Google Git
Sign in
chromium / external / gitster / git-htmldocs / 361c1335bb152db70cab1086e3f8505e63b60354 / . / git-diff-tree.html
blob: b80da2dbeef240a6fb12f7cb8292edf49a6a549f [file] [log] [blame]
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<meta name="generator" content="AsciiDoc 7.0.2" />
<style type="text/css">
/* Debug borders */
p, li, dt, dd, div, pre, h1, h2, h3, h4, h5, h6 {
/*
border: 1px solid red;
*/
}
body {
margin: 1em 5% 1em 5%;
}
a { color: blue; }
a:visited { color: fuchsia; }
em {
font-style: italic;
}
strong {
font-weight: bold;
}
tt {
color: navy;
}
h1, h2, h3, h4, h5, h6 {
color: #527bbd;
font-family: sans-serif;
margin-top: 1.2em;
margin-bottom: 0.5em;
line-height: 1.3;
}
h1 {
border-bottom: 2px solid silver;
}
h2 {
border-bottom: 2px solid silver;
padding-top: 0.5em;
}
div.sectionbody {
font-family: serif;
margin-left: 0;
}
hr {
border: 1px solid silver;
}
p {
margin-top: 0.5em;
margin-bottom: 0.5em;
}
pre {
padding: 0;
margin: 0;
}
span#author {
color: #527bbd;
font-family: sans-serif;
font-weight: bold;
font-size: 1.2em;
}
span#email {
}
span#revision {
font-family: sans-serif;
}
div#footer {
font-family: sans-serif;
font-size: small;
border-top: 2px solid silver;
padding-top: 0.5em;
margin-top: 4.0em;
}
div#footer-text {
float: left;
padding-bottom: 0.5em;
}
div#footer-badges {
float: right;
padding-bottom: 0.5em;
}
div#preamble,
div.tableblock, div.imageblock, div.exampleblock, div.verseblock,
div.quoteblock, div.literalblock, div.listingblock, div.sidebarblock,
div.admonitionblock {
margin-right: 10%;
margin-top: 1.5em;
margin-bottom: 1.5em;
}
div.admonitionblock {
margin-top: 2.5em;
margin-bottom: 2.5em;
}
div.content { /* Block element content. */
padding: 0;
}
/* Block element titles. */
div.title, caption.title {
font-family: sans-serif;
font-weight: bold;
text-align: left;
margin-top: 1.0em;
margin-bottom: 0.5em;
}
div.title + * {
margin-top: 0;
}
td div.title:first-child {
margin-top: 0.0em;
}
div.content div.title:first-child {
margin-top: 0.0em;
}
div.content + div.title {
margin-top: 0.0em;
}
div.sidebarblock > div.content {
background: #ffffee;
border: 1px solid silver;
padding: 0.5em;
}
div.listingblock > div.content {
border: 1px solid silver;
background: #f4f4f4;
padding: 0.5em;
}
div.quoteblock > div.content {
padding-left: 2.0em;
}
div.quoteblock .attribution {
text-align: right;
}
div.admonitionblock .icon {
vertical-align: top;
font-size: 1.1em;
font-weight: bold;
text-decoration: underline;
color: #527bbd;
padding-right: 0.5em;
}
div.admonitionblock td.content {
padding-left: 0.5em;
border-left: 2px solid silver;
}
div.exampleblock > div.content {
border-left: 2px solid silver;
padding: 0.5em;
}
div.verseblock div.content {
white-space: pre;
}
div.imageblock div.content { padding-left: 0; }
div.imageblock img { border: 1px solid silver; }
span.image img { border-style: none; }
dl {
margin-top: 0.8em;
margin-bottom: 0.8em;
}
dt {
margin-top: 0.5em;
margin-bottom: 0;
font-style: italic;
}
dd > *:first-child {
margin-top: 0;
}
ul, ol {
list-style-position: outside;
}
ol.olist2 {
list-style-type: lower-alpha;
}
div.tableblock > table {
border-color: #527bbd;
border-width: 3px;
}
thead {
font-family: sans-serif;
font-weight: bold;
}
tfoot {
font-weight: bold;
}
div.hlist {
margin-top: 0.8em;
margin-bottom: 0.8em;
}
td.hlist1 {
vertical-align: top;
font-style: italic;
padding-right: 0.8em;
}
td.hlist2 {
vertical-align: top;
}
@media print {
div#footer-badges { display: none; }
}
include::./stylesheets/xhtml11-manpage.css[]
/* Workarounds for IE6's broken and incomplete CSS2. */
div.sidebar-content {
background: #ffffee;
border: 1px solid silver;
padding: 0.5em;
}
div.sidebar-title, div.image-title {
font-family: sans-serif;
font-weight: bold;
margin-top: 0.0em;
margin-bottom: 0.5em;
}
div.listingblock div.content {
border: 1px solid silver;
background: #f4f4f4;
padding: 0.5em;
}
div.quoteblock-content {
padding-left: 2.0em;
}
div.exampleblock-content {
border-left: 2px solid silver;
padding-left: 0.5em;
}
</style>
<title>git-diff-tree(1)</title>
</head>
<body>
<div id="header">
<h1>
git-diff-tree(1) Manual Page
</h1>
<h2>NAME</h2>
<div class="sectionbody">
<p>git-diff-tree -
Compares the content and mode of blobs found via two tree objects
</p>
</div>
</div>
<h2>SYNOPSIS</h2>
<div class="sectionbody">
<div class="verseblock">
<div class="content"><em>git-diff-tree</em> [--stdin] [-m] [-s] [-v] [--no-commit-id] [--pretty]
[-t] [-r] [-c | --cc] [--root] [&lt;common diff options&gt;]
&lt;tree-ish&gt; [&lt;tree-ish&gt;] [&lt;path&gt;&#8230;]</div></div>
</div>
<h2>DESCRIPTION</h2>
<div class="sectionbody">
<p>Compares the content and mode of the blobs found via two tree objects.</p>
<p>If there is only one &lt;tree-ish&gt; given, the commit is compared with its parents
(see --stdin below).</p>
<p>Note that "git-diff-tree" can use the tree encapsulated in a commit object.</p>
</div>
<h2>OPTIONS</h2>
<div class="sectionbody">
<dl>
<dt>
-p
</dt>
<dd>
<p>
Generate patch (see section on generating patches)
</p>
</dd>
<dt>
-u
</dt>
<dd>
<p>
Synonym for "-p".
</p>
</dd>
<dt>
-U&lt;n&gt;
</dt>
<dd>
<p>
Shorthand for "--unified=&lt;n&gt;".
</p>
</dd>
<dt>
--unified=&lt;n&gt;
</dt>
<dd>
<p>
Generate diffs with &lt;n&gt; lines of context instead of
the usual three. Implies "-p".
</p>
</dd>
<dt>
--raw
</dt>
<dd>
<p>
Generate the raw format.
</p>
</dd>
<dt>
--patch-with-raw
</dt>
<dd>
<p>
Synonym for "-p --raw".
</p>
</dd>
<dt>
--stat[=width[,name-width]]
</dt>
<dd>
<p>
Generate a diffstat. You can override the default
output width for 80-column terminal by "--stat=width".
The width of the filename part can be controlled by
giving another width to it separated by a comma.
</p>
</dd>
<dt>
--numstat
</dt>
<dd>
<p>
Similar to --stat, but shows number of added and
deleted lines in decimal notation and pathname without
abbreviation, to make it more machine friendly. For
binary files, outputs two <tt>-</tt> instead of saying
<tt>0 0</tt>.
</p>
</dd>
<dt>
--shortstat
</dt>
<dd>
<p>
Output only the last line of the --stat format containing total
number of modified files, as well as number of added and deleted
lines.
</p>
</dd>
<dt>
--summary
</dt>
<dd>
<p>
Output a condensed summary of extended header information
such as creations, renames and mode changes.
</p>
</dd>
<dt>
--patch-with-stat
</dt>
<dd>
<p>
Synonym for "-p --stat".
</p>
</dd>
<dt>
-z
</dt>
<dd>
<p>
NUL-line termination on output. This affects the --raw
output field terminator. Also output from commands such
as "git-log" will be delimited with NUL between commits.
</p>
</dd>
<dt>
--name-only
</dt>
<dd>
<p>
Show only names of changed files.
</p>
</dd>
<dt>
--name-status
</dt>
<dd>
<p>
Show only names and status of changed files.
</p>
</dd>
<dt>
--color
</dt>
<dd>
<p>
Show colored diff.
</p>
</dd>
<dt>
--no-color
</dt>
<dd>
<p>
Turn off colored diff, even when the configuration file
gives the default to color output.
</p>
</dd>
<dt>
--color-words
</dt>
<dd>
<p>
Show colored word diff, i.e. color words which have changed.
</p>
</dd>
<dt>
--no-renames
</dt>
<dd>
<p>
Turn off rename detection, even when the configuration
file gives the default to do so.
</p>
</dd>
<dt>
--check
</dt>
<dd>
<p>
Warn if changes introduce trailing whitespace
or an indent that uses a space before a tab.
</p>
</dd>
<dt>
--full-index
</dt>
<dd>
<p>
Instead of the first handful characters, show full
object name of pre- and post-image blob on the "index"
line when generating a patch format output.
</p>
</dd>
<dt>
--binary
</dt>
<dd>
<p>
In addition to --full-index, output "binary diff" that
can be applied with "git apply".
</p>
</dd>
<dt>
--abbrev[=&lt;n&gt;]
</dt>
<dd>
<p>
Instead of showing the full 40-byte hexadecimal object
name in diff-raw format output and diff-tree header
lines, show only handful hexdigits prefix. This is
independent of --full-index option above, which controls
the diff-patch output format. Non default number of
digits can be specified with --abbrev=&lt;n&gt;.
</p>
</dd>
<dt>
-B
</dt>
<dd>
<p>
Break complete rewrite changes into pairs of delete and create.
</p>
</dd>
<dt>
-M
</dt>
<dd>
<p>
Detect renames.
</p>
</dd>
<dt>
-C
</dt>
<dd>
<p>
Detect copies as well as renames. See also <tt>--find-copies-harder</tt>.
</p>
</dd>
<dt>
--diff-filter=[ACDMRTUXB*]
</dt>
<dd>
<p>
Select only files that are Added (<tt>A</tt>), Copied (<tt>C</tt>),
Deleted (<tt>D</tt>), Modified (<tt>M</tt>), Renamed (<tt>R</tt>), have their
type (mode) changed (<tt>T</tt>), are Unmerged (<tt>U</tt>), are
Unknown (<tt>X</tt>), or have had their pairing Broken (<tt>B</tt>).
Any combination of the filter characters may be used.
When <tt>*</tt> (All-or-none) is added to the combination, all
paths are selected if there is any file that matches
other criteria in the comparison; if there is no file
that matches other criteria, nothing is selected.
</p>
</dd>
<dt>
--find-copies-harder
</dt>
<dd>
<p>
For performance reasons, by default, <tt>-C</tt> option finds copies only
if the original file of the copy was modified in the same
changeset. This flag makes the command
inspect unmodified files as candidates for the source of
copy. This is a very expensive operation for large
projects, so use it with caution. Giving more than one
<tt>-C</tt> option has the same effect.
</p>
</dd>
<dt>
-l&lt;num&gt;
</dt>
<dd>
<p>
-M and -C options require O(n^2) processing time where n
is the number of potential rename/copy targets. This
option prevents rename/copy detection from running if
the number of rename/copy targets exceeds the specified
number.
</p>
</dd>
<dt>
-S&lt;string&gt;
</dt>
<dd>
<p>
Look for differences that contain the change in &lt;string&gt;.
</p>
</dd>
<dt>
--pickaxe-all
</dt>
<dd>
<p>
When -S finds a change, show all the changes in that
changeset, not just the files that contain the change
in &lt;string&gt;.
</p>
</dd>
<dt>
--pickaxe-regex
</dt>
<dd>
<p>
Make the &lt;string&gt; not a plain string but an extended POSIX
regex to match.
</p>
</dd>
<dt>
-O&lt;orderfile&gt;
</dt>
<dd>
<p>
Output the patch in the order specified in the
&lt;orderfile&gt;, which has one shell glob pattern per line.
</p>
</dd>
<dt>
-R
</dt>
<dd>
<p>
Swap two inputs; that is, show differences from index or
on-disk file to tree contents.
</p>
</dd>
<dt>
--text
</dt>
<dd>
<p>
Treat all files as text.
</p>
</dd>
<dt>
-a
</dt>
<dd>
<p>
Shorthand for "--text".
</p>
</dd>
<dt>
--ignore-space-at-eol
</dt>
<dd>
<p>
Ignore changes in white spaces at EOL.
</p>
</dd>
<dt>
--ignore-space-change
</dt>
<dd>
<p>
Ignore changes in amount of white space. This ignores white
space at line end, and consider all other sequences of one or
more white space characters to be equivalent.
</p>
</dd>
<dt>
-b
</dt>
<dd>
<p>
Shorthand for "--ignore-space-change".
</p>
</dd>
<dt>
--ignore-all-space
</dt>
<dd>
<p>
Ignore white space when comparing lines. This ignores
difference even if one line has white space where the other
line has none.
</p>
</dd>
<dt>
-w
</dt>
<dd>
<p>
Shorthand for "--ignore-all-space".
</p>
</dd>
<dt>
--exit-code
</dt>
<dd>
<p>
Make the program exit with codes similar to diff(1).
That is, it exits with 1 if there were differences and
0 means no differences.
</p>
</dd>
<dt>
--quiet
</dt>
<dd>
<p>
Disable all output of the program. Implies --exit-code.
</p>
</dd>
<dt>
--ext-diff
</dt>
<dd>
<p>
Allow an external diff helper to be executed. If you set an
external diff driver with <a href="gitattributes.html">gitattributes(5)</a>, you need
to use this option with <a href="git-log.html">git-log(1)</a> and friends.
</p>
</dd>
<dt>
--no-ext-diff
</dt>
<dd>
<p>
Disallow external diff drivers.
</p>
</dd>
</dl>
<p>For more detailed explanation on these common options, see also
<a href="diffcore.html">diffcore documentation</a>.</p>
<dl>
<dt>
&lt;tree-ish&gt;
</dt>
<dd>
<p>
The id of a tree object.
</p>
</dd>
<dt>
&lt;path&gt;&#8230;
</dt>
<dd>
<p>
If provided, the results are limited to a subset of files
matching one of these prefix strings.
i.e., file matches <tt>/^&lt;pattern1&gt;|&lt;pattern2&gt;|&#8230;/</tt>
Note that this parameter does not provide any wildcard or regexp
features.
</p>
</dd>
<dt>
-r
</dt>
<dd>
<p>
recurse into sub-trees
</p>
</dd>
<dt>
-t
</dt>
<dd>
<p>
show tree entry itself as well as subtrees. Implies -r.
</p>
</dd>
<dt>
--root
</dt>
<dd>
<p>
When <em>--root</em> is specified the initial commit will be showed as a big
creation event. This is equivalent to a diff against the NULL tree.
</p>
</dd>
<dt>
--stdin
</dt>
<dd>
<p>
When <em>--stdin</em> is specified, the command does not take
&lt;tree-ish&gt; arguments from the command line. Instead, it
reads either one &lt;commit&gt; or a pair of &lt;tree-ish&gt;
separated with a single space from its standard input.
</p>
<p>When a single commit is given on one line of such input, it compares
the commit with its parents. The following flags further affects its
behavior. This does not apply to the case where two &lt;tree-ish&gt;
separated with a single space are given.</p>
</dd>
<dt>
-m
</dt>
<dd>
<p>
By default, "git-diff-tree --stdin" does not show
differences for merge commits. With this flag, it shows
differences to that commit from all of its parents. See
also <em>-c</em>.
</p>
</dd>
<dt>
-s
</dt>
<dd>
<p>
By default, "git-diff-tree --stdin" shows differences,
either in machine-readable form (without <em>-p</em>) or in patch
form (with <em>-p</em>). This output can be suppressed. It is
only useful with <em>-v</em> flag.
</p>
</dd>
<dt>
-v
</dt>
<dd>
<p>
This flag causes "git-diff-tree --stdin" to also show
the commit message before the differences.
</p>
</dd>
<dt>
--pretty[=<em>&lt;format&gt;</em>]
</dt>
<dd>
<p>
Pretty-print the contents of the commit logs in a given format,
where <em>&lt;format&gt;</em> can be one of <em>oneline</em>, <em>short</em>, <em>medium</em>,
<em>full</em>, <em>fuller</em>, <em>email</em>, <em>raw</em> and <em>format:&lt;string&gt;</em>.
When omitted, the format defaults to <em>medium</em>.
</p>
</dd>
<dt>
--abbrev-commit
</dt>
<dd>
<p>
Instead of showing the full 40-byte hexadecimal commit object
name, show only handful hexdigits prefix. Non default number of
digits can be specified with "--abbrev=&lt;n&gt;" (which also modifies
diff output, if it is displayed).
</p>
<p>This should make "--pretty=oneline" a whole lot more readable for
people using 80-column terminals.</p>
</dd>
<dt>
--encoding[=&lt;encoding&gt;]
</dt>
<dd>
<p>
The commit objects record the encoding used for the log message
in their encoding header; this option can be used to tell the
command to re-code the commit log message in the encoding
preferred by the user. For non plumbing commands this
defaults to UTF-8.
</p>
</dd>
<dt>
--no-commit-id
</dt>
<dd>
<p>
git-diff-tree outputs a line with the commit ID when
applicable. This flag suppressed the commit ID output.
</p>
</dd>
<dt>
-c
</dt>
<dd>
<p>
This flag changes the way a merge commit is displayed
(which means it is useful only when the command is given
one &lt;tree-ish&gt;, or <em>--stdin</em>). It shows the differences
from each of the parents to the merge result simultaneously
instead of showing pairwise diff between a parent and the
result one at a time (which is what the <em>-m</em> option does).
Furthermore, it lists only files which were modified
from all parents.
</p>
</dd>
<dt>
--cc
</dt>
<dd>
<p>
This flag changes the way a merge commit patch is displayed,
in a similar way to the <em>-c</em> option. It implies the <em>-c</em>
and <em>-p</em> options and further compresses the patch output
by omitting hunks that show differences from only one
parent, or show the same change from all but one parent
for an Octopus merge. When this optimization makes all
hunks disappear, the commit itself and the commit log
message is not shown, just like in any other "empty diff" case.
</p>
</dd>
<dt>
--always
</dt>
<dd>
<p>
Show the commit itself and the commit log message even
if the diff itself is empty.
</p>
</dd>
</dl>
</div>
<h2>PRETTY FORMATS</h2>
<div class="sectionbody">
<p>If the commit is a merge, and if the pretty-format
is not <em>oneline</em>, <em>email</em> or <em>raw</em>, an additional line is
inserted before the <em>Author:</em> line. This line begins with
"Merge: " and the sha1s of ancestral commits are printed,
separated by spaces. Note that the listed commits may not
necessarily be the list of the <strong>direct</strong> parent commits if you
have limited your view of history: for example, if you are
only interested in changes related to a certain directory or
file.</p>
<p>Here are some additional details for each format:</p>
<ul>
<li>
<p>
<em>oneline</em>
</p>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;sha1&gt; &lt;title line&gt;</tt></pre>
</div></div>
<p>This is designed to be as compact as possible.</p>
</li>
<li>
<p>
<em>short</em>
</p>
<div class="literalblock">
<div class="content">
<pre><tt>commit &lt;sha1&gt;
Author: &lt;author&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;title line&gt;</tt></pre>
</div></div>
</li>
<li>
<p>
<em>medium</em>
</p>
<div class="literalblock">
<div class="content">
<pre><tt>commit &lt;sha1&gt;
Author: &lt;author&gt;
Date: &lt;date&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;title line&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;full commit message&gt;</tt></pre>
</div></div>
</li>
<li>
<p>
<em>full</em>
</p>
<div class="literalblock">
<div class="content">
<pre><tt>commit &lt;sha1&gt;
Author: &lt;author&gt;
Commit: &lt;committer&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;title line&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;full commit message&gt;</tt></pre>
</div></div>
</li>
<li>
<p>
<em>fuller</em>
</p>
<div class="literalblock">
<div class="content">
<pre><tt>commit &lt;sha1&gt;
Author: &lt;author&gt;
AuthorDate: &lt;date &amp; time&gt;
Commit: &lt;committer&gt;
CommitDate: &lt;date &amp; time&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;title line&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;full commit message&gt;</tt></pre>
</div></div>
</li>
<li>
<p>
<em>email</em>
</p>
<div class="literalblock">
<div class="content">
<pre><tt>From &lt;sha1&gt; &lt;date&gt;
From: &lt;author&gt;
Date: &lt;date &amp; time&gt;
Subject: [PATCH] &lt;title line&gt;</tt></pre>
</div></div>
<div class="literalblock">
<div class="content">
<pre><tt>&lt;full commit message&gt;</tt></pre>
</div></div>
</li>
<li>
<p>
<em>raw</em>
</p>
<p>The <em>raw</em> format shows the entire commit exactly as
stored in the commit object. Notably, the SHA1s are
displayed in full, regardless of whether --abbrev or
--no-abbrev are used, and <em>parents</em> information show the
true parent commits, without taking grafts nor history
simplification into account.</p>
</li>
<li>
<p>
<em>format:</em>
</p>
<p>The <em>format:</em> format allows you to specify which information
you want to show. It works a little bit like printf format,
with the notable exception that you get a newline with <em>%n</em>
instead of <em>\n</em>.</p>
<p>E.g, <em>format:"The author of %h was %an, %ar%nThe title was &gt;&gt;%s&lt;&lt;%n"</em>
would show something like this:</p>
<div class="listingblock">
<div class="content">
<pre><tt>The author of fe6e0ee was Junio C Hamano, 23 hours ago
The title was &gt;&gt;t4119: test autocomputing -p&lt;n&gt; for traditional diff input.&lt;&lt;
</tt></pre>
</div></div>
<p>The placeholders are:</p>
<ul>
<li>
<p>
<em>%H</em>: commit hash
</p>
</li>
<li>
<p>
<em>%h</em>: abbreviated commit hash
</p>
</li>
<li>
<p>
<em>%T</em>: tree hash
</p>
</li>
<li>
<p>
<em>%t</em>: abbreviated tree hash
</p>
</li>
<li>
<p>
<em>%P</em>: parent hashes
</p>
</li>
<li>
<p>
<em>%p</em>: abbreviated parent hashes
</p>
</li>
<li>
<p>
<em>%an</em>: author name
</p>
</li>
<li>
<p>
<em>%ae</em>: author email
</p>
</li>
<li>
<p>
<em>%ad</em>: author date
</p>
</li>
<li>
<p>
<em>%aD</em>: author date, RFC2822 style
</p>
</li>
<li>
<p>
<em>%ar</em>: author date, relative
</p>
</li>
<li>
<p>
<em>%at</em>: author date, UNIX timestamp
</p>
</li>
<li>
<p>
<em>%ai</em>: author date, ISO 8601 format
</p>
</li>
<li>
<p>
<em>%cn</em>: committer name
</p>
</li>
<li>
<p>
<em>%ce</em>: committer email
</p>
</li>
<li>
<p>
<em>%cd</em>: committer date
</p>
</li>
<li>
<p>
<em>%cD</em>: committer date, RFC2822 style
</p>
</li>
<li>
<p>
<em>%cr</em>: committer date, relative
</p>
</li>
<li>
<p>
<em>%ct</em>: committer date, UNIX timestamp
</p>
</li>
<li>
<p>
<em>%ci</em>: committer date, ISO 8601 format
</p>
</li>
<li>
<p>
<em>%e</em>: encoding
</p>
</li>
<li>
<p>
<em>%s</em>: subject
</p>
</li>
<li>
<p>
<em>%b</em>: body
</p>
</li>
<li>
<p>
<em>%Cred</em>: switch color to red
</p>
</li>
<li>
<p>
<em>%Cgreen</em>: switch color to green
</p>
</li>
<li>
<p>
<em>%Cblue</em>: switch color to blue
</p>
</li>
<li>
<p>
<em>%Creset</em>: reset color
</p>
</li>
<li>
<p>
<em>%m</em>: left, right or boundary mark
</p>
</li>
<li>
<p>
<em>%n</em>: newline
</p>
</li>
</ul>
</li>
</ul>
</div>
<h2>Limiting Output</h2>
<div class="sectionbody">
<p>If you're only interested in differences in a subset of files, for
example some architecture-specific files, you might do:</p>
<div class="literalblock">
<div class="content">
<pre><tt>git-diff-tree -r &lt;tree-ish&gt; &lt;tree-ish&gt; arch/ia64 include/asm-ia64</tt></pre>
</div></div>
<p>and it will only show you what changed in those two directories.</p>
<p>Or if you are searching for what changed in just <tt>kernel/sched.c</tt>, just do</p>
<div class="literalblock">
<div class="content">
<pre><tt>git-diff-tree -r &lt;tree-ish&gt; &lt;tree-ish&gt; kernel/sched.c</tt></pre>
</div></div>
<p>and it will ignore all differences to other files.</p>
<p>The pattern is always the prefix, and is matched exactly. There are no
wildcards. Even stricter, it has to match a complete path component.
I.e. "foo" does not pick up <tt>foobar.h</tt>. "foo" does match <tt>foo/bar.h</tt>
so it can be used to name subdirectories.</p>
<p>An example of normal usage is:</p>
<div class="literalblock">
<div class="content">
<pre><tt>torvalds@ppc970:~/git&gt; git-diff-tree 5319e4......
*100664-&gt;100664 blob ac348b.......-&gt;a01513....... git-fsck-objects.c</tt></pre>
</div></div>
<p>which tells you that the last commit changed just one file (it's from
this one:</p>
<div class="listingblock">
<div class="content">
<pre><tt>commit 3c6f7ca19ad4043e9e72fa94106f352897e651a8
tree 5319e4d609cdd282069cc4dce33c1db559539b03
parent b4e628ea30d5ab3606119d2ea5caeab141d38df7
author Linus Torvalds &lt;torvalds@ppc970.osdl.org&gt; Sat Apr 9 12:02:30 2005
committer Linus Torvalds &lt;torvalds@ppc970.osdl.org&gt; Sat Apr 9 12:02:30 2005
Make "git-fsck-objects" print out all the root commits it finds.
Once I do the reference tracking, I'll also make it print out all the
HEAD commits it finds, which is even more interesting.</tt></pre>
</div></div>
<p>in case you care).</p>
</div>
<h2>Output format</h2>
<div class="sectionbody">
<p>The output format from "git-diff-index", "git-diff-tree",
"git-diff-files" and "git diff --raw" are very similar.</p>
<p>These commands all compare two sets of things; what is
compared differs:</p>
<dl>
<dt>
git-diff-index &lt;tree-ish&gt;
</dt>
<dd>
<p>
compares the &lt;tree-ish&gt; and the files on the filesystem.
</p>
</dd>
<dt>
git-diff-index --cached &lt;tree-ish&gt;
</dt>
<dd>
<p>
compares the &lt;tree-ish&gt; and the index.
</p>
</dd>
<dt>
git-diff-tree [-r] &lt;tree-ish-1&gt; &lt;tree-ish-2&gt; [&lt;pattern&gt;&#8230;]
</dt>
<dd>
<p>
compares the trees named by the two arguments.
</p>
</dd>
<dt>
git-diff-files [&lt;pattern&gt;&#8230;]
</dt>
<dd>
<p>
compares the index and the files on the filesystem.
</p>
</dd>
</dl>
<p>An output line is formatted this way:</p>
<div class="listingblock">
<div class="content">
<pre><tt>in-place edit :100644 100644 bcd1234... 0123456... M file0
copy-edit :100644 100644 abcd123... 1234567... C68 file1 file2
rename-edit :100644 100644 abcd123... 1234567... R86 file1 file3
create :000000 100644 0000000... 1234567... A file4
delete :100644 000000 1234567... 0000000... D file5
unmerged :000000 000000 0000000... 0000000... U file6</tt></pre>
</div></div>
<p>That is, from the left to the right:</p>
<ol>
<li>
<p>
a colon.
</p>
</li>
<li>
<p>
mode for "src"; 000000 if creation or unmerged.
</p>
</li>
<li>
<p>
a space.
</p>
</li>
<li>
<p>
mode for "dst"; 000000 if deletion or unmerged.
</p>
</li>
<li>
<p>
a space.
</p>
</li>
<li>
<p>
sha1 for "src"; 0{40} if creation or unmerged.
</p>
</li>
<li>
<p>
a space.
</p>
</li>
<li>
<p>
sha1 for "dst"; 0{40} if creation, unmerged or "look at work tree".
</p>
</li>
<li>
<p>
a space.
</p>
</li>
<li>
<p>
status, followed by optional "score" number.
</p>
</li>
<li>
<p>
a tab or a NUL when <em>-z</em> option is used.
</p>
</li>
<li>
<p>
path for "src"
</p>
</li>
<li>
<p>
a tab or a NUL when <em>-z</em> option is used; only exists for C or R.
</p>
</li>
<li>
<p>
path for "dst"; only exists for C or R.
</p>
</li>
<li>
<p>
an LF or a NUL when <em>-z</em> option is used, to terminate the record.
</p>
</li>
</ol>
<p>&lt;sha1&gt; is shown as all 0's if a file is new on the filesystem
and it is out of sync with the index.</p>
<p>Example:</p>
<div class="listingblock">
<div class="content">
<pre><tt>:100644 100644 5be4a4...... 000000...... M file.c</tt></pre>
</div></div>
<p>When <tt>-z</tt> option is not used, TAB, LF, and backslash characters
in pathnames are represented as <tt>\t</tt>, <tt>\n</tt>, and <tt>\\</tt>,
respectively.</p>
</div>
<h2>diff format for merges</h2>
<div class="sectionbody">
<p>"git-diff-tree", "git-diff-files" and "git-diff --raw"
can take <em>-c</em> or <em>--cc</em> option
to generate diff output also for merge commits. The output differs
from the format described above in the following way:</p>
<ol>
<li>
<p>
there is a colon for each parent
</p>
</li>
<li>
<p>
there are more "src" modes and "src" sha1
</p>
</li>
<li>
<p>
status is concatenated status characters for each parent
</p>
</li>
<li>
<p>
no optional "score" number
</p>
</li>
<li>
<p>
single path, only for "dst"
</p>
</li>
</ol>
<p>Example:</p>
<div class="listingblock">
<div class="content">
<pre><tt>::100644 100644 100644 fabadb8... cc95eb0... 4866510... MM describe.c</tt></pre>
</div></div>
<p>Note that <em>combined diff</em> lists only files which were modified from
all parents.</p>
</div>
<h2>Generating patches with -p</h2>
<div class="sectionbody">
<p>When "git-diff-index", "git-diff-tree", or "git-diff-files" are run
with a <em>-p</em> option, or "git diff" without the <em>--raw</em> option, they
do not produce the output described above; instead they produce a
patch file. You can customize the creation of such patches via the
GIT_EXTERNAL_DIFF and the GIT_DIFF_OPTS environment variables.</p>
<p>What the -p option produces is slightly different from the traditional
diff format.</p>
<ol>
<li>
<p>
It is preceded with a "git diff" header, that looks like
this:
</p>
<div class="literalblock">
<div class="content">
<pre><tt>diff --git a/file1 b/file2</tt></pre>
</div></div>
<p>The <tt>a/</tt> and <tt>b/</tt> filenames are the same unless rename/copy is
involved. Especially, even for a creation or a deletion,
<tt>/dev/null</tt> is _not_ used in place of <tt>a/</tt> or <tt>b/</tt> filenames.</p>
<p>When rename/copy is involved, <tt>file1</tt> and <tt>file2</tt> show the
name of the source file of the rename/copy and the name of
the file that rename/copy produces, respectively.</p>
</li>
<li>
<p>
It is followed by one or more extended header lines:
</p>
<div class="literalblock">
<div class="content">
<pre><tt>old mode &lt;mode&gt;
new mode &lt;mode&gt;
deleted file mode &lt;mode&gt;
new file mode &lt;mode&gt;
copy from &lt;path&gt;
copy to &lt;path&gt;
rename from &lt;path&gt;
rename to &lt;path&gt;
similarity index &lt;number&gt;
dissimilarity index &lt;number&gt;
index &lt;hash&gt;..&lt;hash&gt; &lt;mode&gt;</tt></pre>
</div></div>
</li>
<li>
<p>
TAB, LF, double quote and backslash characters in pathnames
are represented as <tt>\t</tt>, <tt>\n</tt>, <tt>\"</tt> and <tt>\\</tt>, respectively.
If there is need for such substitution then the whole
pathname is put in double quotes.
</p>
</li>
</ol>
<p>The similarity index is the percentage of unchanged lines, and
the dissimilarity index is the percentage of changed lines. It
is a rounded down integer, followed by a percent sign. The
similarity index value of 100% is thus reserved for two equal
files, while 100% dissimilarity means that no line from the old
file made it into the new one.</p>
</div>
<h2>combined diff format</h2>
<div class="sectionbody">
<p>"git-diff-tree", "git-diff-files" and "git-diff" can take <em>-c</em> or
<em>--cc</em> option to produce <em>combined diff</em>, which looks like this:</p>
<div class="listingblock">
<div class="content">
<pre><tt>diff --combined describe.c
index fabadb8,cc95eb0..4866510
--- a/describe.c
+++ b/describe.c
@@@ -98,20 -98,12 +98,20 @@@
return (a_date &gt; b_date) ? -1 : (a_date == b_date) ? 0 : 1;
}
- static void describe(char *arg)
-static void describe(struct commit *cmit, int last_one)
++static void describe(char *arg, int last_one)
{
+ unsigned char sha1[20];
+ struct commit *cmit;
struct commit_list *list;
static int initialized = 0;
struct commit_name *n;
+ if (get_sha1(arg, sha1) &lt; 0)
+ usage(describe_usage);
+ cmit = lookup_commit_reference(sha1);
+ if (!cmit)
+ usage(describe_usage);
+
if (!initialized) {
initialized = 1;
for_each_ref(get_name);</tt></pre>
</div></div>
<ol>
<li>
<p>
It is preceded with a "git diff" header, that looks like
this (when <em>-c</em> option is used):
</p>
<div class="literalblock">
<div class="content">
<pre><tt>diff --combined file</tt></pre>
</div></div>
<p>or like this (when <em>--cc</em> option is used):</p>
<div class="literalblock">
<div class="content">
<pre><tt>diff --c file</tt></pre>
</div></div>
</li>
<li>
<p>
It is followed by one or more extended header lines
(this example shows a merge with two parents):
</p>
<div class="literalblock">
<div class="content">
<pre><tt>index &lt;hash&gt;,&lt;hash&gt;..&lt;hash&gt;
mode &lt;mode&gt;,&lt;mode&gt;..&lt;mode&gt;
new file mode &lt;mode&gt;
deleted file mode &lt;mode&gt;,&lt;mode&gt;</tt></pre>
</div></div>
<p>The <tt>mode &lt;mode&gt;,&lt;mode&gt;..&lt;mode&gt;</tt> line appears only if at least one of
the &lt;mode&gt; is different from the rest. Extended headers with
information about detected contents movement (renames and
copying detection) are designed to work with diff of two
&lt;tree-ish&gt; and are not used by combined diff format.</p>
</li>
<li>
<p>
It is followed by two-line from-file/to-file header
</p>
<div class="literalblock">
<div class="content">
<pre><tt>--- a/file
+++ b/file</tt></pre>
</div></div>
<p>Similar to two-line header for traditional <em>unified</em> diff
format, <tt>/dev/null</tt> is used to signal created or deleted
files.</p>
</li>
<li>
<p>
Chunk header format is modified to prevent people from
accidentally feeding it to <tt>patch -p1</tt>. Combined diff format
was created for review of merge commit changes, and was not
meant for apply. The change is similar to the change in the
extended <em>index</em> header:
</p>
<div class="literalblock">
<div class="content">
<pre><tt>@@@ &lt;from-file-range&gt; &lt;from-file-range&gt; &lt;to-file-range&gt; @@@</tt></pre>
</div></div>
<p>There are (number of parents + 1) <tt>@</tt> characters in the chunk
header for combined diff format.</p>
</li>
</ol>
<p>Unlike the traditional <em>unified</em> diff format, which shows two
files A and B with a single column that has <tt>-</tt> (minus &#8212;
appears in A but removed in B), <tt>+</tt> (plus &#8212; missing in A but
added to B), or <tt>" "</tt> (space &#8212; unchanged) prefix, this format
compares two or more files file1, file2,&#8230; with one file X, and
shows how X differs from each of fileN. One column for each of
fileN is prepended to the output line to note how X's line is
different from it.</p>
<p>A <tt>-</tt> character in the column N means that the line appears in
fileN but it does not appear in the result. A <tt>+</tt> character
in the column N means that the line appears in the last file,
and fileN does not have that line (in other words, the line was
added, from the point of view of that parent).</p>
<p>In the above example output, the function signature was changed
from both files (hence two <tt>-</tt> removals from both file1 and
file2, plus <tt>++</tt> to mean one line that was added does not appear
in either file1 nor file2). Also two other lines are the same
from file1 but do not appear in file2 (hence prefixed with <tt> +</tt>).</p>
<p>When shown by <tt>git diff-tree -c</tt>, it compares the parents of a
merge commit with the merge result (i.e. file1..fileN are the
parents). When shown by <tt>git diff-files -c</tt>, it compares the
two unresolved merge parents with the working tree file
(i.e. file1 is stage 2 aka "our version", file2 is stage 3 aka
"their version").</p>
</div>
<h2>Author</h2>
<div class="sectionbody">
<p>Written by Linus Torvalds &lt;torvalds@osdl.org&gt;</p>
</div>
<h2>Documentation</h2>
<div class="sectionbody">
<p>Documentation by David Greaves, Junio C Hamano and the git-list &lt;git@vger.kernel.org&gt;.</p>
</div>
<h2>GIT</h2>
<div class="sectionbody">
<p>Part of the <a href="git.html">git(7)</a> suite</p>
</div>
<div id="footer">
<div id="footer-text">
Last updated 14-Nov-2007 12:11:28 UTC
</div>
</div>
</body>
</html>