Skip to Content Skip to Search

The Prism Ruby parser.

“Parsing Ruby is suddenly manageable!”

- You, hopefully

Here we are reopening the prism module to provide methods on nodes that aren’t templated and are meant as convenience methods.

Namespace
Methods
D
L
P

Constants

BACKEND = :FFI
 
VERSION = LibRubyParser.pm_version.read_string
 

The version constant is set by reading the result of calling pm_version.

Class Public methods

dump(code, **options)

Mirror the Prism.dump API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 182
def dump(code, **options)
  LibRubyParser::PrismBuffer.with do |buffer|
    LibRubyParser.pm_serialize_parse(buffer.pointer, code, code.bytesize, dump_options(options))
    buffer.read
  end
end

dump_file(filepath, **options)

Mirror the Prism.dump_file API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 190
def dump_file(filepath, **options)
  LibRubyParser::PrismString.with(filepath) do |string|
    dump(string.read, **options, filepath: filepath)
  end
end

lex(code, **options)

Mirror the Prism.lex API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 197
def lex(code, **options)
  LibRubyParser::PrismBuffer.with do |buffer|
    LibRubyParser.pm_serialize_lex(buffer.pointer, code, code.bytesize, dump_options(options))
    Serialize.load_tokens(Source.new(code), buffer.read)
  end
end

Prism::lex_compat(source, **options) → ParseResult

Returns a parse result whose value is an array of tokens that closely resembles the return value of Ripper::lex. The main difference is that the ‘:on_sp` token is not emitted.

For supported options, see Prism::parse.

# File ruby/lib/prism.rb, line 46
def self.lex_compat(source, **options)
  LexCompat.new(source, **options).result
end

lex_file(filepath, **options)

Mirror the Prism.lex_file API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 205
def lex_file(filepath, **options)
  LibRubyParser::PrismString.with(filepath) do |string|
    lex(string.read, **options, filepath: filepath)
  end
end

Prism::lex_ripper(source) → Array

This lexes with the Ripper lex. It drops any space events but otherwise returns the same tokens. Raises SyntaxError if the syntax in source is invalid.

# File ruby/lib/prism.rb, line 56
def self.lex_ripper(source)
  LexRipper.new(source).result
end

Prism::load(source, serialized) → ParseResult

Load the serialized AST using the source as a reference into a tree.

# File ruby/lib/prism.rb, line 64
def self.load(source, serialized)
  Serialize.load(source, serialized)
end

parse(code, **options)

Mirror the Prism.parse API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 212
def parse(code, **options)
  Prism.load(code, dump(code, **options))
end

parse_comments(code, **options)

Mirror the Prism.parse_comments API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 226
def parse_comments(code, **options)
  LibRubyParser::PrismBuffer.with do |buffer|
    LibRubyParser.pm_serialize_parse_comments(buffer.pointer, code, code.bytesize, 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
  end
end

Prism::parse_failure?(source, **options) → bool

Returns true if the source parses with errors.

# File ruby/lib/prism.rb, line 72
def self.parse_failure?(source, **options)
  !parse_success?(source, **options)
end

parse_file(filepath, **options)

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.

# File ruby/lib/prism/ffi.rb, line 219
def parse_file(filepath, **options)
  LibRubyParser::PrismString.with(filepath) do |string|
    parse(string.read, **options, filepath: filepath)
  end
end

parse_file_comments(filepath, **options)

Mirror the Prism.parse_file_comments 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.

# File ruby/lib/prism/ffi.rb, line 243
def parse_file_comments(filepath, **options)
  LibRubyParser::PrismString.with(filepath) do |string|
    parse_comments(string.read, **options, filepath: filepath)
  end
end

Prism::parse_file_failure?(filepath, **options) → bool

Returns true if the file at filepath parses with errors.

# File ruby/lib/prism.rb, line 80
def self.parse_file_failure?(filepath, **options)
  !parse_file_success?(filepath, **options)
end

parse_file_success?(filepath, **options)

Mirror the Prism.parse_file_success? API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 278
def parse_file_success?(filepath, **options)
  LibRubyParser::PrismString.with(filepath) do |string|
    parse_success?(string.read, **options, filepath: filepath)
  end
end

parse_lex(code, **options)

Mirror the Prism.parse_lex API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 250
def parse_lex(code, **options)
  LibRubyParser::PrismBuffer.with do |buffer|
    LibRubyParser.pm_serialize_parse_lex(buffer.pointer, code, code.bytesize, 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)
  end
end

parse_lex_file(filepath, **options)

Mirror the Prism.parse_lex_file API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 266
def parse_lex_file(filepath, **options)
  LibRubyParser::PrismString.with(filepath) do |string|
    parse_lex(string.read, **options, filepath: filepath)
  end
end

parse_success?(code, **options)

Mirror the Prism.parse_success? API by using the serialization API.

# File ruby/lib/prism/ffi.rb, line 273
def parse_success?(code, **options)
  LibRubyParser.pm_parse_success_p(code, code.bytesize, dump_options(options))
end