can be summarized as: MUTF-8 |
| 514 | is an encoding format for UTF-16, instead of being a more direct |
| 515 | encoding format for Unicode characters.</p> |
| 516 | |
| 517 | <p>The final two items above make it simultaneously possible to include |
| 518 | the code point <code>U+0000</code> in a string <i>and</i> still manipulate |
| 519 | it as a C-style null-terminated string.</p> |
| 520 | |
| 521 | <p>However, the special encoding of <code>U+0000</code> means that, unlike |
| 522 | normal UTF-8, the result of calling the standard C function |
| 523 | <code>strcmp()</code> on a pair of MUTF-8 strings does not always |
| 524 | indicate the properly signed result of comparison of <i>unequal</i> strings. |
| 525 | When ordering (not just equality) is a concern, the most straightforward |
| 526 | way to compare MUTF-8 strings is to decode them character by character, |
| 527 | and compare the decoded values. (However, more clever implementations are |
| 528 | also possible.)</p> |
| 529 | |
| 530 | <p>Please refer to <a href="http://unicode.org">The Unicode |
| 531 | Standard</a> for further information about character encoding. |
| 532 | MUTF-8 is actually closer to the (relatively less well-known) encoding |
| 533 | <a href="http://www.unicode.org/reports/tr26/">CESU-8</a> than to UTF-8 |
| 534 | per se.</p> |
| 535 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 536 | <h3>encoded_value Encoding</h3> |
| 537 | <h4>embedded in annotation_element and encoded_array_item </h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 538 | |
| 539 | <p>An <code>encoded_value</code> is an encoded piece of (nearly) |
| 540 | arbitrary hierarchically structured data. The encoding is meant to |
| 541 | be both compact and straightforward to parse.</p> |
| 542 | |
| 543 | <table class="format"> |
| 544 | <thead> |
| 545 | <tr> |
| 546 | <th>Name</th> |
| 547 | <th>Format</th> |
| 548 | <th>Description</th> |
| 549 | </tr> |
| 550 | </thead> |
| 551 | <tbody> |
| 552 | <tr> |
| 553 | <td>(value_arg << 5) | value_type</td> |
| 554 | <td>ubyte</td> |
| 555 | <td>byte indicating the type of the immediately subsequent |
| 556 | <code>value</code> along |
| 557 | with an optional clarifying argument in the high-order three bits. |
| 558 | See below for the various <code>value</code> definitions. |
| 559 | In most cases, <code>value_arg</code> encodes the length of |
| 560 | the immediately-subsequent <code>value</code> in bytes, as |
| 561 | <code>(size - 1)</code>, e.g., <code>0</code> means that |
| 562 | the value requires one byte, and <code>7</code> means it requires |
| 563 | eight bytes; however, there are exceptions as noted below. |
| 564 | </td> |
| 565 | </tr> |
| 566 | <tr> |
| 567 | <td>value</td> |
| 568 | <td>ubyte[]</td> |
| 569 | <td>bytes representing the value, variable in length and interpreted |
| 570 | differently for different <code>value_type</code> bytes, though |
| 571 | always little-endian. See the various value definitions below for |
| 572 | details. |
| 573 | </td> |
| 574 | </tr> |
| 575 | </tbody> |
| 576 | </table> |
| 577 | |
| 578 | <h3>Value Formats</h3> |
| 579 | |
| 580 | <table class="encodedValue"> |
| 581 | <thead> |
| 582 | <tr> |
| 583 | <th>Type Name</th> |
| 584 | <th><code>value_type</code></th> |
| 585 | <th><code>value_arg</code> Format</th> |
| 586 | <th><code>value</code> Format</th> |
| 587 | <th>Description</th> |
| 588 | </tr> |
| 589 | </thead> |
| 590 | <tbody> |
| 591 | <tr> |
| 592 | <td>VALUE_BYTE</td> |
| 593 | <td>0x00</td> |
| 594 | <td><i>(none; must be <code>0</code>)</i></td> |
| 595 | <td>ubyte[1]</td> |
| 596 | <td>signed one-byte integer value</td> |
| 597 | </tr> |
| 598 | <tr> |
| 599 | <td>VALUE_SHORT</td> |
| 600 | <td>0x02</td> |
| 601 | <td>size - 1 (0…1)</td> |
| 602 | <td>ubyte[size]</td> |
| 603 | <td>signed two-byte integer value, sign-extended</td> |
| 604 | </tr> |
| 605 | <tr> |
| 606 | <td>VALUE_CHAR</td> |
| 607 | <td>0x03</td> |
| 608 | <td>size - 1 (0…1)</td> |
| 609 | <td>ubyte[size]</td> |
| 610 | <td>unsigned two-byte integer value, zero-extended</td> |
| 611 | </tr> |
| 612 | <tr> |
| 613 | <td>VALUE_INT</td> |
| 614 | <td>0x04</td> |
| 615 | <td>size - 1 (0…3)</td> |
| 616 | <td>ubyte[size]</td> |
| 617 | <td>signed four-byte integer value, sign-extended</td> |
| 618 | </tr> |
| 619 | <tr> |
| 620 | <td>VALUE_LONG</td> |
| 621 | <td>0x06</td> |
| 622 | <td>size - 1 (0…7)</td> |
| 623 | <td>ubyte[size]</td> |
| 624 | <td>signed eight-byte integer value, sign-extended</td> |
| 625 | </tr> |
| 626 | <tr> |
| 627 | <td>VALUE_FLOAT</td> |
| 628 | <td>0x10</td> |
| 629 | <td>size - 1 (0…3)</td> |
| 630 | <td>ubyte[size]</td> |
| 631 | <td>four-byte bit pattern, zero-extended <i>to the right</i>, and |
| 632 | interpreted as an IEEE754 32-bit floating point value |
| 633 | </td> |
| 634 | </tr> |
| 635 | <tr> |
| 636 | <td>VALUE_DOUBLE</td> |
| 637 | <td>0x11</td> |
| 638 | <td>size - 1 (0…7)</td> |
| 639 | <td>ubyte[size]</td> |
| 640 | <td>eight-byte bit pattern, zero-extended <i>to the right</i>, and |
| 641 | interpreted as an IEEE754 64-bit floating point value |
| 642 | </td> |
| 643 | </tr> |
| 644 | <tr> |
| 645 | <td>VALUE_STRING</td> |
| 646 | <td>0x17</td> |
| 647 | <td>size - 1 (0…3)</td> |
| 648 | <td>ubyte[size]</td> |
| 649 | <td>unsigned (zero-extended) four-byte integer value, |
| 650 | interpreted as an index into |
| 651 | the <code>string_ids</code> section and representing a string value |
| 652 | </td> |
| 653 | </tr> |
| 654 | <tr> |
| 655 | <td>VALUE_TYPE</td> |
| 656 | <td>0x18</td> |
| 657 | <td>size - 1 (0…3)</td> |
| 658 | <td>ubyte[size]</td> |
| 659 | <td>unsigned (zero-extended) four-byte integer value, |
| 660 | interpreted as an index into |
| 661 | the <code>type_ids</code> section and representing a reflective |
| 662 | type/class value |
| 663 | </td> |
| 664 | </tr> |
| 665 | <tr> |
| 666 | <td>VALUE_FIELD</td> |
| 667 | <td>0x19</td> |
| 668 | <td>size - 1 (0…3)</td> |
| 669 | <td>ubyte[size]</td> |
| 670 | <td>unsigned (zero-extended) four-byte integer value, |
| 671 | interpreted as an index into |
| 672 | the <code>field_ids</code> section and representing a reflective |
| 673 | field value |
| 674 | </td> |
| 675 | </tr> |
| 676 | <tr> |
| 677 | <td>VALUE_METHOD</td> |
| 678 | <td>0x1a</td> |
| 679 | <td>size - 1 (0…3)</td> |
| 680 | <td>ubyte[size]</td> |
| 681 | <td>unsigned (zero-extended) four-byte integer value, |
| 682 | interpreted as an index into |
| 683 | the <code>method_ids</code> section and representing a reflective |
| 684 | method value |
| 685 | </td> |
| 686 | </tr> |
| 687 | <tr> |
| 688 | <td>VALUE_ENUM</td> |
| 689 | <td>0x1b</td> |
| 690 | <td>size - 1 (0…3)</td> |
| 691 | <td>ubyte[size]</td> |
| 692 | <td>unsigned (zero-extended) four-byte integer value, |
| 693 | interpreted as an index into |
| 694 | the <code>field_ids</code> section and representing the value of |
| 695 | an enumerated type constant |
| 696 | </td> |
| 697 | </tr> |
| 698 | <tr> |
| 699 | <td>VALUE_ARRAY</td> |
| 700 | <td>0x1c</td> |
| 701 | <td><i>(none; must be <code>0</code>)</i></td> |
| 702 | <td>encoded_array</td> |
| 703 | <td>an array of values, in the format specified by |
| 704 | "<code>encoded_array</code> Format" below. The size |
| 705 | of the <code>value</code> is implicit in the encoding. |
| 706 | </td> |
| 707 | </tr> |
| 708 | <tr> |
| 709 | <td>VALUE_ANNOTATION</td> |
| 710 | <td>0x1d</td> |
| 711 | <td><i>(none; must be <code>0</code>)</i></td> |
| 712 | <td>encoded_annotation</td> |
| 713 | <td>a sub-annotation, in the format specified by |
| 714 | "<code>encoded_annotation</code> Format" below. The size |
| 715 | of the <code>value</code> is implicit in the encoding. |
| 716 | </td> |
| 717 | </tr> |
| 718 | <tr> |
| 719 | <td>VALUE_NULL</td> |
| 720 | <td>0x1e</td> |
| 721 | <td><i>(none; must be <code>0</code>)</i></td> |
| 722 | <td><i>(none)</i></td> |
| 723 | <td><code>null</code> reference value</td> |
| 724 | </tr> |
| 725 | <tr> |
| 726 | <td>VALUE_BOOLEAN</td> |
| 727 | <td>0x1f</td> |
| 728 | <td>boolean (0…1)</td> |
| 729 | <td><i>(none)</i></td> |
| 730 | <td>one-bit value; <code>0</code> for <code>false</code> and |
| 731 | <code>1</code> for <code>true</code>. The bit is represented in the |
| 732 | <code>value_arg</code>. |
| 733 | </td> |
| 734 | </tr> |
| 735 | </tbody> |
| 736 | </table> |
| 737 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 738 | <h3>encoded_array Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 739 | |
| 740 | <table class="format"> |
| 741 | <thead> |
| 742 | <tr> |
| 743 | <th>Name</th> |
| 744 | <th>Format</th> |
| 745 | <th>Description</th> |
| 746 | </tr> |
| 747 | </thead> |
| 748 | <tbody> |
| 749 | <tr> |
| 750 | <td>size</td> |
| 751 | <td>uleb128</td> |
| 752 | <td>number of elements in the array</td> |
| 753 | </tr> |
| 754 | <tr> |
| 755 | <td>values</td> |
| 756 | <td>encoded_value[size]</td> |
| 757 | <td>a series of <code>size</code> <code>encoded_value</code> byte |
| 758 | sequences in the format specified by this section, concatenated |
| 759 | sequentially. |
| 760 | </td> |
| 761 | </tr> |
| 762 | </tbody> |
| 763 | </table> |
| 764 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 765 | <h3>encoded_annotation Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 766 | |
| 767 | <table class="format"> |
| 768 | <thead> |
| 769 | <tr> |
| 770 | <th>Name</th> |
| 771 | <th>Format</th> |
| 772 | <th>Description</th> |
| 773 | </tr> |
| 774 | </thead> |
| 775 | <tbody> |
| 776 | <tr> |
| 777 | <td>type_idx</td> |
| 778 | <td>uleb128</td> |
| 779 | <td>type of the annotation. This must be a class (not array or primitive) |
| 780 | type. |
| 781 | </td> |
| 782 | </tr> |
| 783 | <tr> |
| 784 | <td>size</td> |
| 785 | <td>uleb128</td> |
| 786 | <td>number of name-value mappings in this annotation</td> |
| 787 | </tr> |
| 788 | <tr> |
| 789 | <td>elements</td> |
| 790 | <td>annotation_element[size]</td> |
| 791 | <td>elements of the annotataion, represented directly in-line (not as |
| 792 | offsets). Elements must be sorted in increasing order by |
| 793 | <code>string_id</code> index. |
| 794 | </td> |
| 795 | </tr> |
| 796 | </tbody> |
| 797 | </table> |
| 798 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 799 | <h3>annotation_element Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 800 | |
| 801 | <table class="format"> |
| 802 | <thead> |
| 803 | <tr> |
| 804 | <th>Name</th> |
| 805 | <th>Format</th> |
| 806 | <th>Description</th> |
| 807 | </tr> |
| 808 | </thead> |
| 809 | <tbody> |
| 810 | <tr> |
| 811 | <td>name_idx</td> |
| 812 | <td>uleb128</td> |
| 813 | <td>element name, represented as an index into the |
| 814 | <code>string_ids</code> section. The string must conform to the |
| 815 | syntax for <i>MemberName</i>, defined above. |
| 816 | </td> |
| 817 | </tr> |
| 818 | <tr> |
| 819 | <td>value</td> |
| 820 | <td>encoded_value</td> |
| 821 | <td>element value</td> |
| 822 | </tr> |
| 823 | </tbody> |
| 824 | </table> |
| 825 | |
| 826 | <h2>String Syntax</h2> |
| 827 | |
| 828 | <p>There are several kinds of item in a <code>.dex</code> file which |
| 829 | ultimately refer to a string. The following BNF-style definitions |
| 830 | indicate the acceptable syntax for these strings.</p> |
| 831 | |
| 832 | <h3><i>SimpleName</i></h3> |
| 833 | |
| 834 | <p>A <i>SimpleName</i> is the basis for the syntax of the names of other |
| 835 | things. The <code>.dex</code> format allows a fair amount of latitude |
| 836 | here (much more than most common source languages). In brief, a simple |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 837 | name consists of any low-ASCII alphabetic character or digit, a few |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 838 | specific low-ASCII symbols, and most non-ASCII code points that are not |
| 839 | control, space, or special characters. Note that surrogate code points |
| 840 | (in the range <code>U+d800</code> … <code>U+dfff</code>) are not |
| 841 | considered valid name characters, per se, but Unicode supplemental |
| 842 | characters <i>are</i> valid (which are represented by the final |
| 843 | alternative of the rule for <i>SimpleNameChar</i>), and they should be |
| 844 | represented in a file as pairs of surrogate code points in the MUTF-8 |
| 845 | encoding.</p> |
| 846 | |
| 847 | <table class="bnf"> |
| 848 | <tr><td colspan="2" class="def"><i>SimpleName</i> →</td></tr> |
| 849 | <tr> |
| 850 | <td/> |
| 851 | <td><i>SimpleNameChar</i> (<i>SimpleNameChar</i>)*</td> |
| 852 | </tr> |
| 853 | |
| 854 | <tr><td colspan="2" class="def"><i>SimpleNameChar</i> →</td></tr> |
| 855 | <tr> |
| 856 | <td/> |
| 857 | <td><code>'A'</code> … <code>'Z'</code></td> |
| 858 | </tr> |
| 859 | <tr> |
| 860 | <td class="bar">|</td> |
| 861 | <td><code>'a'</code> … <code>'z'</code></td> |
| 862 | </tr> |
| 863 | <tr> |
| 864 | <td class="bar">|</td> |
| 865 | <td><code>'0'</code> … <code>'9'</code></td> |
| 866 | </tr> |
| 867 | <tr> |
| 868 | <td class="bar">|</td> |
| 869 | <td><code>'$'</code></td> |
| 870 | </tr> |
| 871 | <tr> |
| 872 | <td class="bar">|</td> |
| 873 | <td><code>'-'</code></td> |
| 874 | </tr> |
| 875 | <tr> |
| 876 | <td class="bar">|</td> |
| 877 | <td><code>'_'</code></td> |
| 878 | </tr> |
| 879 | <tr> |
| 880 | <td class="bar">|</td> |
| 881 | <td><code>U+00a1</code> … <code>U+1fff</code></td> |
| 882 | </tr> |
| 883 | <tr> |
| 884 | <td class="bar">|</td> |
| 885 | <td><code>U+2010</code> … <code>U+2027</code></td> |
| 886 | </tr> |
| 887 | <tr> |
| 888 | <td class="bar">|</td> |
| 889 | <td><code>U+2030</code> … <code>U+d7ff</code></td> |
| 890 | </tr> |
| 891 | <tr> |
| 892 | <td class="bar">|</td> |
| 893 | <td><code>U+e000</code> … <code>U+ffef</code></td> |
| 894 | </tr> |
| 895 | <tr> |
| 896 | <td class="bar">|</td> |
| 897 | <td><code>U+10000</code> … <code>U+10ffff</code></td> |
| 898 | </tr> |
| 899 | </table> |
| 900 | |
| 901 | <h3><i>MemberName</i></h3> |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 902 | <h4>used by field_id_item and method_id_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 903 | |
| 904 | <p>A <i>MemberName</i> is the name of a member of a class, members being |
| 905 | fields, methods, and inner classes.</p> |
| 906 | |
| 907 | <table class="bnf"> |
| 908 | <tr><td colspan="2" class="def"><i>MemberName</i> →</td></tr> |
| 909 | <tr> |
| 910 | <td/> |
| 911 | <td><i>SimpleName</i></td> |
| 912 | </tr> |
| 913 | <tr> |
| 914 | <td class="bar">|</td> |
| 915 | <td><code>'<'</code> <i>SimpleName</i> <code>'>'</code></td> |
| 916 | </tr> |
| 917 | </table> |
| 918 | |
| 919 | <h3><i>FullClassName</i></h3> |
| 920 | |
| 921 | <p>A <i>FullClassName</i> is a fully-qualified class name, including an |
| 922 | optional package specifier followed by a required name.</p> |
| 923 | |
| 924 | <table class="bnf"> |
| 925 | <tr><td colspan="2" class="def"><i>FullClassName</i> →</td></tr> |
| 926 | <tr> |
| 927 | <td/> |
| 928 | <td><i>OptionalPackagePrefix</i> <i>SimpleName</i></td> |
| 929 | </tr> |
| 930 | |
| 931 | <tr><td colspan="2" class="def"><i>OptionalPackagePrefix</i> →</td></tr> |
| 932 | <tr> |
| 933 | <td/> |
| 934 | <td>(<i>SimpleName</i> <code>'/'</code>)*</td> |
| 935 | </tr> |
| 936 | </table> |
| 937 | |
| 938 | <h3><i>TypeDescriptor</i></h3> |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 939 | <h4>used by type_id_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 940 | |
| 941 | <p>A <i>TypeDescriptor</i> is the representation of any type, including |
| 942 | primitives, classes, arrays, and <code>void</code>. See below for |
| 943 | the meaning of the various versions.</p> |
| 944 | |
| 945 | <table class="bnf"> |
| 946 | <tr><td colspan="2" class="def"><i>TypeDescriptor</i> →</td></tr> |
| 947 | <tr> |
| 948 | <td/> |
| 949 | <td><code>'V'</code></td> |
| 950 | </tr> |
| 951 | <tr> |
| 952 | <td class="bar">|</td> |
| 953 | <td><i>FieldTypeDescriptor</i></td> |
| 954 | </tr> |
| 955 | |
| 956 | <tr><td colspan="2" class="def"><i>FieldTypeDescriptor</i> →</td></tr> |
| 957 | <tr> |
| 958 | <td/> |
| 959 | <td><i>NonArrayFieldTypeDescriptor</i></td> |
| 960 | </tr> |
| 961 | <tr> |
| 962 | <td class="bar">|</td> |
| 963 | <td>(<code>'['</code> * 1…255) |
| 964 | <i>NonArrayFieldTypeDescriptor</i></td> |
| 965 | </tr> |
| 966 | |
| 967 | <tr> |
| 968 | <td colspan="2" class="def"><i>NonArrayFieldTypeDescriptor</i>→</td> |
| 969 | </tr> |
| 970 | <tr> |
| 971 | <td/> |
| 972 | <td><code>'Z'</code></td> |
| 973 | </tr> |
| 974 | <tr> |
| 975 | <td class="bar">|</td> |
| 976 | <td><code>'B'</code></td> |
| 977 | </tr> |
| 978 | <tr> |
| 979 | <td class="bar">|</td> |
| 980 | <td><code>'S'</code></td> |
| 981 | </tr> |
| 982 | <tr> |
| 983 | <td class="bar">|</td> |
| 984 | <td><code>'C'</code></td> |
| 985 | </tr> |
| 986 | <tr> |
| 987 | <td class="bar">|</td> |
| 988 | <td><code>'I'</code></td> |
| 989 | </tr> |
| 990 | <tr> |
| 991 | <td class="bar">|</td> |
| 992 | <td><code>'J'</code></td> |
| 993 | </tr> |
| 994 | <tr> |
| 995 | <td class="bar">|</td> |
| 996 | <td><code>'F'</code></td> |
| 997 | </tr> |
| 998 | <tr> |
| 999 | <td class="bar">|</td> |
| 1000 | <td><code>'D'</code></td> |
| 1001 | </tr> |
| 1002 | <tr> |
| 1003 | <td class="bar">|</td> |
| 1004 | <td><code>'L'</code> <i>FullClassName</i> <code>';'</code></td> |
| 1005 | </tr> |
| 1006 | </table> |
| 1007 | |
| 1008 | <h3><i>ShortyDescriptor</i></h3> |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1009 | <h4>used by proto_id_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1010 | |
| 1011 | <p>A <i>ShortyDescriptor</i> is the short form representation of a method |
| 1012 | prototype, including return and parameter types, except that there is |
| 1013 | no distinction between various reference (class or array) types. Instead, |
| 1014 | all reference types are represented by a single <code>'L'</code> character.</p> |
| 1015 | |
| 1016 | <table class="bnf"> |
| 1017 | <tr><td colspan="2" class="def"><i>ShortyDescriptor</i> →</td></tr> |
| 1018 | <tr> |
| 1019 | <td/> |
| 1020 | <td><i>ShortyReturnType</i> (<i>ShortyFieldType</i>)*</td> |
| 1021 | </tr> |
| 1022 | |
| 1023 | <tr><td colspan="2" class="def"><i>ShortyReturnType</i> →</td></tr> |
| 1024 | <tr> |
| 1025 | <td/> |
| 1026 | <td><code>'V'</code></td> |
| 1027 | </tr> |
| 1028 | <tr> |
| 1029 | <td class="bar">|</td> |
| 1030 | <td><i>ShortyFieldType</i></td> |
| 1031 | </tr> |
| 1032 | |
| 1033 | <tr><td colspan="2" class="def"><i>ShortyFieldType</i> →</td></tr> |
| 1034 | <tr> |
| 1035 | <td/> |
| 1036 | <td><code>'Z'</code></td> |
| 1037 | </tr> |
| 1038 | <tr> |
| 1039 | <td class="bar">|</td> |
| 1040 | <td><code>'B'</code></td> |
| 1041 | </tr> |
| 1042 | <tr> |
| 1043 | <td class="bar">|</td> |
| 1044 | <td><code>'S'</code></td> |
| 1045 | </tr> |
| 1046 | <tr> |
| 1047 | <td class="bar">|</td> |
| 1048 | <td><code>'C'</code></td> |
| 1049 | </tr> |
| 1050 | <tr> |
| 1051 | <td class="bar">|</td> |
| 1052 | <td><code>'I'</code></td> |
| 1053 | </tr> |
| 1054 | <tr> |
| 1055 | <td class="bar">|</td> |
| 1056 | <td><code>'J'</code></td> |
| 1057 | </tr> |
| 1058 | <tr> |
| 1059 | <td class="bar">|</td> |
| 1060 | <td><code>'F'</code></td> |
| 1061 | </tr> |
| 1062 | <tr> |
| 1063 | <td class="bar">|</td> |
| 1064 | <td><code>'D'</code></td> |
| 1065 | </tr> |
| 1066 | <tr> |
| 1067 | <td class="bar">|</td> |
| 1068 | <td><code>'L'</code></td> |
| 1069 | </tr> |
| 1070 | </table> |
| 1071 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 1072 | <h3><i>TypeDescriptor</i> Semantics</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1073 | |
| 1074 | <p>This is the meaning of each of the variants of <i>TypeDescriptor</i>.</p> |
| 1075 | |
| 1076 | <table class="descriptor"> |
| 1077 | <thead> |
| 1078 | <tr> |
| 1079 | <th>Syntax</th> |
| 1080 | <th>Meaning</th> |
| 1081 | </tr> |
| 1082 | </thead> |
| 1083 | <tbody> |
| 1084 | <tr> |
| 1085 | <td>V</td> |
| 1086 | <td><code>void</code>; only valid for return types</td> |
| 1087 | </tr> |
| 1088 | <tr> |
| 1089 | <td>Z</td> |
| 1090 | <td><code>boolean</code></td> |
| 1091 | </tr> |
| 1092 | <tr> |
| 1093 | <td>B</td> |
| 1094 | <td><code>byte</code></td> |
| 1095 | </tr> |
| 1096 | <tr> |
| 1097 | <td>S</td> |
| 1098 | <td><code>short</code></td> |
| 1099 | </tr> |
| 1100 | <tr> |
| 1101 | <td>C</td> |
| 1102 | <td><code>char</code></td> |
| 1103 | </tr> |
| 1104 | <tr> |
| 1105 | <td>I</td> |
| 1106 | <td><code>int</code></td> |
| 1107 | </tr> |
| 1108 | <tr> |
| 1109 | <td>J</td> |
| 1110 | <td><code>long</code></td> |
| 1111 | </tr> |
| 1112 | <tr> |
| 1113 | <td>F</td> |
| 1114 | <td><code>float</code></td> |
| 1115 | </tr> |
| 1116 | <tr> |
| 1117 | <td>D</td> |
| 1118 | <td><code>double</code></td> |
| 1119 | </tr> |
| 1120 | <tr> |
| 1121 | <td>L<i>fully/qualified/Name</i>;</td> |
| 1122 | <td>the class <code><i>fully.qualified.Name</i></code></td> |
| 1123 | </tr> |
| 1124 | <tr> |
| 1125 | <td>[<i>descriptor</i></td> |
| 1126 | <td>array of <code><i>descriptor</i></code>, usable recursively for |
| 1127 | arrays-of-arrays, though it is invalid to have more than 255 |
| 1128 | dimensions. |
| 1129 | </td> |
| 1130 | </tr> |
| 1131 | </tbody> |
| 1132 | </table> |
| 1133 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 1134 | <h2>Items and Related Structures</h2> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1135 | |
| 1136 | <p>This section includes definitions for each of the top-level items that |
| 1137 | may appear in a <code>.dex</code> file. |
| 1138 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1139 | <h3>header_item</h3> |
| 1140 | <h4>appears in the header section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1141 | <h4>alignment: 4 bytes</h4> |
| 1142 | |
| 1143 | <table class="format"> |
| 1144 | <thead> |
| 1145 | <tr> |
| 1146 | <th>Name</th> |
| 1147 | <th>Format</th> |
| 1148 | <th>Description</th> |
| 1149 | </tr> |
| 1150 | </thead> |
| 1151 | <tbody> |
| 1152 | <tr> |
| 1153 | <td>magic</td> |
| 1154 | <td>ubyte[8] = DEX_FILE_MAGIC</td> |
| 1155 | <td>magic value. See discussion above under "<code>DEX_FILE_MAGIC</code>" |
| 1156 | for more details. |
| 1157 | </td> |
| 1158 | </tr> |
| 1159 | <tr> |
| 1160 | <td>checksum</td> |
| 1161 | <td>uint</td> |
| 1162 | <td>adler32 checksum of the rest of the file (everything but |
| 1163 | <code>magic</code> and this field); used to detect file corruption |
| 1164 | </td> |
| 1165 | </tr> |
| 1166 | <tr> |
| 1167 | <td>signature</td> |
| 1168 | <td>ubyte[20]</td> |
| 1169 | <td>SHA-1 signature (hash) of the rest of the file (everything but |
| 1170 | <code>magic</code>, <code>checksum</code>, and this field); used |
| 1171 | to uniquely identify files |
| 1172 | </td> |
| 1173 | </tr> |
| 1174 | <tr> |
| 1175 | <td>file_size</td> |
| 1176 | <td>uint</td> |
| 1177 | <td>size of the entire file (including the header), in bytes |
| 1178 | </tr> |
| 1179 | <tr> |
| 1180 | <td>header_size</td> |
| 1181 | <td>uint = 0x70</td> |
| 1182 | <td>size of the header (this entire section), in bytes. This allows for at |
| 1183 | least a limited amount of backwards/forwards compatibility without |
| 1184 | invalidating the format. |
| 1185 | </td> |
| 1186 | </tr> |
| 1187 | <tr> |
| 1188 | <td>endian_tag</td> |
| 1189 | <td>uint = ENDIAN_CONSTANT</td> |
| 1190 | <td>endianness tag. See discussion above under "<code>ENDIAN_CONSTANT</code> |
| 1191 | and <code>REVERSE_ENDIAN_CONSTANT</code>" for more details. |
| 1192 | </td> |
| 1193 | </tr> |
| 1194 | <tr> |
| 1195 | <td>link_size</td> |
| 1196 | <td>uint</td> |
| 1197 | <td>size of the link section, or <code>0</code> if this file isn't |
| 1198 | statically linked</td> |
| 1199 | </tr> |
| 1200 | <tr> |
| 1201 | <td>link_off</td> |
| 1202 | <td>uint</td> |
| 1203 | <td>offset from the start of the file to the link section, or |
| 1204 | <code>0</code> if <code>link_size == 0</code>. The offset, if non-zero, |
| 1205 | should be to an offset into the <code>link_data</code> section. The |
| 1206 | format of the data pointed at is left unspecified by this document; |
| 1207 | this header field (and the previous) are left as hooks for use by |
| 1208 | runtime implementations. |
| 1209 | </td> |
| 1210 | </tr> |
| 1211 | <tr> |
| 1212 | <td>map_off</td> |
| 1213 | <td>uint</td> |
| 1214 | <td>offset from the start of the file to the map item, or |
| 1215 | <code>0</code> if this file has no map. The offset, if non-zero, |
| 1216 | should be to an offset into the <code>data</code> section, |
| 1217 | and the data should be in the format specified by "<code>map_list</code>" |
| 1218 | below. |
| 1219 | </td> |
| 1220 | </tr> |
| 1221 | <tr> |
| 1222 | <td>string_ids_size</td> |
| 1223 | <td>uint</td> |
| 1224 | <td>count of strings in the string identifiers list</td> |
| 1225 | </tr> |
| 1226 | <tr> |
| 1227 | <td>string_ids_off</td> |
| 1228 | <td>uint</td> |
| 1229 | <td>offset from the start of the file to the string identifiers list, or |
| 1230 | <code>0</code> if <code>string_ids_size == 0</code> (admittedly a |
| 1231 | strange edge case). The offset, if non-zero, |
| 1232 | should be to the start of the <code>string_ids</code> section. |
| 1233 | </td> |
| 1234 | </tr> |
| 1235 | <tr> |
| 1236 | <td>type_ids_size</td> |
| 1237 | <td>uint</td> |
| 1238 | <td>count of elements in the type identifiers list</td> |
| 1239 | </tr> |
| 1240 | <tr> |
| 1241 | <td>type_ids_off</td> |
| 1242 | <td>uint</td> |
| 1243 | <td>offset from the start of the file to the type identifiers list, or |
| 1244 | <code>0</code> if <code>type_ids_size == 0</code> (admittedly a |
| 1245 | strange edge case). The offset, if non-zero, |
| 1246 | should be to the start of the <code>type_ids</code> |
| 1247 | section. |
| 1248 | </td> |
| 1249 | </tr> |
| 1250 | <tr> |
| 1251 | <td>proto_ids_size</td> |
| 1252 | <td>uint</td> |
| 1253 | <td>count of elements in the prototype identifiers list</td> |
| 1254 | </tr> |
| 1255 | <tr> |
| 1256 | <td>proto_ids_off</td> |
| 1257 | <td>uint</td> |
| 1258 | <td>offset from the start of the file to the prototype identifiers list, or |
| 1259 | <code>0</code> if <code>proto_ids_size == 0</code> (admittedly a |
| 1260 | strange edge case). The offset, if non-zero, |
| 1261 | should be to the start of the <code>proto_ids</code> |
| 1262 | section. |
| 1263 | </td> |
| 1264 | </tr> |
| 1265 | <tr> |
| 1266 | <td>field_ids_size</td> |
| 1267 | <td>uint</td> |
| 1268 | <td>count of elements in the field identifiers list</td> |
| 1269 | </tr> |
| 1270 | <tr> |
| 1271 | <td>field_ids_off</td> |
| 1272 | <td>uint</td> |
| 1273 | <td>offset from the start of the file to the field identifiers list, or |
| 1274 | <code>0</code> if <code>field_ids_size == 0</code>. The offset, if |
| 1275 | non-zero, should be to the start of the <code>field_ids</code> |
| 1276 | section.</td> |
| 1277 | </td> |
| 1278 | </tr> |
| 1279 | <tr> |
| 1280 | <td>method_ids_size</td> |
| 1281 | <td>uint</td> |
| 1282 | <td>count of elements in the method identifiers list</td> |
| 1283 | </tr> |
| 1284 | <tr> |
| 1285 | <td>method_ids_off</td> |
| 1286 | <td>uint</td> |
| 1287 | <td>offset from the start of the file to the method identifiers list, or |
| 1288 | <code>0</code> if <code>method_ids_size == 0</code>. The offset, if |
| 1289 | non-zero, should be to the start of the <code>method_ids</code> |
| 1290 | section.</td> |
| 1291 | </tr> |
| 1292 | <tr> |
| 1293 | <td>class_defs_size</td> |
| 1294 | <td>uint</td> |
| 1295 | <td>count of elements in the class definitions list</td> |
| 1296 | </tr> |
| 1297 | <tr> |
| 1298 | <td>class_defs_off</td> |
| 1299 | <td>uint</td> |
| 1300 | <td>offset from the start of the file to the class definitions list, or |
| 1301 | <code>0</code> if <code>class_defs_size == 0</code> (admittedly a |
| 1302 | strange edge case). The offset, if non-zero, |
| 1303 | should be to the start of the <code>class_defs</code> section. |
| 1304 | </td> |
| 1305 | </tr> |
| 1306 | <tr> |
| 1307 | <td>data_size</td> |
| 1308 | <td>uint</td> |
| 1309 | <td>Size of <code>data</code> section in bytes. Must be an even |
| 1310 | multiple of sizeof(uint).</td> |
| 1311 | </tr> |
| 1312 | <tr> |
| 1313 | <td>data_off</td> |
| 1314 | <td>uint</td> |
| 1315 | <td>offset from the start of the file to the start of the |
| 1316 | <code>data</code> section. |
| 1317 | </td> |
| 1318 | </tr> |
| 1319 | </tbody> |
| 1320 | </table> |
| 1321 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1322 | <h3>map_list</h3> |
| 1323 | <h4>appears in the data section</h4> |
| 1324 | <h4>referenced from header_item</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1325 | <h4>alignment: 4 bytes</h4> |
| 1326 | |
| 1327 | <p>This is a list of the entire contents of a file, in order. It |
| 1328 | contains some redundancy with respect to the <code>header_item</code> |
| 1329 | but is intended to be an easy form to use to iterate over an entire |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 1330 | file. A given type must appear at most once in a map, but there is no |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1331 | restriction on what order types may appear in, other than the |
| 1332 | restrictions implied by the rest of the format (e.g., a |
| 1333 | <code>header</code> section must appear first, followed by a |
| 1334 | <code>string_ids</code> section, etc.). Additionally, the map entries must |
| 1335 | be ordered by initial offset and must not overlap.</p> |
| 1336 | |
| 1337 | <table class="format"> |
| 1338 | <thead> |
| 1339 | <tr> |
| 1340 | <th>Name</th> |
| 1341 | <th>Format</th> |
| 1342 | <th>Description</th> |
| 1343 | </tr> |
| 1344 | </thead> |
| 1345 | <tbody> |
| 1346 | <tr> |
| 1347 | <td>size</td> |
| 1348 | <td>uint</td> |
| 1349 | <td>size of the list, in entries</td> |
| 1350 | </tr> |
| 1351 | <tr> |
| 1352 | <td>list</td> |
| 1353 | <td>map_item[size]</td> |
| 1354 | <td>elements of the list</td> |
| 1355 | </tr> |
| 1356 | </tbody> |
| 1357 | </table> |
| 1358 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1359 | <h3>map_item Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1360 | |
| 1361 | <table class="format"> |
| 1362 | <thead> |
| 1363 | <tr> |
| 1364 | <th>Name</th> |
| 1365 | <th>Format</th> |
| 1366 | <th>Description</th> |
| 1367 | </tr> |
| 1368 | </thead> |
| 1369 | <tbody> |
| 1370 | <tr> |
| 1371 | <td>type</td> |
| 1372 | <td>ushort</td> |
| 1373 | <td>type of the items; see table below</td> |
| 1374 | </tr> |
| 1375 | <tr> |
| 1376 | <td>unused</td> |
| 1377 | <td>ushort</td> |
| 1378 | <td><i>(unused)</i></td> |
| 1379 | </tr> |
| 1380 | <tr> |
| 1381 | <td>size</td> |
| 1382 | <td>uint</td> |
| 1383 | <td>count of the number of items to be found at the indicated offset</td> |
| 1384 | </tr> |
| 1385 | <tr> |
| 1386 | <td>offset</td> |
| 1387 | <td>uint</td> |
| 1388 | <td>offset from the start of the file to the items in question</td> |
| 1389 | </tr> |
| 1390 | </tbody> |
| 1391 | </table> |
| 1392 | |
| 1393 | |
| 1394 | <h3>Type Codes</h3> |
| 1395 | |
| 1396 | <table class="typeCodes"> |
| 1397 | <thead> |
| 1398 | <tr> |
| 1399 | <th>Item Type</th> |
| 1400 | <th>Constant</th> |
| 1401 | <th>Value</th> |
| 1402 | <th>Item Size In Bytes</th> |
| 1403 | </tr> |
| 1404 | </thead> |
| 1405 | <tbody> |
| 1406 | <tr> |
| 1407 | <td>header_item</td> |
| 1408 | <td>TYPE_HEADER_ITEM</td> |
| 1409 | <td>0x0000</td> |
| 1410 | <td>0x70</td> |
| 1411 | </tr> |
| 1412 | <tr> |
| 1413 | <td>string_id_item</td> |
| 1414 | <td>TYPE_STRING_ID_ITEM</td> |
| 1415 | <td>0x0001</td> |
| 1416 | <td>0x04</td> |
| 1417 | </tr> |
| 1418 | <tr> |
| 1419 | <td>type_id_item</td> |
| 1420 | <td>TYPE_TYPE_ID_ITEM</td> |
| 1421 | <td>0x0002</td> |
| 1422 | <td>0x04</td> |
| 1423 | </tr> |
| 1424 | <tr> |
| 1425 | <td>proto_id_item</td> |
| 1426 | <td>TYPE_PROTO_ID_ITEM</td> |
| 1427 | <td>0x0003</td> |
| 1428 | <td>0x0c</td> |
| 1429 | </tr> |
| 1430 | <tr> |
| 1431 | <td>field_id_item</td> |
| 1432 | <td>TYPE_FIELD_ID_ITEM</td> |
| 1433 | <td>0x0004</td> |
| 1434 | <td>0x08</td> |
| 1435 | </tr> |
| 1436 | <tr> |
| 1437 | <td>method_id_item</td> |
| 1438 | <td>TYPE_METHOD_ID_ITEM</td> |
| 1439 | <td>0x0005</td> |
| 1440 | <td>0x08</td> |
| 1441 | </tr> |
| 1442 | <tr> |
| 1443 | <td>class_def_item</td> |
| 1444 | <td>TYPE_CLASS_DEF_ITEM</td> |
| 1445 | <td>0x0006</td> |
| 1446 | <td>0x20</td> |
| 1447 | </tr> |
| 1448 | <tr> |
| 1449 | <td>map_list</td> |
| 1450 | <td>TYPE_MAP_LIST</td> |
| 1451 | <td>0x1000</td> |
| 1452 | <td>4 + (item.size * 12)</td> |
| 1453 | </tr> |
| 1454 | <tr> |
| 1455 | <td>type_list</td> |
| 1456 | <td>TYPE_TYPE_LIST</td> |
| 1457 | <td>0x1001</td> |
| 1458 | <td>4 + (item.size * 2)</td> |
| 1459 | </tr> |
| 1460 | <tr> |
| 1461 | <td>annotation_set_ref_list</td> |
| 1462 | <td>TYPE_ANNOTATION_SET_REF_LIST</td> |
| 1463 | <td>0x1002</td> |
| 1464 | <td>4 + (item.size * 4)</td> |
| 1465 | </tr> |
| 1466 | <tr> |
| 1467 | <td>annotation_set_item</td> |
| 1468 | <td>TYPE_ANNOTATION_SET_ITEM</td> |
| 1469 | <td>0x1003</td> |
| 1470 | <td>4 + (item.size * 4)</td> |
| 1471 | </tr> |
| 1472 | <tr> |
| 1473 | <td>class_data_item</td> |
| 1474 | <td>TYPE_CLASS_DATA_ITEM</td> |
| 1475 | <td>0x2000</td> |
| 1476 | <td><i>implicit; must parse</i></td> |
| 1477 | </tr> |
| 1478 | <tr> |
| 1479 | <td>code_item</td> |
| 1480 | <td>TYPE_CODE_ITEM</td> |
| 1481 | <td>0x2001</td> |
| 1482 | <td><i>implicit; must parse</i></td> |
| 1483 | </tr> |
| 1484 | <tr> |
| 1485 | <td>string_data_item</td> |
| 1486 | <td>TYPE_STRING_DATA_ITEM</td> |
| 1487 | <td>0x2002</td> |
| 1488 | <td><i>implicit; must parse</i></td> |
| 1489 | </tr> |
| 1490 | <tr> |
| 1491 | <td>debug_info_item</td> |
| 1492 | <td>TYPE_DEBUG_INFO_ITEM</td> |
| 1493 | <td>0x2003</td> |
| 1494 | <td><i>implicit; must parse</i></td> |
| 1495 | </tr> |
| 1496 | <tr> |
| 1497 | <td>annotation_item</td> |
| 1498 | <td>TYPE_ANNOTATION_ITEM</td> |
| 1499 | <td>0x2004</td> |
| 1500 | <td><i>implicit; must parse</i></td> |
| 1501 | </tr> |
| 1502 | <tr> |
| 1503 | <td>encoded_array_item</td> |
| 1504 | <td>TYPE_ENCODED_ARRAY_ITEM</td> |
| 1505 | <td>0x2005</td> |
| 1506 | <td><i>implicit; must parse</i></td> |
| 1507 | </tr> |
| 1508 | <tr> |
| 1509 | <td>annotations_directory_item</td> |
| 1510 | <td>TYPE_ANNOTATIONS_DIRECTORY_ITEM</td> |
| 1511 | <td>0x2006</td> |
| 1512 | <td><i>implicit; must parse</i></td> |
| 1513 | </tr> |
| 1514 | </tbody> |
| 1515 | </table> |
| 1516 | |
| 1517 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1518 | <h3>string_id_item</h3> |
| 1519 | <h4>appears in the string_ids section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1520 | <h4>alignment: 4 bytes</h4> |
| 1521 | |
| 1522 | <table class="format"> |
| 1523 | <thead> |
| 1524 | <tr> |
| 1525 | <th>Name</th> |
| 1526 | <th>Format</th> |
| 1527 | <th>Description</th> |
| 1528 | </tr> |
| 1529 | </thead> |
| 1530 | <tbody> |
| 1531 | <tr> |
| 1532 | <td>string_data_off</td> |
| 1533 | <td>uint</td> |
| 1534 | <td>offset from the start of the file to the string data for this |
| 1535 | item. The offset should be to a location |
| 1536 | in the <code>data</code> section, and the data should be in the |
| 1537 | format specified by "<code>string_data_item</code>" below. |
| 1538 | There is no alignment requirement for the offset. |
| 1539 | </td> |
| 1540 | </tr> |
| 1541 | </tbody> |
| 1542 | </table> |
| 1543 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1544 | <h3>string_data_item</h3> |
| 1545 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1546 | <h4>alignment: none (byte-aligned)</h4> |
| 1547 | |
| 1548 | <table class="format"> |
| 1549 | <thead> |
| 1550 | <tr> |
| 1551 | <th>Name</th> |
| 1552 | <th>Format</th> |
| 1553 | <th>Description</th> |
| 1554 | </tr> |
| 1555 | </thead> |
| 1556 | <tbody> |
| 1557 | <tr> |
| 1558 | <td>utf16_size</td> |
| 1559 | <td>uleb128</td> |
| 1560 | <td>size of this string, in UTF-16 code units (which is the "string |
| 1561 | length" in many systems). That is, this is the decoded length of |
| 1562 | the string. (The encoded length is implied by the position of |
| 1563 | the <code>0</code> byte.)</td> |
| 1564 | </tr> |
| 1565 | <tr> |
| 1566 | <td>data</td> |
| 1567 | <td>ubyte[]</td> |
| 1568 | <td>a series of MUTF-8 code units (a.k.a. octets, a.k.a. bytes) |
| 1569 | followed by a byte of value <code>0</code>. See |
| 1570 | "MUTF-8 (Modified UTF-8) Encoding" above for details and |
| 1571 | discussion about the data format. |
| 1572 | <p><b>Note:</b> It is acceptable to have a string which includes |
| 1573 | (the encoded form of) UTF-16 surrogate code units (that is, |
| 1574 | <code>U+d800</code> … <code>U+dfff</code>) |
| 1575 | either in isolation or out-of-order with respect to the usual |
| 1576 | encoding of Unicode into UTF-16. It is up to higher-level uses of |
| 1577 | strings to reject such invalid encodings, if appropriate.</p> |
| 1578 | </td> |
| 1579 | </tr> |
| 1580 | </tbody> |
| 1581 | </table> |
| 1582 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1583 | <h3>type_id_item</h3> |
| 1584 | <h4>appears in the type_ids section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1585 | <h4>alignment: 4 bytes</h4> |
| 1586 | |
| 1587 | <table class="format"> |
| 1588 | <thead> |
| 1589 | <tr> |
| 1590 | <th>Name</th> |
| 1591 | <th>Format</th> |
| 1592 | <th>Description</th> |
| 1593 | </tr> |
| 1594 | </thead> |
| 1595 | <tbody> |
| 1596 | <tr> |
| 1597 | <td>descriptor_idx</td> |
| 1598 | <td>uint</td> |
| 1599 | <td>index into the <code>string_ids</code> list for the descriptor |
| 1600 | string of this type. The string must conform to the syntax for |
| 1601 | <i>TypeDescriptor</i>, defined above. |
| 1602 | </td> |
| 1603 | </tr> |
| 1604 | </tbody> |
| 1605 | </table> |
| 1606 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1607 | <h3>proto_id_item</h3> |
| 1608 | <h4>appears in the proto_ids section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1609 | <h4>alignment: 4 bytes</h4> |
| 1610 | |
| 1611 | <table class="format"> |
| 1612 | <thead> |
| 1613 | <tr> |
| 1614 | <th>Name</th> |
| 1615 | <th>Format</th> |
| 1616 | <th>Description</th> |
| 1617 | </tr> |
| 1618 | </thead> |
| 1619 | <tbody> |
| 1620 | <tr> |
| 1621 | <td>shorty_idx</td> |
| 1622 | <td>uint</td> |
| 1623 | <td>index into the <code>string_ids</code> list for the short-form |
| 1624 | descriptor string of this prototype. The string must conform to the |
| 1625 | syntax for <i>ShortyDescriptor</i>, defined above, and must correspond |
| 1626 | to the return type and parameters of this item. |
| 1627 | </td> |
| 1628 | </tr> |
| 1629 | <tr> |
| 1630 | <td>return_type_idx</td> |
| 1631 | <td>uint</td> |
| 1632 | <td>index into the <code>type_ids</code> list for the return type |
| 1633 | of this prototype |
| 1634 | </td> |
| 1635 | </tr> |
| 1636 | <tr> |
| 1637 | <td>parameters_off</td> |
| 1638 | <td>uint</td> |
| 1639 | <td>offset from the start of the file to the list of parameter types |
| 1640 | for this prototype, or <code>0</code> if this prototype has no |
| 1641 | parameters. This offset, if non-zero, should be in the |
| 1642 | <code>data</code> section, and the data there should be in the |
| 1643 | format specified by <code>"type_list"</code> below. Additionally, there |
| 1644 | should be no reference to the type <code>void</code> in the list. |
| 1645 | </td> |
| 1646 | </tr> |
| 1647 | </tbody> |
| 1648 | </table> |
| 1649 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1650 | <h3>field_id_item</h3> |
| 1651 | <h4>appears in the field_ids section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1652 | <h4>alignment: 4 bytes</h4> |
| 1653 | |
| 1654 | <table class="format"> |
| 1655 | <thead> |
| 1656 | <tr> |
| 1657 | <th>Name</th> |
| 1658 | <th>Format</th> |
| 1659 | <th>Description</th> |
| 1660 | </tr> |
| 1661 | </thead> |
| 1662 | <tbody> |
| 1663 | <tr> |
| 1664 | <td>class_idx</td> |
| 1665 | <td>ushort</td> |
| 1666 | <td>index into the <code>type_ids</code> list for the definer of this |
| 1667 | field. This must be a class type, and not an array or primitive type. |
| 1668 | </td> |
| 1669 | </tr> |
| 1670 | <tr> |
| 1671 | <td>type_idx</td> |
| 1672 | <td>ushort</td> |
| 1673 | <td>index into the <code>type_ids</code> list for the type of |
| 1674 | this field |
| 1675 | </td> |
| 1676 | </tr> |
| 1677 | <tr> |
| 1678 | <td>name_idx</td> |
| 1679 | <td>uint</td> |
| 1680 | <td>index into the <code>string_ids</code> list for the name of this |
| 1681 | field. The string must conform to the syntax for <i>MemberName</i>, |
| 1682 | defined above. |
| 1683 | </td> |
| 1684 | </tr> |
| 1685 | </tbody> |
| 1686 | </table> |
| 1687 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1688 | <h3>method_id_item</h3> |
| 1689 | <h4>appears in the method_ids section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1690 | <h4>alignment: 4 bytes</h4> |
| 1691 | |
| 1692 | <table class="format"> |
| 1693 | <thead> |
| 1694 | <tr> |
| 1695 | <th>Name</th> |
| 1696 | <th>Format</th> |
| 1697 | <th>Description</th> |
| 1698 | </tr> |
| 1699 | </thead> |
| 1700 | <tbody> |
| 1701 | <tr> |
| 1702 | <td>class_idx</td> |
| 1703 | <td>ushort</td> |
| 1704 | <td>index into the <code>type_ids</code> list for the definer of this |
| 1705 | method. This must be a class or array type, and not a primitive type. |
| 1706 | </td> |
| 1707 | </tr> |
| 1708 | <tr> |
| 1709 | <td>proto_idx</td> |
| 1710 | <td>ushort</td> |
| 1711 | <td>index into the <code>proto_ids</code> list for the prototype of |
| 1712 | this method |
| 1713 | </td> |
| 1714 | </tr> |
| 1715 | <tr> |
| 1716 | <td>name_idx</td> |
| 1717 | <td>uint</td> |
| 1718 | <td>index into the <code>string_ids</code> list for the name of this |
| 1719 | method. The string must conform to the syntax for <i>MemberName</i>, |
| 1720 | defined above. |
| 1721 | </td> |
| 1722 | </tr> |
| 1723 | </tbody> |
| 1724 | </table> |
| 1725 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1726 | <h3>class_def_item</h3> |
| 1727 | <h4>appears in the class_defs section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1728 | <h4>alignment: 4 bytes</h4> |
| 1729 | |
| 1730 | <table class="format"> |
| 1731 | <thead> |
| 1732 | <tr> |
| 1733 | <th>Name</th> |
| 1734 | <th>Format</th> |
| 1735 | <th>Description</th> |
| 1736 | </tr> |
| 1737 | </thead> |
| 1738 | <tbody> |
| 1739 | <tr> |
| 1740 | <td>class_idx</td> |
| 1741 | <td>uint</td> |
| 1742 | <td>index into the <code>type_ids</code> list for this class. |
| 1743 | This must be a class type, and not an array or primitive type. |
| 1744 | </td> |
| 1745 | </tr> |
| 1746 | <tr> |
| 1747 | <td>access_flags</td> |
| 1748 | <td>uint</td> |
| 1749 | <td>access flags for the class (<code>public</code>, <code>final</code>, |
| 1750 | etc.). See "<code>access_flags</code> Definitions" for details. |
| 1751 | </td> |
| 1752 | </tr> |
| 1753 | <tr> |
| 1754 | <td>superclass_idx</td> |
| 1755 | <td>uint</td> |
| 1756 | <td>index into the <code>type_ids</code> list for the superclass, or |
| 1757 | the constant value <code>NO_INDEX</code> if this class has no |
| 1758 | superclass (i.e., it is a root class such as <code>Object</code>). |
| 1759 | If present, this must be a class type, and not an array or primitive type. |
| 1760 | </td> |
| 1761 | </tr> |
| 1762 | <tr> |
| 1763 | <td>interfaces_off</td> |
| 1764 | <td>uint</td> |
| 1765 | <td>offset from the start of the file to the list of interfaces, or |
| 1766 | <code>0</code> if there are none. This offset |
| 1767 | should be in the <code>data</code> section, and the data |
| 1768 | there should be in the format specified by |
| 1769 | "<code>type_list</code>" below. Each of the elements of the list |
| 1770 | must be a class type (not an array or primitive type), and there |
| 1771 | must not be any duplicates. |
| 1772 | </td> |
| 1773 | </tr> |
| 1774 | <tr> |
| 1775 | <td>source_file_idx</td> |
| 1776 | <td>uint</td> |
| 1777 | <td>index into the <code>string_ids</code> list for the name of the |
| 1778 | file containing the original source for (at least most of) this class, |
| 1779 | or the special value <code>NO_INDEX</code> to represent a lack of |
| 1780 | this information. The <code>debug_info_item</code> of any given method |
| 1781 | may override this source file, but the expectation is that most classes |
| 1782 | will only come from one source file. |
| 1783 | </td> |
| 1784 | </tr> |
| 1785 | <tr> |
| 1786 | <td>annotations_off</td> |
| 1787 | <td>uint</td> |
| 1788 | <td>offset from the start of the file to the annotations structure |
| 1789 | for this class, or <code>0</code> if there are no annotations on |
| 1790 | this class. This offset, if non-zero, should be in the |
| 1791 | <code>data</code> section, and the data there should be in |
| 1792 | the format specified by "<code>annotations_directory_item</code>" below, |
| 1793 | with all items referring to this class as the definer. |
| 1794 | </td> |
| 1795 | </tr> |
| 1796 | <tr> |
| 1797 | <td>class_data_off</td> |
| 1798 | <td>uint</td> |
| 1799 | <td>offset from the start of the file to the associated |
| 1800 | class data for this item, or <code>0</code> if there is no class |
| 1801 | data for this class. (This may be the case, for example, if this class |
| 1802 | is a marker interface.) The offset, if non-zero, should be in the |
| 1803 | <code>data</code> section, and the data there should be in the |
| 1804 | format specified by "<code>class_data_item</code>" below, with all |
| 1805 | items referring to this class as the definer. |
| 1806 | </td> |
| 1807 | </tr> |
| 1808 | <tr> |
| 1809 | <td>static_values_off</td> |
| 1810 | <td>uint</td> |
| 1811 | <td>offset from the start of the file to the list of initial |
| 1812 | values for <code>static</code> fields, or <code>0</code> if there |
| 1813 | are none (and all <code>static</code> fields are to be initialized with |
| 1814 | <code>0</code> or <code>null</code>). This offset should be in the |
| 1815 | <code>data</code> section, and the data there should be in the |
| 1816 | format specified by "<code>encoded_array_item</code>" below. The size |
| 1817 | of the array must be no larger than the number of <code>static</code> |
| 1818 | fields declared by this class, and the elements correspond to the |
| 1819 | <code>static</code> fields in the same order as declared in the |
| 1820 | corresponding <code>field_list</code>. The type of each array |
| 1821 | element must match the declared type of its corresponding field. |
| 1822 | If there are fewer elements in the array than there are |
| 1823 | <code>static</code> fields, then the leftover fields are initialized |
| 1824 | with a type-appropriate <code>0</code> or <code>null</code>. |
| 1825 | </td> |
| 1826 | </tr> |
| 1827 | </tbody> |
| 1828 | </table> |
| 1829 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1830 | <h3>class_data_item</h3> |
| 1831 | <h4>referenced from class_def_item</h4> |
| 1832 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1833 | <h4>alignment: none (byte-aligned)</h4> |
| 1834 | |
| 1835 | <table class="format"> |
| 1836 | <thead> |
| 1837 | <tr> |
| 1838 | <th>Name</th> |
| 1839 | <th>Format</th> |
| 1840 | <th>Description</th> |
| 1841 | </tr> |
| 1842 | </thead> |
| 1843 | <tbody> |
| 1844 | <tr> |
| 1845 | <td>static_fields_size</td> |
| 1846 | <td>uleb128</td> |
| 1847 | <td>the number of static fields defined in this item</td> |
| 1848 | </tr> |
| 1849 | <tr> |
| 1850 | <td>instance_fields_size</td> |
| 1851 | <td>uleb128</td> |
| 1852 | <td>the number of instance fields defined in this item</td> |
| 1853 | </tr> |
| 1854 | <tr> |
| 1855 | <td>direct_methods_size</td> |
| 1856 | <td>uleb128</td> |
| 1857 | <td>the number of direct methods defined in this item</td> |
| 1858 | </tr> |
| 1859 | <tr> |
| 1860 | <td>virtual_methods_size</td> |
| 1861 | <td>uleb128</td> |
| 1862 | <td>the number of virtual methods defined in this item</td> |
| 1863 | </tr> |
| 1864 | <tr> |
| 1865 | <td>static_fields</td> |
| 1866 | <td>encoded_field[static_fields_size]</td> |
| 1867 | <td>the defined static fields, represented as a sequence of |
| 1868 | encoded elements. The fields must be sorted by |
| 1869 | <code>field_idx</code> in increasing order. |
| 1870 | </td> |
| 1871 | </tr> |
| 1872 | <tr> |
| 1873 | <td>instance_fields</td> |
| 1874 | <td>encoded_field[instance_fields_size]</td> |
| 1875 | <td>the defined instance fields, represented as a sequence of |
| 1876 | encoded elements. The fields must be sorted by |
| 1877 | <code>field_idx</code> in increasing order. |
| 1878 | </td> |
| 1879 | </tr> |
| 1880 | <tr> |
| 1881 | <td>direct_methods</td> |
| 1882 | <td>encoded_method[direct_methods_size]</td> |
| 1883 | <td>the defined direct (any of <code>static</code>, <code>private</code>, |
| 1884 | or constructor) methods, represented as a sequence of |
| 1885 | encoded elements. The methods must be sorted by |
| 1886 | <code>method_idx</code> in increasing order. |
| 1887 | </td> |
| 1888 | </tr> |
| 1889 | <tr> |
| 1890 | <td>virtual_methods</td> |
| 1891 | <td>encoded_method[virtual_methods_size]</td> |
| 1892 | <td>the defined virtual (none of <code>static</code>, <code>private</code>, |
| 1893 | or constructor) methods, represented as a sequence of |
| 1894 | encoded elements. This list should <i>not</i> include inherited |
| 1895 | methods unless overridden by the class that this item represents. The |
| 1896 | methods must be sorted by <code>method_idx</code> in increasing order. |
| 1897 | </td> |
| 1898 | </tr> |
| 1899 | </tbody> |
| 1900 | </table> |
| 1901 | |
| 1902 | <p><b>Note:</b> All elements' <code>field_id</code>s and |
| 1903 | <code>method_id</code>s must refer to the same defining class.</p> |
| 1904 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1905 | <h3>encoded_field Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1906 | |
| 1907 | <table class="format"> |
| 1908 | <thead> |
| 1909 | <tr> |
| 1910 | <th>Name</th> |
| 1911 | <th>Format</th> |
| 1912 | <th>Description</th> |
| 1913 | </tr> |
| 1914 | </thead> |
| 1915 | <tbody> |
| 1916 | <tr> |
| 1917 | <td>field_idx_diff</td> |
| 1918 | <td>uleb128</td> |
| 1919 | <td>index into the <code>field_ids</code> list for the identity of this |
| 1920 | field (includes the name and descriptor), represented as a difference |
| 1921 | from the index of previous element in the list. The index of the |
| 1922 | first element in a list is represented directly. |
| 1923 | </td> |
| 1924 | </tr> |
| 1925 | <tr> |
| 1926 | <td>access_flags</td> |
| 1927 | <td>uleb128</td> |
| 1928 | <td>access flags for the field (<code>public</code>, <code>final</code>, |
| 1929 | etc.). See "<code>access_flags</code> Definitions" for details. |
| 1930 | </td> |
| 1931 | </tr> |
| 1932 | </tbody> |
| 1933 | </table> |
| 1934 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1935 | <h3>encoded_method Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1936 | |
| 1937 | <table class="format"> |
| 1938 | <thead> |
| 1939 | <tr> |
| 1940 | <th>Name</th> |
| 1941 | <th>Format</th> |
| 1942 | <th>Description</th> |
| 1943 | </tr> |
| 1944 | </thead> |
| 1945 | <tbody> |
| 1946 | <tr> |
| 1947 | <td>method_idx_diff</td> |
| 1948 | <td>uleb128</td> |
| 1949 | <td>index into the <code>method_ids</code> list for the identity of this |
| 1950 | method (includes the name and descriptor), represented as a difference |
| 1951 | from the index of previous element in the list. The index of the |
| 1952 | first element in a list is represented directly. |
| 1953 | </td> |
| 1954 | </tr> |
| 1955 | <tr> |
| 1956 | <td>access_flags</td> |
| 1957 | <td>uleb128</td> |
| 1958 | <td>access flags for the method (<code>public</code>, <code>final</code>, |
| 1959 | etc.). See "<code>access_flags</code> Definitions" for details. |
| 1960 | </td> |
| 1961 | </tr> |
| 1962 | <tr> |
| 1963 | <td>code_off</td> |
| 1964 | <td>uleb128</td> |
| 1965 | <td>offset from the start of the file to the code structure for this |
| 1966 | method, or <code>0</code> if this method is either <code>abstract</code> |
| 1967 | or <code>native</code>. The offset should be to a location in the |
| 1968 | <code>data</code> section. The format of the data is specified by |
| 1969 | "<code>code_item</code>" below. |
| 1970 | </td> |
| 1971 | </tr> |
| 1972 | </tbody> |
| 1973 | </table> |
| 1974 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 1975 | <h3>type_list</h3> |
| 1976 | <h4>referenced from class_def_item and proto_id_item</h4> |
| 1977 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 1978 | <h4>alignment: 4 bytes</h4> |
| 1979 | |
| 1980 | <table class="format"> |
| 1981 | <thead> |
| 1982 | <tr> |
| 1983 | <th>Name</th> |
| 1984 | <th>Format</th> |
| 1985 | <th>Description</th> |
| 1986 | </tr> |
| 1987 | </thead> |
| 1988 | <tbody> |
| 1989 | <tr> |
| 1990 | <td>size</td> |
| 1991 | <td>uint</td> |
| 1992 | <td>size of the list, in entries</td> |
| 1993 | </tr> |
| 1994 | <tr> |
| 1995 | <td>list</td> |
| 1996 | <td>type_item[size]</td> |
| 1997 | <td>elements of the list</td> |
| 1998 | </tr> |
| 1999 | </tbody> |
| 2000 | </table> |
| 2001 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2002 | <h3>type_item Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2003 | |
| 2004 | <table class="format"> |
| 2005 | <thead> |
| 2006 | <tr> |
| 2007 | <th>Name</th> |
| 2008 | <th>Format</th> |
| 2009 | <th>Description</th> |
| 2010 | </tr> |
| 2011 | </thead> |
| 2012 | <tbody> |
| 2013 | <tr> |
| 2014 | <td>type_idx</td> |
| 2015 | <td>ushort</td> |
| 2016 | <td>index into the <code>type_ids</code> list</td> |
| 2017 | </tr> |
| 2018 | </tbody> |
| 2019 | </table> |
| 2020 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2021 | <h3>code_item</h3> |
| 2022 | <h4>referenced from encoded_method</h4> |
| 2023 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2024 | <h4>alignment: 4 bytes</h4> |
| 2025 | |
| 2026 | <table class="format"> |
| 2027 | <thead> |
| 2028 | <tr> |
| 2029 | <th>Name</th> |
| 2030 | <th>Format</th> |
| 2031 | <th>Description</th> |
| 2032 | </tr> |
| 2033 | </thead> |
| 2034 | <tbody> |
| 2035 | <tr> |
| 2036 | <td>registers_size</td> |
| 2037 | <td>ushort</td> |
| 2038 | <td>the number of registers used by this code</td> |
| 2039 | </tr> |
| 2040 | <tr> |
| 2041 | <td>ins_size</td> |
| 2042 | <td>ushort</td> |
| 2043 | <td>the number of words of incoming arguments to the method that this |
| 2044 | code is for</td> |
| 2045 | </tr> |
| 2046 | <tr> |
| 2047 | <td>outs_size</td> |
| 2048 | <td>ushort</td> |
| 2049 | <td>the number of words of outgoing argument space required by this |
| 2050 | code for method invocation |
| 2051 | </td> |
| 2052 | </tr> |
| 2053 | <tr> |
| 2054 | <td>tries_size</td> |
| 2055 | <td>ushort</td> |
| 2056 | <td>the number of <code>try_item</code>s for this instance. If non-zero, |
| 2057 | then these appear as the <code>tries</code> array just after the |
| 2058 | <code>insns</code> in this instance. |
| 2059 | </td> |
| 2060 | </tr> |
| 2061 | <tr> |
| 2062 | <td>debug_info_off</td> |
| 2063 | <td>uint</td> |
| 2064 | <td>offset from the start of the file to the debug info (line numbers + |
| 2065 | local variable info) sequence for this code, or <code>0</code> if |
| 2066 | there simply is no information. The offset, if non-zero, should be |
| 2067 | to a location in the <code>data</code> section. The format of |
| 2068 | the data is specified by "<code>debug_info_item</code>" below. |
| 2069 | </td> |
| 2070 | </tr> |
| 2071 | <tr> |
| 2072 | <td>insns_size</td> |
| 2073 | <td>uint</td> |
| 2074 | <td>size of the instructions list, in 16-bit code units</td> |
| 2075 | </tr> |
| 2076 | <tr> |
| 2077 | <td>insns</td> |
| 2078 | <td>ushort[insns_size]</td> |
| 2079 | <td>actual array of bytecode. The format of code in an <code>insns</code> |
| 2080 | array is specified by the companion document |
| 2081 | <a href="dalvik-bytecode.html">"Bytecode for the Dalvik VM"</a>. Note |
| 2082 | that though this is defined as an array of <code>ushort</code>, there |
| 2083 | are some internal structures that prefer four-byte alignment. Also, |
| 2084 | if this happens to be in an endian-swapped file, then the swapping is |
| 2085 | <i>only</i> done on individual <code>ushort</code>s and not on the |
| 2086 | larger internal structures. |
| 2087 | </td> |
| 2088 | </tr> |
| 2089 | <tr> |
| 2090 | <td>padding</td> |
| 2091 | <td>ushort <i>(optional)</i> = 0</td> |
| 2092 | <td>two bytes of padding to make <code>tries</code> four-byte aligned. |
| 2093 | This element is only present if <code>tries_size</code> is non-zero |
| 2094 | and <code>insns_size</code> is odd. |
| 2095 | </td> |
| 2096 | </tr> |
| 2097 | <tr> |
| 2098 | <td>tries</td> |
| 2099 | <td>try_item[tries_size] <i>(optional)</i></td> |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 2100 | <td>array indicating where in the code exceptions are caught and |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2101 | how to handle them. Elements of the array must be non-overlapping in |
| 2102 | range and in order from low to high address. This element is only |
| 2103 | present if <code>tries_size</code> is non-zero. |
| 2104 | </td> |
| 2105 | </tr> |
| 2106 | <tr> |
| 2107 | <td>handlers</td> |
| 2108 | <td>encoded_catch_handler_list <i>(optional)</i></td> |
| 2109 | <td>bytes representing a list of lists of catch types and associated |
| 2110 | handler addresses. Each <code>try_item</code> has a byte-wise offset |
| 2111 | into this structure. This element is only present if |
| 2112 | <code>tries_size</code> is non-zero. |
| 2113 | </td> |
| 2114 | </tr> |
| 2115 | </tbody> |
| 2116 | </table> |
| 2117 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2118 | <h3>try_item Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2119 | |
| 2120 | <table class="format"> |
| 2121 | <thead> |
| 2122 | <tr> |
| 2123 | <th>Name</th> |
| 2124 | <th>Format</th> |
| 2125 | <th>Description</th> |
| 2126 | </tr> |
| 2127 | </thead> |
| 2128 | <tbody> |
| 2129 | <tr> |
| 2130 | <td>start_addr</td> |
| 2131 | <td>uint</td> |
| 2132 | <td>start address of the block of code covered by this entry. The address |
| 2133 | is a count of 16-bit code units to the start of the first covered |
| 2134 | instruction. |
| 2135 | </td> |
| 2136 | </tr> |
| 2137 | <tr> |
| 2138 | <td>insn_count</td> |
| 2139 | <td>ushort</td> |
| 2140 | <td>number of 16-bit code units covered by this entry. The last code |
| 2141 | unit covered (inclusive) is <code>start_addr + insn_count - 1</code>. |
| 2142 | </td> |
| 2143 | </tr> |
| 2144 | <tr> |
| 2145 | <td>handler_off</td> |
| 2146 | <td>ushort</td> |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 2147 | <td>offset in bytes from the start of the associated |
| 2148 | <code>encoded_catch_hander_list</code> to the |
| 2149 | <code>encoded_catch_handler</code> for this entry. This must be an |
| 2150 | offset to the start of an <code>encoded_catch_handler</code>. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2151 | </td> |
| 2152 | </tr> |
| 2153 | </tbody> |
| 2154 | </table> |
| 2155 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2156 | <h3>encoded_catch_handler_list Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2157 | |
| 2158 | <table class="format"> |
| 2159 | <thead> |
| 2160 | <tr> |
| 2161 | <th>Name</th> |
| 2162 | <th>Format</th> |
| 2163 | <th>Description</th> |
| 2164 | </tr> |
| 2165 | </thead> |
| 2166 | <tbody> |
| 2167 | <tr> |
| 2168 | <td>size</td> |
| 2169 | <td>uleb128</td> |
| 2170 | <td>size of this list, in entries</td> |
| 2171 | </tr> |
| 2172 | <tr> |
| 2173 | <td>list</td> |
| 2174 | <td>encoded_catch_handler[handlers_size]</td> |
| 2175 | <td>actual list of handler lists, represented directly (not as offsets), |
| 2176 | and concatenated sequentially</td> |
| 2177 | </tr> |
| 2178 | </tbody> |
| 2179 | </table> |
| 2180 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2181 | <h3>encoded_catch_handler Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2182 | |
| 2183 | <table class="format"> |
| 2184 | <thead> |
| 2185 | <tr> |
| 2186 | <th>Name</th> |
| 2187 | <th>Format</th> |
| 2188 | <th>Description</th> |
| 2189 | </tr> |
| 2190 | </thead> |
| 2191 | <tbody> |
| 2192 | <tr> |
| 2193 | <td>size</td> |
| 2194 | <td>sleb128</td> |
| 2195 | <td>number of catch types in this list. If non-positive, then this is |
| 2196 | the negative of the number of catch types, and the catches are followed |
| 2197 | by a catch-all handler. For example: A <code>size</code> of <code>0</code> |
| 2198 | means that there is a catch-all but no explicitly typed catches. |
| 2199 | A <code>size</code> of <code>2</code> means that there are two explicitly |
| 2200 | typed catches and no catch-all. And a <code>size</code> of <code>-1</code> |
| 2201 | means that there is one typed catch along with a catch-all. |
| 2202 | </td> |
| 2203 | </tr> |
| 2204 | <tr> |
| 2205 | <td>handlers</td> |
| 2206 | <td>encoded_type_addr_pair[abs(size)]</td> |
| 2207 | <td>stream of <code>abs(size)</code> encoded items, one for each caught |
| 2208 | type, in the order that the types should be tested. |
| 2209 | </td> |
| 2210 | </tr> |
| 2211 | <tr> |
| 2212 | <td>catch_all_addr</td> |
| 2213 | <td>uleb128 <i>(optional)</i></td> |
| 2214 | <td>bytecode address of the catch-all handler. This element is only |
| 2215 | present if <code>size</code> is non-positive. |
| 2216 | </td> |
| 2217 | </tr> |
| 2218 | </tbody> |
| 2219 | </table> |
| 2220 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2221 | <h3>encoded_type_addr_pair Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2222 | |
| 2223 | <table class="format"> |
| 2224 | <thead> |
| 2225 | <tr> |
| 2226 | <th>Name</th> |
| 2227 | <th>Format</th> |
| 2228 | <th>Description</th> |
| 2229 | </tr> |
| 2230 | </thead> |
| 2231 | <tbody> |
| 2232 | <tr> |
| 2233 | <td>type_idx</td> |
| 2234 | <td>uleb128</td> |
| 2235 | <td>index into the <code>type_ids</code> list for the type of the |
| 2236 | exception to catch |
| 2237 | </td> |
| 2238 | </tr> |
| 2239 | <tr> |
| 2240 | <td>addr</td> |
| 2241 | <td>uleb128</td> |
| 2242 | <td>bytecode address of the associated exception handler</td> |
| 2243 | </tr> |
| 2244 | </tbody> |
| 2245 | </table> |
| 2246 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2247 | <h3>debug_info_item</h3> |
| 2248 | <h4>referenced from code_item</h4> |
| 2249 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2250 | <h4>alignment: none (byte-aligned)</h4> |
| 2251 | |
| 2252 | <p>Each <code>debug_info_item</code> defines a DWARF3-inspired byte-coded |
| 2253 | state machine that, when interpreted, emits the positions |
| 2254 | table and (potentially) the local variable information for a |
| 2255 | <code>code_item</code>. The sequence begins with a variable-length |
| 2256 | header (the length of which depends on the number of method |
| 2257 | parameters), is followed by the state machine bytecodes, and ends |
| 2258 | with an <code>DBG_END_SEQUENCE</code> byte.</p> |
| 2259 | |
| 2260 | <p>The state machine consists of five registers. The |
| 2261 | <code>address</code> register represents the instruction offset in the |
| 2262 | associated <code>insns_item</code> in 16-bit code units. The |
| 2263 | <code>address</code> register starts at <code>0</code> at the beginning of each |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 2264 | <code>debug_info</code> sequence and must only monotonically increase. |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2265 | The <code>line</code> register represents what source line number |
| 2266 | should be associated with the next positions table entry emitted by |
| 2267 | the state machine. It is initialized in the sequence header, and may |
| 2268 | change in positive or negative directions but must never be less than |
| 2269 | <code>1</code>. The <code>source_file</code> register represents the |
| 2270 | source file that the line number entries refer to. It is initialized to |
| 2271 | the value of <code>source_file_idx</code> in <code>class_def_item</code>. |
| 2272 | The other two variables, <code>prologue_end</code> and |
| 2273 | <code>epilogue_begin</code>, are boolean flags (initialized to |
| 2274 | <code>false</code>) that indicate whether the next position emitted |
| 2275 | should be considered a method prologue or epilogue. The state machine |
| 2276 | must also track the name and type of the last local variable live in |
| 2277 | each register for the <code>DBG_RESTART_LOCAL</code> code.</p> |
| 2278 | |
| 2279 | <p>The header is as follows:</p> |
| 2280 | |
| 2281 | <table class="format"> |
| 2282 | <thead> |
| 2283 | <tr> |
| 2284 | <th>Name</th> |
| 2285 | <th>Format</th> |
| 2286 | <th>Description</th> |
| 2287 | </tr> |
| 2288 | </thead> |
| 2289 | <tbody> |
| 2290 | <tr> |
| 2291 | <td>line_start</td> |
| 2292 | <td>uleb128</td> |
| 2293 | <td>the initial value for the state machine's <code>line</code> register. |
| 2294 | Does not represent an actual positions entry. |
| 2295 | </td> |
| 2296 | </tr> |
| 2297 | <tr> |
| 2298 | <td>parameters_size</td> |
| 2299 | <td>uleb128</td> |
| 2300 | <td>the number of parameter names that are encoded. There should be |
| 2301 | one per method parameter, excluding an instance method's <code>this</code>, |
| 2302 | if any. |
| 2303 | </td> |
| 2304 | </tr> |
| 2305 | <tr> |
| 2306 | <td>parameter_names</td> |
| 2307 | <td>uleb128p1[parameters_size]</td> |
| 2308 | <td>string index of the method parameter name. An encoded value of |
| 2309 | <code>NO_INDEX</code> indicates that no name |
| 2310 | is available for the associated parameter. The type descriptor |
| 2311 | and signature are implied from the method descriptor and signature. |
| 2312 | </td> |
| 2313 | </tr> |
| 2314 | </tbody> |
| 2315 | </table> |
| 2316 | |
| 2317 | <p>The byte code values are as follows:</p> |
| 2318 | |
| 2319 | <table class="debugByteCode"> |
| 2320 | <thead> |
| 2321 | <tr> |
| 2322 | <th>Name</th> |
| 2323 | <th>Value</th> |
| 2324 | <th>Format</th> |
| 2325 | <th>Arguments</th> |
| 2326 | <th>Description</th> |
| 2327 | </tr> |
| 2328 | </thead> |
| 2329 | <tbody> |
| 2330 | <tr> |
| 2331 | <td>DBG_END_SEQUENCE</td> |
| 2332 | <td>0x00</td> |
| 2333 | <td></td> |
| 2334 | <td><i>(none)</i></td> |
| 2335 | <td>terminates a debug info sequence for a <code>code_item</code></td> |
| 2336 | </tr> |
| 2337 | <tr> |
| 2338 | <td>DBG_ADVANCE_PC</td> |
| 2339 | <td>0x01</td> |
| 2340 | <td>uleb128 addr_diff</td> |
| 2341 | <td><code>addr_diff</code>: amount to add to address register</td> |
| 2342 | <td>advances the address register without emitting a positions entry</td> |
| 2343 | </tr> |
| 2344 | <tr> |
| 2345 | <td>DBG_ADVANCE_LINE</td> |
| 2346 | <td>0x02</td> |
| 2347 | <td>sleb128 line_diff</td> |
| 2348 | <td><code>line_diff</code>: amount to change line register by</td> |
| 2349 | <td>advances the line register without emitting a positions entry</td> |
| 2350 | </tr> |
| 2351 | <tr> |
| 2352 | <td>DBG_START_LOCAL</td> |
| 2353 | <td>0x03</td> |
| 2354 | <td>uleb128 register_num<br/> |
| 2355 | uleb128p1 name_idx<br/> |
| 2356 | uleb128p1 type_idx |
| 2357 | </td> |
| 2358 | <td><code>register_num</code>: register that will contain local<br/> |
| 2359 | <code>name_idx</code>: string index of the name<br/> |
| 2360 | <code>type_idx</code>: type index of the type |
| 2361 | </td> |
| 2362 | <td>introduces a local variable at the current address. Either |
| 2363 | <code>name_idx</code> or <code>type_idx</code> may be |
| 2364 | <code>NO_INDEX</code> to indicate that that value is unknown. |
| 2365 | </td> |
| 2366 | </tr> |
| 2367 | <tr> |
| 2368 | <td>DBG_START_LOCAL_EXTENDED</td> |
| 2369 | <td>0x04</td> |
| 2370 | <td>uleb128 register_num<br/> |
| 2371 | uleb128p1 name_idx<br/> |
| 2372 | uleb128p1 type_idx<br/> |
| 2373 | uleb128p1 sig_idx |
| 2374 | </td> |
| 2375 | <td><code>register_num</code>: register that will contain local<br/> |
| 2376 | <code>name_idx</code>: string index of the name<br/> |
| 2377 | <code>type_idx</code>: type index of the type<br/> |
| 2378 | <code>sig_idx</code>: string index of the type signature |
| 2379 | </td> |
| 2380 | <td>introduces a local with a type signature at the current address. |
| 2381 | Any of <code>name_idx</code>, <code>type_idx</code>, or |
| 2382 | <code>sig_idx</code> may be <code>NO_INDEX</code> |
| 2383 | to indicate that that value is unknown. (If <code>sig_idx</code> is |
| 2384 | <code>-1</code>, though, the same data could be represented more |
| 2385 | efficiently using the opcode <code>DBG_START_LOCAL</code>.) |
| 2386 | <p><b>Note:</b> See the discussion under |
| 2387 | "<code>dalvik.annotation.Signature</code>" below for caveats about |
| 2388 | handling signatures.</p> |
| 2389 | </td> |
| 2390 | </tr> |
| 2391 | <tr> |
| 2392 | <td>DBG_END_LOCAL</td> |
| 2393 | <td>0x05</td> |
| 2394 | <td>uleb128 register_num</td> |
| 2395 | <td><code>register_num</code>: register that contained local</td> |
| 2396 | <td>marks a currently-live local variable as out of scope at the current |
| 2397 | address |
| 2398 | </td> |
| 2399 | </tr> |
| 2400 | <tr> |
| 2401 | <td>DBG_RESTART_LOCAL</td> |
| 2402 | <td>0x06</td> |
| 2403 | <td>uleb128 register_num</td> |
| 2404 | <td><code>register_num</code>: register to restart</td> |
| 2405 | <td>re-introduces a local variable at the current address. The name |
| 2406 | and type are the same as the last local that was live in the specified |
| 2407 | register. |
| 2408 | </td> |
| 2409 | </tr> |
| 2410 | <tr> |
| 2411 | <td>DBG_SET_PROLOGUE_END</td> |
| 2412 | <td>0x07</td> |
| 2413 | <td></td> |
| 2414 | <td><i>(none)</i></td> |
| 2415 | <td>sets the <code>prologue_end</code> state machine register, |
| 2416 | indicating that the next position entry that is added should be |
| 2417 | considered the end of a method prologue (an appropriate place for |
| 2418 | a method breakpoint). The <code>prologue_end</code> register is |
| 2419 | cleared by any special (<code>>= 0x0a</code>) opcode. |
| 2420 | </td> |
| 2421 | </tr> |
| 2422 | <tr> |
| 2423 | <td>DBG_SET_EPILOGUE_BEGIN</td> |
| 2424 | <td>0x08</td> |
| 2425 | <td></td> |
| 2426 | <td><i>(none)</i></td> |
| 2427 | <td>sets the <code>epilogue_begin</code> state machine register, |
| 2428 | indicating that the next position entry that is added should be |
| 2429 | considered the beginning of a method epilogue (an appropriate place |
| 2430 | to suspend execution before method exit). |
| 2431 | The <code>epilogue_begin</code> register is cleared by any special |
| 2432 | (<code>>= 0x0a</code>) opcode. |
| 2433 | </td> |
| 2434 | </tr> |
| 2435 | <tr> |
| 2436 | <td>DBG_SET_FILE</td> |
| 2437 | <td>0x09</td> |
| 2438 | <td>uleb128p1 name_idx</td> |
| 2439 | <td><code>name_idx</code>: string index of source file name; |
| 2440 | <code>NO_INDEX</code> if unknown |
| 2441 | </td> |
| 2442 | <td>indicates that all subsequent line number entries make reference to this |
| 2443 | source file name, instead of the default name specified in |
| 2444 | <code>code_item</code> |
| 2445 | </td> |
| 2446 | </tr> |
| 2447 | <tr> |
| 2448 | <td><i>Special Opcodes</i></td> |
| 2449 | <!-- When updating the range below, make sure to search for other |
| 2450 | instances of 0x0a in this section. --> |
| 2451 | <td>0x0a…0xff</td> |
| 2452 | <td></td> |
| 2453 | <td><i>(none)</i></td> |
| 2454 | <td>advances the <code>line</code> and <code>address</code> registers, |
| 2455 | emits a position entry, and clears <code>prologue_end</code> and |
| 2456 | <code>epilogue_begin</code>. See below for description. |
| 2457 | </td> |
| 2458 | </tr> |
| 2459 | </tbody> |
| 2460 | </table> |
| 2461 | |
| 2462 | <h3>Special Opcodes</h3> |
| 2463 | |
| 2464 | <p>Opcodes with values between <code>0x0a</code> and <code>0xff</code> |
| 2465 | (inclusive) move both the <code>line</code> and <code>address</code> |
| 2466 | registers by a small amount and then emit a new position table entry. |
| 2467 | The formula for the increments are as follows:</p> |
| 2468 | |
| 2469 | <pre> |
| 2470 | DBG_FIRST_SPECIAL = 0x0a // the smallest special opcode |
| 2471 | DBG_LINE_BASE = -4 // the smallest line number increment |
| 2472 | DBG_LINE_RANGE = 15 // the number of line increments represented |
| 2473 | |
| 2474 | adjusted_opcode = opcode - DBG_FIRST_SPECIAL |
| 2475 | |
| 2476 | line += DBG_LINE_BASE + (adjusted_opcode % DBG_LINE_RANGE) |
| 2477 | address += (adjusted_opcode / DBG_LINE_RANGE) |
| 2478 | </pre> |
| 2479 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2480 | <h3>annotations_directory_item</h3> |
| 2481 | <h4>referenced from class_def_item</h4> |
| 2482 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2483 | <h4>alignment: 4 bytes</h4> |
| 2484 | |
| 2485 | <table class="format"> |
| 2486 | <thead> |
| 2487 | <tr> |
| 2488 | <th>Name</th> |
| 2489 | <th>Format</th> |
| 2490 | <th>Description</th> |
| 2491 | </tr> |
| 2492 | </thead> |
| 2493 | <tbody> |
| 2494 | <tr> |
| 2495 | <td>class_annotations_off</td> |
| 2496 | <td>uint</td> |
| 2497 | <td>offset from the start of the file to the annotations made directly |
| 2498 | on the class, or <code>0</code> if the class has no direct annotations. |
| 2499 | The offset, if non-zero, should be to a location in the |
| 2500 | <code>data</code> section. The format of the data is specified |
| 2501 | by "<code>annotation_set_item</code>" below. |
| 2502 | </td> |
| 2503 | </tr> |
| 2504 | <tr> |
| 2505 | <td>fields_size</td> |
| 2506 | <td>uint</td> |
| 2507 | <td>count of fields annotated by this item</td> |
| 2508 | </tr> |
| 2509 | <tr> |
| 2510 | <td>annotated_methods_size</td> |
| 2511 | <td>uint</td> |
| 2512 | <td>count of methods annotated by this item</td> |
| 2513 | </tr> |
| 2514 | <tr> |
| 2515 | <td>annotated_parameters_size</td> |
| 2516 | <td>uint</td> |
| 2517 | <td>count of method parameter lists annotated by this item</td> |
| 2518 | </tr> |
| 2519 | <tr> |
| 2520 | <td>field_annotations</td> |
| 2521 | <td>field_annotation[fields_size] <i>(optional)</i></td> |
| 2522 | <td>list of associated field annotations. The elements of the list must |
| 2523 | be sorted in increasing order, by <code>field_idx</code>. |
| 2524 | </td> |
| 2525 | </tr> |
| 2526 | <tr> |
| 2527 | <td>method_annotations</td> |
| 2528 | <td>method_annotation[methods_size] <i>(optional)</i></td> |
| 2529 | <td>list of associated method annotations. The elements of the list must |
| 2530 | be sorted in increasing order, by <code>method_idx</code>. |
| 2531 | </td> |
| 2532 | </tr> |
| 2533 | <tr> |
| 2534 | <td>parameter_annotations</td> |
| 2535 | <td>parameter_annotation[parameters_size] <i>(optional)</i></td> |
| 2536 | <td>list of associated method parameter annotations. The elements of the |
| 2537 | list must be sorted in increasing order, by <code>method_idx</code>. |
| 2538 | </td> |
| 2539 | </tr> |
| 2540 | </tbody> |
| 2541 | </table> |
| 2542 | |
| 2543 | <p><b>Note:</b> All elements' <code>field_id</code>s and |
| 2544 | <code>method_id</code>s must refer to the same defining class.</p> |
| 2545 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2546 | <h3>field_annotation Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2547 | |
| 2548 | <table class="format"> |
| 2549 | <thead> |
| 2550 | <tr> |
| 2551 | <th>Name</th> |
| 2552 | <th>Format</th> |
| 2553 | <th>Description</th> |
| 2554 | </tr> |
| 2555 | </thead> |
| 2556 | <tbody> |
| 2557 | <tr> |
| 2558 | <td>field_idx</td> |
| 2559 | <td>uint</td> |
| 2560 | <td>index into the <code>field_ids</code> list for the identity of the |
| 2561 | field being annotated |
| 2562 | </td> |
| 2563 | </tr> |
| 2564 | <tr> |
| 2565 | <td>annotations_off</td> |
| 2566 | <td>uint</td> |
| 2567 | <td>offset from the start of the file to the list of annotations for |
| 2568 | the field. The offset should be to a location in the <code>data</code> |
| 2569 | section. The format of the data is specified by |
| 2570 | "<code>annotation_set_item</code>" below. |
| 2571 | </td> |
| 2572 | </tr> |
| 2573 | </tbody> |
| 2574 | </table> |
| 2575 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2576 | <h3>method_annotation Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2577 | |
| 2578 | <table class="format"> |
| 2579 | <thead> |
| 2580 | <tr> |
| 2581 | <th>Name</th> |
| 2582 | <th>Format</th> |
| 2583 | <th>Description</th> |
| 2584 | </tr> |
| 2585 | </thead> |
| 2586 | <tbody> |
| 2587 | <tr> |
| 2588 | <td>method_idx</td> |
| 2589 | <td>uint</td> |
| 2590 | <td>index into the <code>method_ids</code> list for the identity of the |
| 2591 | method being annotated |
| 2592 | </td> |
| 2593 | </tr> |
| 2594 | <tr> |
| 2595 | <td>annotations_off</td> |
| 2596 | <td>uint</td> |
| 2597 | <td>offset from the start of the file to the list of annotations for |
| 2598 | the method. The offset should be to a location in the |
| 2599 | <code>data</code> section. The format of the data is specified by |
| 2600 | "<code>annotation_set_item</code>" below. |
| 2601 | </td> |
| 2602 | </tr> |
| 2603 | </tbody> |
| 2604 | </table> |
| 2605 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2606 | <h3>parameter_annotation Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2607 | |
| 2608 | <table class="format"> |
| 2609 | <thead> |
| 2610 | <tr> |
| 2611 | <th>Name</th> |
| 2612 | <th>Format</th> |
| 2613 | <th>Description</th> |
| 2614 | </tr> |
| 2615 | </thead> |
| 2616 | <tbody> |
| 2617 | <tr> |
| 2618 | <td>method_idx</td> |
| 2619 | <td>uint</td> |
| 2620 | <td>index into the <code>method_ids</code> list for the identity of the |
| 2621 | method whose parameters are being annotated |
| 2622 | </td> |
| 2623 | </tr> |
| 2624 | <tr> |
| 2625 | <td>annotations_off</td> |
| 2626 | <td>uint</td> |
| 2627 | <td>offset from the start of the file to the list of annotations for |
| 2628 | the method parameters. The offset should be to a location in the |
| 2629 | <code>data</code> section. The format of the data is specified by |
| 2630 | "<code>annotation_set_ref_list</code>" below. |
| 2631 | </td> |
| 2632 | </tr> |
| 2633 | </tbody> |
| 2634 | </table> |
| 2635 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2636 | <h3>annotation_set_ref_list</h3> |
| 2637 | <h4>referenced from parameter_annotations_item</h4> |
| 2638 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2639 | <h4>alignment: 4 bytes</h4> |
| 2640 | |
| 2641 | <table class="format"> |
| 2642 | <thead> |
| 2643 | <tr> |
| 2644 | <th>Name</th> |
| 2645 | <th>Format</th> |
| 2646 | <th>Description</th> |
| 2647 | </tr> |
| 2648 | </thead> |
| 2649 | <tbody> |
| 2650 | <tr> |
| 2651 | <td>size</td> |
| 2652 | <td>uint</td> |
| 2653 | <td>size of the list, in entries</td> |
| 2654 | </tr> |
| 2655 | <tr> |
| 2656 | <td>list</td> |
| 2657 | <td>annotation_set_ref_item[size]</td> |
| 2658 | <td>elements of the list</td> |
| 2659 | </tr> |
| 2660 | </tbody> |
| 2661 | </table> |
| 2662 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2663 | <h3>annotation_set_ref_item Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2664 | |
| 2665 | <table class="format"> |
| 2666 | <thead> |
| 2667 | <tr> |
| 2668 | <th>Name</th> |
| 2669 | <th>Format</th> |
| 2670 | <th>Description</th> |
| 2671 | </tr> |
| 2672 | </thead> |
| 2673 | <tbody> |
| 2674 | <tr> |
| 2675 | <td>annotations_off</td> |
| 2676 | <td>uint</td> |
| 2677 | <td>offset from the start of the file to the referenced annotation set |
| 2678 | or <code>0</code> if there are no annotations for this element. |
| 2679 | The offset, if non-zero, should be to a location in the <code>data</code> |
| 2680 | section. The format of the data is specified by |
| 2681 | "<code>annotation_set_item</code>" below. |
| 2682 | </td> |
| 2683 | </tr> |
| 2684 | </tbody> |
| 2685 | </table> |
| 2686 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2687 | <h3>annotation_set_item</h3> |
| 2688 | <h4>referenced from annotations_directory_item, field_annotations_item, |
| 2689 | method_annotations_item, and annotation_set_ref_item</h4> |
| 2690 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2691 | <h4>alignment: 4 bytes</h4> |
| 2692 | |
| 2693 | <table class="format"> |
| 2694 | <thead> |
| 2695 | <tr> |
| 2696 | <th>Name</th> |
| 2697 | <th>Format</th> |
| 2698 | <th>Description</th> |
| 2699 | </tr> |
| 2700 | </thead> |
| 2701 | <tbody> |
| 2702 | <tr> |
| 2703 | <td>size</td> |
| 2704 | <td>uint</td> |
| 2705 | <td>size of the set, in entries</td> |
| 2706 | </tr> |
| 2707 | <tr> |
| 2708 | <td>entries</td> |
| 2709 | <td>annotation_off_item[size]</td> |
| 2710 | <td>elements of the set. The elements must be sorted in increasing order, |
| 2711 | by <code>type_idx</code>. |
| 2712 | </td> |
| 2713 | </tr> |
| 2714 | </tbody> |
| 2715 | </table> |
| 2716 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2717 | <h3>annotation_off_item Format</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2718 | |
| 2719 | <table class="format"> |
| 2720 | <thead> |
| 2721 | <tr> |
| 2722 | <th>Name</th> |
| 2723 | <th>Format</th> |
| 2724 | <th>Description</th> |
| 2725 | </tr> |
| 2726 | </thead> |
| 2727 | <tbody> |
| 2728 | <tr> |
| 2729 | <td>annotation_off</td> |
| 2730 | <td>uint</td> |
| 2731 | <td>offset from the start of the file to an annotation. |
| 2732 | The offset should be to a location in the <code>data</code> section, |
| 2733 | and the format of the data at that location is specified by |
| 2734 | "<code>annotation_item</code>" below. |
| 2735 | </td> |
| 2736 | </tr> |
| 2737 | </tbody> |
| 2738 | </table> |
| 2739 | |
| 2740 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2741 | <h3>annotation_item</h3> |
| 2742 | <h4>referenced from annotation_set_item</h4> |
| 2743 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2744 | <h4>alignment: none (byte-aligned)</h4> |
| 2745 | |
| 2746 | <table class="format"> |
| 2747 | <thead> |
| 2748 | <tr> |
| 2749 | <th>Name</th> |
| 2750 | <th>Format</th> |
| 2751 | <th>Description</th> |
| 2752 | </tr> |
| 2753 | </thead> |
| 2754 | <tbody> |
| 2755 | <tr> |
| 2756 | <td>visibility</td> |
| 2757 | <td>ubyte</td> |
| 2758 | <td>intended visibility of this annotation (see below)</td> |
| 2759 | </tr> |
| 2760 | <tr> |
| 2761 | <td>annotation</td> |
| 2762 | <td>encoded_annotation</td> |
| 2763 | <td>encoded annotation contents, in the format described by |
| 2764 | "<code>encoded_annotation</code> Format" under |
| 2765 | "<code>encoded_value</code> Encoding" above. |
| 2766 | </td> |
| 2767 | </tr> |
| 2768 | </tbody> |
| 2769 | </table> |
| 2770 | |
| 2771 | <h3>Visibility values</h3> |
| 2772 | |
| 2773 | <p>These are the options for the <code>visibility</code> field in an |
| 2774 | <code>annotation_item</code>:</p> |
| 2775 | |
| 2776 | <table class="format"> |
| 2777 | <thead> |
| 2778 | <tr> |
| 2779 | <th>Name</th> |
| 2780 | <th>Value</th> |
| 2781 | <th>Description</th> |
| 2782 | </tr> |
| 2783 | </thead> |
| 2784 | <tbody> |
| 2785 | <tr> |
| 2786 | <td>VISIBILITY_BUILD</td> |
| 2787 | <td>0x00</td> |
| 2788 | <td>intended only to be visible at build time (e.g., during compilation |
| 2789 | of other code) |
| 2790 | </td> |
| 2791 | </tr> |
| 2792 | <tr> |
| 2793 | <td>VISIBILITY_RUNTIME</td> |
| 2794 | <td>0x01</td> |
| 2795 | <td>intended to visible at runtime</td> |
| 2796 | </tr> |
| 2797 | <tr> |
| 2798 | <td>VISIBILITY_SYSTEM</td> |
| 2799 | <td>0x02</td> |
| 2800 | <td>intended to visible at runtime, but only to the underlying system |
| 2801 | (and not to regular user code) |
| 2802 | </td> |
| 2803 | </tr> |
| 2804 | </tbody> |
| 2805 | </table> |
| 2806 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2807 | <h3>encoded_array_item</h3> |
| 2808 | <h4>referenced from class_def_item</h4> |
| 2809 | <h4>appears in the data section</h4> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2810 | <h4>alignment: none (byte-aligned)</h4> |
| 2811 | |
| 2812 | <table class="format"> |
| 2813 | <thead> |
| 2814 | <tr> |
| 2815 | <th>Name</th> |
| 2816 | <th>Format</th> |
| 2817 | <th>Description</th> |
| 2818 | </tr> |
| 2819 | </thead> |
| 2820 | <tbody> |
| 2821 | <tr> |
| 2822 | <td>value</td> |
| 2823 | <td>encoded_array</td> |
| 2824 | <td>bytes representing the encoded array value, in the format specified |
| 2825 | by "<code>encoded_array</code> Format" under "<code>encoded_value</code> |
| 2826 | Encoding" above. |
| 2827 | </td> |
| 2828 | </tr> |
| 2829 | </tbody> |
| 2830 | </table> |
| 2831 | |
Clay Murphy | 414d471 | 2013-06-03 18:44:16 -0700 | [diff] [blame] | 2832 | <h2>System Annotations</h2> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2833 | |
| 2834 | <p>System annotations are used to represent various pieces of reflective |
| 2835 | information about classes (and methods and fields). This information is |
| 2836 | generally only accessed indirectly by client (non-system) code.</p> |
| 2837 | |
| 2838 | <p>System annotations are represented in <code>.dex</code> files as |
| 2839 | annotations with visibility set to <code>VISIBILITY_SYSTEM</code>. |
| 2840 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2841 | <h3>dalvik.annotation.AnnotationDefault</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2842 | <h4>appears on methods in annotation interfaces</h4> |
| 2843 | |
| 2844 | <p>An <code>AnnotationDefault</code> annotation is attached to each |
| 2845 | annotation interface which wishes to indicate default bindings.</p> |
| 2846 | |
| 2847 | <table class="format"> |
| 2848 | <thead> |
| 2849 | <tr> |
| 2850 | <th>Name</th> |
| 2851 | <th>Format</th> |
| 2852 | <th>Description</th> |
| 2853 | </tr> |
| 2854 | </thead> |
| 2855 | <tbody> |
| 2856 | <tr> |
| 2857 | <td>value</td> |
| 2858 | <td>Annotation</td> |
| 2859 | <td>the default bindings for this annotation, represented as an annotation |
| 2860 | of this type. The annotation need not include all names defined by the |
| 2861 | annotation; missing names simply do not have defaults. |
| 2862 | </td> |
| 2863 | </tr> |
| 2864 | </tbody> |
| 2865 | </table> |
| 2866 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2867 | <h3>dalvik.annotation.EnclosingClass</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2868 | <h4>appears on classes</h4> |
| 2869 | |
| 2870 | <p>An <code>EnclosingClass</code> annotation is attached to each class |
| 2871 | which is either defined as a member of another class, per se, or is |
| 2872 | anonymous but not defined within a method body (e.g., a synthetic |
| 2873 | inner class). Every class that has this annotation must also have an |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 2874 | <code>InnerClass</code> annotation. Additionally, a class must not have |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2875 | both an <code>EnclosingClass</code> and an |
| 2876 | <code>EnclosingMethod</code> annotation.</p> |
| 2877 | |
| 2878 | <table class="format"> |
| 2879 | <thead> |
| 2880 | <tr> |
| 2881 | <th>Name</th> |
| 2882 | <th>Format</th> |
| 2883 | <th>Description</th> |
| 2884 | </tr> |
| 2885 | </thead> |
| 2886 | <tbody> |
| 2887 | <tr> |
| 2888 | <td>value</td> |
| 2889 | <td>Class</td> |
| 2890 | <td>the class which most closely lexically scopes this class</td> |
| 2891 | </tr> |
| 2892 | </tbody> |
| 2893 | </table> |
| 2894 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2895 | <h3>dalvik.annotation.EnclosingMethod</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2896 | <h4>appears on classes</h4> |
| 2897 | |
| 2898 | <p>An <code>EnclosingMethod</code> annotation is attached to each class |
| 2899 | which is defined inside a method body. Every class that has this |
| 2900 | annotation must also have an <code>InnerClass</code> annotation. |
Elliott Hughes | 8d77794 | 2012-01-05 17:27:02 -0800 | [diff] [blame] | 2901 | Additionally, a class must not have both an <code>EnclosingClass</code> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2902 | and an <code>EnclosingMethod</code> annotation.</p> |
| 2903 | |
| 2904 | <table class="format"> |
| 2905 | <thead> |
| 2906 | <tr> |
| 2907 | <th>Name</th> |
| 2908 | <th>Format</th> |
| 2909 | <th>Description</th> |
| 2910 | </tr> |
| 2911 | </thead> |
| 2912 | <tbody> |
| 2913 | <tr> |
| 2914 | <td>value</td> |
| 2915 | <td>Method</td> |
| 2916 | <td>the method which most closely lexically scopes this class</td> |
| 2917 | </tr> |
| 2918 | </tbody> |
| 2919 | </table> |
| 2920 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2921 | <h3>dalvik.annotation.InnerClass</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2922 | <h4>appears on classes</h4> |
| 2923 | |
| 2924 | <p>An <code>InnerClass</code> annotation is attached to each class |
| 2925 | which is defined in the lexical scope of another class's definition. |
| 2926 | Any class which has this annotation must also have <i>either</i> an |
| 2927 | <code>EnclosingClass</code> annotation <i>or</i> an |
| 2928 | <code>EnclosingMethod</code> annotation.</p> |
| 2929 | |
| 2930 | <table class="format"> |
| 2931 | <thead> |
| 2932 | <tr> |
| 2933 | <th>Name</th> |
| 2934 | <th>Format</th> |
| 2935 | <th>Description</th> |
| 2936 | </tr> |
| 2937 | </thead> |
| 2938 | <tbody> |
| 2939 | <tr> |
| 2940 | <td>name</td> |
| 2941 | <td>String</td> |
| 2942 | <td>the originally declared simple name of this class (not including any |
| 2943 | package prefix). If this class is anonymous, then the name is |
| 2944 | <code>null</code>. |
| 2945 | </td> |
| 2946 | </tr> |
| 2947 | <tr> |
| 2948 | <td>accessFlags</td> |
| 2949 | <td>int</td> |
| 2950 | <td>the originally declared access flags of the class (which may differ |
| 2951 | from the effective flags because of a mismatch between the execution |
| 2952 | models of the source language and target virtual machine) |
| 2953 | </td> |
| 2954 | </tr> |
| 2955 | </tbody> |
| 2956 | </table> |
| 2957 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2958 | <h3>dalvik.annotation.MemberClasses</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2959 | <h4>appears on classes</h4> |
| 2960 | |
| 2961 | <p>A <code>MemberClasses</code> annotation is attached to each class |
| 2962 | which declares member classes. (A member class is a direct inner class |
| 2963 | that has a name.)</p> |
| 2964 | |
| 2965 | <table class="format"> |
| 2966 | <thead> |
| 2967 | <tr> |
| 2968 | <th>Name</th> |
| 2969 | <th>Format</th> |
| 2970 | <th>Description</th> |
| 2971 | </tr> |
| 2972 | </thead> |
| 2973 | <tbody> |
| 2974 | <tr> |
| 2975 | <td>value</td> |
| 2976 | <td>Class[]</td> |
| 2977 | <td>array of the member classes</td> |
| 2978 | </tr> |
| 2979 | </tbody> |
| 2980 | </table> |
| 2981 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 2982 | <h3>dalvik.annotation.Signature</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 2983 | <h4>appears on classes, fields, and methods</h4> |
| 2984 | |
| 2985 | <p>A <code>Signature</code> annotation is attached to each class, |
| 2986 | field, or method which is defined in terms of a more complicated type |
| 2987 | than is representable by a <code>type_id_item</code>. The |
| 2988 | <code>.dex</code> format does not define the format for signatures; it |
| 2989 | is merely meant to be able to represent whatever signatures a source |
| 2990 | language requires for successful implementation of that language's |
| 2991 | semantics. As such, signatures are not generally parsed (or verified) |
| 2992 | by virtual machine implementations. The signatures simply get handed |
| 2993 | off to higher-level APIs and tools (such as debuggers). Any use of a |
| 2994 | signature, therefore, should be written so as not to make any |
| 2995 | assumptions about only receiving valid signatures, explicitly guarding |
| 2996 | itself against the possibility of coming across a syntactically |
| 2997 | invalid signature.</p> |
| 2998 | |
| 2999 | <p>Because signature strings tend to have a lot of duplicated content, |
| 3000 | a <code>Signature</code> annotation is defined as an <i>array</i> of |
| 3001 | strings, where duplicated elements naturally refer to the same |
| 3002 | underlying data, and the signature is taken to be the concatenation of |
| 3003 | all the strings in the array. There are no rules about how to pull |
| 3004 | apart a signature into separate strings; that is entirely up to the |
| 3005 | tools that generate <code>.dex</code> files.</p> |
| 3006 | |
| 3007 | <table class="format"> |
| 3008 | <thead> |
| 3009 | <tr> |
| 3010 | <th>Name</th> |
| 3011 | <th>Format</th> |
| 3012 | <th>Description</th> |
| 3013 | </tr> |
| 3014 | </thead> |
| 3015 | <tbody> |
| 3016 | <tr> |
| 3017 | <td>value</td> |
| 3018 | <td>String[]</td> |
| 3019 | <td>the signature of this class or member, as an array of strings that |
| 3020 | is to be concatenated together</td> |
| 3021 | </tr> |
| 3022 | </tbody> |
| 3023 | </table> |
| 3024 | |
Clay Murphy | 945af1a | 2013-07-01 17:31:13 -0700 | [diff] [blame] | 3025 | <h3>dalvik.annotation.Throws</h3> |
Dan Bornstein | 25705bc | 2011-04-12 16:23:13 -0700 | [diff] [blame] | 3026 | <h4>appears on methods</h4> |
| 3027 | |
| 3028 | <p>A <code>Throws</code> annotation is attached to each method which is |
| 3029 | declared to throw one or more exception types.</p> |
| 3030 | |
| 3031 | <table class="format"> |
| 3032 | <thead> |
| 3033 | <tr> |
| 3034 | <th>Name</th> |
| 3035 | <th>Format</th> |
| 3036 | <th>Description</th> |
| 3037 | </tr> |
| 3038 | </thead> |
| 3039 | <tbody> |
| 3040 | <tr> |
| 3041 | <td>value</td> |
| 3042 | <td>Class[]</td> |
| 3043 | <td>the array of exception types thrown</td> |
| 3044 | </tr> |
| 3045 | </tbody> |
| 3046 | </table> |