jruby-prism-parser 0.24.0-java → 1.4.0-java

data/lib/prism/ffi.rb

@@ -1,4 +1,5 @@
 # frozen_string_literal: true
+# typed: ignore
 
 # This file is responsible for mirroring the API provided by the C extension by
 # using FFI to call into the shared library.
@@ -7,14 +8,21 @@ require "rbconfig"
 require "ffi"
 
 module Prism
-  BACKEND = :FFI
-
   module LibRubyParser # :nodoc:
     extend FFI::Library
 
     # Define the library that we will be pulling functions from. Note that this
     # must align with the build shared library from make/rake.
-    ffi_lib File.expand_path("../../build/libprism.#{RbConfig::CONFIG["SOEXT"]}", __dir__)
+    libprism_in_build = File.expand_path("../../build/libprism.#{RbConfig::CONFIG["SOEXT"]}", __dir__)
+    libprism_in_libdir = "#{RbConfig::CONFIG["libdir"]}/prism/libprism.#{RbConfig::CONFIG["SOEXT"]}"
+
+    if File.exist?(libprism_in_build)
+      INCLUDE_DIR = File.expand_path("../../include", __dir__)
+      ffi_lib libprism_in_build
+    else
+      INCLUDE_DIR = "#{RbConfig::CONFIG["libdir"]}/prism/include"
+      ffi_lib libprism_in_libdir
+    end
 
     # Convert a native C type declaration into a symbol that FFI understands.
     # For example:
@@ -24,16 +32,22 @@ module Prism
     #     size_t       -> :size_t
     #     void         -> :void
     #
-    def self.resolve_type(type)
+    def self.resolve_type(type, callbacks)
       type = type.strip
-      type.end_with?("*") ? :pointer : type.delete_prefix("const ").to_sym
+
+      if !type.end_with?("*")
+        type.delete_prefix("const ").to_sym
+      else
+        type = type.delete_suffix("*").rstrip
+        callbacks.include?(type.to_sym) ? type.to_sym : :pointer
+      end
     end
 
     # Read through the given header file and find the declaration of each of the
     # given functions. For each one, define a function with the same name and
     # signature as the C function.
-    def self.load_exported_functions_from(header, *functions)
-      File.foreach(File.expand_path("../../include/#{header}", __dir__)) do |line|
+    def self.load_exported_functions_from(header, *functions, callbacks)
+      File.foreach("#{INCLUDE_DIR}/#{header}") do |line|
         # We only want to attempt to load exported functions.
         next unless line.start_with?("PRISM_EXPORTED_FUNCTION ")
 
@@ -56,24 +70,33 @@ module Prism
 
         # Resolve the type of the argument by dropping the name of the argument
         # first if it is present.
-        arg_types.map! { |type| resolve_type(type.sub(/\w+$/, "")) }
+        arg_types.map! { |type| resolve_type(type.sub(/\w+$/, ""), callbacks) }
 
         # Attach the function using the FFI library.
-        attach_function name, arg_types, resolve_type(return_type)
+        attach_function name, arg_types, resolve_type(return_type, [])
       end
 
       # If we didn't find all of the functions, raise an error.
       raise "Could not find functions #{functions.inspect}" unless functions.empty?
     end
 
+    callback :pm_parse_stream_fgets_t, [:pointer, :int, :pointer], :pointer
+    enum :pm_string_init_result_t, %i[PM_STRING_INIT_SUCCESS PM_STRING_INIT_ERROR_GENERIC PM_STRING_INIT_ERROR_DIRECTORY]
+    enum :pm_string_query_t, [:PM_STRING_QUERY_ERROR, -1, :PM_STRING_QUERY_FALSE, :PM_STRING_QUERY_TRUE]
+
     load_exported_functions_from(
       "prism.h",
       "pm_version",
       "pm_serialize_parse",
+      "pm_serialize_parse_stream",
       "pm_serialize_parse_comments",
       "pm_serialize_lex",
       "pm_serialize_parse_lex",
-      "pm_parse_success_p"
+      "pm_parse_success_p",
+      "pm_string_query_local",
+      "pm_string_query_constant",
+      "pm_string_query_method_name",
+      [:pm_parse_stream_fgets_t]
     )
 
     load_exported_functions_from(
@@ -82,7 +105,8 @@ module Prism
       "pm_buffer_init",
       "pm_buffer_value",
       "pm_buffer_length",
-      "pm_buffer_free"
+      "pm_buffer_free",
+      []
     )
 
     load_exported_functions_from(
@@ -91,7 +115,8 @@ module Prism
       "pm_string_free",
       "pm_string_source",
       "pm_string_length",
-      "pm_string_sizeof"
+      "pm_string_sizeof",
+      []
     )
 
     # This object represents a pm_buffer_t. We only use it as an opaque pointer,
@@ -165,13 +190,26 @@ module Prism
       def self.with_file(filepath)
         raise TypeError unless filepath.is_a?(String)
 
+        # On Windows and Mac, it's expected that filepaths will be encoded in
+        # UTF-8. If they are not, we need to convert them to UTF-8 before
+        # passing them into pm_string_mapped_init.
+        if RbConfig::CONFIG["host_os"].match?(/bccwin|cygwin|djgpp|mingw|mswin|wince|darwin/i) &&
+           (encoding = filepath.encoding) != Encoding::ASCII_8BIT && encoding != Encoding::UTF_8
+          filepath = filepath.encode(Encoding::UTF_8)
+        end
+
         FFI::MemoryPointer.new(SIZEOF) do |pm_string|
-          if LibRubyParser.pm_string_mapped_init(pm_string, filepath)
+          case (result = LibRubyParser.pm_string_mapped_init(pm_string, filepath))
+          when :PM_STRING_INIT_SUCCESS
             pointer = LibRubyParser.pm_string_source(pm_string)
             length = LibRubyParser.pm_string_length(pm_string)
             return yield new(pointer, length, false)
-          else
+          when :PM_STRING_INIT_ERROR_GENERIC
             raise SystemCallError.new(filepath, FFI.errno)
+          when :PM_STRING_INIT_ERROR_DIRECTORY
+            raise Errno::EISDIR.new(filepath)
+          else
+            raise "Unknown error initializing pm_string_t: #{result.inspect}"
           end
         ensure
           LibRubyParser.pm_string_free(pm_string)
@@ -189,8 +227,8 @@ module Prism
 
   class << self
     # Mirror the Prism.dump API by using the serialization API.
-    def dump(code, **options)
-      LibRubyParser::PrismString.with_string(code) { |string| dump_common(string, options) }
+    def dump(source, **options)
+      LibRubyParser::PrismString.with_string(source) { |string| dump_common(string, options) }
     end
 
     # Mirror the Prism.dump_file API by using the serialization API.
@@ -216,13 +254,36 @@ module Prism
     end
 
     # Mirror the Prism.parse_file API by using the serialization API. This uses
-    # native strings instead of Ruby strings because it allows us to use mmap when
-    # it is available.
+    # native strings instead of Ruby strings because it allows us to use mmap
+    # when it is available.
     def parse_file(filepath, **options)
       options[:filepath] = filepath
       LibRubyParser::PrismString.with_file(filepath) { |string| parse_common(string, string.read, options) }
     end
 
+    # Mirror the Prism.parse_stream API by using the serialization API.
+    def parse_stream(stream, **options)
+      LibRubyParser::PrismBuffer.with do |buffer|
+        source = +""
+        callback = -> (string, size, _) {
+          raise "Expected size to be >= 0, got: #{size}" if size <= 0
+
+          if !(line = stream.gets(size - 1)).nil?
+            source << line
+            string.write_string("#{line}\x00", line.bytesize + 1)
+          end
+        }
+
+        # In the pm_serialize_parse_stream function it accepts a pointer to the
+        # IO object as a void* and then passes it through to the callback as the
+        # third argument, but it never touches it itself. As such, since we have
+        # access to the IO object already through the closure of the lambda, we
+        # can pass a null pointer here and not worry.
+        LibRubyParser.pm_serialize_parse_stream(buffer.pointer, nil, callback, dump_options(options))
+        Prism.load(source, buffer.read, options.fetch(:freeze, false))
+      end
+    end
+
     # Mirror the Prism.parse_comments API by using the serialization API.
     def parse_comments(code, **options)
       LibRubyParser::PrismString.with_string(code) { |string| parse_comments_common(string, code, options) }
@@ -252,61 +313,79 @@ module Prism
       LibRubyParser::PrismString.with_string(code) { |string| parse_file_success_common(string, options) }
     end
 
+    # Mirror the Prism.parse_failure? API by using the serialization API.
+    def parse_failure?(code, **options)
+      !parse_success?(code, **options)
+    end
+
     # Mirror the Prism.parse_file_success? API by using the serialization API.
     def parse_file_success?(filepath, **options)
       options[:filepath] = filepath
       LibRubyParser::PrismString.with_file(filepath) { |string| parse_file_success_common(string, options) }
     end
 
+    # Mirror the Prism.parse_file_failure? API by using the serialization API.
+    def parse_file_failure?(filepath, **options)
+      !parse_file_success?(filepath, **options)
+    end
+
+    # Mirror the Prism.profile API by using the serialization API.
+    def profile(source, **options)
+      LibRubyParser::PrismString.with_string(source) do |string|
+        LibRubyParser::PrismBuffer.with do |buffer|
+          LibRubyParser.pm_serialize_parse(buffer.pointer, string.pointer, string.length, dump_options(options))
+          nil
+        end
+      end
+    end
+
+    # Mirror the Prism.profile_file API by using the serialization API.
+    def profile_file(filepath, **options)
+      LibRubyParser::PrismString.with_file(filepath) do |string|
+        LibRubyParser::PrismBuffer.with do |buffer|
+          options[:filepath] = filepath
+          LibRubyParser.pm_serialize_parse(buffer.pointer, string.pointer, string.length, dump_options(options))
+          nil
+        end
+      end
+    end
+
     private
 
     def dump_common(string, options) # :nodoc:
       LibRubyParser::PrismBuffer.with do |buffer|
         LibRubyParser.pm_serialize_parse(buffer.pointer, string.pointer, string.length, dump_options(options))
-        buffer.read
+
+        dumped = buffer.read
+        dumped.freeze if options.fetch(:freeze, false)
+
+        dumped
       end
     end
 
     def lex_common(string, code, options) # :nodoc:
-      serialized = LibRubyParser::PrismBuffer.with do |buffer|
+      LibRubyParser::PrismBuffer.with do |buffer|
         LibRubyParser.pm_serialize_lex(buffer.pointer, string.pointer, string.length, dump_options(options))
-        buffer.read
+        Serialize.load_lex(code, buffer.read, options.fetch(:freeze, false))
       end
-
-      Serialize.load_tokens(Source.new(code), serialized)
     end
 
     def parse_common(string, code, options) # :nodoc:
       serialized = dump_common(string, options)
-      Prism.load(code, serialized)
+      Serialize.load_parse(code, serialized, options.fetch(:freeze, false))
     end
 
     def parse_comments_common(string, code, options) # :nodoc:
       LibRubyParser::PrismBuffer.with do |buffer|
         LibRubyParser.pm_serialize_parse_comments(buffer.pointer, string.pointer, string.length, dump_options(options))
-
-        source = Source.new(code)
-        loader = Serialize::Loader.new(source, buffer.read)
-
-        loader.load_header
-        loader.load_encoding
-        loader.load_start_line
-        loader.load_comments
+        Serialize.load_parse_comments(code, buffer.read, options.fetch(:freeze, false))
       end
     end
 
     def parse_lex_common(string, code, options) # :nodoc:
       LibRubyParser::PrismBuffer.with do |buffer|
         LibRubyParser.pm_serialize_parse_lex(buffer.pointer, string.pointer, string.length, dump_options(options))
-
-        source = Source.new(code)
-        loader = Serialize::Loader.new(source, buffer.read)
-
-        tokens = loader.load_tokens
-        node, comments, magic_comments, data_loc, errors, warnings = loader.load_nodes
-        tokens.each { |token,| token.value.force_encoding(loader.encoding) }
-
-        ParseResult.new([node, tokens], comments, magic_comments, data_loc, errors, warnings, source)
+        Serialize.load_parse_lex(code, buffer.read, options.fetch(:freeze, false))
       end
     end
 
@@ -314,6 +393,40 @@ module Prism
       LibRubyParser.pm_parse_success_p(string.pointer, string.length, dump_options(options))
     end
 
+    # Return the value that should be dumped for the command_line option.
+    def dump_options_command_line(options)
+      command_line = options.fetch(:command_line, "")
+      raise ArgumentError, "command_line must be a string" unless command_line.is_a?(String)
+
+      command_line.each_char.inject(0) do |value, char|
+        case char
+        when "a" then value | 0b000001
+        when "e" then value | 0b000010
+        when "l" then value | 0b000100
+        when "n" then value | 0b001000
+        when "p" then value | 0b010000
+        when "x" then value | 0b100000
+        else raise ArgumentError, "invalid command_line option: #{char}"
+        end
+      end
+    end
+
+    # Return the value that should be dumped for the version option.
+    def dump_options_version(version)
+      case version
+      when nil, "latest"
+        0
+      when /\A3\.3(\.\d+)?\z/
+        1
+      when /\A3\.4(\.\d+)?\z/
+        2
+      when /\A3\.5(\.\d+)?\z/
+        0
+      else
+        raise ArgumentError, "invalid version: #{version}"
+      end
+    end
+
     # Convert the given options into a serialized options string.
     def dump_options(options)
       template = +""
@@ -332,7 +445,7 @@ module Prism
 
       template << "L"
       if (encoding = options[:encoding])
-        name = encoding.name
+        name = encoding.is_a?(Encoding) ? encoding.name : encoding
         values.push(name.bytesize, name.b)
         template << "A*"
       else
@@ -343,17 +456,57 @@ module Prism
       values << (options.fetch(:frozen_string_literal, false) ? 1 : 0)
 
       template << "C"
-      values << { nil => 0, "3.3.0" => 1, "3.4.0" => 0, "latest" => 0 }.fetch(options[:version])
+      values << dump_options_command_line(options)
+
+      template << "C"
+      values << dump_options_version(options[:version])
+
+      template << "C"
+      values << (options[:encoding] == false ? 1 : 0)
+
+      template << "C"
+      values << (options.fetch(:main_script, false) ? 1 : 0)
+
+      template << "C"
+      values << (options.fetch(:partial_script, false) ? 1 : 0)
+
+      template << "C"
+      values << (options.fetch(:freeze, false) ? 1 : 0)
 
       template << "L"
       if (scopes = options[:scopes])
         values << scopes.length
 
         scopes.each do |scope|
+          locals = nil
+          forwarding = 0
+
+          case scope
+          when Array
+            locals = scope
+          when Scope
+            locals = scope.locals
+
+            scope.forwarding.each do |forward|
+              case forward
+              when :*     then forwarding |= 0x1
+              when :**    then forwarding |= 0x2
+              when :&     then forwarding |= 0x4
+              when :"..." then forwarding |= 0x8
+              else raise ArgumentError, "invalid forwarding value: #{forward}"
+              end
+            end
+          else
+            raise TypeError, "wrong argument type #{scope.class.inspect} (expected Array or Prism::Scope)"
+          end
+
           template << "L"
-          values << scope.length
+          values << locals.length
 
-          scope.each do |local|
+          template << "C"
+          values << forwarding
+
+          locals.each do |local|
             name = local.name
             template << "L"
             values << name.bytesize
@@ -369,4 +522,39 @@ module Prism
       values.pack(template)
     end
   end
+
+  # Here we are going to patch StringQuery to put in the class-level methods so
+  # that it can maintain a consistent interface
+  class StringQuery
+    class << self
+      # Mirrors the C extension's StringQuery::local? method.
+      def local?(string)
+        query(LibRubyParser.pm_string_query_local(string, string.bytesize, string.encoding.name))
+      end
+
+      # Mirrors the C extension's StringQuery::constant? method.
+      def constant?(string)
+        query(LibRubyParser.pm_string_query_constant(string, string.bytesize, string.encoding.name))
+      end
+
+      # Mirrors the C extension's StringQuery::method_name? method.
+      def method_name?(string)
+        query(LibRubyParser.pm_string_query_method_name(string, string.bytesize, string.encoding.name))
+      end
+
+      private
+
+      # Parse the enum result and return an appropriate boolean.
+      def query(result)
+        case result
+        when :PM_STRING_QUERY_ERROR
+          raise ArgumentError, "Invalid or non ascii-compatible encoding"
+        when :PM_STRING_QUERY_FALSE
+          false
+        when :PM_STRING_QUERY_TRUE
+          true
+        end
+      end
+    end
+  end
 end