| 1 |
from (https://wiki.mobileread.com/wiki/MOBI) |
| 2 |
|
| 3 |
About |
| 4 |
----- |
| 5 |
|
| 6 |
MOBI is the format used by the MobiPocket Reader. It may have a .mobi |
| 7 |
extension or it may have a .prc extension. The extension can be changed by the |
| 8 |
user to either of the accepted forms. In either case it may be DRM protected or |
| 9 |
non-DRM. The .prc extension is used because the PalmOS doesn't support any file |
| 10 |
extensions except .prc or .pdb. Note that Mobipocket prohibits their DRM format |
| 11 |
to be used on dedicated eBook readers that support other DRM formats. |
| 12 |
|
| 13 |
|
| 14 |
Description |
| 15 |
----------- |
| 16 |
|
| 17 |
MOBI format was originally an extension of the PalmDOC format by adding |
| 18 |
certain HTML like tags to the data. Many MOBI formatted documents still use |
| 19 |
this form. However there is also a high compression version of this file format |
| 20 |
that compresses data to a larger degree in a proprietary manner. There are some |
| 21 |
third party programs that can read the eBooks in the original MOBI format but |
| 22 |
there are only a few third party program that can read the eBooks in the new |
| 23 |
compressed form. The higher compression mode is using a huffman coding scheme |
| 24 |
that has been called the Huff/cdic algorithm. |
| 25 |
|
| 26 |
From time to time features have been added to the format so new files may have |
| 27 |
problems if you try and read them with a down level reader. Currently the |
| 28 |
source files follow the guidelines in the Open eBook format. |
| 29 |
|
| 30 |
Note that AZW for the Amazon Kindle is the same format as MOBI except that it |
| 31 |
uses a slightly different DRM scheme. |
| 32 |
|
| 33 |
|
| 34 |
Format |
| 35 |
------ |
| 36 |
|
| 37 |
Like PalmDOC, the Mobipocket file format is that of a standard Palm Database |
| 38 |
Format file. The header of that format includes the name of the database |
| 39 |
(usually the book title and sometimes a portion of the authors name) which is |
| 40 |
up to 31 bytes of data. The files are identified as Creator ID of MOBI and a |
| 41 |
Type of BOOK. |
| 42 |
|
| 43 |
|
| 44 |
PalmDOC Header |
| 45 |
-------------- |
| 46 |
|
| 47 |
The first record in the Palm Database Format gives more information about the |
| 48 |
Mobipocket file. The first 16 bytes are almost identical to the first sixteen |
| 49 |
bytes of a PalmDOC format file. |
| 50 |
|
| 51 |
bytes content comments |
| 52 |
2 Compression 1 == no compression, 2 = PalmDOC compression, |
| 53 |
17480 = HUFF/CDIC compression. |
| 54 |
2 Unused Always zero |
| 55 |
4 text length Uncompressed length of the entire text of the book |
| 56 |
2 record count Number of PDB records used for the text of the book. |
| 57 |
2 record size Maximum size of each record containing text, always |
| 58 |
4096. |
| 59 |
4 Current Position Current reading position, as an offset into the |
| 60 |
uncompressed text |
| 61 |
|
| 62 |
There are two differences from a Palm DOC file. There's an additional |
| 63 |
compression type (17480), and the Current Position bytes are used for a |
| 64 |
different purpose: |
| 65 |
|
| 66 |
bytes content comments |
| 67 |
2 Encryption Type 0 == no encryption, 1 = Old Mobipocket Encryption, |
| 68 |
2 = Mobipocket Encryption. |
| 69 |
2 Unknown Usually zero |
| 70 |
|
| 71 |
The old Mobipocket Encryption scheme only allows the file to be registered |
| 72 |
with one PID, unlike the current encryption scheme that allows multiple PIDs to |
| 73 |
be used in a single file. Unless specifically mentioned, all the encryption |
| 74 |
information on this page refers to the current scheme. |
| 75 |
|
| 76 |
|
| 77 |
MOBI Header |
| 78 |
----------- |
| 79 |
|
| 80 |
Most Mobipocket file also have a MOBI header in record 0 that follows these |
| 81 |
16 bytes, and newer formats also have an EXTH header following the MOBI header, |
| 82 |
again all in record 0 of the PDB file format. |
| 83 |
|
| 84 |
The MOBI header is of variable length and is not documented. Some fields have |
| 85 |
been tentatively identified as follows: |
| 86 |
|
| 87 |
offset bytes content comments |
| 88 |
16 4 identifier The characters M O B I |
| 89 |
20 4 header length The length of the MOBI header, including |
| 90 |
the previous 4 bytes |
| 91 |
24 4 Mobi type The kind of Mobipocket file this is |
| 92 |
2 Mobipocket Book |
| 93 |
3 PalmDoc Book |
| 94 |
4 Audio |
| 95 |
257 News |
| 96 |
258 News_Feed |
| 97 |
259 News_Magazine |
| 98 |
513 PICS |
| 99 |
514 WORD |
| 100 |
515 XLS |
| 101 |
516 PPT |
| 102 |
517 TEXT |
| 103 |
518 HTML |
| 104 |
28 4 text Encoding 1252 = CP1252 (WinLatin1); 65001 = UTF-8 |
| 105 |
32 4 Unique-ID Some kind of unique ID number (random?) |
| 106 |
36 4 Generator version Potentially the version of the |
| 107 |
Mobipocket-generation tool. Always >= |
| 108 |
the value of the "format version" field |
| 109 |
and <= the version of mobigen used to |
| 110 |
produce the file. |
| 111 |
40 40 Reserved All 0xFF. In case of a dictionary, or |
| 112 |
some newer file formats, a few bytes are |
| 113 |
used from this range of 40 0xFFs |
| 114 |
80 4 First Non-book index? First record number (starting with 0) |
| 115 |
that's not the book's text |
| 116 |
84 4 Full Name Offset Offset in record 0 (not from start of |
| 117 |
file) of the full name of the book |
| 118 |
88 4 Full Name Length Length in bytes of the full name of the |
| 119 |
book |
| 120 |
92 4 Language Book language code. Low byte is main |
| 121 |
language 09= English, next byte is |
| 122 |
dialect, 08 = British, 04 = US |
| 123 |
96 4 Input Language Input language for a dictionary |
| 124 |
100 4 Output Language Output language for a dictionary |
| 125 |
104 4 Format version Potentially the version of the |
| 126 |
Mobipocket format used in this file. |
| 127 |
Always >= 1 and <= the value of the |
| 128 |
"generator version" field. |
| 129 |
108 4 First Image record First record number (starting with 0) |
| 130 |
that contains an image. Image records |
| 131 |
should be sequential. If there are |
| 132 |
no images this will be 0xffffffff. |
| 133 |
112 4 HUFF record Record containing Huff information |
| 134 |
used in HUFF/CDIC decompression. |
| 135 |
116 4 HUFF count Number of Huff records. |
| 136 |
122 4 DATP record Unknown: Records starts with DATP. |
| 137 |
124 4 DATP count Number of DATP records. |
| 138 |
128 4 EXTH flags Bitfield. if bit 6, 0x40 is set, then |
| 139 |
there's an EXTH record |
| 140 |
The following records are only present if the mobi header is long enough. |
| 141 |
132 36 ? 32 unknown bytes, if MOBI is long enough |
| 142 |
168 4 DRM Offset Offset to DRM key info in DRMed files. |
| 143 |
0xFFFFFFFF if no DRM |
| 144 |
172 4 DRM Count Number of entries in DRM info. |
| 145 |
174 4 DRM Size Number of bytes in DRM info. |
| 146 |
176 4 DRM Flags Some flags concerning the DRM info. |
| 147 |
180 6 ? |
| 148 |
186 2 Last Image record Possible value with the last image |
| 149 |
record. If there are no images in the |
| 150 |
book this will be 0xffff. |
| 151 |
188 4 ? |
| 152 |
192 4 FCIS record Unknown. Record starts with FCIS. |
| 153 |
196 4 ? |
| 154 |
200 4 FLIS record Unknown. Records starts with FLIS. |
| 155 |
204 ? ? Bytes to the end of the MOBI header, |
| 156 |
including the following if the header |
| 157 |
length >= 228. ( 244 from start of |
| 158 |
record) |
| 159 |
242 2 Extra Data Flags A set of binary flags, some of which |
| 160 |
indicate extra data at the end of each |
| 161 |
text block. This only seems to be valid |
| 162 |
for Mobipocket format version 5 and 6 |
| 163 |
(and higher?), when the header length |
| 164 |
is 228 (0xE4) or 232 (0xE8). |
| 165 |
|
| 166 |
|
| 167 |
EXTH Header |
| 168 |
----------- |
| 169 |
|
| 170 |
If the MOBI header indicates that there's an EXTH header, it follows immediately |
| 171 |
after the MOBI header. since the MOBI header is of variable length, this isn't |
| 172 |
at any fixed offset in record 0. Note that some readers will ignore any EXTH |
| 173 |
header info if the mobipocket version number specified in the MOBI header is 2 |
| 174 |
or less (perhaps 3 or less). |
| 175 |
|
| 176 |
The EXTH header is also undocumented, so some of this is guesswork. |
| 177 |
|
| 178 |
bytes content comments |
| 179 |
4 identifier the characters E X T H |
| 180 |
4 header length the length of the EXTH header, including the previous 4 bytes |
| 181 |
4 record Count The number of records in the EXTH header. the rest of the EXTH header consists of repeated EXTH records to the end of the EXTH length. |
| 182 |
EXTH record start Repeat until done. |
| 183 |
4 record type Exth Record type. Just a number identifying what's stored in the record |
| 184 |
4 record length length of EXTH record = L , including the 8 bytes in the type and length fields |
| 185 |
L-8 record data Data. |
| 186 |
EXTH record end Repeat until done. |
| 187 |
|
| 188 |
There are lots of different EXTH Records types. Ones found so far in Mobipocket |
| 189 |
files are listed here, with possible meanings. Hopefully the table will be |
| 190 |
filled in as more information comes to light. |
| 191 |
|
| 192 |
record type usual length name comments |
| 193 |
1 drm_server_id |
| 194 |
2 drm_commerce_id |
| 195 |
3 drm_ebookbase_book_id |
| 196 |
100 author |
| 197 |
101 publisher |
| 198 |
102 imprint |
| 199 |
103 description |
| 200 |
104 isbn |
| 201 |
105 subject |
| 202 |
106 publishingdate |
| 203 |
107 review |
| 204 |
108 contributor |
| 205 |
109 rights |
| 206 |
110 subjectcode |
| 207 |
111 type |
| 208 |
112 source |
| 209 |
113 asin |
| 210 |
114 versionnumber |
| 211 |
115 sample |
| 212 |
116 startreading |
| 213 |
117 3 adult Mobipocket Creator adds this if Adult only is checked; contents: "yes" |
| 214 |
118 retail price As text, e.g. "4.99" |
| 215 |
119 retail price currency As text, e.g. "USD" |
| 216 |
201 4 coveroffset Add to first image field in Mobi Header to find PDB record containing the cover image |
| 217 |
202 4 thumboffset Add to first image field in Mobi Header to find PDB record containing the thumbnail cover image |
| 218 |
203 hasfakecover |
| 219 |
204 4 Creator Software Records 204-207 are usually the same for all books from a certain source, e.g. 1-6-2-41 for Baen and 201-1-0-85 for project gutenberg, 200-1-0-85 for amazon when converted to a 32 bit integer. |
| 220 |
205 4 Creator Major Version |
| 221 |
206 4 Creator Minor Version |
| 222 |
207 4 Creator Build Number |
| 223 |
208 watermark |
| 224 |
209 tamper proof keys Used by the Kindle (and Android app) for generating book-specific PIDs. |
| 225 |
300 fontsignature |
| 226 |
401 1 clippinglimit |
| 227 |
402 publisherlimit |
| 228 |
403 403 Unknown 1 - Text to Speech disabled; 0 - Text to Speech enabled |
| 229 |
404 1 404 ttsflag |
| 230 |
501 4 cdetype PDOC - Personal Doc; |
| 231 |
EBOK - ebook; |
| 232 |
502 lastupdatetime |
| 233 |
503 updatedtitle |
| 234 |
|
| 235 |
And now, at the end of Record 0 of the PDB file format, we usually get the full |
| 236 |
file name, the offset of which is given in the MOBI header. |
| 237 |
|
| 238 |
|
| 239 |
Variable-width integers |
| 240 |
----------------------- |
| 241 |
|
| 242 |
Some parts of the Mobipocket format encode data as variable-width integers. |
| 243 |
These integers are represented big-endian with 7 bits per byte in bits 1-7. They |
| 244 |
may be either forward-encoded, in which case only the LSB has bit 8 set, or |
| 245 |
backward-encoded, in which case only the MSB has bit 8 set. For example, the |
| 246 |
number 0x11111 would be represented forward-encoded as: |
| 247 |
|
| 248 |
0x04 0x22 0x91 |
| 249 |
|
| 250 |
And backward-encoded as: |
| 251 |
|
| 252 |
0x84 0x22 0x11 |
| 253 |
|
| 254 |
|
| 255 |
Trailing entries |
| 256 |
---------------- |
| 257 |
|
| 258 |
The Extra Data Flags field of the MOBI header indicates which, if any, trailing |
| 259 |
entries are appended to the end of each text record. Each set bit in the field |
| 260 |
indicates a trailing entry. The entries appear to occur in bit-order; e.g., |
| 261 |
trailing entry 1 immediately follows the text content and entry 16 occurs at |
| 262 |
the very end of the record. The effect and exact details of most of these |
| 263 |
entries is unknown. The trailing entries indicated by bits 2-16 appear to |
| 264 |
follow a common format. That format is: |
| 265 |
|
| 266 |
<data><size> |
| 267 |
|
| 268 |
Where <size> is the size of the entire trailing entry (including the size of |
| 269 |
<size>) as a backward-encoded Mobipocket variable-width integer. |
| 270 |
|
| 271 |
Only a few bits have been identified |
| 272 |
|
| 273 |
bit Data at end of records |
| 274 |
0x0001 Multi-byte character overlaps |
| 275 |
0x0002 Some data to help with indexing |
| 276 |
0x0004 Some data about uncrossable breaks |
| 277 |
|
| 278 |
|
| 279 |
Multibyte character overlap |
| 280 |
--------------------------- |
| 281 |
|
| 282 |
When bit 1 of the Extra Data Flags field is set, each record is followed by a |
| 283 |
trailing entry containing any extra bytes necessary to complete a multibyte |
| 284 |
character which crosses the record boundary. The bytes do not participate in |
| 285 |
compression regardless which compression scheme is used for the file. However, |
| 286 |
unlike the trailing data bytes, the multibytes (including the count byte) do |
| 287 |
get included in any encryption. The overlapping bytes then re-appear as normal |
| 288 |
content at the beginning of the following record. The trailing entry ends with |
| 289 |
a byte containing a count of the overlapping bytes plus additional flags. |
| 290 |
|
| 291 |
offset bytes content comments |
| 292 |
0 0-3 N terminal bytes |
| 293 |
of a multibyte |
| 294 |
character |
| 295 |
N 1 Size & flags bits 1-2 encode N, use of bits 3-8 is unknown |
| 296 |
|
| 297 |
|
| 298 |
PalmDOC Compression |
| 299 |
------------------- |
| 300 |
|
| 301 |
PalmDOC uses LZ77 compression techniques. DOC files can contain only compressed |
| 302 |
text. The format does not allow for any text formatting. This keeps files small, |
| 303 |
in keeping with the Palm philosophy. However, extensions to the format can use |
| 304 |
tags, such as HTML or PML, to include formatting within text. These extensions |
| 305 |
to PalmDoc are not interchangeable and are the basis for most eBook Reader |
| 306 |
formats on Palm devices. |
| 307 |
|
| 308 |
LZ77 algorithms achieve compression by replacing portions of the data with |
| 309 |
references to matching data that has already passed through both encoder and |
| 310 |
decoder. A match is encoded by a pair of numbers called a length-distance pair, |
| 311 |
which is equivalent to the statement "each of the next length characters is |
| 312 |
equal to the character exactly distance characters behind it in the uncompressed |
| 313 |
stream." (The "distance" is sometimes called the "offset" instead.) |
| 314 |
|
| 315 |
In the PalmDoc format, a length-distance pair is always encoded by a two-byte |
| 316 |
sequence. Of the 16 bits that make up these two bytes, 11 bits go to encoding |
| 317 |
the distance, 3 go to encoding the length, and the remaining two are used to |
| 318 |
make sure the decoder can identify the first byte as the beginning of such a |
| 319 |
two-byte sequence. The exact algorithm needed to decode the compressed text can |
| 320 |
be found on the PalmDOC page. |
| 321 |
|
| 322 |
PalmDOC data is always divided into 4096 byte blocks and the blocks are acted |
| 323 |
upon independently. |
| 324 |
|
| 325 |
PalmDOC does have support for bookmarks. These pointers are named and refer to |
| 326 |
an offset location in a file. If the file is edited these locations may no |
| 327 |
longer refer to the correct locations. Some reading programs allow the user to |
| 328 |
enter or edit these bookmarks while others treat them as a TOC. Some reading |
| 329 |
programs may ignore them entirely. They are stored at the end of the file itself |
| 330 |
so the full file needs to be scanned when loaded to find them. |
| 331 |
|
| 332 |
|
| 333 |
Image Records |
| 334 |
------------- |
| 335 |
|
| 336 |
If the file contains images, they follow the text blocks, with each image using a |
| 337 |
single block. The 4096-byte record size in the PalmDoc header applies only to |
| 338 |
text records; image records may be larger. |
| 339 |
|
| 340 |
|
| 341 |
Magic Records |
| 342 |
------------- |
| 343 |
|
| 344 |
In some cases, MobiPocket Creator adds a 2-zero-byte record after the text |
| 345 |
records in a file. This record is not included in the "record count" of text |
| 346 |
records in the PalmDoc header, and is also not used as the "first non-book |
| 347 |
index" in the MOBI header. (If the 2-zero-byte record is present, the index of |
| 348 |
the following block is used as the "first non-book index".) |
| 349 |
|
| 350 |
MobiPocket Creator also ends files with three records: 'FLIS', 'FCIS', and |
| 351 |
'end-of-file', in that order. The 'FLIS' and 'FCIS' records do not seem to be |
| 352 |
necessary for MobiPocket Reader or the Amazon Kindle 2 to read the file. The |
| 353 |
'end-of-file' record might be necessary. |
| 354 |
|
| 355 |
|
| 356 |
FLIS Record |
| 357 |
----------- |
| 358 |
|
| 359 |
The FLIS record appears to have a fixed value. The meaning of the values is not known. |
| 360 |
|
| 361 |
offset bytes content comments |
| 362 |
0 4 identifier the characters F L I S (0x46 0x4c 0x49 0x53) |
| 363 |
4 4 ? fixed value: 8 |
| 364 |
8 2 ? fixed value: 65 |
| 365 |
10 2 ? fixed value: 0 |
| 366 |
12 4 ? fixed value: 0 |
| 367 |
16 4 ? fixed value: -1 |
| 368 |
20 2 ? fixed value: 1 |
| 369 |
22 2 ? fixed value: 3 |
| 370 |
24 4 ? fixed value: 3 |
| 371 |
28 4 ? fixed value: 1 |
| 372 |
32 4 ? fixed value: -1 |
| 373 |
|
| 374 |
|
| 375 |
FCIS Record |
| 376 |
----------- |
| 377 |
|
| 378 |
The FCIS record appears to have mostly fixed values. |
| 379 |
|
| 380 |
offset bytes content comments |
| 381 |
0 4 identifier the characters F C I S (0x46 0x43 0x49 0x53) |
| 382 |
4 4 ? fixed value: 20 |
| 383 |
8 4 ? fixed value: 16 |
| 384 |
12 4 ? fixed value: 1 |
| 385 |
16 4 ? fixed value: 0 |
| 386 |
20 4 ? text length (the same value as "text length" in the PalmDoc header) |
| 387 |
24 4 ? fixed value: 0 |
| 388 |
28 4 ? fixed value: 32 |
| 389 |
32 4 ? fixed value: 8 |
| 390 |
36 2 ? fixed value: 1 |
| 391 |
38 2 ? fixed value: 1 |
| 392 |
40 4 ? fixed value: 0 |
| 393 |
|
| 394 |
|
| 395 |
End-of-file Record |
| 396 |
------------------ |
| 397 |
|
| 398 |
The end-of-file record is a fixed 4-byte record. While the last two bytes |
| 399 |
appear to be a CRLF marker, the meaning of the first two bytes is unknown. |
| 400 |
|
| 401 |
offset bytes content comments |
| 402 |
0 1 ? fixed value: 233 (0xe9) |
| 403 |
1 1 ? fixed value: 142 (0x8e) |
| 404 |
2 1 ? fixed value: 13 (0x0d) |
| 405 |
3 1 ? fixed value: 10 (0x0a) |
| 406 |
|
| 407 |
|
| 408 |
SRCS Record |
| 409 |
----------- |
| 410 |
|
| 411 |
kindlegen creates a record whose content is a zip archive of all source files |
| 412 |
(i.e., .opf, .ncx, .htm, .jpg, ...) given to the command and puts it in the |
| 413 |
generated MOBI file. The record begins with the "SRCS" signature and is |
| 414 |
located just before the #End-of-file Record. |
| 415 |
|
| 416 |
MOBI files created with Mobipocket creator, Amazon's Personal Document Service, |
| 417 |
or Kindle Direct Publishing (former Amazon DTP) don't include SRCS record. |
| 418 |
In a past, kindlegen had an undocumented option to suppress this record, but |
| 419 |
the option was removed in 2010. |
| 420 |
|
| 421 |
offset bytes content comments |
| 422 |
0 4 identifier "SRCS" (0x53 0x52 0x43 0x53) |
| 423 |
4 4 ? fixed value(?): 0x00000010 |
| 424 |
8 4 ? fixed value(?): 0x0000002f |
| 425 |
12 4 ? fixed value(?): 0x00000001 |
| 426 |
16 zip The zip archive continues to the end of this record |
| 427 |
|
| 428 |
|
| 429 |
MBP |
| 430 |
--- |
| 431 |
|
| 432 |
This is the extension used on a side file (auxiliary) for MOBI formatted eBooks. |
| 433 |
It is used to store metadata used by the library software and also to store |
| 434 |
user entered data like bookmarks, annotations, last read position. This file is |
| 435 |
created automatically by the reader program when the eBook is first opened and |
| 436 |
has a .mbp extension. The Library management software in MobiPocket uses this |
| 437 |
file to get information displayed in the library window such as title and author |
| 438 |
so that it won't have to open the larger eBook file. |
| 439 |
|