Rollup merge of #79809 - Eric-Arellano:split-once, r=matklad

Dogfood `str_split_once()`

Part of https://github.com/rust-lang/rust/issues/74773.

Beyond increased clarity, this fixes some instances of a common confusion with how `splitn(2)` behaves: the first element will always be `Some()`, regardless of the delimiter, and even if the value is empty.

Given this code:

```rust
fn main() {
    let val = "...";
    let mut iter = val.splitn(2, '=');
    println!("Input: {:?}, first: {:?}, second: {:?}", val, iter.next(), iter.next());
}
```

We get:

```
Input: "no_delimiter", first: Some("no_delimiter"), second: None
Input: "k=v", first: Some("k"), second: Some("v")
Input: "=", first: Some(""), second: Some("")
```

Using `str_split_once()` makes more clear what happens when the delimiter is not found.

11 files changed, 92 insertions, 77 deletions

diff --git a/src/librustdoc/config.rs b/src/librustdoc/config.rs
index e60970af0d3..2d58614b139 100644
--- a/src/librustdoc/config.rs
+++ b/src/librustdoc/config.rs
@@ -397,12 +397,9 @@ impl Options {
             matches
                 .opt_strs("default-setting")
                 .iter()
-                .map(|s| {
-                    let mut kv = s.splitn(2, '=');
-                    // never panics because `splitn` always returns at least one element
-                    let k = kv.next().unwrap().to_string();
-                    let v = kv.next().unwrap_or("true").to_string();
-                    (k, v)
+                .map(|s| match s.split_once('=') {
+                    None => (s.clone(), "true".to_string()),
+                    Some((k, v)) => (k.to_string(), v.to_string()),
                 })
                 .collect(),
         ];
@@ -707,11 +704,9 @@ fn parse_extern_html_roots(
 ) -> Result<BTreeMap<String, String>, &'static str> {
     let mut externs = BTreeMap::new();
     for arg in &matches.opt_strs("extern-html-root-url") {
-        let mut parts = arg.splitn(2, '=');
-        let name = parts.next().ok_or("--extern-html-root-url must not be empty")?;
-        let url = parts.next().ok_or("--extern-html-root-url must be of the form name=url")?;
+        let (name, url) =
+            arg.split_once('=').ok_or("--extern-html-root-url must be of the form name=url")?;
         externs.insert(name.to_string(), url.to_string());
     }
-
     Ok(externs)
 }
diff --git a/src/librustdoc/html/render/mod.rs b/src/librustdoc/html/render/mod.rs
index 901f00b21da..efee4c0be06 100644
--- a/src/librustdoc/html/render/mod.rs
+++ b/src/librustdoc/html/render/mod.rs
@@ -167,10 +167,8 @@ impl Context {
         // `style-suffix.min.css`.  Path::extension would just return `css`
         // which would result in `style.min-suffix.css` which isn't what we
         // want.
-        let mut iter = filename.splitn(2, '.');
-        let base = iter.next().unwrap();
-        let ext = iter.next().unwrap();
-        let filename = format!("{}{}.{}", base, self.shared.resource_suffix, ext,);
+        let (base, ext) = filename.split_once('.').unwrap();
+        let filename = format!("{}{}.{}", base, self.shared.resource_suffix, ext);
         self.dst.join(&filename)
     }
 }
diff --git a/src/librustdoc/lib.rs b/src/librustdoc/lib.rs
index 26bf4b569ff..f851d1a2372 100644
--- a/src/librustdoc/lib.rs
+++ b/src/librustdoc/lib.rs
@@ -16,6 +16,7 @@
 #![feature(once_cell)]
 #![feature(type_ascription)]
 #![feature(split_inclusive)]
+#![feature(str_split_once)]
 #![recursion_limit = "256"]
 
 #[macro_use]
diff --git a/src/librustdoc/passes/collect_intra_doc_links.rs b/src/librustdoc/passes/collect_intra_doc_links.rs
index 551c086a8d4..5ce64c4cd83 100644
--- a/src/librustdoc/passes/collect_intra_doc_links.rs
+++ b/src/librustdoc/passes/collect_intra_doc_links.rs
@@ -435,8 +435,10 @@ impl<'a, 'tcx> LinkCollector<'a, 'tcx> {
 
         // Try looking for methods and associated items.
         let mut split = path_str.rsplitn(2, "::");
-        // this can be an `unwrap()` because we ensure the link is never empty
-        let (item_str, item_name) = split.next().map(|i| (i, Symbol::intern(i))).unwrap();
+        // NB: `split`'s first element is always defined, even if the delimiter was not present.
+        let item_str = split.next().unwrap();
+        assert!(!item_str.is_empty());
+        let item_name = Symbol::intern(item_str);
         let path_root = split
             .next()
             .map(|f| f.to_owned())
diff --git a/src/tools/linkchecker/main.rs b/src/tools/linkchecker/main.rs
index f213944e0ab..dcfe1bb803f 100644
--- a/src/tools/linkchecker/main.rs
+++ b/src/tools/linkchecker/main.rs
@@ -14,6 +14,8 @@
 //! A few exceptions are allowed as there's known bugs in rustdoc, but this
 //! should catch the majority of "broken link" cases.
 
+#![feature(str_split_once)]
+
 use std::collections::hash_map::Entry;
 use std::collections::{HashMap, HashSet};
 use std::env;
@@ -232,11 +234,12 @@ fn check(cache: &mut Cache, root: &Path, file: &Path, errors: &mut bool) -> Opti
         {
             return;
         }
-        let mut parts = url.splitn(2, '#');
-        let url = parts.next().unwrap();
-        let fragment = parts.next();
-        let mut parts = url.splitn(2, '?');
-        let url = parts.next().unwrap();
+        let (url, fragment) = match url.split_once('#') {
+            None => (url, None),
+            Some((url, fragment)) => (url, Some(fragment)),
+        };
+        // NB: the `splitn` always succeeds, even if the delimiter is not present.
+        let url = url.splitn(2, '?').next().unwrap();
 
         // Once we've plucked out the URL, parse it using our base url and
         // then try to extract a file path.
diff --git a/src/tools/tidy/src/cargo.rs b/src/tools/tidy/src/cargo.rs
index 7bdd78a91e7..e06616a59f3 100644
--- a/src/tools/tidy/src/cargo.rs
+++ b/src/tools/tidy/src/cargo.rs
@@ -59,11 +59,10 @@ fn verify(tomlfile: &Path, libfile: &Path, bad: &mut bool) {
             break;
         }
 
-        let mut parts = line.splitn(2, '=');
-        let krate = parts.next().unwrap().trim();
-        if parts.next().is_none() {
-            continue;
-        }
+        let krate = match line.split_once('=') {
+            None => continue,
+            Some((krate, _)) => krate.trim(),
+        };
 
         // Don't worry about depending on core/std while not writing `extern crate
         // core/std` -- that's intentional.
diff --git a/src/tools/tidy/src/error_codes_check.rs b/src/tools/tidy/src/error_codes_check.rs
index 82a5234ac5b..a7199fdfce6 100644
--- a/src/tools/tidy/src/error_codes_check.rs
+++ b/src/tools/tidy/src/error_codes_check.rs
@@ -85,47 +85,61 @@ fn extract_error_codes(
     for line in f.lines() {
         let s = line.trim();
         if !reached_no_explanation && s.starts_with('E') && s.contains("include_str!(\"") {
-            if let Some(err_code) = s.splitn(2, ':').next() {
-                let err_code = err_code.to_owned();
-                if !error_codes.contains_key(&err_code) {
-                    error_codes.insert(err_code.clone(), false);
-                }
-                // Now we extract the tests from the markdown file!
-                let md = some_or_continue!(s.splitn(2, "include_str!(\"").nth(1));
-                let md_file_name = some_or_continue!(md.splitn(2, "\")").next());
-                let path = some_or_continue!(path.parent())
-                    .join(md_file_name)
-                    .canonicalize()
-                    .expect("failed to canonicalize error explanation file path");
-                match read_to_string(&path) {
-                    Ok(content) => {
-                        if !IGNORE_EXPLANATION_CHECK.contains(&err_code.as_str())
-                            && !check_if_error_code_is_test_in_explanation(&content, &err_code)
-                        {
-                            errors.push(format!(
-                                "`{}` doesn't use its own error code in compile_fail example",
-                                path.display(),
-                            ));
-                        }
-                        if check_error_code_explanation(&content, error_codes, err_code) {
-                            errors.push(format!(
-                                "`{}` uses invalid tag `compile-fail` instead of `compile_fail`",
-                                path.display(),
-                            ));
-                        }
+            let err_code = s
+                .split_once(':')
+                .expect(
+                    format!(
+                        "Expected a line with the format `E0xxx: include_str!(\"..\")`, but got {} without a `:` delimiter",
+                        s,
+                    ).as_str()
+                )
+                .0
+                .to_owned();
+            if !error_codes.contains_key(&err_code) {
+                error_codes.insert(err_code.clone(), false);
+            }
+            // Now we extract the tests from the markdown file!
+            let md_file_name = match s.split_once("include_str!(\"") {
+                None => continue,
+                Some((_, md)) => match md.split_once("\")") {
+                    None => continue,
+                    Some((file_name, _)) => file_name,
+                },
+            };
+            let path = some_or_continue!(path.parent())
+                .join(md_file_name)
+                .canonicalize()
+                .expect("failed to canonicalize error explanation file path");
+            match read_to_string(&path) {
+                Ok(content) => {
+                    if !IGNORE_EXPLANATION_CHECK.contains(&err_code.as_str())
+                        && !check_if_error_code_is_test_in_explanation(&content, &err_code)
+                    {
+                        errors.push(format!(
+                            "`{}` doesn't use its own error code in compile_fail example",
+                            path.display(),
+                        ));
                     }
-                    Err(e) => {
-                        eprintln!("Couldn't read `{}`: {}", path.display(), e);
+                    if check_error_code_explanation(&content, error_codes, err_code) {
+                        errors.push(format!(
+                            "`{}` uses invalid tag `compile-fail` instead of `compile_fail`",
+                            path.display(),
+                        ));
                     }
                 }
+                Err(e) => {
+                    eprintln!("Couldn't read `{}`: {}", path.display(), e);
+                }
             }
         } else if reached_no_explanation && s.starts_with('E') {
-            if let Some(err_code) = s.splitn(2, ',').next() {
-                let err_code = err_code.to_owned();
-                if !error_codes.contains_key(&err_code) {
-                    // this check should *never* fail!
-                    error_codes.insert(err_code, false);
-                }
+            let err_code = match s.split_once(',') {
+                None => s,
+                Some((err_code, _)) => err_code,
+            }
+            .to_string();
+            if !error_codes.contains_key(&err_code) {
+                // this check should *never* fail!
+                error_codes.insert(err_code, false);
             }
         } else if s == ";" {
             reached_no_explanation = true;
@@ -137,12 +151,15 @@ fn extract_error_codes_from_tests(f: &str, error_codes: &mut HashMap<String, boo
     for line in f.lines() {
         let s = line.trim();
         if s.starts_with("error[E") || s.starts_with("warning[E") {
-            if let Some(err_code) = s.splitn(2, ']').next() {
-                if let Some(err_code) = err_code.splitn(2, '[').nth(1) {
-                    let nb = error_codes.entry(err_code.to_owned()).or_insert(false);
-                    *nb = true;
-                }
-            }
+            let err_code = match s.split_once(']') {
+                None => continue,
+                Some((err_code, _)) => match err_code.split_once('[') {
+                    None => continue,
+                    Some((_, err_code)) => err_code,
+                },
+            };
+            let nb = error_codes.entry(err_code.to_owned()).or_insert(false);
+            *nb = true;
         }
     }
 }
diff --git a/src/tools/tidy/src/extdeps.rs b/src/tools/tidy/src/extdeps.rs
index 1cf0d24e26f..93d4d3d8047 100644
--- a/src/tools/tidy/src/extdeps.rs
+++ b/src/tools/tidy/src/extdeps.rs
@@ -23,7 +23,7 @@ pub fn check(root: &Path, bad: &mut bool) {
         }
 
         // Extract source value.
-        let source = line.splitn(2, '=').nth(1).unwrap().trim();
+        let source = line.split_once('=').unwrap().1.trim();
 
         // Ensure source is allowed.
         if !ALLOWED_SOURCES.contains(&&*source) {
diff --git a/src/tools/tidy/src/features.rs b/src/tools/tidy/src/features.rs
index d8029ea04f0..3c2880d0d5e 100644
--- a/src/tools/tidy/src/features.rs
+++ b/src/tools/tidy/src/features.rs
@@ -112,6 +112,7 @@ pub fn check(
                 let gate_test_str = "gate-test-";
 
                 let feature_name = match line.find(gate_test_str) {
+                    // NB: the `splitn` always succeeds, even if the delimiter is not present.
                     Some(i) => line[i + gate_test_str.len()..].splitn(2, ' ').next().unwrap(),
                     None => continue,
                 };
diff --git a/src/tools/tidy/src/lib.rs b/src/tools/tidy/src/lib.rs
index cc4c43f0468..e11d293210b 100644
--- a/src/tools/tidy/src/lib.rs
+++ b/src/tools/tidy/src/lib.rs
@@ -3,6 +3,8 @@
 //! This library contains the tidy lints and exposes it
 //! to be used by tools.
 
+#![feature(str_split_once)]
+
 use std::fs::File;
 use std::io::Read;
 use walkdir::{DirEntry, WalkDir};
diff --git a/src/tools/tidy/src/ui_tests.rs b/src/tools/tidy/src/ui_tests.rs
index 47b328dae47..03f4efea983 100644
--- a/src/tools/tidy/src/ui_tests.rs
+++ b/src/tools/tidy/src/ui_tests.rs
@@ -19,14 +19,11 @@ pub fn check(path: &Path, bad: &mut bool) {
                     //
                     // For now, just make sure that there is a corresponding
                     // `$testname.rs` file.
-                    let testname = file_path
-                        .file_name()
-                        .unwrap()
-                        .to_str()
-                        .unwrap()
-                        .splitn(2, '.')
-                        .next()
-                        .unwrap();
+                    //
+                    // NB: We do not use file_stem() as some file names have multiple `.`s and we
+                    // must strip all of them.
+                    let testname =
+                        file_path.file_name().unwrap().to_str().unwrap().split_once('.').unwrap().0;
                     if !file_path.with_file_name(testname).with_extension("rs").exists() {
                         println!("Stray file with UI testing output: {:?}", file_path);
                         *bad = true;