rubydex 0.2.6 → 0.2.7

data/ext/rubydex/signature.c

@@ -1,6 +1,11 @@
 #include "signature.h"
 #include "location.h"
 
+/*
+ * RDoc parser workaround for https://github.com/ruby/rdoc/issues/1744:
+ * mRubydex = rb_define_module("Rubydex")
+ */
+
 static VALUE empty_params = Qundef;
 
 VALUE cSignature;

data/ext/rubydex/utils.c

@@ -39,6 +39,19 @@ void rdxi_check_array_of_strings(VALUE array) {
     }
 }
 
+// Convert a Rust-owned C string to a Ruby string and release it with free_c_string.
+// Returns nil when the Rust side returned NULL.
+VALUE rdxi_owned_c_string_to_ruby(const char *string) {
+    if (string == NULL) {
+        return Qnil;
+    }
+
+    VALUE value = rb_utf8_str_new_cstr(string);
+    free_c_string(string);
+
+    return value;
+}
+
 // Yield body for iterating over declarations
 VALUE rdxi_declarations_yield(VALUE args) {
     VALUE self = rb_ary_entry(args, 0);

data/ext/rubydex/utils.h

@@ -13,6 +13,10 @@ void rdxi_free_str_array(char **array, size_t length);
 // Verify that the Ruby object is an array of strings or raise `TypeError`
 void rdxi_check_array_of_strings(VALUE array);
 
+// Convert a Rust-owned C string to a Ruby string and release it with free_c_string.
+// Returns nil when the Rust side returned NULL.
+VALUE rdxi_owned_c_string_to_ruby(const char *string);
+
 // Yield body for iterating over declarations
 VALUE rdxi_declarations_yield(VALUE args);
 

data/lib/rubydex/errors.rb

@@ -5,4 +5,7 @@ module Rubydex
 
   # Raised when `MethodAliasDefinition#target` walks an alias chain that loops back on itself.
   class AliasCycleError < Error; end
+
+  # Raised by `Graph#load_config` when the requested config file does not exist, cannot be read, or is malformed
+  class ConfigError < Error; end
 end

data/lib/rubydex/graph.rb

@@ -5,49 +5,30 @@ module Rubydex
   #
   # Note: this class is partially defined in C to integrate with the Rust backend
   class Graph
-    IGNORED_DIRECTORIES = [
-      ".bundle",
-      ".claude",
-      ".git",
-      ".github",
-      ".ruby-lsp",
-      ".vscode",
-      "log",
-      "node_modules",
-      "tmp",
-    ].freeze
-
     INDEXABLE_EXTENSIONS = [".rb", ".rake", ".rbs", ".ru"].freeze
 
-    #: String
-    attr_accessor :workspace_path
-
-    #: (?workspace_path: String) -> void
-    def initialize(workspace_path: Dir.pwd)
-      @workspace_path = workspace_path
-
-      exclude_paths(IGNORED_DIRECTORIES.map { |dir| File.join(@workspace_path, dir) })
+    #: (?workspace_path: String?) -> void
+    def initialize(workspace_path: nil)
+      self.workspace_path = workspace_path if workspace_path
     end
 
-    # Index all files and dependencies of the workspace that exists in `@workspace_path`
+    # Index all files and dependencies of the workspace that exists in `workspace_path`
     #: -> Array[String]
     def index_workspace
       index_all(workspace_paths)
     end
 
-    # Returns all workspace paths that should be indexed, excluding directories that we don't need to descend into such
-    # as `.git`, `node_modules`. Also includes any top level Ruby files
+    # Returns all workspace paths that should be indexed
     #
     #: -> Array[String]
     def workspace_paths
       paths = []
+      root = workspace_path
 
-      Dir.each_child(@workspace_path) do |entry|
-        full_path = File.join(@workspace_path, entry)
+      Dir.each_child(root) do |entry|
+        full_path = File.join(root, entry)
 
-        if File.directory?(full_path)
-          paths << full_path unless IGNORED_DIRECTORIES.include?(entry)
-        elsif INDEXABLE_EXTENSIONS.include?(File.extname(entry))
+        if File.directory?(full_path) || INDEXABLE_EXTENSIONS.include?(File.extname(entry))
           paths << full_path
         end
       end

data/lib/rubydex/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Rubydex
-  VERSION = "0.2.6"
+  VERSION = "0.2.7"
 end

data/rbi/rubydex.rbi

@@ -174,10 +174,12 @@ class Rubydex::ConstantDefinition < Rubydex::Definition; end
 class Rubydex::GlobalVariableAliasDefinition < Rubydex::Definition; end
 class Rubydex::GlobalVariableDefinition < Rubydex::Definition; end
 class Rubydex::InstanceVariableDefinition < Rubydex::Definition; end
+
 class Rubydex::MethodAliasDefinition < Rubydex::Definition
   sig { returns(T.nilable(Rubydex::Method)) }
   def target; end
 end
+
 class Rubydex::MethodDefinition < Rubydex::Definition; end
 
 class Rubydex::ModuleDefinition < Rubydex::Definition
@@ -264,6 +266,7 @@ end
 
 class Rubydex::Error < StandardError; end
 class Rubydex::AliasCycleError < Rubydex::Error; end
+class Rubydex::ConfigError < Rubydex::Error; end
 
 class Rubydex::Failure
   sig { params(message: String).void }
@@ -276,8 +279,6 @@ end
 class Rubydex::IntegrityFailure < Rubydex::Failure; end
 
 class Rubydex::Graph
-  IGNORED_DIRECTORIES = T.let(T.unsafe(nil), T::Array[String])
-
   sig { params(workspace_path: T.nilable(String)).void }
   def initialize(workspace_path: nil); end
 
@@ -308,10 +309,17 @@ class Rubydex::Graph
   sig { params(uri: String, source: String, language_id: String).void }
   def index_source(uri, source, language_id); end
 
-  # Index all files and dependencies of the workspace that exists in `@workspace_path`
+  # Index all files and dependencies of the workspace that exists in `workspace_path`
   sig { returns(T::Array[String]) }
   def index_workspace; end
 
+  # Loads configuration, merging its excluded paths into the graph's configuration (the workspace path is never
+  # overridden). With `config_path` (resolved relative to the workspace path), an explicitly named file that does not
+  # exist raises `Rubydex::ConfigError`. With no argument, the default `.rubydex` is loaded if present and ignored if
+  # missing. Raises `Rubydex::ConfigError` if a file cannot be read or is malformed.
+  sig { params(config_path: T.nilable(String)).void }
+  def load_config(config_path = nil); end
+
   sig { returns(T::Enumerable[Rubydex::MethodReference]) }
   def method_references; end
 

data/rust/Cargo.lock

@@ -180,7 +180,7 @@ dependencies = [
  "serde_json",
  "syn",
  "tempfile",
- "toml",
+ "toml 0.8.23",
 ]
 
 [[package]]
@@ -536,9 +536,9 @@ checksum = "a8d1add55171497b4705a648c6b583acafb01d58050a51727785f0b2c8e0a2b2"
 
 [[package]]
 name = "hashbrown"
-version = "0.15.4"
+version = "0.17.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5971ac85611da7067dbfcabef3c70ebb5606018acd9e2a3903a0da507521e0d5"
+checksum = "ed5909b6e89a2db4456e54cd5f673791d7eca6732202bbf2a9cc504fe2f9b84a"
 
 [[package]]
 name = "heck"
@@ -685,9 +685,9 @@ dependencies = [
 
 [[package]]
 name = "indexmap"
-version = "2.10.0"
+version = "2.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fe4cd85333e22411419a0bcae1297d25e58c9443848b11dc6a86fefe8c78a661"
+checksum = "d466e9454f08e4a911e14806c24e16fba1b4c121d1ea474396f396069cf949d9"
 dependencies = [
  "equivalent",
  "hashbrown",
@@ -823,6 +823,12 @@ version = "1.70.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a4895175b425cb1f87721b59f0f286c2092bd4af812243672510e1ac53e2e0ad"
 
+[[package]]
+name = "paste"
+version = "1.0.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "57c0d7b74b563b49d38dae00a0c37d4d6de9b432382b2892f0574ddcae73fd0a"
+
 [[package]]
 name = "pastey"
 version = "0.2.1"
@@ -1065,7 +1071,11 @@ dependencies = [
  "ruby-prism",
  "ruby-rbs",
  "rubydex",
+ "serde",
  "tempfile",
+ "tikv-jemalloc-ctl",
+ "tikv-jemallocator",
+ "toml 1.1.2+spec-1.1.0",
  "url",
  "xxhash-rust",
 ]
@@ -1155,18 +1165,28 @@ dependencies = [
 
 [[package]]
 name = "serde"
-version = "1.0.219"
+version = "1.0.228"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9a8e94ea7f378bd32cbbd37198a4a91436180c5bb472411e48b5ec2e2124ae9e"
+dependencies = [
+ "serde_core",
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_core"
+version = "1.0.228"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5f0e2c6ed6606019b4e29e69dbaba95b11854410e5347d525002456dbbb786b6"
+checksum = "41d385c7d4ca58e59fc732af25c3983b67ac852c1a25000afe1175de458b67ad"
 dependencies = [
  "serde_derive",
 ]
 
 [[package]]
 name = "serde_derive"
-version = "1.0.219"
+version = "1.0.228"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5b0276cf7f2c73365f7157c8123c21cd9a50fbbd844757af28ca1f5925fc2a00"
+checksum = "d540f220d3187173da220f885ab66608367b6574e925011a9353e4badda91d79"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -1205,6 +1225,15 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "serde_spanned"
+version = "1.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6662b5879511e06e8999a8a235d848113e942c9124f211511b16466ee2995f26"
+dependencies = [
+ "serde_core",
+]
+
 [[package]]
 name = "serde_yaml"
 version = "0.9.34+deprecated"
@@ -1315,6 +1344,37 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "tikv-jemalloc-ctl"
+version = "0.7.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3a184c43b8ab2f41df2733b55556e3f5f632f4aeaa205b1bb018f574b7f5f142"
+dependencies = [
+ "libc",
+ "paste",
+ "tikv-jemalloc-sys",
+]
+
+[[package]]
+name = "tikv-jemalloc-sys"
+version = "0.7.1+5.3.1-0-g81034ce1f1373e37dc865038e1bc8eeecf559ce8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1a2825c78386b4ae0314074867860ba9577875de945f05992c38815cbec327f0"
+dependencies = [
+ "cc",
+ "libc",
+]
+
+[[package]]
+name = "tikv-jemallocator"
+version = "0.7.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "249f09e49ab1609436f34c776e84231bead18d6a955f119f939bdc1d847561bd"
+dependencies = [
+ "libc",
+ "tikv-jemalloc-sys",
+]
+
 [[package]]
 name = "tinystr"
 version = "0.8.1"
@@ -1367,11 +1427,26 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "dc1beb996b9d83529a9e75c17a1686767d148d70663143c7854d8b4a09ced362"
 dependencies = [
  "serde",
- "serde_spanned",
- "toml_datetime",
+ "serde_spanned 0.6.9",
+ "toml_datetime 0.6.11",
  "toml_edit",
 ]
 
+[[package]]
+name = "toml"
+version = "1.1.2+spec-1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "81f3d15e84cbcd896376e6730314d59fb5a87f31e4b038454184435cd57defee"
+dependencies = [
+ "indexmap",
+ "serde_core",
+ "serde_spanned 1.1.1",
+ "toml_datetime 1.1.1+spec-1.1.0",
+ "toml_parser",
+ "toml_writer",
+ "winnow 1.0.3",
+]
+
 [[package]]
 name = "toml_datetime"
 version = "0.6.11"
@@ -1381,6 +1456,15 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "toml_datetime"
+version = "1.1.1+spec-1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3165f65f62e28e0115a00b2ebdd37eb6f3b641855f9d636d3cd4103767159ad7"
+dependencies = [
+ "serde_core",
+]
+
 [[package]]
 name = "toml_edit"
 version = "0.22.27"
@@ -1389,10 +1473,19 @@ checksum = "41fe8c660ae4257887cf66394862d21dbca4a6ddd26f04a3560410406a2f819a"
 dependencies = [
  "indexmap",
  "serde",
- "serde_spanned",
- "toml_datetime",
+ "serde_spanned 0.6.9",
+ "toml_datetime 0.6.11",
  "toml_write",
- "winnow",
+ "winnow 0.7.12",
+]
+
+[[package]]
+name = "toml_parser"
+version = "1.1.2+spec-1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a2abe9b86193656635d2411dc43050282ca48aa31c2451210f4202550afb7526"
+dependencies = [
+ "winnow 1.0.3",
 ]
 
 [[package]]
@@ -1401,6 +1494,12 @@ version = "0.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5d99f8c9a7727884afe522e9bd5edbfc91a3312b36a77b5fb8926e4c31a41801"
 
+[[package]]
+name = "toml_writer"
+version = "1.1.1+spec-1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "756daf9b1013ebe47a8776667b466417e2d4c5679d441c26230efd9ef78692db"
+
 [[package]]
 name = "tracing"
 version = "0.1.44"
@@ -1751,6 +1850,12 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "winnow"
+version = "1.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0592e1c9d151f854e6fd382574c3a0855250e1d9b2f99d9281c6e6391af352f1"
+
 [[package]]
 name = "wit-bindgen-rt"
 version = "0.39.0"

data/rust/rubydex/Cargo.toml

@@ -19,6 +19,11 @@ path = "src/main.rs"
 crate-type = ["rlib"]
 
 [features]
+jemalloc = ["dep:tikv-jemallocator"]
+# The Ruby native extension uses a dlopen to load the rubydex shared object. In these scenarios, jemalloc must be linked
+# dynamically or else the process crashes. This has a small performance cost, which we don't need to pay when using the
+# Rust crate by itself, so we gate it with a feature
+jemalloc_dylib = ["jemalloc", "tikv-jemallocator/disable_initial_exec_tls"]
 test_utils = ["dep:tempfile"]
 
 [dependencies]
@@ -32,17 +37,30 @@ bitflags = "2.9"
 bytecount = "0.6.9"
 libc = "0.2"
 line-index = "0.1.2"
+serde = { version = "1.0", features = ["derive"] }
 tempfile = { version = "3.0", optional = true }
 crossbeam-deque = "0.8"
 crossbeam-utils = "0.8"
 crossbeam-channel = "0.5"
+toml = "1.1.2"
+
+[target.'cfg(not(target_os = "windows"))'.dependencies]
+tikv-jemallocator = { version = "0.7.0", optional = true }
 
 [dev-dependencies]
-rubydex = { path = ".", features = ["test_utils"] }
+rubydex = { path = ".", features = ["test_utils", "jemalloc"] }
 tempfile = "3.0"
 assert_cmd = "2.0"
 predicates = "3.1"
 regex = "1.10"
 
+[target.'cfg(not(target_os = "windows"))'.dev-dependencies]
+tikv-jemallocator = { version = "0.7.0", features = ["stats"] }
+tikv-jemalloc-ctl = { version = "0.7.0", features = ["stats"] }
+
+[[bench]]
+name = "graph_memory"
+harness = false
+
 [lints]
 workspace = true

data/rust/rubydex/benches/graph_memory.rs

@@ -0,0 +1,46 @@
+#[cfg(all(feature = "jemalloc", not(target_os = "windows")))]
+mod imp {
+    use std::collections::HashSet;
+
+    use rubydex::{
+        indexing::{self, IndexerBackend},
+        listing,
+        model::graph::Graph,
+        resolution::Resolver,
+    };
+    use tikv_jemalloc_ctl::{epoch, stats};
+
+    /// Advance the jemalloc epoch (stats are cached between epochs) and read the
+    /// number of bytes currently allocated by the application.
+    fn allocated_bytes() -> usize {
+        epoch::advance().expect("failed to advance jemalloc epoch");
+        stats::allocated::read().expect("failed to read stats.allocated (is the `stats` feature on?)")
+    }
+
+    pub fn run() {
+        let paths: Vec<String> = std::env::args().skip(1).collect();
+        let paths = if paths.is_empty() { vec![".".to_string()] } else { paths };
+        let (file_paths, _) = listing::collect_file_paths(paths, &HashSet::new());
+
+        let mut graph = Graph::new();
+        let _ = indexing::index_files(&mut graph, file_paths, IndexerBackend::RubyIndexer);
+        Resolver::new(&mut graph).resolve();
+
+        // Compare the total memory used in the allocator before and after dropping the graph
+        let before_drop = allocated_bytes();
+        drop(graph);
+        let after_drop = allocated_bytes();
+
+        let total_graph_memory = before_drop.saturating_sub(after_drop);
+
+        #[allow(clippy::cast_precision_loss)]
+        let mega_bytes = total_graph_memory as f64 / 1024.0 / 1024.0;
+
+        println!("Total graph memory: {mega_bytes:.2} MB");
+    }
+}
+
+fn main() {
+    #[cfg(all(feature = "jemalloc", not(target_os = "windows")))]
+    imp::run();
+}