<title>Fossil Delta Encoding Algorithm</title>
<nowiki>
<h2>Abstract</h2>
<p>A key component for the efficient storage of multiple revisions of
a file in fossil repositories is the use of delta-compression, i.e. to
store only the changes between revisions instead of the whole
file.</p>
<p>This document describes the encoding algorithm used by Fossil to
generate deltas. It is targeted at developers working on either
<a href="index.wiki">fossil</a> itself, or on tools compatible with
it. The exact format of the generated byte-sequences, while in general
not necessary to understand encoder operation, can be found in the
companion specification titled "<a href="delta_format.wiki">Fossil
Delta Format</a>".
</p>
<p>The algorithm is inspired
by <a href="http://samba.anu.edu.au/rsync/">rsync</a>.</p>
<a name="argresparam"></a><h2>1.0 Arguments, Results, and Parameters</h2>
<p>The encoder takes two byte-sequences as input, the "original", and
the "target", and returns a single byte-sequence containing the
"delta" which transforms the original into the target upon its
application.</p>
<p>Note that the data of a "byte-sequence" includes its length,
i.e. the number of bytes contained in the sequence.</p>
<p>The algorithm has one parameter named "NHASH", the size of the
"sliding window" for the "rolling hash", in bytes. These two terms are
explained in the next section. The value of this parameter has to be a
power of two for the algorithm to work. For Fossil the value of this
parameter is set to "16".</p>
<a name="operation"></a><h2>2.0 Operation</h2>
<p>The algorithm is split into three phases which generate
the <a href="delta_format.wiki#header">header</a>,
<a href="delta_format.wiki#slist">segment list</a>,
and <a href="delta_format.wiki#trailer">trailer</a> of the delta, per
its general <a href="delta_format.wiki#structure">structure</a>.</p>
<p>The two phases generating header and trailer are not covered here
as their implementation trivially follows directly from the
specification of the <a href="delta_format.wiki">delta format</a>.</p>
<p>This leaves the segment-list. Its generation is done in two phases,
a pre-processing step operating on the "original" byte-sequence,
followed by the processing of the "target" byte-sequence using the
information gathered by the first step.</p>
<a name="preprocessing"></a><h3>2.1 Preprocessing the original</h3>
<p>A major part of the processing of the "target" is to find a range
in the "original" which contains the same content as found at the
current location in the "target".</p>
<p>A naive approach to this would be to search the whole "original"
for such content. This however is very inefficient as it would search
the same parts of the "original" over and over. What is done instead
is to sample the "original" at regular intervals, compute signatures
for the sampled locations and store them in a hash table keyed by
these signatures.</p>
<p>That is what happens in this step. The following processing step
can then the compute signature for its current location and then has
to search only a narrow set of locations in the "original" for
possible matches, namely those which have the same signature.</p>
<p>In detail:</p>
<ol>
<li>The "original" is split into chunks of NHASH bytes. Note that a
partial chunk of less than NHASH bytes at the end of "original" is
ignored.
</li>
<li>The <a href="#rollhash">rolling hash</a> of each chunk is
computed.
</li>
<li>A hash table is filled, mapping from the hashes of the chunks to
the list of chunk locations having this hash.
</li>
</ol>
<a name="processing"></a><h3>2.1 Processing the target</h3>
<p>This, the main phase of the encoder, processes the target in a loop
from beginning to end. The state of the encoder is captured by two
locations, the "base" and the "slide". "base" points to the first byte
of the target for which no delta output has been generated yet, and
"slide" is the location of the window used to look in the "origin" for
commonalities. This window is NHASH bytes long.</p>
<p>Initially both "base" and "slide" point to the beginning of the
"target". In each iteration of the loop the encoder decides whether to
<ul>
<li>emit a single instruction
to <a href="delta_format.wiki#copyrange">copy a range</a>, or
</li>
<li>emit two instructions, first
to <a href="delta_format.wiki#insertlit">insert a literal</a>, then
to <a href="delta_format.wiki#copyrange">copy a range</a>, or
</li>
<li>move the window forward one byte.
</li>
</ul>
</p>
<img src="encode10.gif" align="right" hspace="10">
<p>To make this decision the encoder first computes the hash value for
the NHASH bytes in the window and then looks at all the locations in
the "origin" which have the same signature. This part uses the hash
table created by the pre-processing step to efficiently find these
locations.</p>
<p>For each of the possible candidates the encoder finds the maximal
range of bytes common to both "origin" and "target", going forward and
backward from "slide" in the "target", and the candidate location in
the "origin". This search is constrained on the side of the "target"
by the "base" (backward search), and the end of the "target" (forward
search), and on the side of the "origin" by the beginning and end of
the "origin", respectively.</p>
<p>There are input files for which the hash chains generated by the
pre-processing step can become very long, leading to long search times
and affecting the performance of the delta generator. To limit the
effect such long chains can have the actual search for candidates is
bounded, looking at most N candidates. Currently N is set to 250.</p>
<p>From the ranges for all the candidates the best (= largest) common
range is taken and it is determined how many bytes are needed to
encode the bytes between the "base" and the end of that range. If the
range extended back to the "base" then this can be done in a single
copy instruction. Otherwise, i.e if there is a gap between the "base"
and the beginning of the range then two instructions are needed, one
to insert the bytes in the gap as a literal, and a copy instruction
for the range itself. The general situation at this point can be seen
in the picture to the right.</p>
<p>If the number of bytes needed to encode both gap (if present), and
range is less than the number of bytes we are encoding the encoder
will emit the necessary instructions as described above, set "base"
and "slide" to the end of the encoded range and start the next
iteration at that point.</p>
<p>If, on the other hand, the encoder either did not find candidate
locations in the origin, or the best range coming out of the search
needed more bytes to encode the range than there were bytes in the
range, then no instructions are emitted and the window is moved one
byte forward. The "base" is left unchanged in that case.</p>
<p>The processing loop stops at one of two conditions:
<ol>
<li>The encoder decided to move the window forward, but the end of the
window reached the end of the "target".
</li>
<li>After the emission of instructions the new "base" location is
within NHASH bytes of end of the "target", i.e. there are no more than
at most NHASH bytes left.
</li>
</ol>
</p>
<p>If the processing loop left bytes unencoded, i.e. "base" not
exactly at the end of the "target", as is possible for both end
conditions, then one last insert instruction is emitted to put these
bytes into the delta.<p>
<a name="exceptions"></a><h2>3.0 Exceptions</h2>
<p>If the "original" is at most NHASH bytes long no compression of
changes is possible, and the segment-list of the delta consists of a
single literal which contains the entire "target".</p>
<p>This is actually equivalent to the second end condition of the
processing loop described in the previous section, just checked before
actually entering the loop.</p>
<a name="rollhash"></a><h2>4.0 The rolling hash</h2>
<p>The rolling hash described below and used to compute content
signatures was chosen not only for good hashing properties, but also
to enable the easy (incremental) recalculation of its value for a
sliding window, i.e. where the oldest byte is removed from the window
and a new byte is shifted in.<p>
<a name="rhdef"></a><h3>4.1 Definition</h3>
<p>Assuming an array Z of NHASH bytes (indexing starting at 0) the
hash V is computed via</p>
<p align=center><table><tr><td>
<p><img src="encode1.gif" align="center"></p>
<p><img src="encode2.gif" align="center"></p>
<p><img src="encode3.gif" align="center"></p>
</td></tr></table></p>
where A and B are unsigned 16-bit integers (hence the <u>mod</u>), and
V is a 32-bit unsigned integer with B as MSB, A as LSB.
<a name="rhincr"></a><h3>4.2 Incremental recalculation</h3>
<p>Assuming an array Z of NHASH bytes (indexing starting at 0) with
hash V (and components A and B), the dropped
byte <img src="encode4.gif" align="center">, and the new byte
<img src="encode5.gif" align="center"> , the new hash can
be computed incrementally via: </p>
<p align=center><table><tr><td>
<p><img src="encode6.gif" align="center"></p>
<p><img src="encode7.gif" align="center"></p>
<p><img src="encode8.gif" align="center"></p>
</td></tr></table></p>
<p>For A, the regular sum, it can be seen easily that this the correct
way recomputing that component.</p>
<p>For B, the weighted sum, note first that <img src="encode4.gif"
align="center"> has the weight NHASH in the sum, so that is what has
to be removed. Then adding in <img src="encode9.gif" align="center">
adds one weight factor to all the other values of Z, and at last adds
in <img src="encode5.gif" align="center"> with weight 1, also
generating the correct new sum</p>