refactor(rust): misc fixes & tidying

Author: amaanq

Cargo.lock

@@ -26,6 +26,7 @@ once_cell.workspace = true
 regex.workspace = true
 serde.workspace = true
 serde_json.workspace = true
+tempfile.workspace = true
 
 tree-sitter.workspace = true
 tree-sitter-highlight.workspace = true

cli/loader/Cargo.toml

@@ -130,11 +130,12 @@ pub struct CompileConfig<'a> {
 }
 
 impl<'a> CompileConfig<'a> {
+    #[must_use]
     pub fn new(
         src_path: &'a Path,
         externals: Option<&'a [PathBuf]>,
         output_path: Option<PathBuf>,
-    ) -> CompileConfig<'a> {
+    ) -> Self {
         Self {
             src_path,
             header_paths: vec![src_path],
@@ -449,7 +450,7 @@ impl Loader {
         let parser_path = config.src_path.join("parser.c");
         config.scanner_path = self.get_scanner_path(config.src_path);
 
-        let mut paths_to_check = vec![parser_path.clone()];
+        let mut paths_to_check = vec![parser_path];
 
         if let Some(scanner_path) = config.scanner_path.as_ref() {
             paths_to_check.push(scanner_path.clone());
@@ -488,7 +489,9 @@ impl Loader {
         }
 
         let lock_path = if env::var("CROSS_RUNNER").is_ok() {
-            PathBuf::from("/tmp")
+            tempfile::tempdir()
+                .unwrap()
+                .path()
                 .join("tree-sitter")
                 .join("lock")
                 .join(format!("{}.lock", config.name))
@@ -1021,7 +1024,7 @@ impl Loader {
                         language_name: grammar_json.name.clone(),
                         scope: config_json.scope,
                         language_id,
-                        file_types: config_json.file_types.unwrap_or(Vec::new()),
+                        file_types: config_json.file_types.unwrap_or_default(),
                         content_regex: Self::regex(config_json.content_regex.as_deref()),
                         first_line_regex: Self::regex(config_json.first_line_regex.as_deref()),
                         injection_regex: Self::regex(config_json.injection_regex.as_deref()),
@@ -1048,8 +1051,11 @@ impl Loader {
                             .push(self.language_configurations.len());
                     }
 
-                    self.language_configurations
-                        .push(unsafe { mem::transmute(configuration) });
+                    self.language_configurations.push(unsafe {
+                        mem::transmute::<LanguageConfiguration<'_>, LanguageConfiguration<'static>>(
+                            configuration,
+                        )
+                    });
 
                     if set_current_path_config
                         && self.language_configuration_in_current_path.is_none()
@@ -1088,8 +1094,11 @@ impl Loader {
                 highlight_names: &self.highlight_names,
                 use_all_highlight_names: self.use_all_highlight_names,
             };
-            self.language_configurations
-                .push(unsafe { mem::transmute(configuration) });
+            self.language_configurations.push(unsafe {
+                mem::transmute::<LanguageConfiguration<'_>, LanguageConfiguration<'static>>(
+                    configuration,
+                )
+            });
             self.languages_by_id
                 .push((parser_path.to_owned(), OnceCell::new(), None));
         }
@@ -1327,8 +1336,7 @@ impl<'a> LanguageConfiguration<'a> {
             .unwrap_or_else(|| ranges.last().unwrap());
         error.offset = offset_within_section - range.start;
         error.row = source[range.start..offset_within_section]
-            .chars()
-            .filter(|c| *c == '\n')
+            .matches(|c| c == '\n')
             .count();
         Error::from(error).context(format!("Error in query file {path:?}"))
     }

cli/loader/src/lib.rs

@@ -298,7 +298,7 @@ impl<'a> ParseTableBuilder<'a> {
                         }
                     }
 
-                    reduction_info.precedence = precedence.clone();
+                    reduction_info.precedence.clone_from(precedence);
                     if let Err(i) = reduction_info.symbols.binary_search(&symbol) {
                         reduction_info.symbols.insert(i, symbol);
                     }
@@ -604,13 +604,13 @@ impl<'a> ParseTableBuilder<'a> {
             write!(&mut msg, "  {}", self.symbol_name(symbol)).unwrap();
         }
 
-        write!(
+        writeln!(
             &mut msg,
-            "  •  {}  …\n\n",
+            "  •  {}  …\n",
             self.symbol_name(&conflicting_lookahead)
         )
         .unwrap();
-        write!(&mut msg, "Possible interpretations:\n\n").unwrap();
+        writeln!(&mut msg, "Possible interpretations:\n").unwrap();
 
         let mut interpretations = conflicting_items
             .iter()
@@ -685,7 +685,7 @@ impl<'a> ParseTableBuilder<'a> {
         }
 
         let mut resolution_count = 0;
-        write!(&mut msg, "\nPossible resolutions:\n\n").unwrap();
+        writeln!(&mut msg, "\nPossible resolutions:\n").unwrap();
         let mut shift_items = Vec::new();
         let mut reduce_items = Vec::new();
         for item in conflicting_items {
@@ -961,7 +961,7 @@ fn populate_following_tokens(
             for entry in result.iter_mut() {
                 entry.insert(*extra);
             }
-            result[extra.index] = all_tokens.clone();
+            result[extra.index].clone_from(&all_tokens);
         }
     }
 }

cli/src/generate/build_tables/build_parse_table.rs

@@ -133,7 +133,7 @@ impl<'a> ParseItem<'a> {
 
     /// Create an item like this one, but advanced by one step.
     #[must_use]
-    pub const fn successor(&self) -> ParseItem<'a> {
+    pub const fn successor(&self) -> Self {
         ParseItem {
             variable_index: self.variable_index,
             production: self.production,

cli/src/generate/build_tables/item.rs

@@ -166,19 +166,18 @@ fn parse_rule(json: RuleJSON) -> Rule {
         RuleJSON::PATTERN { value, flags } => Rule::Pattern(
             value,
             flags.map_or(String::new(), |f| {
-                f.chars()
-                    .filter(|c| {
-                        if *c == 'i' {
-                            true
-                        } else {
-                            // silently ignore unicode flags
-                            if *c != 'u' && *c != 'v' {
-                                eprintln!("Warning: unsupported flag {c}");
-                            }
-                            false
+                f.matches(|c| {
+                    if c == 'i' {
+                        true
+                    } else {
+                        // silently ignore unicode flags
+                        if c != 'u' && c != 'v' {
+                            eprintln!("Warning: unsupported flag {c}");
                         }
-                    })
-                    .collect()
+                        false
+                    }
+                })
+                .collect()
             }),
         ),
         RuleJSON::SYMBOL { name } => Rule::NamedSymbol(name),

cli/src/generate/parse_grammar.rs

@@ -154,7 +154,7 @@ impl InlinedProductionMapBuilder {
                 self.productions
                     .iter()
                     .position(|p| *p == production)
-                    .unwrap_or({
+                    .unwrap_or_else(|| {
                         self.productions.push(production);
                         self.productions.len() - 1
                     })

cli/src/generate/prepare_grammar/process_inlines.rs

@@ -29,7 +29,7 @@ macro_rules! add {
 }
 
 macro_rules! add_whitespace {
-    ($this: tt) => {{
+    ($this:tt) => {{
         for _ in 0..$this.indent_level {
             write!(&mut $this.buffer, "  ").unwrap();
         }
@@ -45,13 +45,13 @@ macro_rules! add_line {
 }
 
 macro_rules! indent {
-    ($this: tt) => {
+    ($this:tt) => {
         $this.indent_level += 1;
     };
 }
 
 macro_rules! dedent {
-    ($this: tt) => {
+    ($this:tt) => {
         assert_ne!($this.indent_level, 0);
         $this.indent_level -= 1;
     };
@@ -221,9 +221,8 @@ impl Generator {
                     });
 
                     // Some aliases match an existing symbol in the grammar.
-                    let alias_id;
-                    if let Some(existing_symbol) = existing_symbol {
-                        alias_id = self.symbol_ids[&self.symbol_map[&existing_symbol]].clone();
+                    let alias_id = if let Some(existing_symbol) = existing_symbol {
+                        self.symbol_ids[&self.symbol_map[&existing_symbol]].clone()
                     }
                     // Other aliases don't match any existing symbol, and need their own
                     // identifiers.
@@ -232,12 +231,12 @@ impl Generator {
                             self.unique_aliases.insert(i, alias.clone());
                         }
 
-                        alias_id = if alias.is_named {
+                        if alias.is_named {
                             format!("alias_sym_{}", self.sanitize_identifier(&alias.value))
                         } else {
                             format!("anon_alias_sym_{}", self.sanitize_identifier(&alias.value))
-                        };
-                    }
+                        }
+                    };
 
                     self.alias_ids.entry(alias.clone()).or_insert(alias_id);
                 }

cli/src/generate/render.rs

@@ -870,8 +870,7 @@ fn run() -> Result<()> {
             let open_in_browser = !playground_options.quiet;
             let grammar_path = playground_options
                 .grammar_path
-                .map(PathBuf::from)
-                .unwrap_or(current_dir);
+                .map_or(current_dir, PathBuf::from);
             playground::serve(&grammar_path, open_in_browser)?;
         }
 

cli/src/main.rs

@@ -12,7 +12,7 @@ use tiny_http::{Header, Response, Server};
 use super::wasm;
 
 macro_rules! optional_resource {
-    ($name: tt, $path: tt) => {
+    ($name:tt, $path:tt) => {
         #[cfg(TREE_SITTER_EMBED_WASM_BINDING)]
         fn $name(tree_sitter_dir: Option<&Path>) -> Cow<'static, [u8]> {
             if let Some(tree_sitter_dir) = tree_sitter_dir {

cli/src/playground.rs

@@ -432,9 +432,9 @@ fn write_tests_to_buffer(
         if i > 0 {
             writeln!(buffer)?;
         }
-        write!(
+        writeln!(
             buffer,
-            "{}\n{name}\n{}\n{input}\n{}\n\n{}\n",
+            "{}\n{name}\n{}\n{input}\n{}\n\n{}",
             "=".repeat(*header_delim_len),
             "=".repeat(*header_delim_len),
             "-".repeat(*divider_delim_len),
@@ -662,7 +662,7 @@ fn parse_test_content(name: String, content: &str, file_path: Option<PathBuf>) -
             }
         }
         prev_attributes = attributes;
-        prev_name = test_name.unwrap_or(String::new());
+        prev_name = test_name.unwrap_or_default();
         prev_header_len = header_delim_len;
         prev_header_end = header_range.end;
     }

cli/src/test.rs

@@ -217,7 +217,7 @@ fn test_language_corpus(
 
                 // Perform a random series of edits and reparse.
                 let mut undo_stack = Vec::new();
-                for _ in 0..1 + rand.unsigned(*EDIT_COUNT) {
+                for _ in 0..=rand.unsigned(*EDIT_COUNT) {
                     let edit = get_random_edit(&mut rand, &input);
                     undo_stack.push(invert_edit(&input, &edit));
                     perform_edit(&mut tree, &mut input, &edit).unwrap();

cli/src/tests/corpus_test.rs

@@ -33,25 +33,25 @@ fn detect_language_by_first_line_regex() {
     assert_eq!(config[0].scope.as_ref().unwrap(), "source.strace");
 
     let file_name = strace_dir.path().join("strace.log");
-    std::fs::write(&file_name, "execve\nworld").unwrap();
+    fs::write(&file_name, "execve\nworld").unwrap();
     assert_eq!(
         get_lang_scope(&loader, &file_name),
         Some("source.strace".into())
     );
 
     let file_name = strace_dir.path().join("strace.log");
-    std::fs::write(&file_name, "447845 execve\nworld").unwrap();
+    fs::write(&file_name, "447845 execve\nworld").unwrap();
     assert_eq!(
         get_lang_scope(&loader, &file_name),
         Some("source.strace".into())
     );
 
     let file_name = strace_dir.path().join("strace.log");
-    std::fs::write(&file_name, "hello\nexecve").unwrap();
+    fs::write(&file_name, "hello\nexecve").unwrap();
     assert!(get_lang_scope(&loader, &file_name).is_none());
 
     let file_name = strace_dir.path().join("strace.log");
-    std::fs::write(&file_name, "").unwrap();
+    fs::write(&file_name, "").unwrap();
     assert!(get_lang_scope(&loader, &file_name).is_none());
 
     let dummy_dir = tree_sitter_dir(
@@ -76,7 +76,7 @@ fn detect_language_by_first_line_regex() {
         .find_language_configurations_at_path(dummy_dir.path(), false)
         .unwrap();
     let file_name = dummy_dir.path().join("strace.dummy");
-    std::fs::write(&file_name, "execve").unwrap();
+    fs::write(&file_name, "execve").unwrap();
     assert_eq!(
         get_lang_scope(&loader, &file_name),
         Some("source.dummy".into())
@@ -85,15 +85,14 @@ fn detect_language_by_first_line_regex() {
 
 fn tree_sitter_dir(package_json: &str, name: &str) -> tempfile::TempDir {
     let temp_dir = tempfile::tempdir().unwrap();
-    std::fs::write(temp_dir.path().join("package.json"), package_json).unwrap();
-    std::fs::create_dir(temp_dir.path().join("src")).unwrap();
-    std::fs::create_dir(temp_dir.path().join("src/tree_sitter")).unwrap();
-    std::fs::write(
+    fs::write(temp_dir.path().join("package.json"), package_json).unwrap();
+    fs::create_dir_all(temp_dir.path().join("src/tree_sitter")).unwrap();
+    fs::write(
         temp_dir.path().join("src/grammar.json"),
         format!(r#"{{"name":"{name}"}}"#),
     )
     .unwrap();
-    std::fs::write(
+    fs::write(
         temp_dir.path().join("src/parser.c"),
         format!(
             r##"
@@ -108,15 +107,15 @@ fn tree_sitter_dir(package_json: &str, name: &str) -> tempfile::TempDir {
         ),
     )
     .unwrap();
-    std::fs::write(
+    fs::write(
         temp_dir.path().join("src/tree_sitter/parser.h"),
         include_str!("../../../lib/src/parser.h"),
     )
     .unwrap();
     temp_dir
 }
 
-// if we manage to get the language scope, it means we correctly detected the file-type
+// If we manage to get the language scope, it means we correctly detected the file-type
 fn get_lang_scope(loader: &Loader, file_name: &Path) -> Option<String> {
     loader
         .language_configuration_for_file_name(file_name)

cli/src/tests/detect_language.rs

@@ -31,7 +31,7 @@ impl<'a> ReadRecorder<'a> {
 
     pub fn strings_read(&self) -> Vec<&'a str> {
         let mut result = Vec::new();
-        let mut last_range: Option<Range<usize>> = None;
+        let mut last_range = Option::<Range<usize>>::None;
         for index in &self.indices_read {
             if let Some(ref mut range) = &mut last_range {
                 if range.end == *index {

cli/src/tests/helpers/edits.rs

@@ -15,7 +15,7 @@ impl Rand {
     }
 
     pub fn unsigned(&mut self, max: usize) -> usize {
-        self.0.gen_range(0..max + 1)
+        self.0.gen_range(0..=max)
     }
 
     pub fn words(&mut self, max_count: usize) -> Vec<u8> {

cli/src/tests/helpers/random.rs

@@ -753,8 +753,7 @@ fn to_token_vector<'a>(
                 for (i, l) in s.split('\n').enumerate() {
                     let l = l.trim_end_matches('\r');
                     if i > 0 {
-                        lines.push(line);
-                        line = Vec::new();
+                        lines.push(std::mem::take(&mut line));
                     }
                     if !l.is_empty() {
                         line.push((l, highlights.clone()));