Files
docs-rust/ch04/ch04-03-slices.html
2026-06-22 21:27:36 +05:30

430 lines
24 KiB
HTML
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<title>The Slice Type</title>
</head>
<body>
<h2 id="the-slice-type"><a class="header" href="#the-slice-type">The Slice Type</a></h2>
<p><em>Slices</em> let you reference a contiguous sequence of elements in a
<a href="../ch08/ch08-00-common-collections.html">collection</a><!-- ignore -->. A slice is a kind
of reference, so it does not have ownership.</p>
<p>Heres a small programming problem: Write a function that takes a string of
words separated by spaces and returns the first word it finds in that string.
If the function doesnt find a space in the string, the whole string must be
one word, so the entire string should be returned.</p>
<section class="note" aria-role="note">
<p>Note: For the purposes of introducing slices, we are assuming ASCII only in
this section; a more thorough discussion of UTF-8 handling is in the
<a href="../ch08/ch08-02-strings.html#storing-utf-8-encoded-text-with-strings">“Storing UTF-8 Encoded Text with Strings”</a><!-- ignore --> section
of Chapter 8.</p>
</section>
<p>Lets work through how wed write the signature of this function without using
slices, to understand the problem that slices will solve:</p>
<pre><code class="language-rust ignore">fn first_word(s: &amp;String) -&gt; ?</code></pre>
<p>The <code>first_word</code> function has a parameter of type <code>&amp;String</code>. We dont need
ownership, so this is fine. (In idiomatic Rust, functions do not take ownership
of their arguments unless they need to, and the reasons for that will become
clear as we keep going.) But what should we return? We dont really have a way
to talk about <em>part</em> of a string. However, we could return the index of the end
of the word, indicated by a space. Lets try that, as shown in Listing 4-7.</p>
<figure class="listing" id="listing-4-7">
<span class="file-name">Filename: src/main.rs</span>
<pre class="playground"><code class="language-rust edition2024">fn first_word(s: &amp;String) -&gt; usize {
let bytes = s.as_bytes();
for (i, &amp;item) in bytes.iter().enumerate() {
if item == b' ' {
return i;
}
}
s.len()
}
<span class="boring">
</span><span class="boring">fn main() {}</span></code></pre>
<figcaption><a href="#listing-4-7">Listing 4-7</a>: The <code>first_word</code> function that returns a byte index value into the <code>String</code> parameter</figcaption>
</figure>
<p>Because we need to go through the <code>String</code> element by element and check whether
a value is a space, well convert our <code>String</code> to an array of bytes using the
<code>as_bytes</code> method.</p>
<pre><code class="language-rust ignore"><span class="boring">fn first_word(s: &amp;String) -&gt; usize {
</span> let bytes = s.as_bytes();
<span class="boring">
</span><span class="boring"> for (i, &amp;item) in bytes.iter().enumerate() {
</span><span class="boring"> if item == b' ' {
</span><span class="boring"> return i;
</span><span class="boring"> }
</span><span class="boring"> }
</span><span class="boring">
</span><span class="boring"> s.len()
</span><span class="boring">}
</span><span class="boring">
</span><span class="boring">fn main() {}</span></code></pre>
<p>Next, we create an iterator over the array of bytes using the <code>iter</code> method:</p>
<pre><code class="language-rust ignore"><span class="boring">fn first_word(s: &amp;String) -&gt; usize {
</span><span class="boring"> let bytes = s.as_bytes();
</span><span class="boring">
</span> for (i, &amp;item) in bytes.iter().enumerate() {
<span class="boring"> if item == b' ' {
</span><span class="boring"> return i;
</span><span class="boring"> }
</span><span class="boring"> }
</span><span class="boring">
</span><span class="boring"> s.len()
</span><span class="boring">}
</span><span class="boring">
</span><span class="boring">fn main() {}</span></code></pre>
<p>Well discuss iterators in more detail in <a href="../ch13/ch13-02-iterators.html">Chapter 13</a><!-- ignore -->.
For now, know that <code>iter</code> is a method that returns each element in a collection
and that <code>enumerate</code> wraps the result of <code>iter</code> and returns each element as
part of a tuple instead. The first element of the tuple returned from
<code>enumerate</code> is the index, and the second element is a reference to the element.
This is a bit more convenient than calculating the index ourselves.</p>
<p>Because the <code>enumerate</code> method returns a tuple, we can use patterns to
destructure that tuple. Well be discussing patterns more in <a href="../ch06/ch06-02-match.html#patterns-that-bind-to-values">Chapter
6</a><!-- ignore -->. In the <code>for</code> loop, we specify a pattern that has <code>i</code>
for the index in the tuple and <code>&amp;item</code> for the single byte in the tuple.
Because we get a reference to the element from <code>.iter().enumerate()</code>, we use
<code>&amp;</code> in the pattern.</p>
<p>Inside the <code>for</code> loop, we search for the byte that represents the space by
using the byte literal syntax. If we find a space, we return the position.
Otherwise, we return the length of the string by using <code>s.len()</code>.</p>
<pre><code class="language-rust ignore"><span class="boring">fn first_word(s: &amp;String) -&gt; usize {
</span><span class="boring"> let bytes = s.as_bytes();
</span><span class="boring">
</span><span class="boring"> for (i, &amp;item) in bytes.iter().enumerate() {
</span> if item == b' ' {
return i;
}
}
s.len()
<span class="boring">}
</span><span class="boring">
</span><span class="boring">fn main() {}</span></code></pre>
<p>We now have a way to find out the index of the end of the first word in the
string, but theres a problem. Were returning a <code>usize</code> on its own, but its
only a meaningful number in the context of the <code>&amp;String</code>. In other words,
because its a separate value from the <code>String</code>, theres no guarantee that it
will still be valid in the future. Consider the program in Listing 4-8 that
uses the <code>first_word</code> function from Listing 4-7.</p>
<figure class="listing" id="listing-4-8">
<span class="file-name">Filename: src/main.rs</span>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">fn first_word(s: &amp;String) -&gt; usize {
</span><span class="boring"> let bytes = s.as_bytes();
</span><span class="boring">
</span><span class="boring"> for (i, &amp;item) in bytes.iter().enumerate() {
</span><span class="boring"> if item == b' ' {
</span><span class="boring"> return i;
</span><span class="boring"> }
</span><span class="boring"> }
</span><span class="boring">
</span><span class="boring"> s.len()
</span><span class="boring">}
</span><span class="boring">
</span>fn main() {
let mut s = String::from("hello world");
let word = first_word(&amp;s); // word will get the value 5
s.clear(); // this empties the String, making it equal to ""
// word still has the value 5 here, but s no longer has any content that we
// could meaningfully use with the value 5, so word is now totally invalid!
}</code></pre>
<figcaption><a href="#listing-4-8">Listing 4-8</a>: Storing the result from calling the <code>first_word</code> function and then changing the <code>String</code> contents</figcaption>
</figure>
<p>This program compiles without any errors and would also do so if we used <code>word</code>
after calling <code>s.clear()</code>. Because <code>word</code> isnt connected to the state of <code>s</code>
at all, <code>word</code> still contains the value <code>5</code>. We could use that value <code>5</code> with
the variable <code>s</code> to try to extract the first word out, but this would be a bug
because the contents of <code>s</code> have changed since we saved <code>5</code> in <code>word</code>.</p>
<p>Having to worry about the index in <code>word</code> getting out of sync with the data in
<code>s</code> is tedious and error-prone! Managing these indices is even more brittle if
we write a <code>second_word</code> function. Its signature would have to look like this:</p>
<pre><code class="language-rust ignore">fn second_word(s: &amp;String) -&gt; (usize, usize) {</code></pre>
<p>Now were tracking a starting <em>and</em> an ending index, and we have even more
values that were calculated from data in a particular state but arent tied to
that state at all. We have three unrelated variables floating around that need
to be kept in sync.</p>
<p>Luckily, Rust has a solution to this problem: string slices.</p>
<h3 id="string-slices"><a class="header" href="#string-slices">String Slices</a></h3>
<p>A <em>string slice</em> is a reference to a contiguous sequence of the elements of a
<code>String</code>, and it looks like this:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">fn main() {
</span> let s = String::from("hello world");
let hello = &amp;s[0..5];
let world = &amp;s[6..11];
<span class="boring">}</span></code></pre>
<p>Rather than a reference to the entire <code>String</code>, <code>hello</code> is a reference to a
portion of the <code>String</code>, specified in the extra <code>[0..5]</code> bit. We create slices
using a range within square brackets by specifying
<code>[starting_index..ending_index]</code>, where <em><code>starting_index</code></em> is the first
position in the slice and <em><code>ending_index</code></em> is one more than the last position
in the slice. Internally, the slice data structure stores the starting position
and the length of the slice, which corresponds to <em><code>ending_index</code></em> minus
<em><code>starting_index</code></em>. So, in the case of <code>let world = &amp;s[6..11];</code>, <code>world</code> would
be a slice that contains a pointer to the byte at index 6 of <code>s</code> with a length
value of <code>5</code>.</p>
<p>Figure 4-7 shows this in a diagram.</p>
<p><img alt="Three tables: a table representing the stack data of s, which points
to the byte at index 0 in a table of the string data &quot;hello world&quot; on
the heap. The third table represents the stack data of the slice world, which
has a length value of 5 and points to byte 6 of the heap data table." src="../img/trpl04-07.svg" class="center" style="width: 50%;" /></p>
<p><span class="caption">Figure 4-7: A string slice referring to part of a
<code>String</code></span></p>
<p>With Rusts <code>..</code> range syntax, if you want to start at index 0, you can drop
the value before the two periods. In other words, these are equal:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">#![allow(unused)]
</span><span class="boring">fn main() {
</span>let s = String::from("hello");
let slice = &amp;s[0..2];
let slice = &amp;s[..2];
<span class="boring">}</span></code></pre>
<p>By the same token, if your slice includes the last byte of the <code>String</code>, you
can drop the trailing number. That means these are equal:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">#![allow(unused)]
</span><span class="boring">fn main() {
</span>let s = String::from("hello");
let len = s.len();
let slice = &amp;s[3..len];
let slice = &amp;s[3..];
<span class="boring">}</span></code></pre>
<p>You can also drop both values to take a slice of the entire string. So, these
are equal:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">#![allow(unused)]
</span><span class="boring">fn main() {
</span>let s = String::from("hello");
let len = s.len();
let slice = &amp;s[0..len];
let slice = &amp;s[..];
<span class="boring">}</span></code></pre>
<section class="note" aria-role="note">
<p>Note: String slice range indices must occur at valid UTF-8 character
boundaries. If you attempt to create a string slice in the middle of a
multibyte character, your program will exit with an error.</p>
</section>
<p>With all this information in mind, lets rewrite <code>first_word</code> to return a
slice. The type that signifies “string slice” is written as <code>&amp;str</code>:</p>
<figure class="listing">
<span class="file-name">Filename: src/main.rs</span>
<pre class="playground"><code class="language-rust edition2024">fn first_word(s: &amp;String) -&gt; &amp;str {
let bytes = s.as_bytes();
for (i, &amp;item) in bytes.iter().enumerate() {
if item == b' ' {
return &amp;s[0..i];
}
}
&amp;s[..]
}
<span class="boring">
</span><span class="boring">fn main() {}</span></code></pre>
</figure>
<p>We get the index for the end of the word the same way we did in Listing 4-7, by
looking for the first occurrence of a space. When we find a space, we return a
string slice using the start of the string and the index of the space as the
starting and ending indices.</p>
<p>Now when we call <code>first_word</code>, we get back a single value that is tied to the
underlying data. The value is made up of a reference to the starting point of
the slice and the number of elements in the slice.</p>
<p>Returning a slice would also work for a <code>second_word</code> function:</p>
<pre><code class="language-rust ignore">fn second_word(s: &amp;String) -&gt; &amp;str {</code></pre>
<p>We now have a straightforward API thats much harder to mess up because the
compiler will ensure that the references into the <code>String</code> remain valid.
Remember the bug in the program in Listing 4-8, when we got the index to the
end of the first word but then cleared the string so our index was invalid?
That code was logically incorrect but didnt show any immediate errors. The
problems would show up later if we kept trying to use the first word index with
an emptied string. Slices make this bug impossible and let us know much sooner
that we have a problem with our code. Using the slice version of <code>first_word</code>
will throw a compile-time error:</p>
<figure class="listing">
<span class="file-name">Filename: src/main.rs</span>
<pre><code class="language-rust ignore does_not_compile"><span class="boring">fn first_word(s: &amp;String) -&gt; &amp;str {
</span><span class="boring"> let bytes = s.as_bytes();
</span><span class="boring">
</span><span class="boring"> for (i, &amp;item) in bytes.iter().enumerate() {
</span><span class="boring"> if item == b' ' {
</span><span class="boring"> return &amp;s[0..i];
</span><span class="boring"> }
</span><span class="boring"> }
</span><span class="boring">
</span><span class="boring"> &amp;s[..]
</span><span class="boring">}
</span><span class="boring">
</span>fn main() {
let mut s = String::from("hello world");
let word = first_word(&amp;s);
s.clear(); // error!
println!("the first word is: {word}");
}</code></pre>
</figure>
<p>Heres the compiler error:</p>
<pre><code class="language-console">$ cargo run
Compiling ownership v0.1.0 (file:///projects/ownership)
error[E0502]: cannot borrow `s` as mutable because it is also borrowed as immutable
--&gt; src/main.rs:18:5
|
16 | let word = first_word(&amp;s);
| -- immutable borrow occurs here
17 |
18 | s.clear(); // error!
| ^^^^^^^^^ mutable borrow occurs here
19 |
20 | println!("the first word is: {word}");
| ---- immutable borrow later used here
For more information about this error, try `rustc --explain E0502`.
error: could not compile `ownership` (bin "ownership") due to 1 previous error
</code></pre>
<p>Recall from the borrowing rules that if we have an immutable reference to
something, we cannot also take a mutable reference. Because <code>clear</code> needs to
truncate the <code>String</code>, it needs to get a mutable reference. The <code>println!</code>
after the call to <code>clear</code> uses the reference in <code>word</code>, so the immutable
reference must still be active at that point. Rust disallows the mutable
reference in <code>clear</code> and the immutable reference in <code>word</code> from existing at the
same time, and compilation fails. Not only has Rust made our API easier to use,
but it has also eliminated an entire class of errors at compile time!</p>
<!-- Old headings. Do not remove or links may break. -->
<p><a id="string-literals-are-slices"></a></p>
<h4 id="string-literals-as-slices"><a class="header" href="#string-literals-as-slices">String Literals as Slices</a></h4>
<p>Recall that we talked about string literals being stored inside the binary. Now
that we know about slices, we can properly understand string literals:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">#![allow(unused)]
</span><span class="boring">fn main() {
</span>let s = "Hello, world!";
<span class="boring">}</span></code></pre>
<p>The type of <code>s</code> here is <code>&amp;str</code>: Its a slice pointing to that specific point of
the binary. This is also why string literals are immutable; <code>&amp;str</code> is an
immutable reference.</p>
<h4 id="string-slices-as-parameters"><a class="header" href="#string-slices-as-parameters">String Slices as Parameters</a></h4>
<p>Knowing that you can take slices of literals and <code>String</code> values leads us to
one more improvement on <code>first_word</code>, and thats its signature:</p>
<pre><code class="language-rust ignore">fn first_word(s: &amp;String) -&gt; &amp;str {</code></pre>
<p>A more experienced Rustacean would write the signature shown in Listing 4-9
instead because it allows us to use the same function on both <code>&amp;String</code> values
and <code>&amp;str</code> values.</p>
<figure class="listing" id="listing-4-9">
<pre><code class="language-rust ignore">fn first_word(s: &amp;str) -&gt; &amp;str {
<span class="boring"> let bytes = s.as_bytes();
</span><span class="boring">
</span><span class="boring"> for (i, &amp;item) in bytes.iter().enumerate() {
</span><span class="boring"> if item == b' ' {
</span><span class="boring"> return &amp;s[0..i];
</span><span class="boring"> }
</span><span class="boring"> }
</span><span class="boring">
</span><span class="boring"> &amp;s[..]
</span><span class="boring">}
</span><span class="boring">
</span><span class="boring">fn main() {
</span><span class="boring"> let my_string = String::from("hello world");
</span><span class="boring">
</span><span class="boring"> // `first_word` works on slices of `String`s, whether partial or whole.
</span><span class="boring"> let word = first_word(&amp;my_string[0..6]);
</span><span class="boring"> let word = first_word(&amp;my_string[..]);
</span><span class="boring"> // `first_word` also works on references to `String`s, which are equivalent
</span><span class="boring"> // to whole slices of `String`s.
</span><span class="boring"> let word = first_word(&amp;my_string);
</span><span class="boring">
</span><span class="boring"> let my_string_literal = "hello world";
</span><span class="boring">
</span><span class="boring"> // `first_word` works on slices of string literals, whether partial or
</span><span class="boring"> // whole.
</span><span class="boring"> let word = first_word(&amp;my_string_literal[0..6]);
</span><span class="boring"> let word = first_word(&amp;my_string_literal[..]);
</span><span class="boring">
</span><span class="boring"> // Because string literals *are* string slices already,
</span><span class="boring"> // this works too, without the slice syntax!
</span><span class="boring"> let word = first_word(my_string_literal);
</span><span class="boring">}</span></code></pre>
<figcaption><a href="#listing-4-9">Listing 4-9</a>: Improving the <code>first_word</code> function by using a string slice for the type of the <code>s</code> parameter</figcaption>
</figure>
<p>If we have a string slice, we can pass that directly. If we have a <code>String</code>, we
can pass a slice of the <code>String</code> or a reference to the <code>String</code>. This
flexibility takes advantage of deref coercions, a feature we will cover in
the <a href="../ch15/ch15-02-deref.html#using-deref-coercions-in-functions-and-methods">“Using Deref Coercions in Functions and Methods”</a><!--
ignore --> section of Chapter 15.</p>
<p>Defining a function to take a string slice instead of a reference to a <code>String</code>
makes our API more general and useful without losing any functionality:</p>
<figure class="listing">
<span class="file-name">Filename: src/main.rs</span>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">fn first_word(s: &amp;str) -&gt; &amp;str {
</span><span class="boring"> let bytes = s.as_bytes();
</span><span class="boring">
</span><span class="boring"> for (i, &amp;item) in bytes.iter().enumerate() {
</span><span class="boring"> if item == b' ' {
</span><span class="boring"> return &amp;s[0..i];
</span><span class="boring"> }
</span><span class="boring"> }
</span><span class="boring">
</span><span class="boring"> &amp;s[..]
</span><span class="boring">}
</span><span class="boring">
</span>fn main() {
let my_string = String::from("hello world");
// `first_word` works on slices of `String`s, whether partial or whole.
let word = first_word(&amp;my_string[0..6]);
let word = first_word(&amp;my_string[..]);
// `first_word` also works on references to `String`s, which are equivalent
// to whole slices of `String`s.
let word = first_word(&amp;my_string);
let my_string_literal = "hello world";
// `first_word` works on slices of string literals, whether partial or
// whole.
let word = first_word(&amp;my_string_literal[0..6]);
let word = first_word(&amp;my_string_literal[..]);
// Because string literals *are* string slices already,
// this works too, without the slice syntax!
let word = first_word(my_string_literal);
}</code></pre>
</figure>
<h3 id="other-slices"><a class="header" href="#other-slices">Other Slices</a></h3>
<p>String slices, as you might imagine, are specific to strings. But theres a
more general slice type too. Consider this array:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">#![allow(unused)]
</span><span class="boring">fn main() {
</span>let a = [1, 2, 3, 4, 5];
<span class="boring">}</span></code></pre>
<p>Just as we might want to refer to part of a string, we might want to refer to
part of an array. Wed do so like this:</p>
<pre class="playground"><code class="language-rust edition2024"><span class="boring">#![allow(unused)]
</span><span class="boring">fn main() {
</span>let a = [1, 2, 3, 4, 5];
let slice = &amp;a[1..3];
assert_eq!(slice, &amp;[2, 3]);
<span class="boring">}</span></code></pre>
<p>This slice has the type <code>&amp;[i32]</code>. It works the same way as string slices do, by
storing a reference to the first element and a length. Youll use this kind of
slice for all sorts of other collections. Well discuss these collections in
detail when we talk about vectors in Chapter 8.</p>
<h2 id="summary"><a class="header" href="#summary">Summary</a></h2>
<p>The concepts of ownership, borrowing, and slices ensure memory safety in Rust
programs at compile time. The Rust language gives you control over your memory
usage in the same way as other systems programming languages. But having the
owner of data automatically clean up that data when the owner goes out of scope
means you dont have to write and debug extra code to get this control.</p>
<p>Ownership affects how lots of other parts of Rust work, so well talk about
these concepts further throughout the rest of the book. Lets move on to
Chapter 5 and look at grouping pieces of data together in a <code>struct</code>.</p>
</body>
</html>