Auto merge of #119384 - matthiaskrgr:rollup-hhz9ws0, r=matthiaskrgr

Rollup of 5 pull requests

Successful merges:

 - #119331 (rustdoc-search: count path edits with separate edit limit)
 - #119359 (Simplify Parser::ident_or_error)
 - #119376 (Add regression test for #106630)
 - #119379 (Update `parse_seq` doc)
 - #119380 (Don't suggest writing a bodyless arm if the pattern can never be a never pattern)

r? `@ghost`
`@rustbot` modify labels: rollup

compiler/rustc_parse/src/parser/expr.rs

@@ -2937,7 +2937,13 @@ impl<'a> Parser<'a> {
             let is_almost_fat_arrow = TokenKind::FatArrow
                 .similar_tokens()
                 .is_some_and(|similar_tokens| similar_tokens.contains(&this.token.kind));
-            let mut result = if !is_fat_arrow && !is_almost_fat_arrow {
+
+            // this avoids the compiler saying that a `,` or `}` was expected even though
+            // the pattern isn't a never pattern (and thus an arm body is required)
+            let armless = (!is_fat_arrow && !is_almost_fat_arrow && pat.could_be_never_pattern())
+                || matches!(this.token.kind, token::Comma | token::CloseDelim(Delimiter::Brace));
+
+            let mut result = if armless {
                 // A pattern without a body, allowed for never patterns.
                 arm_body = None;
                 this.expect_one_of(&[token::Comma], &[token::CloseDelim(Delimiter::Brace)]).map(

compiler/rustc_parse/src/parser/mod.rs

@@ -320,9 +320,15 @@ impl TokenType {
     }
 }
 
+/// Used by [`Parser::expect_any_with_type`].
 #[derive(Copy, Clone, Debug)]
 enum TokenExpectType {
+    /// Unencountered tokens are inserted into [`Parser::expected_tokens`].
+    /// See [`Parser::check`].
     Expect,
+
+    /// Unencountered tokens are not inserted into [`Parser::expected_tokens`].
+    /// See [`Parser::check_noexpect`].
     NoExpect,
 }
 
@@ -504,18 +510,10 @@ impl<'a> Parser<'a> {
     }
 
     fn ident_or_err(&mut self, recover: bool) -> PResult<'a, (Ident, /* is_raw */ bool)> {
-        let result = self.token.ident().ok_or_else(|| self.expected_ident_found(recover));
-
-        let (ident, is_raw) = match result {
-            Ok(ident) => ident,
-            Err(err) => match err {
-                // we recovered!
-                Ok(ident) => ident,
-                Err(err) => return Err(err),
-            },
-        };
-
-        Ok((ident, is_raw))
+        match self.token.ident() {
+            Some(ident) => Ok(ident),
+            None => self.expected_ident_found(recover),
+        }
     }
 
     /// Checks if the next token is `tok`, and returns `true` if so.
@@ -766,13 +764,17 @@ impl<'a> Parser<'a> {
         }
     }
 
+    /// Checks if the next token is contained within `kets`, and returns `true` if so.
     fn expect_any_with_type(&mut self, kets: &[&TokenKind], expect: TokenExpectType) -> bool {
         kets.iter().any(|k| match expect {
             TokenExpectType::Expect => self.check(k),
-            TokenExpectType::NoExpect => self.token == **k,
+            TokenExpectType::NoExpect => self.check_noexpect(k),
         })
     }
 
+    /// Parses a sequence until the specified delimiters. The function
+    /// `f` must consume tokens until reaching the next separator or
+    /// closing bracket.
     fn parse_seq_to_before_tokens<T>(
         &mut self,
         kets: &[&TokenKind],
@@ -791,13 +793,15 @@ impl<'a> Parser<'a> {
             }
             if let Some(t) = &sep.sep {
                 if first {
+                    // no separator for the first element
                     first = false;
                 } else {
+                    // check for separator
                     match self.expect(t) {
-                        Ok(false) => {
+                        Ok(false) /* not recovered */ => {
                             self.current_closure.take();
                         }
-                        Ok(true) => {
+                        Ok(true) /* recovered */ => {
                             self.current_closure.take();
                             recovered = true;
                             break;
@@ -965,19 +969,19 @@ impl<'a> Parser<'a> {
         Ok(())
     }
 
-    /// Parses a sequence, not including the closing delimiter. The function
+    /// Parses a sequence, not including the delimiters. The function
     /// `f` must consume tokens until reaching the next separator or
     /// closing bracket.
     fn parse_seq_to_before_end<T>(
         &mut self,
         ket: &TokenKind,
         sep: SeqSep,
         f: impl FnMut(&mut Parser<'a>) -> PResult<'a, T>,
-    ) -> PResult<'a, (ThinVec<T>, bool, bool)> {
+    ) -> PResult<'a, (ThinVec<T>, bool /* trailing */, bool /* recovered */)> {
         self.parse_seq_to_before_tokens(&[ket], sep, TokenExpectType::Expect, f)
     }
 
-    /// Parses a sequence, including the closing delimiter. The function
+    /// Parses a sequence, including only the closing delimiter. The function
     /// `f` must consume tokens until reaching the next separator or
     /// closing bracket.
     fn parse_seq_to_end<T>(
@@ -993,7 +997,7 @@ impl<'a> Parser<'a> {
         Ok((val, trailing))
     }
 
-    /// Parses a sequence, including the closing delimiter. The function
+    /// Parses a sequence, including both delimiters. The function
     /// `f` must consume tokens until reaching the next separator or
     /// closing bracket.
     fn parse_unspanned_seq<T>(
@@ -1002,16 +1006,19 @@ impl<'a> Parser<'a> {
         ket: &TokenKind,
         sep: SeqSep,
         f: impl FnMut(&mut Parser<'a>) -> PResult<'a, T>,
-    ) -> PResult<'a, (ThinVec<T>, bool)> {
+    ) -> PResult<'a, (ThinVec<T>, bool /* trailing */)> {
         self.expect(bra)?;
         self.parse_seq_to_end(ket, sep, f)
     }
 
+    /// Parses a comma-separated sequence, including both delimiters.
+    /// The function `f` must consume tokens until reaching the next separator or
+    /// closing bracket.
     fn parse_delim_comma_seq<T>(
         &mut self,
         delim: Delimiter,
         f: impl FnMut(&mut Parser<'a>) -> PResult<'a, T>,
-    ) -> PResult<'a, (ThinVec<T>, bool)> {
+    ) -> PResult<'a, (ThinVec<T>, bool /* trailing */)> {
         self.parse_unspanned_seq(
             &token::OpenDelim(delim),
             &token::CloseDelim(delim),
@@ -1020,10 +1027,13 @@ impl<'a> Parser<'a> {
         )
     }
 
+    /// Parses a comma-separated sequence delimited by parentheses (e.g. `(x, y)`).
+    /// The function `f` must consume tokens until reaching the next separator or
+    /// closing bracket.
     fn parse_paren_comma_seq<T>(
         &mut self,
         f: impl FnMut(&mut Parser<'a>) -> PResult<'a, T>,
-    ) -> PResult<'a, (ThinVec<T>, bool)> {
+    ) -> PResult<'a, (ThinVec<T>, bool /* trailing */)> {
         self.parse_delim_comma_seq(Delimiter::Parenthesis, f)
     }
 

src/librustdoc/html/static/js/search.js

@@ -1805,11 +1805,20 @@ function initSearch(rawSearchIndex) {
             return unifyFunctionTypes([row], [elem], whereClause, mgens);
         }
 
-        function checkPath(contains, ty, maxEditDistance) {
+        /**
+         * Compute an "edit distance" that ignores missing path elements.
+         * @param {string[]} contains search query path
+         * @param {Row} ty indexed item
+         * @returns {null|number} edit distance
+         */
+        function checkPath(contains, ty) {
             if (contains.length === 0) {
                 return 0;
             }
-            let ret_dist = maxEditDistance + 1;
+            const maxPathEditDistance = Math.floor(
+                contains.reduce((acc, next) => acc + next.length, 0) / 3
+            );
+            let ret_dist = maxPathEditDistance + 1;
             const path = ty.path.split("::");
 
             if (ty.parent && ty.parent.name) {
@@ -1821,15 +1830,23 @@ function initSearch(rawSearchIndex) {
             pathiter: for (let i = length - clength; i >= 0; i -= 1) {
                 let dist_total = 0;
                 for (let x = 0; x < clength; ++x) {
-                    const dist = editDistance(path[i + x], contains[x], maxEditDistance);
-                    if (dist > maxEditDistance) {
-                        continue pathiter;
+                    const [p, c] = [path[i + x], contains[x]];
+                    if (Math.floor((p.length - c.length) / 3) <= maxPathEditDistance &&
+                        p.indexOf(c) !== -1
+                    ) {
+                        // discount distance on substring match
+                        dist_total += Math.floor((p.length - c.length) / 3);
+                    } else {
+                        const dist = editDistance(p, c, maxPathEditDistance);
+                        if (dist > maxPathEditDistance) {
+                            continue pathiter;
+                        }
+                        dist_total += dist;
                     }
-                    dist_total += dist;
                 }
                 ret_dist = Math.min(ret_dist, Math.round(dist_total / clength));
             }
-            return ret_dist;
+            return ret_dist > maxPathEditDistance ? null : ret_dist;
         }
 
         function typePassesFilter(filter, type) {
@@ -2030,8 +2047,8 @@ function initSearch(rawSearchIndex) {
             }
 
             if (elem.fullPath.length > 1) {
-                path_dist = checkPath(elem.pathWithoutLast, row, maxEditDistance);
-                if (path_dist > maxEditDistance) {
+                path_dist = checkPath(elem.pathWithoutLast, row);
+                if (path_dist === null) {
                     return;
                 }
             }
@@ -2045,7 +2062,7 @@ function initSearch(rawSearchIndex) {
 
             const dist = editDistance(row.normalizedName, elem.normalizedPathLast, maxEditDistance);
 
-            if (index === -1 && dist + path_dist > maxEditDistance) {
+            if (index === -1 && dist > maxEditDistance) {
                 return;
             }
 
@@ -2100,13 +2117,9 @@ function initSearch(rawSearchIndex) {
         }
 
         function innerRunQuery() {
-            let queryLen = 0;
-            for (const elem of parsedQuery.elems) {
-                queryLen += elem.name.length;
-            }
-            for (const elem of parsedQuery.returned) {
-                queryLen += elem.name.length;
-            }
+            const queryLen =
+                parsedQuery.elems.reduce((acc, next) => acc + next.pathLast.length, 0) +
+                parsedQuery.returned.reduce((acc, next) => acc + next.pathLast.length, 0);
             const maxEditDistance = Math.floor(queryLen / 3);
 
             /**

tests/rustdoc-js-std/asrawfd.js

@@ -7,7 +7,6 @@ const EXPECTED = {
         // Validate that type alias methods get the correct path.
         { 'path': 'std::os::fd::AsRawFd', 'name': 'as_raw_fd' },
         { 'path': 'std::os::fd::AsRawFd', 'name': 'as_raw_fd' },
-        { 'path': 'std::os::linux::process::PidFd', 'name': 'as_raw_fd' },
         { 'path': 'std::os::fd::RawFd', 'name': 'as_raw_fd' },
     ],
 };

tests/rustdoc-js-std/path-maxeditdistance.js

@@ -0,0 +1,42 @@
+// exact-check
+const FILTER_CRATE = "std";
+const EXPECTED = [
+    {
+        query: 'vec::intoiterator',
+        others: [
+            // trait std::iter::IntoIterator is not the first result
+            { 'path': 'std::vec', 'name': 'IntoIter' },
+            { 'path': 'std::vec::Vec', 'name': 'into_iter' },
+            { 'path': 'std::vec::Drain', 'name': 'into_iter' },
+            { 'path': 'std::vec::IntoIter', 'name': 'into_iter' },
+            { 'path': 'std::vec::ExtractIf', 'name': 'into_iter' },
+            { 'path': 'std::vec::Splice', 'name': 'into_iter' },
+            { 'path': 'std::collections::VecDeque', 'name': 'into_iter' },
+        ],
+    },
+    {
+        query: 'vec::iter',
+        others: [
+            // std::net::ToSocketAttrs::iter should not show up here
+            { 'path': 'std::vec', 'name': 'IntoIter' },
+            { 'path': 'std::vec::Vec', 'name': 'from_iter' },
+            { 'path': 'std::vec::Vec', 'name': 'into_iter' },
+            { 'path': 'std::vec::Drain', 'name': 'into_iter' },
+            { 'path': 'std::vec::IntoIter', 'name': 'into_iter' },
+            { 'path': 'std::vec::ExtractIf', 'name': 'into_iter' },
+            { 'path': 'std::vec::Splice', 'name': 'into_iter' },
+            { 'path': 'std::collections::VecDeque', 'name': 'iter' },
+            { 'path': 'std::collections::VecDeque', 'name': 'iter_mut' },
+            { 'path': 'std::collections::VecDeque', 'name': 'from_iter' },
+            { 'path': 'std::collections::VecDeque', 'name': 'into_iter' },
+        ],
+    },
+    {
+        query: 'slice::itermut',
+        others: [
+            // std::collections::btree_map::itermut should not show up here
+            { 'path': 'std::slice', 'name': 'IterMut' },
+            { 'path': 'std::slice', 'name': 'iter_mut' },
+        ],
+    },
+];

tests/rustdoc-js-std/path-ordering.js

@@ -1,11 +1,20 @@
-const EXPECTED = {
-    query: 'hashset::insert',
-    others: [
-        // ensure hashset::insert comes first
-        { 'path': 'std::collections::hash_set::HashSet', 'name': 'insert' },
-        { 'path': 'std::collections::hash_set::HashSet', 'name': 'get_or_insert' },
-        { 'path': 'std::collections::hash_set::HashSet', 'name': 'get_or_insert_with' },
-        { 'path': 'std::collections::hash_set::HashSet', 'name': 'get_or_insert_owned' },
-        { 'path': 'std::collections::hash_map::HashMap', 'name': 'insert' },
-    ],
-};
+const EXPECTED = [
+    {
+        query: 'hashset::insert',
+        others: [
+            // ensure hashset::insert comes first
+            { 'path': 'std::collections::hash_set::HashSet', 'name': 'insert' },
+            { 'path': 'std::collections::hash_set::HashSet', 'name': 'get_or_insert' },
+            { 'path': 'std::collections::hash_set::HashSet', 'name': 'get_or_insert_with' },
+            { 'path': 'std::collections::hash_set::HashSet', 'name': 'get_or_insert_owned' },
+        ],
+    },
+    {
+        query: 'hash::insert',
+        others: [
+            // ensure hashset/hashmap::insert come first
+            { 'path': 'std::collections::hash_map::HashMap', 'name': 'insert' },
+            { 'path': 'std::collections::hash_set::HashSet', 'name': 'insert' },
+        ],
+    },
+];

tests/rustdoc-js/exact-match.js

@@ -3,6 +3,5 @@ const EXPECTED = {
     'others': [
         { 'path': 'exact_match::Si', 'name': 'pc' },
         { 'path': 'exact_match::Psi', 'name': 'pc' },
-        { 'path': 'exact_match::Si', 'name': 'pa' },
     ],
 };

tests/rustdoc-js/module-substring.js

@@ -1,7 +1,15 @@
-const EXPECTED = {
-    'query': 'ig::pc',
-    'others': [
-        { 'path': 'module_substring::Sig', 'name': 'pc' },
-        { 'path': 'module_substring::Si', 'name': 'pc' },
-    ],
-};
+const EXPECTED = [
+    {
+        'query': 'ig::pc',
+        'others': [
+            { 'path': 'module_substring::Sig', 'name': 'pc' },
+        ],
+    },
+    {
+        'query': 'si::pc',
+        'others': [
+            { 'path': 'module_substring::Si', 'name': 'pc' },
+            { 'path': 'module_substring::Sig', 'name': 'pc' },
+        ],
+    },
+];

tests/rustdoc-js/path-maxeditdistance.js

@@ -0,0 +1,35 @@
+// exact-check
+
+const EXPECTED = [
+    {
+        'query': 'xxxxxxxxxxx::hocuspocusprestidigitation',
+        // do not match abracadabra::hocuspocusprestidigitation
+        'others': [],
+    },
+    {
+        // exact match
+        'query': 'abracadabra::hocuspocusprestidigitation',
+        'others': [
+            { 'path': 'abracadabra', 'name': 'HocusPocusPrestidigitation' },
+        ],
+    },
+    {
+        // swap br/rb; that's edit distance 2, where maxPathEditDistance = 3 (11 / 3)
+        'query': 'arbacadarba::hocuspocusprestidigitation',
+        'others': [
+            { 'path': 'abracadabra', 'name': 'HocusPocusPrestidigitation' },
+        ],
+    },
+    {
+        // truncate 5 chars, where maxEditDistance = 7 (21 / 3)
+        'query': 'abracadarba::hocusprestidigitation',
+        'others': [
+            { 'path': 'abracadabra', 'name': 'HocusPocusPrestidigitation' },
+        ],
+    },
+    {
+        // truncate 9 chars, where maxEditDistance = 5 (17 / 3)
+        'query': 'abracadarba::hprestidigitation',
+        'others': [],
+    },
+];