bootsnap 1.4.1 → 1.10.3

data/lib/bootsnap/load_path_cache/change_observer.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Bootsnap
   module LoadPathCache
     module ChangeObserver
@@ -14,18 +16,20 @@ module Bootsnap
           @lpc_observer.push_paths(self, *entries.map(&:to_s))
           super
         end
+        alias_method :append, :push
 
         def unshift(*entries)
           @lpc_observer.unshift_paths(self, *entries.map(&:to_s))
           super
         end
+        alias_method :prepend, :unshift
 
         def concat(entries)
           @lpc_observer.push_paths(self, *entries.map(&:to_s))
           super
         end
 
-        # uniq! keeps the first occurance of each path, otherwise preserving
+        # uniq! keeps the first occurrence of each path, otherwise preserving
         # order, preserving the effective load path
         def uniq!(*args)
           ret = super
@@ -54,6 +58,7 @@ module Bootsnap
 
       def self.register(observer, arr)
         return if arr.frozen? # can't register observer, but no need to.
+
         arr.instance_variable_set(:@lpc_observer, observer)
         arr.extend(ArrayMixin)
       end

data/lib/bootsnap/load_path_cache/core_ext/kernel_require.rb

@@ -1,65 +1,54 @@
-module Bootsnap
-  module LoadPathCache
-    module CoreExt
-      def self.make_load_error(path)
-        err = LoadError.new("cannot load such file -- #{path}")
-        err.define_singleton_method(:path) { path }
-        err
-      end
-    end
-  end
-end
+# frozen_string_literal: true
 
 module Kernel
-  module_function # rubocop:disable Style/ModuleFunction
+  module_function
 
   alias_method(:require_without_bootsnap, :require)
 
-  # Note that require registers to $LOADED_FEATURES while load does not.
-  def require_with_bootsnap_lfi(path, resolved = nil)
-    Bootsnap::LoadPathCache.loaded_features_index.register(path, resolved) do
-      require_without_bootsnap(resolved || path)
-    end
-  end
-
   def require(path)
-    return false if Bootsnap::LoadPathCache.loaded_features_index.key?(path)
+    string_path = path.to_s
+    return false if Bootsnap::LoadPathCache.loaded_features_index.key?(string_path)
 
-    if (resolved = Bootsnap::LoadPathCache.load_path_cache.find(path))
-      return require_with_bootsnap_lfi(path, resolved)
+    resolved = Bootsnap::LoadPathCache.load_path_cache.find(string_path)
+    if Bootsnap::LoadPathCache::FALLBACK_SCAN.equal?(resolved)
+      if (cursor = Bootsnap::LoadPathCache.loaded_features_index.cursor(string_path))
+        ret = require_without_bootsnap(path)
+        resolved = Bootsnap::LoadPathCache.loaded_features_index.identify(string_path, cursor)
+        Bootsnap::LoadPathCache.loaded_features_index.register(string_path, resolved)
+        return ret
+      else
+        return require_without_bootsnap(path)
+      end
+    elsif false == resolved
+      return false
+    elsif resolved.nil?
+      error = LoadError.new(+"cannot load such file -- #{path}")
+      error.instance_variable_set(:@path, path)
+      raise error
+    else
+      # Note that require registers to $LOADED_FEATURES while load does not.
+      ret = require_without_bootsnap(resolved)
+      Bootsnap::LoadPathCache.loaded_features_index.register(string_path, resolved)
+      return ret
     end
-
-    raise(Bootsnap::LoadPathCache::CoreExt.make_load_error(path))
-  rescue Bootsnap::LoadPathCache::ReturnFalse
-    false
-  rescue Bootsnap::LoadPathCache::FallbackScan
-    require_with_bootsnap_lfi(path)
   end
 
   alias_method(:require_relative_without_bootsnap, :require_relative)
   def require_relative(path)
+    location = caller_locations(1..1).first
     realpath = Bootsnap::LoadPathCache.realpath_cache.call(
-      caller_locations(1..1).first.absolute_path, path
+      location.absolute_path || location.path, path
     )
     require(realpath)
   end
 
   alias_method(:load_without_bootsnap, :load)
   def load(path, wrap = false)
-    if (resolved = Bootsnap::LoadPathCache.load_path_cache.find(path))
-      return load_without_bootsnap(resolved, wrap)
-    end
-
-    # load also allows relative paths from pwd even when not in $:
-    if File.exist?(relative = File.expand_path(path))
-      return load_without_bootsnap(relative, wrap)
+    if (resolved = Bootsnap::LoadPathCache.load_path_cache.find(path, try_extensions: false))
+      load_without_bootsnap(resolved, wrap)
+    else
+      load_without_bootsnap(path, wrap)
     end
-
-    raise(Bootsnap::LoadPathCache::CoreExt.make_load_error(path))
-  rescue Bootsnap::LoadPathCache::ReturnFalse
-    false
-  rescue Bootsnap::LoadPathCache::FallbackScan
-    load_without_bootsnap(path, wrap)
   end
 end
 
@@ -73,10 +62,13 @@ class Module
     # The challenge is that we don't control the point at which the entry gets
     # added to $LOADED_FEATURES and won't be able to hook that modification
     # since it's done in C-land.
-    autoload_without_bootsnap(const, Bootsnap::LoadPathCache.load_path_cache.find(path) || path)
-  rescue Bootsnap::LoadPathCache::ReturnFalse
-    false
-  rescue Bootsnap::LoadPathCache::FallbackScan
-    autoload_without_bootsnap(const, path)
+    resolved = Bootsnap::LoadPathCache.load_path_cache.find(path)
+    if Bootsnap::LoadPathCache::FALLBACK_SCAN.equal?(resolved)
+      autoload_without_bootsnap(const, path)
+    elsif resolved == false
+      return false
+    else
+      autoload_without_bootsnap(const, resolved || path)
+    end
   end
 end

data/lib/bootsnap/load_path_cache/core_ext/loaded_features.rb

@@ -1,7 +1,19 @@
+# frozen_string_literal: true
+
 class << $LOADED_FEATURES
   alias_method(:delete_without_bootsnap, :delete)
   def delete(key)
     Bootsnap::LoadPathCache.loaded_features_index.purge(key)
     delete_without_bootsnap(key)
   end
+
+  alias_method(:reject_without_bootsnap!, :reject!)
+  def reject!(&block)
+    backup = dup
+
+    # FIXME: if no block is passed we'd need to return a decorated iterator
+    reject_without_bootsnap!(&block)
+
+    Bootsnap::LoadPathCache.loaded_features_index.purge_multi(backup - self)
+  end
 end

data/lib/bootsnap/load_path_cache/loaded_features_index.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Bootsnap
   module LoadPathCache
     # LoadedFeaturesIndex partially mirrors an internal structure in ruby that
@@ -24,21 +26,22 @@ module Bootsnap
     class LoadedFeaturesIndex
       def initialize
         @lfi = {}
-        @mutex = defined?(::Mutex) ? ::Mutex.new : ::Thread::Mutex.new # TODO: Remove once Ruby 2.2 support is dropped.
+        @mutex = Mutex.new
 
         # In theory the user could mutate $LOADED_FEATURES and invalidate our
-        # cache. If this ever comes up in practice — or if you, the
-        # enterprising reader, feels inclined to solve this problem — we could
+        # cache. If this ever comes up in practice - or if you, the
+        # enterprising reader, feels inclined to solve this problem - we could
         # parallel the work done with ChangeObserver on $LOAD_PATH to mirror
         # updates to our @lfi.
         $LOADED_FEATURES.each do |feat|
           hash = feat.hash
           $LOAD_PATH.each do |lpe|
             next unless feat.start_with?(lpe)
+
             # /a/b/lib/my/foo.rb
             #          ^^^^^^^^^
             short = feat[(lpe.length + 1)..-1]
-            stripped = strip_extension(short)
+            stripped = strip_extension_if_elidable(short)
             @lfi[short] = hash
             @lfi[stripped] = hash
           end
@@ -55,15 +58,41 @@ module Bootsnap
         end
       end
 
+      def purge_multi(features)
+        rejected_hashes = features.each_with_object({}) { |f, h| h[f.hash] = true }
+        @mutex.synchronize do
+          @lfi.reject! { |_, hash| rejected_hashes.key?(hash) }
+        end
+      end
+
       def key?(feature)
         @mutex.synchronize { @lfi.key?(feature) }
       end
 
+      def cursor(short)
+        unless Bootsnap.absolute_path?(short.to_s)
+          $LOADED_FEATURES.size
+        end
+      end
+
+      def identify(short, cursor)
+        $LOADED_FEATURES[cursor..-1].detect do |feat|
+          offset = 0
+          while (offset = feat.index(short, offset))
+            if feat.index(".", offset + 1) && !feat.index("/", offset + 2)
+              break true
+            else
+              offset += 1
+            end
+          end
+        end
+      end
+
       # There is a relatively uncommon case where we could miss adding an
       # entry:
       #
       # If the user asked for e.g. `require 'bundler'`, and we went through the
-      # `FallbackScan` pathway in `kernel_require.rb` and therefore did not
+      # `FALLBACK_SCAN` pathway in `kernel_require.rb` and therefore did not
       # pass `long` (the full expanded absolute path), then we did are not able
       # to confidently add the `bundler.rb` form to @lfi.
       #
@@ -73,25 +102,19 @@ module Bootsnap
       #    not quite right; or
       # 2. Inspect $LOADED_FEATURES upon return from yield to find the matching
       #    entry.
-      def register(short, long = nil)
-        if long.nil?
-          pat = %r{/#{Regexp.escape(short)}(\.[^/]+)?$}
-          len = $LOADED_FEATURES.size
-          ret = yield
-          long = $LOADED_FEATURES[len..-1].detect { |feat| feat =~ pat }
-        else
-          ret = yield
-        end
+      def register(short, long)
+        return if Bootsnap.absolute_path?(short)
 
         hash = long.hash
 
-        # do we have 'bundler' or 'bundler.rb'?
-        altname = if File.extname(short) != ''
-          # strip the path from 'bundler.rb' -> 'bundler'
-          strip_extension(short)
-        elsif long && (ext = File.extname(long))
-          # get the extension from the expanded path if given
-          # 'bundler' + '.rb'
+        # Do we have a filename with an elidable extension, e.g.,
+        # 'bundler.rb', or 'libgit2.so'?
+        altname = if extension_elidable?(short)
+          # Strip the extension off, e.g. 'bundler.rb' -> 'bundler'.
+          strip_extension_if_elidable(short)
+        elsif long && (ext = File.extname(long.freeze))
+          # We already know the extension of the actual file this
+          # resolves to, so put that back on.
           short + ext
         end
 
@@ -99,17 +122,37 @@ module Bootsnap
           @lfi[short] = hash
           (@lfi[altname] = hash) if altname
         end
-
-        ret
       end
 
       private
 
-      STRIP_EXTENSION = /\.[^.]*?$/
+      STRIP_EXTENSION = /\.[^.]*?$/.freeze
       private_constant(:STRIP_EXTENSION)
 
-      def strip_extension(f)
-        f.sub(STRIP_EXTENSION, '')
+      # Might Ruby automatically search for this extension if
+      # someone tries to 'require' the file without it? E.g. Ruby
+      # will implicitly try 'x.rb' if you ask for 'x'.
+      #
+      # This is complex and platform-dependent, and the Ruby docs are a little
+      # handwavy about what will be tried when and in what order.
+      # So optimistically pretend that all known elidable extensions
+      # will be tried on all platforms, and that people are unlikely
+      # to name files in a way that assumes otherwise.
+      # (E.g. It's unlikely that someone will know that their code
+      # will _never_ run on MacOS, and therefore think they can get away
+      # with calling a Ruby file 'x.dylib.rb' and then requiring it as 'x.dylib'.)
+      #
+      # See <https://ruby-doc.org/core-2.6.4/Kernel.html#method-i-require>.
+      def extension_elidable?(feature)
+        feature.to_s.end_with?(".rb", ".so", ".o", ".dll", ".dylib")
+      end
+
+      def strip_extension_if_elidable(feature)
+        if extension_elidable?(feature)
+          feature.sub(STRIP_EXTENSION, "")
+        else
+          feature
+        end
       end
     end
   end

data/lib/bootsnap/load_path_cache/path.rb

@@ -1,4 +1,6 @@
-require_relative('path_scanner')
+# frozen_string_literal: true
+
+require_relative("path_scanner")
 
 module Bootsnap
   module LoadPathCache
@@ -20,7 +22,7 @@ module Bootsnap
       attr_reader(:path)
 
       def initialize(path)
-        @path = path.to_s
+        @path = path.to_s.freeze
       end
 
       # True if the path exists, but represents a non-directory object
@@ -42,6 +44,7 @@ module Bootsnap
           # set to zero anyway, just in case we change the stability heuristics.
           _, entries, dirs = store.get(expanded_path)
           return [entries, dirs] if entries # cache hit
+
           entries, dirs = scan!
           store.set(expanded_path, [0, entries, dirs])
           return [entries, dirs]
@@ -59,7 +62,7 @@ module Bootsnap
       end
 
       def expanded_path
-        File.expand_path(path)
+        File.expand_path(path).freeze
       end
 
       private
@@ -92,8 +95,8 @@ module Bootsnap
 
       # Built-in ruby lib stuff doesn't change, but things can occasionally be
       # installed into sitedir, which generally lives under libdir.
-      RUBY_LIBDIR  = RbConfig::CONFIG['libdir']
-      RUBY_SITEDIR = RbConfig::CONFIG['sitedir']
+      RUBY_LIBDIR  = RbConfig::CONFIG["libdir"]
+      RUBY_SITEDIR = RbConfig::CONFIG["sitedir"]
 
       def stability
         @stability ||= begin

data/lib/bootsnap/load_path_cache/path_scanner.rb

@@ -1,47 +1,74 @@
-require_relative('../explicit_require')
+# frozen_string_literal: true
+
+require_relative("../explicit_require")
 
 module Bootsnap
   module LoadPathCache
     module PathScanner
-      ALL_FILES = "/{,**/*/**/}*"
       REQUIRABLE_EXTENSIONS = [DOT_RB] + DL_EXTENSIONS
       NORMALIZE_NATIVE_EXTENSIONS = !DL_EXTENSIONS.include?(LoadPathCache::DOT_SO)
-      ALTERNATIVE_NATIVE_EXTENSIONS_PATTERN = /\.(o|bundle|dylib)\z/
+      ALTERNATIVE_NATIVE_EXTENSIONS_PATTERN = /\.(o|bundle|dylib)\z/.freeze
 
       BUNDLE_PATH = if Bootsnap.bundler?
         (Bundler.bundle_path.cleanpath.to_s << LoadPathCache::SLASH).freeze
       else
-        ''.freeze
+        ""
       end
 
-      def self.call(path)
-        path = path.to_s
-
-        relative_slice = (path.size + 1)..-1
-        # If the bundle path is a descendent of this path, we do additional
-        # checks to prevent recursing into the bundle path as we recurse
-        # through this path. We don't want to scan the bundle path because
-        # anything useful in it will be present on other load path items.
-        #
-        # This can happen if, for example, the user adds '.' to the load path,
-        # and the bundle path is '.bundle'.
-        contains_bundle_path = BUNDLE_PATH.start_with?(path)
-
-        dirs = []
-        requirables = []
-
-        Dir.glob(path + ALL_FILES).each do |absolute_path|
-          next if contains_bundle_path && absolute_path.start_with?(BUNDLE_PATH)
-          relative_path = absolute_path.slice(relative_slice)
-
-          if File.directory?(absolute_path)
-            dirs << relative_path
-          elsif REQUIRABLE_EXTENSIONS.include?(File.extname(relative_path))
-            requirables << relative_path
+      class << self
+        def call(path)
+          path = File.expand_path(path.to_s).freeze
+          return [[], []] unless File.directory?(path)
+
+          # If the bundle path is a descendent of this path, we do additional
+          # checks to prevent recursing into the bundle path as we recurse
+          # through this path. We don't want to scan the bundle path because
+          # anything useful in it will be present on other load path items.
+          #
+          # This can happen if, for example, the user adds '.' to the load path,
+          # and the bundle path is '.bundle'.
+          contains_bundle_path = BUNDLE_PATH.start_with?(path)
+
+          dirs = []
+          requirables = []
+          walk(path, nil) do |relative_path, absolute_path, is_directory|
+            if is_directory
+              dirs << os_path(relative_path)
+              !contains_bundle_path || !absolute_path.start_with?(BUNDLE_PATH)
+            elsif relative_path.end_with?(*REQUIRABLE_EXTENSIONS)
+              requirables << os_path(relative_path)
+            end
           end
+          [requirables, dirs]
         end
 
-        [requirables, dirs]
+        def walk(absolute_dir_path, relative_dir_path, &block)
+          Dir.foreach(absolute_dir_path) do |name|
+            next if name.start_with?(".")
+
+            relative_path = relative_dir_path ? File.join(relative_dir_path, name) : name
+
+            absolute_path = "#{absolute_dir_path}/#{name}"
+            if File.directory?(absolute_path)
+              if yield relative_path, absolute_path, true
+                walk(absolute_path, relative_path, &block)
+              end
+            else
+              yield relative_path, absolute_path, false
+            end
+          end
+        end
+
+        if RUBY_VERSION >= "3.1"
+          def os_path(path)
+            path.freeze
+          end
+        else
+          def os_path(path)
+            path.force_encoding(Encoding::US_ASCII) if path.ascii_only?
+            path.freeze
+          end
+        end
       end
     end
   end

data/lib/bootsnap/load_path_cache/realpath_cache.rb

@@ -15,15 +15,16 @@ module Bootsnap
 
       def realpath(caller_location, path)
         base = File.dirname(caller_location)
-        file = find_file(File.expand_path(path, base))
-        dir = File.dirname(file)
-        File.join(dir, File.basename(file))
+        abspath = File.expand_path(path, base).freeze
+        find_file(abspath)
       end
 
       def find_file(name)
-        ['', *CACHED_EXTENSIONS].each do |ext|
+        return File.realpath(name).freeze if File.exist?(name)
+
+        CACHED_EXTENSIONS.each do |ext|
           filename = "#{name}#{ext}"
-          return File.realpath(filename) if File.exist?(filename)
+          return File.realpath(filename).freeze if File.exist?(filename)
         end
         name
       end

data/lib/bootsnap/load_path_cache/store.rb

@@ -1,17 +1,21 @@
-require_relative('../explicit_require')
+# frozen_string_literal: true
 
-Bootsnap::ExplicitRequire.with_gems('msgpack') { require('msgpack') }
-Bootsnap::ExplicitRequire.from_rubylibdir('fileutils')
+require_relative("../explicit_require")
+
+Bootsnap::ExplicitRequire.with_gems("msgpack") { require("msgpack") }
 
 module Bootsnap
   module LoadPathCache
     class Store
+      VERSION_KEY = "__bootsnap_ruby_version__"
+      CURRENT_VERSION = "#{RUBY_REVISION}-#{RUBY_PLATFORM}".freeze # rubocop:disable Style/RedundantFreeze
+
       NestedTransactionError = Class.new(StandardError)
       SetOutsideTransactionNotAllowed = Class.new(StandardError)
 
       def initialize(store_path)
         @store_path = store_path
-        @in_txn = false
+        @txn_mutex = Mutex.new
         @dirty = false
         load_data
       end
@@ -21,7 +25,8 @@ module Bootsnap
       end
 
       def fetch(key)
-        raise(SetOutsideTransactionNotAllowed) unless @in_txn
+        raise(SetOutsideTransactionNotAllowed) unless @txn_mutex.owned?
+
         v = get(key)
         unless v
           @dirty = true
@@ -32,7 +37,8 @@ module Bootsnap
       end
 
       def set(key, value)
-        raise(SetOutsideTransactionNotAllowed) unless @in_txn
+        raise(SetOutsideTransactionNotAllowed) unless @txn_mutex.owned?
+
         if value != @data[key]
           @dirty = true
           @data[key] = value
@@ -40,12 +46,15 @@ module Bootsnap
       end
 
       def transaction
-        raise(NestedTransactionError) if @in_txn
-        @in_txn = true
-        yield
-      ensure
-        commit_transaction
-        @in_txn = false
+        raise(NestedTransactionError) if @txn_mutex.owned?
+
+        @txn_mutex.synchronize do
+          begin
+            yield
+          ensure
+            commit_transaction
+          end
+        end
       end
 
       private
@@ -59,25 +68,47 @@ module Bootsnap
 
       def load_data
         @data = begin
-          MessagePack.load(File.binread(@store_path))
-                # handle malformed data due to upgrade incompatability
-                rescue Errno::ENOENT, MessagePack::MalformedFormatError, MessagePack::UnknownExtTypeError, EOFError
-                  {}
+          data = File.open(@store_path, encoding: Encoding::BINARY) do |io|
+            MessagePack.load(io)
+          end
+          if data.is_a?(Hash) && data[VERSION_KEY] == CURRENT_VERSION
+            data
+          else
+            default_data
+          end
+        # handle malformed data due to upgrade incompatibility
+        rescue Errno::ENOENT, MessagePack::MalformedFormatError, MessagePack::UnknownExtTypeError, EOFError
+          default_data
+        rescue ArgumentError => error
+          if error.message =~ /negative array size/
+            default_data
+          else
+            raise
+          end
         end
       end
 
       def dump_data
+        require "fileutils" unless defined? FileUtils
+
         # Change contents atomically so other processes can't get invalid
         # caches if they read at an inopportune time.
-        tmp = "#{@store_path}.#{Process.pid}.#{(rand * 100000).to_i}.tmp"
+        tmp = "#{@store_path}.#{Process.pid}.#{(rand * 100_000).to_i}.tmp"
         FileUtils.mkpath(File.dirname(tmp))
         exclusive_write = File::Constants::CREAT | File::Constants::EXCL | File::Constants::WRONLY
         # `encoding:` looks redundant wrt `binwrite`, but necessary on windows
         # because binary is part of mode.
-        File.binwrite(tmp, MessagePack.dump(@data), mode: exclusive_write, encoding: Encoding::BINARY)
+        File.open(tmp, mode: exclusive_write, encoding: Encoding::BINARY) do |io|
+          MessagePack.dump(@data, io, freeze: true)
+        end
         FileUtils.mv(tmp, @store_path)
       rescue Errno::EEXIST
         retry
+      rescue SystemCallError
+      end
+
+      def default_data
+        {VERSION_KEY => CURRENT_VERSION}
       end
     end
   end