1
Fork 0
mirror of git://git.sv.gnu.org/emacs.git synced 2025-12-25 06:50:46 -08:00
emacs/admin/notes/tree-sitter/html-manual/Pattern-Matching.html
2023-01-01 05:31:12 -05:00

450 lines
19 KiB
HTML

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- Created by GNU Texinfo 6.8, https://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<!-- This is the GNU Emacs Lisp Reference Manual
corresponding to Emacs version 29.0.50.
Copyright © 1990-1996, 1998-2023 Free Software Foundation, Inc.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with the
Invariant Sections being "GNU General Public License," with the
Front-Cover Texts being "A GNU Manual," and with the Back-Cover
Texts as in (a) below. A copy of the license is included in the
section entitled "GNU Free Documentation License."
(a) The FSF's Back-Cover Text is: "You have the freedom to copy and
modify this GNU manual. Buying copies from the FSF supports it in
developing GNU and promoting software freedom." -->
<title>Pattern Matching (GNU Emacs Lisp Reference Manual)</title>
<meta name="description" content="Pattern Matching (GNU Emacs Lisp Reference Manual)">
<meta name="keywords" content="Pattern Matching (GNU Emacs Lisp Reference Manual)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<meta name="viewport" content="width=device-width,initial-scale=1">
<link href="index.html" rel="start" title="Top">
<link href="Index.html" rel="index" title="Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Parsing-Program-Source.html" rel="up" title="Parsing Program Source">
<link href="Multiple-Languages.html" rel="next" title="Multiple Languages">
<link href="Accessing-Node-Information.html" rel="prev" title="Accessing Node Information">
<style type="text/css">
<!--
a.copiable-anchor {visibility: hidden; text-decoration: none; line-height: 0em}
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
span:hover a.copiable-anchor {visibility: visible}
ul.no-bullet {list-style: none}
-->
</style>
<link rel="stylesheet" type="text/css" href="./manual.css">
</head>
<body lang="en">
<div class="section" id="Pattern-Matching">
<div class="header">
<p>
Next: <a href="Multiple-Languages.html" accesskey="n" rel="next">Parsing Text in Multiple Languages</a>, Previous: <a href="Accessing-Node-Information.html" accesskey="p" rel="prev">Accessing Node Information</a>, Up: <a href="Parsing-Program-Source.html" accesskey="u" rel="up">Parsing Program Source</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Pattern-Matching-Tree_002dsitter-Nodes"></span><h3 class="section">37.5 Pattern Matching Tree-sitter Nodes</h3>
<span id="index-pattern-matching-with-tree_002dsitter-nodes"></span>
<span id="index-capturing_002c-tree_002dsitter-node"></span>
<p>Tree-sitter lets Lisp programs match patterns using a small
declarative language. This pattern matching consists of two steps:
first tree-sitter matches a <em>pattern</em> against nodes in the syntax
tree, then it <em>captures</em> specific nodes that matched the pattern
and returns the captured nodes.
</p>
<p>We describe first how to write the most basic query pattern and how to
capture nodes in a pattern, then the pattern-matching function, and
finally the more advanced pattern syntax.
</p>
<span id="Basic-query-syntax"></span><h3 class="heading">Basic query syntax</h3>
<span id="index-tree_002dsitter-query-pattern-syntax"></span>
<span id="index-pattern-syntax_002c-tree_002dsitter-query"></span>
<span id="index-query_002c-tree_002dsitter"></span>
<p>A <em>query</em> consists of multiple <em>patterns</em>. Each pattern is an
s-expression that matches a certain node in the syntax node. A
pattern has the form <code>(<var>type</var>&nbsp;(<var>child</var>&hellip;))</code><!-- /@w -->
</p>
<p>For example, a pattern that matches a <code>binary_expression</code> node that
contains <code>number_literal</code> child nodes would look like
</p>
<div class="example">
<pre class="example">(binary_expression (number_literal))
</pre></div>
<p>To <em>capture</em> a node using the query pattern above, append
<code>@<var>capture-name</var></code> after the node pattern you want to
capture. For example,
</p>
<div class="example">
<pre class="example">(binary_expression (number_literal) @number-in-exp)
</pre></div>
<p>captures <code>number_literal</code> nodes that are inside a
<code>binary_expression</code> node with the capture name
<code>number-in-exp</code>.
</p>
<p>We can capture the <code>binary_expression</code> node as well, with, for
example, the capture name <code>biexp</code>:
</p>
<div class="example">
<pre class="example">(binary_expression
(number_literal) @number-in-exp) @biexp
</pre></div>
<span id="Query-function"></span><h3 class="heading">Query function</h3>
<span id="index-query-functions_002c-tree_002dsitter"></span>
<p>Now we can introduce the <em>query functions</em>.
</p>
<dl class="def">
<dt id="index-treesit_002dquery_002dcapture"><span class="category">Function: </span><span><strong>treesit-query-capture</strong> <em>node query &amp;optional beg end node-only</em><a href='#index-treesit_002dquery_002dcapture' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>This function matches patterns in <var>query</var> within <var>node</var>.
The argument <var>query</var> can be either a string, a s-expression, or a
compiled query object. For now, we focus on the string syntax;
s-expression syntax and compiled query are described at the end of the
section.
</p>
<p>The argument <var>node</var> can also be a parser or a language symbol. A
parser means using its root node, a language symbol means find or
create a parser for that language in the current buffer, and use the
root node.
</p>
<p>The function returns all the captured nodes in a list of the form
<code>(<var><span class="nolinebreak">capture_name</span></var>&nbsp;.&nbsp;<var>node</var>)</code><!-- /@w -->. If <var>node-only</var> is
non-<code>nil</code>, it returns the list of nodes instead. By default the
entire text of <var>node</var> is searched, but if <var>beg</var> and <var>end</var>
are both non-<code>nil</code>, they specify the region of buffer text where
this function should match nodes. Any matching node whose span
overlaps with the region between <var>beg</var> and <var>end</var> are captured,
it doesn&rsquo;t have to be completely in the region.
</p>
<span id="index-treesit_002dquery_002derror"></span>
<span id="index-treesit_002dquery_002dvalidate"></span>
<p>This function raises the <code>treesit-query-error</code> error if
<var>query</var> is malformed. The signal data contains a description of
the specific error. You can use <code>treesit-query-validate</code> to
validate and debug the query.
</p></dd></dl>
<p>For example, suppose <var>node</var>&rsquo;s text is <code>1 + 2</code>, and
<var>query</var> is
</p>
<div class="example">
<pre class="example">(setq query
&quot;(binary_expression
(number_literal) @number-in-exp) @biexp&quot;)
</pre></div>
<p>Matching that query would return
</p>
<div class="example">
<pre class="example">(treesit-query-capture node query)
&rArr; ((biexp . <var>&lt;node for &quot;1 + 2&quot;&gt;</var>)
(number-in-exp . <var>&lt;node for &quot;1&quot;&gt;</var>)
(number-in-exp . <var>&lt;node for &quot;2&quot;&gt;</var>))
</pre></div>
<p>As mentioned earlier, <var>query</var> could contain multiple patterns.
For example, it could have two top-level patterns:
</p>
<div class="example">
<pre class="example">(setq query
&quot;(binary_expression) @biexp
(number_literal) @number @biexp&quot;)
</pre></div>
<dl class="def">
<dt id="index-treesit_002dquery_002dstring"><span class="category">Function: </span><span><strong>treesit-query-string</strong> <em>string query language</em><a href='#index-treesit_002dquery_002dstring' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>This function parses <var>string</var> with <var>language</var>, matches its
root node with <var>query</var>, and returns the result.
</p></dd></dl>
<span id="More-query-syntax"></span><h3 class="heading">More query syntax</h3>
<p>Besides node type and capture, tree-sitter&rsquo;s pattern syntax can
express anonymous node, field name, wildcard, quantification,
grouping, alternation, anchor, and predicate.
</p>
<span id="Anonymous-node"></span><h4 class="subheading">Anonymous node</h4>
<p>An anonymous node is written verbatim, surrounded by quotes. A
pattern matching (and capturing) keyword <code>return</code> would be
</p>
<div class="example">
<pre class="example">&quot;return&quot; @keyword
</pre></div>
<span id="Wild-card"></span><h4 class="subheading">Wild card</h4>
<p>In a pattern, &lsquo;<samp>(_)</samp>&rsquo; matches any named node, and &lsquo;<samp>_</samp>&rsquo; matches
any named and anonymous node. For example, to capture any named child
of a <code>binary_expression</code> node, the pattern would be
</p>
<div class="example">
<pre class="example">(binary_expression (_) @in_biexp)
</pre></div>
<span id="Field-name"></span><h4 class="subheading">Field name</h4>
<p>It is possible to capture child nodes that have specific field names.
In the pattern below, <code>declarator</code> and <code>body</code> are field
names, indicated by the colon following them.
</p>
<div class="example">
<pre class="example">(function_definition
declarator: (_) @func-declarator
body: (_) @func-body)
</pre></div>
<p>It is also possible to capture a node that doesn&rsquo;t have a certain
field, say, a <code>function_definition</code> without a <code>body</code> field.
</p>
<div class="example">
<pre class="example">(function_definition !body) @func-no-body
</pre></div>
<span id="Quantify-node"></span><h4 class="subheading">Quantify node</h4>
<span id="index-quantify-node_002c-tree_002dsitter"></span>
<p>Tree-sitter recognizes quantification operators &lsquo;<samp>*</samp>&rsquo;, &lsquo;<samp>+</samp>&rsquo; and
&lsquo;<samp>?</samp>&rsquo;. Their meanings are the same as in regular expressions:
&lsquo;<samp>*</samp>&rsquo; matches the preceding pattern zero or more times, &lsquo;<samp>+</samp>&rsquo;
matches one or more times, and &lsquo;<samp>?</samp>&rsquo; matches zero or one time.
</p>
<p>For example, the following pattern matches <code>type_declaration</code>
nodes that has <em>zero or more</em> <code>long</code> keyword.
</p>
<div class="example">
<pre class="example">(type_declaration &quot;long&quot;*) @long-type
</pre></div>
<p>The following pattern matches a type declaration that has zero or one
<code>long</code> keyword:
</p>
<div class="example">
<pre class="example">(type_declaration &quot;long&quot;?) @long-type
</pre></div>
<span id="Grouping"></span><h4 class="subheading">Grouping</h4>
<p>Similar to groups in regular expression, we can bundle patterns into
groups and apply quantification operators to them. For example, to
express a comma separated list of identifiers, one could write
</p>
<div class="example">
<pre class="example">(identifier) (&quot;,&quot; (identifier))*
</pre></div>
<span id="Alternation"></span><h4 class="subheading">Alternation</h4>
<p>Again, similar to regular expressions, we can express &ldquo;match anyone
from this group of patterns&rdquo; in a pattern. The syntax is a list of
patterns enclosed in square brackets. For example, to capture some
keywords in C, the pattern would be
</p>
<div class="example">
<pre class="example">[
&quot;return&quot;
&quot;break&quot;
&quot;if&quot;
&quot;else&quot;
] @keyword
</pre></div>
<span id="Anchor"></span><h4 class="subheading">Anchor</h4>
<p>The anchor operator &lsquo;<samp>.</samp>&rsquo; can be used to enforce juxtaposition,
i.e., to enforce two things to be directly next to each other. The
two &ldquo;things&rdquo; can be two nodes, or a child and the end of its parent.
For example, to capture the first child, the last child, or two
adjacent children:
</p>
<div class="example">
<pre class="example">;; Anchor the child with the end of its parent.
(compound_expression (_) @last-child .)
</pre><pre class="example">
</pre><pre class="example">;; Anchor the child with the beginning of its parent.
(compound_expression . (_) @first-child)
</pre><pre class="example">
</pre><pre class="example">;; Anchor two adjacent children.
(compound_expression
(_) @prev-child
.
(_) @next-child)
</pre></div>
<p>Note that the enforcement of juxtaposition ignores any anonymous
nodes.
</p>
<span id="Predicate"></span><h4 class="subheading">Predicate</h4>
<p>It is possible to add predicate constraints to a pattern. For
example, with the following pattern:
</p>
<div class="example">
<pre class="example">(
(array . (_) @first (_) @last .)
(#equal @first @last)
)
</pre></div>
<p>tree-sitter only matches arrays where the first element equals to
the last element. To attach a predicate to a pattern, we need to
group them together. A predicate always starts with a &lsquo;<samp>#</samp>&rsquo;.
Currently there are two predicates, <code>#equal</code> and <code>#match</code>.
</p>
<dl class="def">
<dt id="index-equal-1"><span class="category">Predicate: </span><span><strong>equal</strong> <em>arg1 arg2</em><a href='#index-equal-1' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>Matches if <var>arg1</var> equals to <var>arg2</var>. Arguments can be either
strings or capture names. Capture names represent the text that the
captured node spans in the buffer.
</p></dd></dl>
<dl class="def">
<dt id="index-match-1"><span class="category">Predicate: </span><span><strong>match</strong> <em>regexp capture-name</em><a href='#index-match-1' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>Matches if the text that <var>capture-name</var>&rsquo;s node spans in the buffer
matches regular expression <var>regexp</var>. Matching is case-sensitive.
</p></dd></dl>
<p>Note that a predicate can only refer to capture names that appear in
the same pattern. Indeed, it makes little sense to refer to capture
names in other patterns.
</p>
<span id="S_002dexpression-patterns"></span><h3 class="heading">S-expression patterns</h3>
<span id="index-tree_002dsitter-patterns-as-sexps"></span>
<span id="index-patterns_002c-tree_002dsitter_002c-in-sexp-form"></span>
<p>Besides strings, Emacs provides a s-expression based syntax for
tree-sitter patterns. It largely resembles the string-based syntax.
For example, the following query
</p>
<div class="example">
<pre class="example">(treesit-query-capture
node &quot;(addition_expression
left: (_) @left
\&quot;+\&quot; @plus-sign
right: (_) @right) @addition
[\&quot;return\&quot; \&quot;break\&quot;] @keyword&quot;)
</pre></div>
<p>is equivalent to
</p>
<div class="example">
<pre class="example">(treesit-query-capture
node '((addition_expression
left: (_) @left
&quot;+&quot; @plus-sign
right: (_) @right) @addition
[&quot;return&quot; &quot;break&quot;] @keyword))
</pre></div>
<p>Most patterns can be written directly as strange but nevertheless
valid s-expressions. Only a few of them needs modification:
</p>
<ul>
<li> Anchor &lsquo;<samp>.</samp>&rsquo; is written as <code>:anchor</code>.
</li><li> &lsquo;<samp>?</samp>&rsquo; is written as &lsquo;<samp>:?</samp>&rsquo;.
</li><li> &lsquo;<samp>*</samp>&rsquo; is written as &lsquo;<samp>:*</samp>&rsquo;.
</li><li> &lsquo;<samp>+</samp>&rsquo; is written as &lsquo;<samp>:+</samp>&rsquo;.
</li><li> <code>#equal</code> is written as <code>:equal</code>. In general, predicates
change their &lsquo;<samp>#</samp>&rsquo; to &lsquo;<samp>:</samp>&rsquo;.
</li></ul>
<p>For example,
</p>
<div class="example">
<pre class="example">&quot;(
(compound_expression . (_) @first (_)* @rest)
(#match \&quot;love\&quot; @first)
)&quot;
</pre></div>
<p>is written in s-expression as
</p>
<div class="example">
<pre class="example">'((
(compound_expression :anchor (_) @first (_) :* @rest)
(:match &quot;love&quot; @first)
))
</pre></div>
<span id="Compiling-queries"></span><h3 class="heading">Compiling queries</h3>
<span id="index-compiling-tree_002dsitter-queries"></span>
<span id="index-queries_002c-compiling"></span>
<p>If a query is intended to be used repeatedly, especially in tight
loops, it is important to compile that query, because a compiled query
is much faster than an uncompiled one. A compiled query can be used
anywhere a query is accepted.
</p>
<dl class="def">
<dt id="index-treesit_002dquery_002dcompile"><span class="category">Function: </span><span><strong>treesit-query-compile</strong> <em>language query</em><a href='#index-treesit_002dquery_002dcompile' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>This function compiles <var>query</var> for <var>language</var> into a compiled
query object and returns it.
</p>
<p>This function raises the <code>treesit-query-error</code> error if
<var>query</var> is malformed. The signal data contains a description of
the specific error. You can use <code>treesit-query-validate</code> to
validate and debug the query.
</p></dd></dl>
<dl class="def">
<dt id="index-treesit_002dquery_002dlanguage"><span class="category">Function: </span><span><strong>treesit-query-language</strong> <em>query</em><a href='#index-treesit_002dquery_002dlanguage' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>This function return the language of <var>query</var>.
</p></dd></dl>
<dl class="def">
<dt id="index-treesit_002dquery_002dexpand"><span class="category">Function: </span><span><strong>treesit-query-expand</strong> <em>query</em><a href='#index-treesit_002dquery_002dexpand' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>This function converts the s-expression <var>query</var> into the string
format.
</p></dd></dl>
<dl class="def">
<dt id="index-treesit_002dpattern_002dexpand"><span class="category">Function: </span><span><strong>treesit-pattern-expand</strong> <em>pattern</em><a href='#index-treesit_002dpattern_002dexpand' class='copiable-anchor'> &para;</a></span></dt>
<dd><p>This function converts the s-expression <var>pattern</var> into the string
format.
</p></dd></dl>
<p>For more details, read the tree-sitter project&rsquo;s documentation about
pattern-matching, which can be found at
<a href="https://tree-sitter.github.io/tree-sitter/using-parsers#pattern-matching-with-queries">https://tree-sitter.github.io/tree-sitter/using-parsers#pattern-matching-with-queries</a>.
</p>
</div>
<hr>
<div class="header">
<p>
Next: <a href="Multiple-Languages.html">Parsing Text in Multiple Languages</a>, Previous: <a href="Accessing-Node-Information.html">Accessing Node Information</a>, Up: <a href="Parsing-Program-Source.html">Parsing Program Source</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>