AuroraMiddleware/brotli
1
0
Fork 0
mirror of https://github.com/google/brotli.git synced 2024-11-09 21:50:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Some wording changes to Section 2 of the spec.

Browse Source
This commit is contained in:
Zoltan Szabadka 2015-04-07 17:23:37 +02:00
parent f9bb85eb92
commit 92b551734a
2 changed files with 81 additions and 76 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
docs/draft-alakuijala-brotli-02.nroff
View File

@ -235,10 +235,10 @@ point during decoding the stream.
Each meta-block is compressed using a combination of the LZ77
algorithm (Lempel-Ziv 1977, [LZ77]) and Huffman
coding. The Huffman trees for each block are independent of those for
previous or subsequent blocks; the LZ77 algorithm may use a
reference to a duplicated string occurring in a previous meta-block,
up to sliding window size input bytes before.
coding. The Huffman trees for each meta-block are independent of
those for previous or subsequent meta-blocks; the LZ77 algorithm may
use a reference to a duplicated string occurring in a previous
meta-block, up to sliding window size input bytes before.
Each meta-block consists of two parts: a meta-block header that
describes the representation of the compressed data part, and a
@ -250,15 +250,15 @@ represented as a pair <length, backward distance>.
Each command in the compressed data is represented using three kinds
of Huffman codes: one kind of code tree for the literal sequence
lengths (also referred to as literal insertion lengths) and backward
lengths (also referred to as literal insertion lengths) and backward
copy lengths (that is, a single code word represents two lengths,
one of the literal sequence and one of the backward copy), a separate
kind of code tree for literals, and a third kind of code tree for
distances. The code trees for each meta-block appear in a compact
form just before the compressed data in the meta-block header.
The sequence of each type of value in the representation of a command
(insert-and-copy lengths, literals and distances) within a meta-
The sequence of each category of value in the representation of a
command (insert-and-copy lengths, literals and distances) within a meta-
block is further divided into blocks. In the "brotli" format, blocks
are not contiguous chunks of compressed data, but rather the pieces
of compressed data belonging to a block are interleaved with pieces
@ -268,19 +268,24 @@ of literal blocks and a series of distance blocks. These are also
called the three block categories: a meta-block has a series of
blocks for each block category. Note that the physical structure of
the meta-block is a series of commands, while the three series of
blocks is the logical structure. Consider the following example:
blocks is the logical structure.
A block is defined by a type (0-255) and a length. The length is the
amount of huffman symbols of its category, the type dictates which
huffman code is used for these symbols. Consider the following
example:
(IaC0, L0, L1, L2, D0)(IaC1, D1)(IaC2, L3, L4, D2)(IaC3, L5, D3)
The meta-block here has 4 commands, and each three types of symbols
within these commands can be rearranged for example into the
following logical block structure:
The meta-block here has 4 commands, and each of the three categories of
symbols within these commands are part of a logical block structure,
for example the following:
[IaC0, IaC1][IaC2, IaC3] <-- block types 0 and 1
[IaC0, IaC1][IaC2, IaC3] <-- insert-and-copy: block types 0 and 1
[L0, L1][L2, L3, L4][L5] <-- block types 0, 1, and 0
[L0, L1][L2, L3, L4][L5] <-- literals: block types 0, 1, and 0
[D0][D1, D2, D3] <-- block types 0 and 1
[D0][D1, D2, D3] <-- distances: block types 0 and 1
The subsequent blocks within each block category must have different
block types, but blocks further away in the block sequence can have

124
docs/draft-alakuijala-brotli-02.txt
View File

@ -77,7 +77,7 @@ Table of Contents
3.4. Simple Huffman codes . . . . . . . . . . . . . . . . . . 10
3.5. Complex Huffman codes . . . . . . . . . . . . . . . . . . 11
4. Encoding of distances . . . . . . . . . . . . . . . . . . . . 13
5. Encoding of literal insertion lengths and copy lengths . . . . 14
5. Encoding of literal insertion lengths and copy lengths . . . . 15
6. Encoding of block switch commands . . . . . . . . . . . . . . 17
7. Context modeling . . . . . . . . . . . . . . . . . . . . . . . 18
7.1. Context modes and context ID lookup for literals . . . . 18
@ -264,8 +264,8 @@ Internet-Draft Brotli October 2014
Each meta-block is compressed using a combination of the LZ77
algorithm (Lempel-Ziv 1977, [LZ77]) and Huffman coding. The Huffman
trees for each block are independent of those for previous or
subsequent blocks; the LZ77 algorithm may use a reference to a
trees for each meta-block are independent of those for previous or
subsequent meta-blocks; the LZ77 algorithm may use a reference to a
duplicated string occurring in a previous meta-block, up to sliding
window size input bytes before.
@ -293,30 +293,35 @@ Internet-Draft Brotli October 2014
distances. The code trees for each meta-block appear in a compact
form just before the compressed data in the meta-block header.
The sequence of each type of value in the representation of a command
(insert-and-copy lengths, literals and distances) within a meta-
block is further divided into blocks. In the "brotli" format, blocks
are not contiguous chunks of compressed data, but rather the pieces
of compressed data belonging to a block are interleaved with pieces
of data belonging to other blocks. Each meta-block can be logically
decomposed into a series of insert-and-copy length blocks, a series
of literal blocks and a series of distance blocks. These are also
called the three block categories: a meta-block has a series of
The sequence of each category of value in the representation of a
command (insert-and-copy lengths, literals and distances) within a
meta- block is further divided into blocks. In the "brotli" format,
blocks are not contiguous chunks of compressed data, but rather the
pieces of compressed data belonging to a block are interleaved with
pieces of data belonging to other blocks. Each meta-block can be
logically decomposed into a series of insert-and-copy length blocks,
a series of literal blocks and a series of distance blocks. These are
also called the three block categories: a meta-block has a series of
blocks for each block category. Note that the physical structure of
the meta-block is a series of commands, while the three series of
blocks is the logical structure. Consider the following example:
blocks is the logical structure.
A block is defined by a type (0-255) and a length. The length is the
amount of huffman symbols of its category, the type dictates which
huffman code is used for these symbols. Consider the following
example:
(IaC0, L0, L1, L2, D0)(IaC1, D1)(IaC2, L3, L4, D2)(IaC3, L5, D3)
The meta-block here has 4 commands, and each three types of symbols
within these commands can be rearranged for example into the
following logical block structure:
The meta-block here has 4 commands, and each of the three categories
of symbols within these commands are part of a logical block
structure, for example the following:
[IaC0, IaC1][IaC2, IaC3] <-- block types 0 and 1
[IaC0, IaC1][IaC2, IaC3] <-- insert-and-copy: block types 0 and 1
[L0, L1][L2, L3, L4][L5] <-- block types 0, 1, and 0
[L0, L1][L2, L3, L4][L5] <-- literals: block types 0, 1, and 0
[D0][D1, D2, D3] <-- block types 0 and 1
[D0][D1, D2, D3] <-- distances: block types 0 and 1
The subsequent blocks within each block category must have different
block types, but blocks further away in the block sequence can have
@ -327,11 +332,6 @@ Internet-Draft Brotli October 2014
where a block-switch command is a pair <block type, block length>.
The block-switch commands are represented in the compressed data
before the start of each new block using a Huffman code tree for
block types and a separate Huffman code tree for block lengths for
each block category. In the above example the physical layout of the
meta-block is the following:
IaC0 L0 L1 LBlockSwitch(1, 3) L2 D0 IaC1 DBlockSwitch(1, 1) D1
@ -340,6 +340,11 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 6]
Internet-Draft Brotli October 2014
block types and a separate Huffman code tree for block lengths for
each block category. In the above example the physical layout of the
meta-block is the following:
IaC0 L0 L1 LBlockSwitch(1, 3) L2 D0 IaC1 DBlockSwitch(1, 1) D1
IaCBlockSwitch(1, 2) IaC2 L3 L4 D2 IaC3 LBlockSwitch(0, 1) D3
Note that the block switch commands for the first blocks are not part
@ -383,11 +388,6 @@ Internet-Draft Brotli October 2014
We define a prefix code in terms of a binary tree in which the two
edges descending from each non-leaf node are labeled 0 and 1 and in
which the leaf nodes correspond one-for-one with (are labeled with)
the symbols of the alphabet; then the code for a symbol is the
sequence of 0's and 1's on the edges leading from the root to the
leaf labeled with that symbol. For example:
@ -396,6 +396,11 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 7]
Internet-Draft Brotli October 2014
which the leaf nodes correspond one-for-one with (are labeled with)
the symbols of the alphabet; then the code for a symbol is the
sequence of 0's and 1's on the edges leading from the root to the
leaf labeled with that symbol. For example:
/\ Symbol Code
0 1 ------ ----
/ \ A 00
@ -437,12 +442,7 @@ Internet-Draft Brotli October 2014
We could recode the example above to follow this rule as follows,
assuming that the order of the alphabet is ABCD:
Symbol Code
------ ----
A 10
B 0
C 110
D 111
@ -452,6 +452,13 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 8]
Internet-Draft Brotli October 2014
Symbol Code
------ ----
A 10
B 0
C 110
D 111
I.e., 0 precedes 10 which precedes 11x, and 110 and 111 are
lexicographically consecutive.
@ -493,13 +500,6 @@ Internet-Draft Brotli October 2014
Example:
Consider the alphabet ABCDEFGH, with bit lengths (3, 3, 3, 3, 3, 2,
4, 4). After step 1, we have:
@ -508,6 +508,9 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 9]
Internet-Draft Brotli October 2014
Consider the alphabet ABCDEFGH, with bit lengths (3, 3, 3, 3, 3, 2,
4, 4). After step 1, we have:
N bl_count[N]
- -----------
2 1
@ -553,9 +556,6 @@ Internet-Draft Brotli October 2014
value is 1, then a simple Huffman code follows. Otherwise the value
indicates the number of leading zeros.
A simple Huffman code can have only up to four symbols with non- zero
code length. The format of the simple Huffman code is as follows:
@ -564,6 +564,9 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 10]
Internet-Draft Brotli October 2014
A simple Huffman code can have only up to four symbols with non- zero
code length. The format of the simple Huffman code is as follows:
2 bits: value of 1 indicates a simple Huffman code
2 bits: NSYM - 1, where NSYM = # of symbols with non-zero
code length
@ -609,9 +612,6 @@ Internet-Draft Brotli October 2014
follows:
0 - 15: Represent code lengths of 0 - 15
16: Copy the previous non-zero code length 3 - 6 times
The next 2 bits indicate repeat length
(0 = 3, ... , 3 = 6)
@ -620,6 +620,9 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 11]
Internet-Draft Brotli October 2014
16: Copy the previous non-zero code length 3 - 6 times
The next 2 bits indicate repeat length
(0 = 3, ... , 3 = 6)
If this is the first code length, or all previous
code lengths are zero, a code length of 8 is
repeated 3 - 6 times
@ -665,9 +668,6 @@ Internet-Draft Brotli October 2014
8, 9, 10, 11, 12, 13, 14, 15
The code lengths of code length symbols are between 0 and
5 and they are represented with 2 - 5 bits according to
the static Huffman code above. A code length of 0 means
the corresponding code length symbol is not used.
@ -676,6 +676,10 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 12]
Internet-Draft Brotli October 2014
5 and they are represented with 2 - 5 bits according to
the static Huffman code above. A code length of 0 means
the corresponding code length symbol is not used.
If HSKIP is 2 or 3, a respective number of leading code
lengths are implicit zeros and are not present in the
code lengths sequence above. If there are at least two
@ -720,10 +724,6 @@ Internet-Draft Brotli October 2014
0: last distance
1: second last distance
2: third last distance
3: fourth last distance
4: last distance - 1
5: last distance + 1
@ -732,6 +732,10 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 13]
Internet-Draft Brotli October 2014
2: third last distance
3: fourth last distance
4: last distance - 1
5: last distance + 1
6: last distance - 2
7: last distance + 2
8: last distance - 3
@ -777,10 +781,6 @@ Internet-Draft Brotli October 2014
offset = ((2 + (hcode & 1)) << ndistbits) - 4;
distance = ((offset + dextra) << NPOSTFIX) + lcode + NDIRECT + 1
5. Encoding of literal insertion lengths and copy lengths
As described in Section 2, the literal insertion lengths and backward
Alakuijala & Szabadka Expires April 27, 2015 [Page 14]
@ -788,6 +788,9 @@ Alakuijala & Szabadka Expires April 27, 2015 [Page 14]
Internet-Draft Brotli October 2014
5. Encoding of literal insertion lengths and copy lengths
As described in Section 2, the literal insertion lengths and backward
copy lengths are encoded using a single Huffman code. This section
provides the details to this encoding.
@ -836,9 +839,6 @@ Internet-Draft Brotli October 2014
Alakuijala & Szabadka Expires April 27, 2015 [Page 15]
Internet-Draft Brotli October 2014