diff options
Diffstat (limited to 'ext/psych/lib')
37 files changed, 3659 insertions, 0 deletions
diff --git a/ext/psych/lib/psych.rb b/ext/psych/lib/psych.rb new file mode 100644 index 0000000000..850a6d1937 --- /dev/null +++ b/ext/psych/lib/psych.rb @@ -0,0 +1,794 @@ +# frozen_string_literal: true +require 'date' + +require_relative 'psych/versions' +case RUBY_ENGINE +when 'jruby' + require_relative 'psych_jars' + if JRuby::Util.respond_to?(:load_ext) + JRuby::Util.load_ext('org.jruby.ext.psych.PsychLibrary') + else + require 'java'; require 'jruby' + org.jruby.ext.psych.PsychLibrary.new.load(JRuby.runtime, false) + end +else + require 'psych.so' +end +require_relative 'psych/nodes' +require_relative 'psych/streaming' +require_relative 'psych/visitors' +require_relative 'psych/handler' +require_relative 'psych/tree_builder' +require_relative 'psych/parser' +require_relative 'psych/omap' +require_relative 'psych/set' +require_relative 'psych/coder' +require_relative 'psych/stream' +require_relative 'psych/json/tree_builder' +require_relative 'psych/json/stream' +require_relative 'psych/handlers/document_stream' +require_relative 'psych/class_loader' + +### +# = Overview +# +# Psych is a YAML parser and emitter. +# Psych leverages libyaml [Home page: https://pyyaml.org/wiki/LibYAML] +# or [git repo: https://github.com/yaml/libyaml] for its YAML parsing +# and emitting capabilities. In addition to wrapping libyaml, Psych also +# knows how to serialize and de-serialize most Ruby objects to and from +# the YAML format. +# +# = I NEED TO PARSE OR EMIT YAML RIGHT NOW! +# +# # Parse some YAML +# Psych.load("--- foo") # => "foo" +# +# # Emit some YAML +# Psych.dump("foo") # => "--- foo\n...\n" +# { :a => 'b'}.to_yaml # => "---\n:a: b\n" +# +# Got more time on your hands? Keep on reading! +# +# == YAML Parsing +# +# Psych provides a range of interfaces for parsing a YAML document ranging from +# low level to high level, depending on your parsing needs. At the lowest +# level, is an event based parser. Mid level is access to the raw YAML AST, +# and at the highest level is the ability to unmarshal YAML to Ruby objects. +# +# == YAML Emitting +# +# Psych provides a range of interfaces ranging from low to high level for +# producing YAML documents. Very similar to the YAML parsing interfaces, Psych +# provides at the lowest level, an event based system, mid-level is building +# a YAML AST, and the highest level is converting a Ruby object straight to +# a YAML document. +# +# == High-level API +# +# === Parsing +# +# The high level YAML parser provided by Psych simply takes YAML as input and +# returns a Ruby data structure. For information on using the high level parser +# see Psych.load +# +# ==== Reading from a string +# +# Psych.safe_load("--- a") # => 'a' +# Psych.safe_load("---\n - a\n - b") # => ['a', 'b'] +# # From a trusted string: +# Psych.load("--- !ruby/range\nbegin: 0\nend: 42\nexcl: false\n") # => 0..42 +# +# ==== Reading from a file +# +# Psych.safe_load_file("data.yml", permitted_classes: [Date]) +# Psych.load_file("trusted_database.yml") +# +# ==== \Exception handling +# +# begin +# # The second argument changes only the exception contents +# Psych.parse("--- `", "file.txt") +# rescue Psych::SyntaxError => ex +# ex.file # => 'file.txt' +# ex.message # => "(file.txt): found character that cannot start any token" +# end +# +# === Emitting +# +# The high level emitter has the easiest interface. Psych simply takes a Ruby +# data structure and converts it to a YAML document. See Psych.dump for more +# information on dumping a Ruby data structure. +# +# ==== Writing to a string +# +# # Dump an array, get back a YAML string +# Psych.dump(['a', 'b']) # => "---\n- a\n- b\n" +# +# # Dump an array to an IO object +# Psych.dump(['a', 'b'], StringIO.new) # => #<StringIO:0x000001009d0890> +# +# # Dump an array with indentation set +# Psych.dump(['a', ['b']], :indentation => 3) # => "---\n- a\n- - b\n" +# +# # Dump an array to an IO with indentation set +# Psych.dump(['a', ['b']], StringIO.new, :indentation => 3) +# +# ==== Writing to a file +# +# Currently there is no direct API for dumping Ruby structure to file: +# +# File.open('database.yml', 'w') do |file| +# file.write(Psych.dump(['a', 'b'])) +# end +# +# == Mid-level API +# +# === Parsing +# +# Psych provides access to an AST produced from parsing a YAML document. This +# tree is built using the Psych::Parser and Psych::TreeBuilder. The AST can +# be examined and manipulated freely. Please see Psych::parse_stream, +# Psych::Nodes, and Psych::Nodes::Node for more information on dealing with +# YAML syntax trees. +# +# ==== Reading from a string +# +# # Returns Psych::Nodes::Stream +# Psych.parse_stream("---\n - a\n - b") +# +# # Returns Psych::Nodes::Document +# Psych.parse("---\n - a\n - b") +# +# ==== Reading from a file +# +# # Returns Psych::Nodes::Stream +# Psych.parse_stream(File.read('database.yml')) +# +# # Returns Psych::Nodes::Document +# Psych.parse_file('database.yml') +# +# ==== \Exception handling +# +# begin +# # The second argument changes only the exception contents +# Psych.parse("--- `", "file.txt") +# rescue Psych::SyntaxError => ex +# ex.file # => 'file.txt' +# ex.message # => "(file.txt): found character that cannot start any token" +# end +# +# === Emitting +# +# At the mid level is building an AST. This AST is exactly the same as the AST +# used when parsing a YAML document. Users can build an AST by hand and the +# AST knows how to emit itself as a YAML document. See Psych::Nodes, +# Psych::Nodes::Node, and Psych::TreeBuilder for more information on building +# a YAML AST. +# +# ==== Writing to a string +# +# # We need Psych::Nodes::Stream (not Psych::Nodes::Document) +# stream = Psych.parse_stream("---\n - a\n - b") +# +# stream.to_yaml # => "---\n- a\n- b\n" +# +# ==== Writing to a file +# +# # We need Psych::Nodes::Stream (not Psych::Nodes::Document) +# stream = Psych.parse_stream(File.read('database.yml')) +# +# File.open('database.yml', 'w') do |file| +# file.write(stream.to_yaml) +# end +# +# == Low-level API +# +# === Parsing +# +# The lowest level parser should be used when the YAML input is already known, +# and the developer does not want to pay the price of building an AST or +# automatic detection and conversion to Ruby objects. See Psych::Parser for +# more information on using the event based parser. +# +# ==== Reading to Psych::Nodes::Stream structure +# +# parser = Psych::Parser.new(TreeBuilder.new) # => #<Psych::Parser> +# parser = Psych.parser # it's an alias for the above +# +# parser.parse("---\n - a\n - b") # => #<Psych::Parser> +# parser.handler # => #<Psych::TreeBuilder> +# parser.handler.root # => #<Psych::Nodes::Stream> +# +# ==== Receiving an events stream +# +# recorder = Psych::Handlers::Recorder.new +# parser = Psych::Parser.new(recorder) +# +# parser.parse("---\n - a\n - b") +# recorder.events # => [list of [event, args] lists] +# # event is one of: Psych::Handler::EVENTS +# # args are the arguments passed to the event +# +# === Emitting +# +# The lowest level emitter is an event based system. Events are sent to a +# Psych::Emitter object. That object knows how to convert the events to a YAML +# document. This interface should be used when document format is known in +# advance or speed is a concern. See Psych::Emitter for more information. +# +# ==== Writing to a Ruby structure +# +# Psych.parser.parse("--- a") # => #<Psych::Parser> +# +# parser.handler.first # => #<Psych::Nodes::Stream> +# parser.handler.first.to_ruby # => ["a"] +# +# parser.handler.root.first # => #<Psych::Nodes::Document> +# parser.handler.root.first.to_ruby # => "a" +# +# # You can instantiate an Emitter manually +# Psych::Visitors::ToRuby.new.accept(parser.handler.root.first) +# # => "a" + +module Psych + # The version of libyaml Psych is using + LIBYAML_VERSION = Psych.libyaml_version.join('.').freeze + + ### + # Load +yaml+ in to a Ruby data structure. If multiple documents are + # provided, the object contained in the first document will be returned. + # +filename+ will be used in the exception message if any exception + # is raised while parsing. If +yaml+ is empty, it returns + # the specified +fallback+ return value, which defaults to +false+. + # + # Raises a Psych::SyntaxError when a YAML syntax error is detected. + # + # Example: + # + # Psych.unsafe_load("--- a") # => 'a' + # Psych.unsafe_load("---\n - a\n - b") # => ['a', 'b'] + # + # begin + # Psych.unsafe_load("--- `", filename: "file.txt") + # rescue Psych::SyntaxError => ex + # ex.file # => 'file.txt' + # ex.message # => "(file.txt): found character that cannot start any token" + # end + # + # When the optional +symbolize_names+ keyword argument is set to a + # true value, returns symbols for keys in Hash objects (default: strings). + # + # Psych.unsafe_load("---\n foo: bar") # => {"foo"=>"bar"} + # Psych.unsafe_load("---\n foo: bar", symbolize_names: true) # => {:foo=>"bar"} + # + # Raises a TypeError when `yaml` parameter is NilClass + # + # NOTE: This method *should not* be used to parse untrusted documents, such as + # YAML documents that are supplied via user input. Instead, please use the + # load method or the safe_load method. + # + def self.unsafe_load yaml, filename: nil, fallback: false, symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true + result = parse(yaml, filename: filename) + return fallback unless result + result.to_ruby(symbolize_names: symbolize_names, freeze: freeze, strict_integer: strict_integer, parse_symbols: parse_symbols) + end + + ### + # Safely load the yaml string in +yaml+. By default, only the following + # classes are allowed to be deserialized: + # + # * TrueClass + # * FalseClass + # * NilClass + # * Integer + # * Float + # * String + # * Array + # * Hash + # + # Recursive data structures are not allowed by default. Arbitrary classes + # can be allowed by adding those classes to the +permitted_classes+ keyword argument. They are + # additive. For example, to allow Date deserialization: + # + # Psych.safe_load(yaml, permitted_classes: [Date]) + # + # Now the Date class can be loaded in addition to the classes listed above. + # + # Aliases can be explicitly allowed by changing the +aliases+ keyword argument. + # For example: + # + # x = [] + # x << x + # yaml = Psych.dump x + # Psych.safe_load yaml # => raises an exception + # Psych.safe_load yaml, aliases: true # => loads the aliases + # + # A Psych::DisallowedClass exception will be raised if the yaml contains a + # class that isn't in the +permitted_classes+ list. + # + # A Psych::AliasesNotEnabled exception will be raised if the yaml contains aliases + # but the +aliases+ keyword argument is set to false. + # + # +filename+ will be used in the exception message if any exception is raised + # while parsing. + # + # When the optional +symbolize_names+ keyword argument is set to a + # true value, returns symbols for keys in Hash objects (default: strings). + # + # Psych.safe_load("---\n foo: bar") # => {"foo"=>"bar"} + # Psych.safe_load("---\n foo: bar", symbolize_names: true) # => {:foo=>"bar"} + # + def self.safe_load yaml, permitted_classes: [], permitted_symbols: [], aliases: false, filename: nil, fallback: nil, symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true + result = parse(yaml, filename: filename) + return fallback unless result + + class_loader = ClassLoader::Restricted.new(permitted_classes.map(&:to_s), + permitted_symbols.map(&:to_s)) + scanner = ScalarScanner.new class_loader, strict_integer: strict_integer, parse_symbols: parse_symbols + visitor = if aliases + Visitors::ToRuby.new scanner, class_loader, symbolize_names: symbolize_names, freeze: freeze + else + Visitors::NoAliasRuby.new scanner, class_loader, symbolize_names: symbolize_names, freeze: freeze + end + result = visitor.accept result + result + end + + ### + # Load +yaml+ in to a Ruby data structure. If multiple documents are + # provided, the object contained in the first document will be returned. + # +filename+ will be used in the exception message if any exception + # is raised while parsing. If +yaml+ is empty, it returns + # the specified +fallback+ return value, which defaults to +nil+. + # + # Raises a Psych::SyntaxError when a YAML syntax error is detected. + # + # Example: + # + # Psych.load("--- a") # => 'a' + # Psych.load("---\n - a\n - b") # => ['a', 'b'] + # + # begin + # Psych.load("--- `", filename: "file.txt") + # rescue Psych::SyntaxError => ex + # ex.file # => 'file.txt' + # ex.message # => "(file.txt): found character that cannot start any token" + # end + # + # When the optional +symbolize_names+ keyword argument is set to a + # true value, returns symbols for keys in Hash objects (default: strings). + # + # Psych.load("---\n foo: bar") # => {"foo"=>"bar"} + # Psych.load("---\n foo: bar", symbolize_names: true) # => {:foo=>"bar"} + # + # Raises a TypeError when `yaml` parameter is NilClass. This method is + # similar to `safe_load` except that `Symbol` objects are allowed by default. + # + def self.load yaml, permitted_classes: [Symbol], permitted_symbols: [], aliases: false, filename: nil, fallback: nil, symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true + safe_load yaml, permitted_classes: permitted_classes, + permitted_symbols: permitted_symbols, + aliases: aliases, + filename: filename, + fallback: fallback, + symbolize_names: symbolize_names, + freeze: freeze, + strict_integer: strict_integer, + parse_symbols: parse_symbols + end + + ### + # Parse a YAML string in +yaml+. Returns the Psych::Nodes::Document. + # +filename+ is used in the exception message if a Psych::SyntaxError is + # raised. + # + # Raises a Psych::SyntaxError when a YAML syntax error is detected. + # + # Example: + # + # Psych.parse("---\n - a\n - b") # => #<Psych::Nodes::Document:0x00> + # + # begin + # Psych.parse("--- `", filename: "file.txt") + # rescue Psych::SyntaxError => ex + # ex.file # => 'file.txt' + # ex.message # => "(file.txt): found character that cannot start any token" + # end + # + # See Psych::Nodes for more information about YAML AST. + def self.parse yaml, filename: nil + parse_stream(yaml, filename: filename) do |node| + return node + end + + false + end + + ### + # Parse a file at +filename+. Returns the Psych::Nodes::Document. + # + # Raises a Psych::SyntaxError when a YAML syntax error is detected. + def self.parse_file filename, fallback: false + result = File.open filename, 'r:bom|utf-8' do |f| + parse f, filename: filename + end + result || fallback + end + + ### + # Returns a default parser + def self.parser + Psych::Parser.new(TreeBuilder.new) + end + + ### + # Parse a YAML string in +yaml+. Returns the Psych::Nodes::Stream. + # This method can handle multiple YAML documents contained in +yaml+. + # +filename+ is used in the exception message if a Psych::SyntaxError is + # raised. + # + # If a block is given, a Psych::Nodes::Document node will be yielded to the + # block as it's being parsed. + # + # Raises a Psych::SyntaxError when a YAML syntax error is detected. + # + # Example: + # + # Psych.parse_stream("---\n - a\n - b") # => #<Psych::Nodes::Stream:0x00> + # + # Psych.parse_stream("--- a\n--- b") do |node| + # node # => #<Psych::Nodes::Document:0x00> + # end + # + # begin + # Psych.parse_stream("--- `", filename: "file.txt") + # rescue Psych::SyntaxError => ex + # ex.file # => 'file.txt' + # ex.message # => "(file.txt): found character that cannot start any token" + # end + # + # Raises a TypeError when NilClass is passed. + # + # See Psych::Nodes for more information about YAML AST. + def self.parse_stream yaml, filename: nil, &block + if block_given? + parser = Psych::Parser.new(Handlers::DocumentStream.new(&block)) + parser.parse yaml, filename + else + parser = self.parser + parser.parse yaml, filename + parser.handler.root + end + end + + ### + # call-seq: + # Psych.dump(o) -> string of yaml + # Psych.dump(o, options) -> string of yaml + # Psych.dump(o, io) -> io object passed in + # Psych.dump(o, io, options) -> io object passed in + # + # Dump Ruby object +o+ to a YAML string. Optional +options+ may be passed in + # to control the output format. If an IO object is passed in, the YAML will + # be dumped to that IO object. + # + # Currently supported options are: + # + # [<tt>:indentation</tt>] Number of space characters used to indent. + # Acceptable value should be in <tt>0..9</tt> range, + # otherwise option is ignored. + # + # Default: <tt>2</tt>. + # [<tt>:line_width</tt>] Max character to wrap line at. + # For unlimited line width use <tt>-1</tt>. + # + # Default: <tt>0</tt> (meaning "wrap at 81"). + # [<tt>:canonical</tt>] Write "canonical" YAML form (very verbose, yet + # strictly formal). + # + # Default: <tt>false</tt>. + # [<tt>:header</tt>] Write <tt>%YAML [version]</tt> at the beginning of document. + # + # Default: <tt>false</tt>. + # + # [<tt>:stringify_names</tt>] Dump symbol keys in Hash objects as string. + # + # Default: <tt>false</tt>. + # + # Example: + # + # # Dump an array, get back a YAML string + # Psych.dump(['a', 'b']) # => "---\n- a\n- b\n" + # + # # Dump an array to an IO object + # Psych.dump(['a', 'b'], StringIO.new) # => #<StringIO:0x000001009d0890> + # + # # Dump an array with indentation set + # Psych.dump(['a', ['b']], indentation: 3) # => "---\n- a\n- - b\n" + # + # # Dump an array to an IO with indentation set + # Psych.dump(['a', ['b']], StringIO.new, indentation: 3) + # + # # Dump hash with symbol keys as string + # Psych.dump({a: "b"}, stringify_names: true) # => "---\na: b\n" + def self.dump o, io = nil, options = {} + if Hash === io + options = io + io = nil + end + + visitor = Psych::Visitors::YAMLTree.create options + visitor << o + visitor.tree.yaml io, options + end + + ### + # call-seq: + # Psych.safe_dump(o) -> string of yaml + # Psych.safe_dump(o, options) -> string of yaml + # Psych.safe_dump(o, io) -> io object passed in + # Psych.safe_dump(o, io, options) -> io object passed in + # + # Safely dump Ruby object +o+ to a YAML string. Optional +options+ may be passed in + # to control the output format. If an IO object is passed in, the YAML will + # be dumped to that IO object. By default, only the following + # classes are allowed to be serialized: + # + # * TrueClass + # * FalseClass + # * NilClass + # * Integer + # * Float + # * String + # * Array + # * Hash + # + # Arbitrary classes can be allowed by adding those classes to the +permitted_classes+ + # keyword argument. They are additive. For example, to allow Date serialization: + # + # Psych.safe_dump(yaml, permitted_classes: [Date]) + # + # Now the Date class can be dumped in addition to the classes listed above. + # + # A Psych::DisallowedClass exception will be raised if the object contains a + # class that isn't in the +permitted_classes+ list. + # + # Currently supported options are: + # + # [<tt>:indentation</tt>] Number of space characters used to indent. + # Acceptable value should be in <tt>0..9</tt> range, + # otherwise option is ignored. + # + # Default: <tt>2</tt>. + # [<tt>:line_width</tt>] Max character to wrap line at. + # For unlimited line width use <tt>-1</tt>. + # + # Default: <tt>0</tt> (meaning "wrap at 81"). + # [<tt>:canonical</tt>] Write "canonical" YAML form (very verbose, yet + # strictly formal). + # + # Default: <tt>false</tt>. + # [<tt>:header</tt>] Write <tt>%YAML [version]</tt> at the beginning of document. + # + # Default: <tt>false</tt>. + # + # [<tt>:stringify_names</tt>] Dump symbol keys in Hash objects as string. + # + # Default: <tt>false</tt>. + # + # Example: + # + # # Dump an array, get back a YAML string + # Psych.safe_dump(['a', 'b']) # => "---\n- a\n- b\n" + # + # # Dump an array to an IO object + # Psych.safe_dump(['a', 'b'], StringIO.new) # => #<StringIO:0x000001009d0890> + # + # # Dump an array with indentation set + # Psych.safe_dump(['a', ['b']], indentation: 3) # => "---\n- a\n- - b\n" + # + # # Dump an array to an IO with indentation set + # Psych.safe_dump(['a', ['b']], StringIO.new, indentation: 3) + # + # # Dump hash with symbol keys as string + # Psych.dump({a: "b"}, stringify_names: true) # => "---\na: b\n" + def self.safe_dump o, io = nil, options = {} + if Hash === io + options = io + io = nil + end + + visitor = Psych::Visitors::RestrictedYAMLTree.create options + visitor << o + visitor.tree.yaml io, options + end + + ### + # Dump a list of objects as separate documents to a document stream. + # + # Example: + # + # Psych.dump_stream("foo\n ", {}) # => "--- ! \"foo\\n \"\n--- {}\n" + def self.dump_stream *objects + visitor = Psych::Visitors::YAMLTree.create({}) + objects.each do |o| + visitor << o + end + visitor.tree.yaml + end + + ### + # Dump Ruby +object+ to a JSON string. + def self.to_json object + visitor = Psych::Visitors::JSONTree.create + visitor << object + visitor.tree.yaml + end + + ### + # Load multiple documents given in +yaml+. Returns the parsed documents + # as a list. If a block is given, each document will be converted to Ruby + # and passed to the block during parsing + # + # Example: + # + # Psych.load_stream("--- foo\n...\n--- bar\n...") # => ['foo', 'bar'] + # + # list = [] + # Psych.load_stream("--- foo\n...\n--- bar\n...") do |ruby| + # list << ruby + # end + # list # => ['foo', 'bar'] + # + def self.load_stream yaml, filename: nil, fallback: [], **kwargs + result = if block_given? + parse_stream(yaml, filename: filename) do |node| + yield node.to_ruby(**kwargs) + end + else + parse_stream(yaml, filename: filename).children.map { |node| node.to_ruby(**kwargs) } + end + + return fallback if result.is_a?(Array) && result.empty? + result + end + + ### + # Load multiple documents given in +yaml+. Returns the parsed documents + # as a list. + # + # Example: + # + # Psych.safe_load_stream("--- foo\n...\n--- bar\n...") # => ['foo', 'bar'] + # + # list = [] + # Psych.safe_load_stream("--- foo\n...\n--- bar\n...") do |ruby| + # list << ruby + # end + # list # => ['foo', 'bar'] + # + def self.safe_load_stream yaml, filename: nil, permitted_classes: [], aliases: false + documents = parse_stream(yaml, filename: filename).children.map do |child| + stream = Psych::Nodes::Stream.new + stream.children << child + safe_load(stream.to_yaml, permitted_classes: permitted_classes, aliases: aliases) + end + + if block_given? + documents.each { |doc| yield doc } + nil + else + documents + end + end + + ### + # Load the document contained in +filename+. Returns the yaml contained in + # +filename+ as a Ruby object, or if the file is empty, it returns + # the specified +fallback+ return value, which defaults to +false+. + # + # NOTE: This method *should not* be used to parse untrusted documents, such as + # YAML documents that are supplied via user input. Instead, please use the + # safe_load_file method. + def self.unsafe_load_file filename, **kwargs + File.open(filename, 'r:bom|utf-8') { |f| + self.unsafe_load f, filename: filename, **kwargs + } + end + + ### + # Safely loads the document contained in +filename+. Returns the yaml contained in + # +filename+ as a Ruby object, or if the file is empty, it returns + # the specified +fallback+ return value, which defaults to +nil+. + # See safe_load for options. + def self.safe_load_file filename, **kwargs + File.open(filename, 'r:bom|utf-8') { |f| + self.safe_load f, filename: filename, **kwargs + } + end + + ### + # Loads the document contained in +filename+. Returns the yaml contained in + # +filename+ as a Ruby object, or if the file is empty, it returns + # the specified +fallback+ return value, which defaults to +nil+. + # See load for options. + def self.load_file filename, **kwargs + File.open(filename, 'r:bom|utf-8') { |f| + self.load f, filename: filename, **kwargs + } + end + + # :stopdoc: + def self.add_domain_type domain, type_tag, &block + key = ['tag', domain, type_tag].join ':' + domain_types[key] = [key, block] + domain_types["tag:#{type_tag}"] = [key, block] + end + + def self.add_builtin_type type_tag, &block + domain = 'yaml.org,2002' + key = ['tag', domain, type_tag].join ':' + domain_types[key] = [key, block] + end + + def self.remove_type type_tag + domain_types.delete type_tag + end + + def self.add_tag tag, klass + load_tags[tag] = klass.name + dump_tags[klass] = tag + end + + class << self + if defined?(Ractor) + class Config + attr_accessor :load_tags, :dump_tags, :domain_types + def initialize + @load_tags = {} + @dump_tags = {} + @domain_types = {} + end + end + + def config + Ractor.current[:PsychConfig] ||= Config.new + end + + def load_tags + config.load_tags + end + + def dump_tags + config.dump_tags + end + + def domain_types + config.domain_types + end + + def load_tags=(value) + config.load_tags = value + end + + def dump_tags=(value) + config.dump_tags = value + end + + def domain_types=(value) + config.domain_types = value + end + else + attr_accessor :load_tags + attr_accessor :dump_tags + attr_accessor :domain_types + end + end + self.load_tags = {} + self.dump_tags = {} + self.domain_types = {} + # :startdoc: +end + +require_relative 'psych/core_ext' diff --git a/ext/psych/lib/psych/class_loader.rb b/ext/psych/lib/psych/class_loader.rb new file mode 100644 index 0000000000..c8f509720a --- /dev/null +++ b/ext/psych/lib/psych/class_loader.rb @@ -0,0 +1,105 @@ +# frozen_string_literal: true +require_relative 'omap' +require_relative 'set' + +module Psych + class ClassLoader # :nodoc: + BIG_DECIMAL = 'BigDecimal' + COMPLEX = 'Complex' + DATA = 'Data' unless RUBY_VERSION < "3.2" + DATE = 'Date' + DATE_TIME = 'DateTime' + EXCEPTION = 'Exception' + OBJECT = 'Object' + PSYCH_OMAP = 'Psych::Omap' + PSYCH_SET = 'Psych::Set' + RANGE = 'Range' + RATIONAL = 'Rational' + REGEXP = 'Regexp' + STRUCT = 'Struct' + SYMBOL = 'Symbol' + + def initialize + @cache = CACHE.dup + end + + def load klassname + return nil if !klassname || klassname.empty? + + find klassname + end + + def symbolize sym + symbol + sym.to_sym + end + + constants.each do |const| + konst = const_get const + class_eval <<~RUBY, __FILE__, __LINE__ + 1 + def #{const.to_s.downcase} + load #{konst.inspect} + end + RUBY + end + + private + + def find klassname + @cache[klassname] ||= resolve(klassname) + end + + def resolve klassname + name = klassname + retried = false + + begin + path2class(name) + rescue ArgumentError, NameError => ex + unless retried + name = "Struct::#{name}" + retried = ex + retry + end + raise retried + end + end + + CACHE = Hash[constants.map { |const| + val = const_get const + begin + [val, ::Object.const_get(val)] + rescue + nil + end + }.compact].freeze + + class Restricted < ClassLoader + def initialize classes, symbols + @classes = classes + @symbols = symbols + super() + end + + def symbolize sym + return super if @symbols.empty? + + if @symbols.include? sym + super + else + raise DisallowedClass.new('load', 'Symbol') + end + end + + private + + def find klassname + if @classes.include? klassname + super + else + raise DisallowedClass.new('load', klassname) + end + end + end + end +end diff --git a/ext/psych/lib/psych/coder.rb b/ext/psych/lib/psych/coder.rb new file mode 100644 index 0000000000..96a9c3fbad --- /dev/null +++ b/ext/psych/lib/psych/coder.rb @@ -0,0 +1,95 @@ +# frozen_string_literal: true +module Psych + ### + # If an object defines +encode_with+, then an instance of Psych::Coder will + # be passed to the method when the object is being serialized. The Coder + # automatically assumes a Psych::Nodes::Mapping is being emitted. Other + # objects like Sequence and Scalar may be emitted if +seq=+ or +scalar=+ are + # called, respectively. + class Coder + attr_accessor :tag, :style, :implicit, :object + attr_reader :type, :seq + + def initialize tag + @map = {} + @seq = [] + @implicit = false + @type = :map + @tag = tag + @style = Psych::Nodes::Mapping::BLOCK + @scalar = nil + @object = nil + end + + def scalar *args + if args.length > 0 + warn "#{caller[0]}: Coder#scalar(a,b,c) is deprecated" if $VERBOSE + @tag, @scalar, _ = args + @type = :scalar + end + @scalar + end + + # Emit a map. The coder will be yielded to the block. + def map tag = @tag, style = @style + @tag = tag + @style = style + yield self if block_given? + @map + end + + # Emit a scalar with +value+ and +tag+ + def represent_scalar tag, value + self.tag = tag + self.scalar = value + end + + # Emit a sequence with +list+ and +tag+ + def represent_seq tag, list + @tag = tag + self.seq = list + end + + # Emit a sequence with +map+ and +tag+ + def represent_map tag, map + @tag = tag + self.map = map + end + + # Emit an arbitrary object +obj+ and +tag+ + def represent_object tag, obj + @tag = tag + @type = :object + @object = obj + end + + # Emit a scalar with +value+ + def scalar= value + @type = :scalar + @scalar = value + end + + # Emit a map with +value+ + def map= map + @type = :map + @map = map + end + + def []= k, v + @type = :map + @map[k] = v + end + alias :add :[]= + + def [] k + @type = :map + @map[k] + end + + # Emit a sequence of +list+ + def seq= list + @type = :seq + @seq = list + end + end +end diff --git a/ext/psych/lib/psych/core_ext.rb b/ext/psych/lib/psych/core_ext.rb new file mode 100644 index 0000000000..6dfd0f1696 --- /dev/null +++ b/ext/psych/lib/psych/core_ext.rb @@ -0,0 +1,36 @@ +# frozen_string_literal: true +class Object + def self.yaml_tag url + Psych.add_tag(url, self) + end + + ### + # call-seq: to_yaml(options = {}) + # + # Convert an object to YAML. See Psych.dump for more information on the + # available +options+. + def to_yaml options = {} + Psych.dump self, options + end +end + +# Up to Ruby 3.4, Set was a regular object and was dumped as such +# by Pysch. +# Starting from Ruby 4.0 it's a core class written in C, so we have to implement +# #encode_with / #init_with to preserve backward compatibility. +if defined?(::Set) && Set.new.instance_variables.empty? + class Set + def encode_with(coder) + hash = {} + each do |m| + hash[m] = true + end + coder["hash"] = hash + coder + end + + def init_with(coder) + replace(coder["hash"].keys) + end + end +end diff --git a/ext/psych/lib/psych/exception.rb b/ext/psych/lib/psych/exception.rb new file mode 100644 index 0000000000..d7469a4b30 --- /dev/null +++ b/ext/psych/lib/psych/exception.rb @@ -0,0 +1,28 @@ +# frozen_string_literal: true +module Psych + class Exception < RuntimeError + end + + class BadAlias < Exception + end + + # Subclasses `BadAlias` for backwards compatibility + class AliasesNotEnabled < BadAlias + def initialize + super "Alias parsing was not enabled. To enable it, pass `aliases: true` to `Psych::load` or `Psych::safe_load`." + end + end + + # Subclasses `BadAlias` for backwards compatibility + class AnchorNotDefined < BadAlias + def initialize anchor_name + super "An alias referenced an unknown anchor: #{anchor_name}" + end + end + + class DisallowedClass < Exception + def initialize action, klass_name + super "Tried to #{action} unspecified class: #{klass_name}" + end + end +end diff --git a/ext/psych/lib/psych/handler.rb b/ext/psych/lib/psych/handler.rb new file mode 100644 index 0000000000..ad7249ff77 --- /dev/null +++ b/ext/psych/lib/psych/handler.rb @@ -0,0 +1,255 @@ +# frozen_string_literal: true +module Psych + ### + # Psych::Handler is an abstract base class that defines the events used + # when dealing with Psych::Parser. Clients who want to use Psych::Parser + # should implement a class that inherits from Psych::Handler and define + # events that they can handle. + # + # Psych::Handler defines all events that Psych::Parser can possibly send to + # event handlers. + # + # See Psych::Parser for more details + class Handler + ### + # Configuration options for dumping YAML. + class DumperOptions + attr_accessor :line_width, :indentation, :canonical + + def initialize + @line_width = 0 + @indentation = 2 + @canonical = false + end + end + + # Default dumping options + OPTIONS = DumperOptions.new + + # Events that a Handler should respond to. + EVENTS = [ :alias, + :empty, + :end_document, + :end_mapping, + :end_sequence, + :end_stream, + :scalar, + :start_document, + :start_mapping, + :start_sequence, + :start_stream ] + + ### + # Called with +encoding+ when the YAML stream starts. This method is + # called once per stream. A stream may contain multiple documents. + # + # See the constants in Psych::Parser for the possible values of +encoding+. + def start_stream encoding + end + + ### + # Called when the document starts with the declared +version+, + # +tag_directives+, if the document is +implicit+. + # + # +version+ will be an array of integers indicating the YAML version being + # dealt with, +tag_directives+ is a list of tuples indicating the prefix + # and suffix of each tag, and +implicit+ is a boolean indicating whether + # the document is started implicitly. + # + # === Example + # + # Given the following YAML: + # + # %YAML 1.1 + # %TAG ! tag:tenderlovemaking.com,2009: + # --- !squee + # + # The parameters for start_document must be this: + # + # version # => [1, 1] + # tag_directives # => [["!", "tag:tenderlovemaking.com,2009:"]] + # implicit # => false + def start_document version, tag_directives, implicit + end + + ### + # Called with the document ends. +implicit+ is a boolean value indicating + # whether or not the document has an implicit ending. + # + # === Example + # + # Given the following YAML: + # + # --- + # hello world + # + # +implicit+ will be true. Given this YAML: + # + # --- + # hello world + # ... + # + # +implicit+ will be false. + def end_document implicit + end + + ### + # Called when an alias is found to +anchor+. +anchor+ will be the name + # of the anchor found. + # + # === Example + # + # Here we have an example of an array that references itself in YAML: + # + # --- &ponies + # - first element + # - *ponies + # + # &ponies is the anchor, *ponies is the alias. In this case, alias is + # called with "ponies". + def alias anchor + end + + ### + # Called when a scalar +value+ is found. The scalar may have an + # +anchor+, a +tag+, be implicitly +plain+ or implicitly +quoted+ + # + # +value+ is the string value of the scalar + # +anchor+ is an associated anchor or nil + # +tag+ is an associated tag or nil + # +plain+ is a boolean value + # +quoted+ is a boolean value + # +style+ is an integer indicating the string style + # + # See the constants in Psych::Nodes::Scalar for the possible values of + # +style+ + # + # === Example + # + # Here is a YAML document that exercises most of the possible ways this + # method can be called: + # + # --- + # - !str "foo" + # - &anchor fun + # - many + # lines + # - | + # many + # newlines + # + # The above YAML document contains a list with four strings. Here are + # the parameters sent to this method in the same order: + # + # # value anchor tag plain quoted style + # ["foo", nil, "!str", false, false, 3 ] + # ["fun", "anchor", nil, true, false, 1 ] + # ["many lines", nil, nil, true, false, 1 ] + # ["many\nnewlines\n", nil, nil, false, true, 4 ] + # + def scalar value, anchor, tag, plain, quoted, style + end + + ### + # Called when a sequence is started. + # + # +anchor+ is the anchor associated with the sequence or nil. + # +tag+ is the tag associated with the sequence or nil. + # +implicit+ a boolean indicating whether or not the sequence was implicitly + # started. + # +style+ is an integer indicating the list style. + # + # See the constants in Psych::Nodes::Sequence for the possible values of + # +style+. + # + # === Example + # + # Here is a YAML document that exercises most of the possible ways this + # method can be called: + # + # --- + # - !!seq [ + # a + # ] + # - &pewpew + # - b + # + # The above YAML document consists of three lists, an outer list that + # contains two inner lists. Here is a matrix of the parameters sent + # to represent these lists: + # + # # anchor tag implicit style + # [nil, nil, true, 1 ] + # [nil, "tag:yaml.org,2002:seq", false, 2 ] + # ["pewpew", nil, true, 1 ] + + def start_sequence anchor, tag, implicit, style + end + + ### + # Called when a sequence ends. + def end_sequence + end + + ### + # Called when a map starts. + # + # +anchor+ is the anchor associated with the map or +nil+. + # +tag+ is the tag associated with the map or +nil+. + # +implicit+ is a boolean indicating whether or not the map was implicitly + # started. + # +style+ is an integer indicating the mapping style. + # + # See the constants in Psych::Nodes::Mapping for the possible values of + # +style+. + # + # === Example + # + # Here is a YAML document that exercises most of the possible ways this + # method can be called: + # + # --- + # k: !!map { hello: world } + # v: &pewpew + # hello: world + # + # The above YAML document consists of three maps, an outer map that contains + # two inner maps. Below is a matrix of the parameters sent in order to + # represent these three maps: + # + # # anchor tag implicit style + # [nil, nil, true, 1 ] + # [nil, "tag:yaml.org,2002:map", false, 2 ] + # ["pewpew", nil, true, 1 ] + + def start_mapping anchor, tag, implicit, style + end + + ### + # Called when a map ends + def end_mapping + end + + ### + # Called when an empty event happens. (Which, as far as I can tell, is + # never). + def empty + end + + ### + # Called when the YAML stream ends + def end_stream + end + + ### + # Called before each event with line/column information. + def event_location(start_line, start_column, end_line, end_column) + end + + ### + # Is this handler a streaming handler? + def streaming? + false + end + end +end diff --git a/ext/psych/lib/psych/handlers/document_stream.rb b/ext/psych/lib/psych/handlers/document_stream.rb new file mode 100644 index 0000000000..b77115d074 --- /dev/null +++ b/ext/psych/lib/psych/handlers/document_stream.rb @@ -0,0 +1,23 @@ +# frozen_string_literal: true +require_relative '../tree_builder' + +module Psych + module Handlers + class DocumentStream < Psych::TreeBuilder # :nodoc: + def initialize &block + super + @block = block + end + + def start_document version, tag_directives, implicit + n = Nodes::Document.new version, tag_directives, implicit + push n + end + + def end_document implicit_end = !streaming? + @last.implicit_end = implicit_end + @block.call pop + end + end + end +end diff --git a/ext/psych/lib/psych/handlers/recorder.rb b/ext/psych/lib/psych/handlers/recorder.rb new file mode 100644 index 0000000000..c98724cb76 --- /dev/null +++ b/ext/psych/lib/psych/handlers/recorder.rb @@ -0,0 +1,40 @@ +# frozen_string_literal: true +require_relative '../handler' + +module Psych + module Handlers + ### + # This handler will capture an event and record the event. Recorder events + # are available vial Psych::Handlers::Recorder#events. + # + # For example: + # + # recorder = Psych::Handlers::Recorder.new + # parser = Psych::Parser.new recorder + # parser.parse '--- foo' + # + # recorder.events # => [list of events] + # + # # Replay the events + # + # emitter = Psych::Emitter.new $stdout + # recorder.events.each do |m, args| + # emitter.send m, *args + # end + + class Recorder < Psych::Handler + attr_reader :events + + def initialize + @events = [] + super + end + + EVENTS.each do |event| + define_method event do |*args| + @events << [event, args] + end + end + end + end +end diff --git a/ext/psych/lib/psych/json/ruby_events.rb b/ext/psych/lib/psych/json/ruby_events.rb new file mode 100644 index 0000000000..17b7ddc386 --- /dev/null +++ b/ext/psych/lib/psych/json/ruby_events.rb @@ -0,0 +1,20 @@ +# frozen_string_literal: true +module Psych + module JSON + module RubyEvents # :nodoc: + def visit_Time o + formatted = format_time o + @emitter.scalar formatted, nil, nil, false, true, Nodes::Scalar::DOUBLE_QUOTED + end + + def visit_DateTime o + visit_Time o.to_time + end + + def visit_String o + @emitter.scalar o.to_s, nil, nil, false, true, Nodes::Scalar::DOUBLE_QUOTED + end + alias :visit_Symbol :visit_String + end + end +end diff --git a/ext/psych/lib/psych/json/stream.rb b/ext/psych/lib/psych/json/stream.rb new file mode 100644 index 0000000000..24dd4b9baf --- /dev/null +++ b/ext/psych/lib/psych/json/stream.rb @@ -0,0 +1,17 @@ +# frozen_string_literal: true +require_relative 'ruby_events' +require_relative 'yaml_events' + +module Psych + module JSON + class Stream < Psych::Visitors::JSONTree + include Psych::JSON::RubyEvents + include Psych::Streaming + extend Psych::Streaming::ClassMethods + + class Emitter < Psych::Stream::Emitter # :nodoc: + include Psych::JSON::YAMLEvents + end + end + end +end diff --git a/ext/psych/lib/psych/json/tree_builder.rb b/ext/psych/lib/psych/json/tree_builder.rb new file mode 100644 index 0000000000..9a45f6b94c --- /dev/null +++ b/ext/psych/lib/psych/json/tree_builder.rb @@ -0,0 +1,13 @@ +# frozen_string_literal: true +require_relative 'yaml_events' + +module Psych + module JSON + ### + # Psych::JSON::TreeBuilder is an event based AST builder. Events are sent + # to an instance of Psych::JSON::TreeBuilder and a JSON AST is constructed. + class TreeBuilder < Psych::TreeBuilder + include Psych::JSON::YAMLEvents + end + end +end diff --git a/ext/psych/lib/psych/json/yaml_events.rb b/ext/psych/lib/psych/json/yaml_events.rb new file mode 100644 index 0000000000..eb973f5361 --- /dev/null +++ b/ext/psych/lib/psych/json/yaml_events.rb @@ -0,0 +1,30 @@ +# frozen_string_literal: true +module Psych + module JSON + module YAMLEvents # :nodoc: + def start_document version, tag_directives, implicit + super(version, tag_directives, !streaming?) + end + + def end_document implicit_end = !streaming? + super(implicit_end) + end + + def start_mapping anchor, tag, implicit, style + super(anchor, nil, true, Nodes::Mapping::FLOW) + end + + def start_sequence anchor, tag, implicit, style + super(anchor, nil, true, Nodes::Sequence::FLOW) + end + + def scalar value, anchor, tag, plain, quoted, style + if "tag:yaml.org,2002:null" == tag + super('null', nil, nil, true, false, Nodes::Scalar::PLAIN) + else + super + end + end + end + end +end diff --git a/ext/psych/lib/psych/nodes.rb b/ext/psych/lib/psych/nodes.rb new file mode 100644 index 0000000000..2fa52e0055 --- /dev/null +++ b/ext/psych/lib/psych/nodes.rb @@ -0,0 +1,78 @@ +# frozen_string_literal: true +require_relative 'nodes/node' +require_relative 'nodes/stream' +require_relative 'nodes/document' +require_relative 'nodes/sequence' +require_relative 'nodes/scalar' +require_relative 'nodes/mapping' +require_relative 'nodes/alias' + +module Psych + ### + # = Overview + # + # When using Psych.load to deserialize a YAML document, the document is + # translated to an intermediary AST. That intermediary AST is then + # translated in to a Ruby object graph. + # + # In the opposite direction, when using Psych.dump, the Ruby object graph is + # translated to an intermediary AST which is then converted to a YAML + # document. + # + # Psych::Nodes contains all of the classes that make up the nodes of a YAML + # AST. You can manually build an AST and use one of the visitors (see + # Psych::Visitors) to convert that AST to either a YAML document or to a + # Ruby object graph. + # + # Here is an example of building an AST that represents a list with one + # scalar: + # + # # Create our nodes + # stream = Psych::Nodes::Stream.new + # doc = Psych::Nodes::Document.new + # seq = Psych::Nodes::Sequence.new + # scalar = Psych::Nodes::Scalar.new('foo') + # + # # Build up our tree + # stream.children << doc + # doc.children << seq + # seq.children << scalar + # + # The stream is the root of the tree. We can then convert the tree to YAML: + # + # stream.to_yaml => "---\n- foo\n" + # + # Or convert it to Ruby: + # + # stream.to_ruby => [["foo"]] + # + # == YAML AST Requirements + # + # A valid YAML AST *must* have one Psych::Nodes::Stream at the root. A + # Psych::Nodes::Stream node must have 1 or more Psych::Nodes::Document nodes + # as children. + # + # Psych::Nodes::Document nodes must have one and *only* one child. That child + # may be one of: + # + # * Psych::Nodes::Sequence + # * Psych::Nodes::Mapping + # * Psych::Nodes::Scalar + # + # Psych::Nodes::Sequence and Psych::Nodes::Mapping nodes may have many + # children, but Psych::Nodes::Mapping nodes should have an even number of + # children. + # + # All of these are valid children for Psych::Nodes::Sequence and + # Psych::Nodes::Mapping nodes: + # + # * Psych::Nodes::Sequence + # * Psych::Nodes::Mapping + # * Psych::Nodes::Scalar + # * Psych::Nodes::Alias + # + # Psych::Nodes::Scalar and Psych::Nodes::Alias are both terminal nodes and + # should not have any children. + module Nodes + end +end diff --git a/ext/psych/lib/psych/nodes/alias.rb b/ext/psych/lib/psych/nodes/alias.rb new file mode 100644 index 0000000000..6da655f0fd --- /dev/null +++ b/ext/psych/lib/psych/nodes/alias.rb @@ -0,0 +1,21 @@ +# frozen_string_literal: true +module Psych + module Nodes + ### + # This class represents a {YAML Alias}[http://yaml.org/spec/1.1/#alias]. + # It points to an +anchor+. + # + # A Psych::Nodes::Alias is a terminal node and may have no children. + class Alias < Psych::Nodes::Node + # The anchor this alias links to + attr_accessor :anchor + + # Create a new Alias that points to an +anchor+ + def initialize anchor + @anchor = anchor + end + + def alias?; true; end + end + end +end diff --git a/ext/psych/lib/psych/nodes/document.rb b/ext/psych/lib/psych/nodes/document.rb new file mode 100644 index 0000000000..f57410d636 --- /dev/null +++ b/ext/psych/lib/psych/nodes/document.rb @@ -0,0 +1,63 @@ +# frozen_string_literal: true +module Psych + module Nodes + ### + # This represents a YAML Document. This node must be a child of + # Psych::Nodes::Stream. A Psych::Nodes::Document must have one child, + # and that child may be one of the following: + # + # * Psych::Nodes::Sequence + # * Psych::Nodes::Mapping + # * Psych::Nodes::Scalar + class Document < Psych::Nodes::Node + # The version of the YAML document + attr_accessor :version + + # A list of tag directives for this document + attr_accessor :tag_directives + + # Was this document implicitly created? + attr_accessor :implicit + + # Is the end of the document implicit? + attr_accessor :implicit_end + + ### + # Create a new Psych::Nodes::Document object. + # + # +version+ is a list indicating the YAML version. + # +tags_directives+ is a list of tag directive declarations + # +implicit+ is a flag indicating whether the document will be implicitly + # started. + # + # == Example: + # This creates a YAML document object that represents a YAML 1.1 document + # with one tag directive, and has an implicit start: + # + # Psych::Nodes::Document.new( + # [1,1], + # [["!", "tag:tenderlovemaking.com,2009:"]], + # true + # ) + # + # == See Also + # See also Psych::Handler#start_document + def initialize version = [], tag_directives = [], implicit = false + super() + @version = version + @tag_directives = tag_directives + @implicit = implicit + @implicit_end = true + end + + ### + # Returns the root node. A Document may only have one root node: + # http://yaml.org/spec/1.1/#id898031 + def root + children.first + end + + def document?; true; end + end + end +end diff --git a/ext/psych/lib/psych/nodes/mapping.rb b/ext/psych/lib/psych/nodes/mapping.rb new file mode 100644 index 0000000000..d49678cb0e --- /dev/null +++ b/ext/psych/lib/psych/nodes/mapping.rb @@ -0,0 +1,59 @@ +# frozen_string_literal: true +module Psych + module Nodes + ### + # This class represents a {YAML Mapping}[http://yaml.org/spec/1.1/#mapping]. + # + # A Psych::Nodes::Mapping node may have 0 or more children, but must have + # an even number of children. Here are the valid children a + # Psych::Nodes::Mapping node may have: + # + # * Psych::Nodes::Sequence + # * Psych::Nodes::Mapping + # * Psych::Nodes::Scalar + # * Psych::Nodes::Alias + class Mapping < Psych::Nodes::Node + # Any Map Style + ANY = 0 + + # Block Map Style + BLOCK = 1 + + # Flow Map Style + FLOW = 2 + + # The optional anchor for this mapping + attr_accessor :anchor + + # The optional tag for this mapping + attr_accessor :tag + + # Is this an implicit mapping? + attr_accessor :implicit + + # The style of this mapping + attr_accessor :style + + ### + # Create a new Psych::Nodes::Mapping object. + # + # +anchor+ is the anchor associated with the map or +nil+. + # +tag+ is the tag associated with the map or +nil+. + # +implicit+ is a boolean indicating whether or not the map was implicitly + # started. + # +style+ is an integer indicating the mapping style. + # + # == See Also + # See also Psych::Handler#start_mapping + def initialize anchor = nil, tag = nil, implicit = true, style = BLOCK + super() + @anchor = anchor + @tag = tag + @implicit = implicit + @style = style + end + + def mapping?; true; end + end + end +end diff --git a/ext/psych/lib/psych/nodes/node.rb b/ext/psych/lib/psych/nodes/node.rb new file mode 100644 index 0000000000..fc27448f2e --- /dev/null +++ b/ext/psych/lib/psych/nodes/node.rb @@ -0,0 +1,76 @@ +# frozen_string_literal: true +require_relative '../class_loader' +require_relative '../scalar_scanner' + +module Psych + module Nodes + ### + # The base class for any Node in a YAML parse tree. This class should + # never be instantiated. + class Node + include Enumerable + + # The children of this node + attr_reader :children + + # An associated tag + attr_reader :tag + + # The line number where this node start + attr_accessor :start_line + + # The column number where this node start + attr_accessor :start_column + + # The line number where this node ends + attr_accessor :end_line + + # The column number where this node ends + attr_accessor :end_column + + # Create a new Psych::Nodes::Node + def initialize + @children = [] + end + + ### + # Iterate over each node in the tree. Yields each node to +block+ depth + # first. + def each &block + return enum_for :each unless block_given? + Visitors::DepthFirst.new(block).accept self + end + + ### + # Convert this node to Ruby. + # + # See also Psych::Visitors::ToRuby + def to_ruby(symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true) + Visitors::ToRuby.create(symbolize_names: symbolize_names, freeze: freeze, strict_integer: strict_integer, parse_symbols: parse_symbols).accept(self) + end + alias :transform :to_ruby + + ### + # Convert this node to YAML. + # + # See also Psych::Visitors::Emitter + def yaml io = nil, options = {} + require "stringio" unless defined?(StringIO) + + real_io = io || StringIO.new(''.encode('utf-8')) + + Visitors::Emitter.new(real_io, options).accept self + return real_io.string unless io + io + end + alias :to_yaml :yaml + + def alias?; false; end + def document?; false; end + def mapping?; false; end + def scalar?; false; end + def sequence?; false; end + def stream?; false; end + end + end +end diff --git a/ext/psych/lib/psych/nodes/scalar.rb b/ext/psych/lib/psych/nodes/scalar.rb new file mode 100644 index 0000000000..5550b616a3 --- /dev/null +++ b/ext/psych/lib/psych/nodes/scalar.rb @@ -0,0 +1,70 @@ +# frozen_string_literal: true +module Psych + module Nodes + ### + # This class represents a {YAML Scalar}[http://yaml.org/spec/1.1/#id858081]. + # + # This node type is a terminal node and should not have any children. + class Scalar < Psych::Nodes::Node + # Any style scalar, the emitter chooses + ANY = 0 + + # Plain scalar style + PLAIN = 1 + + # Single quoted style + SINGLE_QUOTED = 2 + + # Double quoted style + DOUBLE_QUOTED = 3 + + # Literal style + LITERAL = 4 + + # Folded style + FOLDED = 5 + + # The scalar value + attr_accessor :value + + # The anchor value (if there is one) + attr_accessor :anchor + + # The tag value (if there is one) + attr_accessor :tag + + # Is this a plain scalar? + attr_accessor :plain + + # Is this scalar quoted? + attr_accessor :quoted + + # The style of this scalar + attr_accessor :style + + ### + # Create a new Psych::Nodes::Scalar object. + # + # +value+ is the string value of the scalar + # +anchor+ is an associated anchor or nil + # +tag+ is an associated tag or nil + # +plain+ is a boolean value + # +quoted+ is a boolean value + # +style+ is an integer indicating the string style + # + # == See Also + # + # See also Psych::Handler#scalar + def initialize value, anchor = nil, tag = nil, plain = true, quoted = false, style = ANY + @value = value + @anchor = anchor + @tag = tag + @plain = plain + @quoted = quoted + @style = style + end + + def scalar?; true; end + end + end +end diff --git a/ext/psych/lib/psych/nodes/sequence.rb b/ext/psych/lib/psych/nodes/sequence.rb new file mode 100644 index 0000000000..740f1938a4 --- /dev/null +++ b/ext/psych/lib/psych/nodes/sequence.rb @@ -0,0 +1,84 @@ +# frozen_string_literal: true +module Psych + module Nodes + ### + # This class represents a + # {YAML sequence}[http://yaml.org/spec/1.1/#sequence/syntax]. + # + # A YAML sequence is basically a list, and looks like this: + # + # %YAML 1.1 + # --- + # - I am + # - a Sequence + # + # A YAML sequence may have an anchor like this: + # + # %YAML 1.1 + # --- + # &A [ + # "This sequence", + # "has an anchor" + # ] + # + # A YAML sequence may also have a tag like this: + # + # %YAML 1.1 + # --- + # !!seq [ + # "This sequence", + # "has a tag" + # ] + # + # This class represents a sequence in a YAML document. A + # Psych::Nodes::Sequence node may have 0 or more children. Valid children + # for this node are: + # + # * Psych::Nodes::Sequence + # * Psych::Nodes::Mapping + # * Psych::Nodes::Scalar + # * Psych::Nodes::Alias + class Sequence < Psych::Nodes::Node + # Any Styles, emitter chooses + ANY = 0 + + # Block style sequence + BLOCK = 1 + + # Flow style sequence + FLOW = 2 + + # The anchor for this sequence (if any) + attr_accessor :anchor + + # The tag name for this sequence (if any) + attr_accessor :tag + + # Is this sequence started implicitly? + attr_accessor :implicit + + # The sequence style used + attr_accessor :style + + ### + # Create a new object representing a YAML sequence. + # + # +anchor+ is the anchor associated with the sequence or nil. + # +tag+ is the tag associated with the sequence or nil. + # +implicit+ a boolean indicating whether or not the sequence was + # implicitly started. + # +style+ is an integer indicating the list style. + # + # See Psych::Handler#start_sequence + def initialize anchor = nil, tag = nil, implicit = true, style = BLOCK + super() + @anchor = anchor + @tag = tag + @implicit = implicit + @style = style + end + + def sequence?; true; end + end + end +end diff --git a/ext/psych/lib/psych/nodes/stream.rb b/ext/psych/lib/psych/nodes/stream.rb new file mode 100644 index 0000000000..b525217821 --- /dev/null +++ b/ext/psych/lib/psych/nodes/stream.rb @@ -0,0 +1,40 @@ +# frozen_string_literal: true +module Psych + module Nodes + ### + # Represents a YAML stream. This is the root node for any YAML parse + # tree. This node must have one or more child nodes. The only valid + # child node for a Psych::Nodes::Stream node is Psych::Nodes::Document. + class Stream < Psych::Nodes::Node + + # Encodings supported by Psych (and libyaml) + + # Any encoding + ANY = Psych::Parser::ANY + + # UTF-8 encoding + UTF8 = Psych::Parser::UTF8 + + # UTF-16LE encoding + UTF16LE = Psych::Parser::UTF16LE + + # UTF-16BE encoding + UTF16BE = Psych::Parser::UTF16BE + + # The encoding used for this stream + attr_accessor :encoding + + ### + # Create a new Psych::Nodes::Stream node with an +encoding+ that + # defaults to Psych::Nodes::Stream::UTF8. + # + # See also Psych::Handler#start_stream + def initialize encoding = UTF8 + super() + @encoding = encoding + end + + def stream?; true; end + end + end +end diff --git a/ext/psych/lib/psych/omap.rb b/ext/psych/lib/psych/omap.rb new file mode 100644 index 0000000000..29cde0be50 --- /dev/null +++ b/ext/psych/lib/psych/omap.rb @@ -0,0 +1,5 @@ +# frozen_string_literal: true +module Psych + class Omap < ::Hash + end +end diff --git a/ext/psych/lib/psych/parser.rb b/ext/psych/lib/psych/parser.rb new file mode 100644 index 0000000000..2181c730e5 --- /dev/null +++ b/ext/psych/lib/psych/parser.rb @@ -0,0 +1,65 @@ +# frozen_string_literal: true +module Psych + ### + # YAML event parser class. This class parses a YAML document and calls + # events on the handler that is passed to the constructor. The events can + # be used for things such as constructing a YAML AST or deserializing YAML + # documents. It can even be fed back to Psych::Emitter to emit the same + # document that was parsed. + # + # See Psych::Handler for documentation on the events that Psych::Parser emits. + # + # Here is an example that prints out ever scalar found in a YAML document: + # + # # Handler for detecting scalar values + # class ScalarHandler < Psych::Handler + # def scalar value, anchor, tag, plain, quoted, style + # puts value + # end + # end + # + # parser = Psych::Parser.new(ScalarHandler.new) + # parser.parse(yaml_document) + # + # Here is an example that feeds the parser back in to Psych::Emitter. The + # YAML document is read from STDIN and written back out to STDERR: + # + # parser = Psych::Parser.new(Psych::Emitter.new($stderr)) + # parser.parse($stdin) + # + # Psych uses Psych::Parser in combination with Psych::TreeBuilder to + # construct an AST of the parsed YAML document. + + class Parser + class Mark < Struct.new(:index, :line, :column) + end + + # The handler on which events will be called + attr_accessor :handler + + # Set the encoding for this parser to +encoding+ + attr_writer :external_encoding + + ### + # Creates a new Psych::Parser instance with +handler+. YAML events will + # be called on +handler+. See Psych::Parser for more details. + + def initialize handler = Handler.new + @handler = handler + @external_encoding = ANY + end + + ### + # call-seq: + # parser.parse(yaml) + # + # Parse the YAML document contained in +yaml+. Events will be called on + # the handler set on the parser instance. + # + # See Psych::Parser and Psych::Parser#handler + + def parse yaml, path = yaml.respond_to?(:path) ? yaml.path : "<unknown>" + _native_parse @handler, yaml, path + end + end +end diff --git a/ext/psych/lib/psych/scalar_scanner.rb b/ext/psych/lib/psych/scalar_scanner.rb new file mode 100644 index 0000000000..6a556fb3b8 --- /dev/null +++ b/ext/psych/lib/psych/scalar_scanner.rb @@ -0,0 +1,142 @@ +# frozen_string_literal: true + +module Psych + ### + # Scan scalars for built in types + class ScalarScanner + # Taken from http://yaml.org/type/timestamp.html + TIME = /^-?\d{4}-\d{1,2}-\d{1,2}(?:[Tt]|\s+)\d{1,2}:\d\d:\d\d(?:\.\d*)?(?:\s*(?:Z|[-+]\d{1,2}:?(?:\d\d)?))?$/ + + # Taken from http://yaml.org/type/float.html + # Base 60, [-+]inf and NaN are handled separately + FLOAT = /^(?:[-+]?([0-9][0-9_,]*)?\.[0-9]*([eE][-+][0-9]+)?(?# base 10))$/x + + # Taken from http://yaml.org/type/int.html and modified to ensure at least one numerical symbol exists + INTEGER_STRICT = /^(?:[-+]?0b[_]*[0-1][0-1_]* (?# base 2) + |[-+]?0[_]*[0-7][0-7_]* (?# base 8) + |[-+]?(0|[1-9][0-9_]*) (?# base 10) + |[-+]?0x[_]*[0-9a-fA-F][0-9a-fA-F_]* (?# base 16))$/x + + # Same as above, but allows commas. + # Not to YML spec, but kept for backwards compatibility + INTEGER_LEGACY = /^(?:[-+]?0b[_,]*[0-1][0-1_,]* (?# base 2) + |[-+]?0[_,]*[0-7][0-7_,]* (?# base 8) + |[-+]?(?:0|[1-9](?:[0-9]|,[0-9]|_[0-9])*) (?# base 10) + |[-+]?0x[_,]*[0-9a-fA-F][0-9a-fA-F_,]* (?# base 16))$/x + + attr_reader :class_loader + + # Create a new scanner + def initialize class_loader, strict_integer: false, parse_symbols: true + @symbol_cache = {} + @class_loader = class_loader + @strict_integer = strict_integer + @parse_symbols = parse_symbols + end + + # Tokenize +string+ returning the Ruby object + def tokenize string + return nil if string.empty? + return @symbol_cache[string] if @symbol_cache.key?(string) + integer_regex = @strict_integer ? INTEGER_STRICT : INTEGER_LEGACY + # Check for a String type, being careful not to get caught by hash keys, hex values, and + # special floats (e.g., -.inf). + if string.match?(%r{^[^\d.:-]?[[:alpha:]_\s!@#$%\^&*(){}<>|/\\~;=]+}) || string.match?(/\n/) + return string if string.length > 5 + + if string.match?(/^[^ytonf~]/i) + string + elsif string == '~' || string.match?(/^null$/i) + nil + elsif string.match?(/^(yes|true|on)$/i) + true + elsif string.match?(/^(no|false|off)$/i) + false + else + string + end + elsif string.match?(TIME) + begin + parse_time string + rescue ArgumentError + string + end + elsif string.match?(/^\d{4}-(?:1[012]|0\d|\d)-(?:[12]\d|3[01]|0\d|\d)$/) + begin + class_loader.date.strptime(string, '%F', Date::GREGORIAN) + rescue ArgumentError + string + end + elsif string.match?(/^\+?\.inf$/i) + Float::INFINITY + elsif string.match?(/^-\.inf$/i) + -Float::INFINITY + elsif string.match?(/^\.nan$/i) + Float::NAN + elsif @parse_symbols && string.match?(/^:./) + if string =~ /^:(["'])(.*)\1/ + @symbol_cache[string] = class_loader.symbolize($2.sub(/^:/, '')) + else + @symbol_cache[string] = class_loader.symbolize(string.sub(/^:/, '')) + end + elsif string.match?(/^[-+]?[0-9][0-9_]*(:[0-5]?[0-9]){1,2}$/) + i = 0 + string.split(':').each_with_index do |n,e| + i += (n.to_i * 60 ** (e - 2).abs) + end + i + elsif string.match?(/^[-+]?[0-9][0-9_]*(:[0-5]?[0-9]){1,2}\.[0-9_]*$/) + i = 0 + string.split(':').each_with_index do |n,e| + i += (n.to_f * 60 ** (e - 2).abs) + end + i + elsif string.match?(FLOAT) + if string.match?(/\A[-+]?\.\Z/) + string + else + Float(string.delete(',_').gsub(/\.([Ee]|$)/, '\1')) + end + elsif string.match?(integer_regex) + parse_int string + else + string + end + end + + ### + # Parse and return an int from +string+ + def parse_int string + Integer(string.delete(',_')) + end + + ### + # Parse and return a Time from +string+ + def parse_time string + klass = class_loader.load 'Time' + + date, time = *(string.split(/[ tT]/, 2)) + (yy, m, dd) = date.match(/^(-?\d{4})-(\d{1,2})-(\d{1,2})/).captures.map { |x| x.to_i } + md = time.match(/(\d+:\d+:\d+)(?:\.(\d*))?\s*(Z|[-+]\d+(:\d\d)?)?/) + + (hh, mm, ss) = md[1].split(':').map { |x| x.to_i } + us = (md[2] ? Rational("0.#{md[2]}") : 0) * 1000000 + + time = klass.utc(yy, m, dd, hh, mm, ss, us) + + return time if 'Z' == md[3] + return klass.at(time.to_i, us) unless md[3] + + tz = md[3].match(/^([+\-]?\d{1,2})\:?(\d{1,2})?$/)[1..-1].compact.map { |digit| Integer(digit, 10) } + offset = tz.first * 3600 + + if offset < 0 + offset -= ((tz[1] || 0) * 60) + else + offset += ((tz[1] || 0) * 60) + end + + klass.new(yy, m, dd, hh, mm, ss+us/(1_000_000r), offset) + end + end +end diff --git a/ext/psych/lib/psych/set.rb b/ext/psych/lib/psych/set.rb new file mode 100644 index 0000000000..760d217098 --- /dev/null +++ b/ext/psych/lib/psych/set.rb @@ -0,0 +1,5 @@ +# frozen_string_literal: true +module Psych + class Set < ::Hash + end +end diff --git a/ext/psych/lib/psych/stream.rb b/ext/psych/lib/psych/stream.rb new file mode 100644 index 0000000000..24e45afc3b --- /dev/null +++ b/ext/psych/lib/psych/stream.rb @@ -0,0 +1,38 @@ +# frozen_string_literal: true +module Psych + ### + # Psych::Stream is a streaming YAML emitter. It will not buffer your YAML, + # but send it straight to an IO. + # + # Here is an example use: + # + # stream = Psych::Stream.new($stdout) + # stream.start + # stream.push({:foo => 'bar'}) + # stream.finish + # + # YAML will be immediately emitted to $stdout with no buffering. + # + # Psych::Stream#start will take a block and ensure that Psych::Stream#finish + # is called, so you can do this form: + # + # stream = Psych::Stream.new($stdout) + # stream.start do |em| + # em.push(:foo => 'bar') + # end + # + class Stream < Psych::Visitors::YAMLTree + class Emitter < Psych::Emitter # :nodoc: + def end_document implicit_end = !streaming? + super + end + + def streaming? + true + end + end + + include Psych::Streaming + extend Psych::Streaming::ClassMethods + end +end diff --git a/ext/psych/lib/psych/streaming.rb b/ext/psych/lib/psych/streaming.rb new file mode 100644 index 0000000000..eb19792ad0 --- /dev/null +++ b/ext/psych/lib/psych/streaming.rb @@ -0,0 +1,28 @@ +# frozen_string_literal: true +module Psych + module Streaming + module ClassMethods + ### + # Create a new streaming emitter. Emitter will print to +io+. See + # Psych::Stream for an example. + def new io + emitter = const_get(:Emitter).new(io) + class_loader = ClassLoader.new + ss = ScalarScanner.new class_loader + super(emitter, ss, {}) + end + end + + ### + # Start streaming using +encoding+ + def start encoding = Nodes::Stream::UTF8 + super.tap { yield self if block_given? } + ensure + finish if block_given? + end + + private + def register target, obj + end + end +end diff --git a/ext/psych/lib/psych/syntax_error.rb b/ext/psych/lib/psych/syntax_error.rb new file mode 100644 index 0000000000..a4c9c4a376 --- /dev/null +++ b/ext/psych/lib/psych/syntax_error.rb @@ -0,0 +1,22 @@ +# frozen_string_literal: true +require_relative 'exception' + +module Psych + class SyntaxError < Psych::Exception + attr_reader :file, :line, :column, :offset, :problem, :context + + def initialize file, line, col, offset, problem, context + err = [problem, context].compact.join ' ' + filename = file || '<unknown>' + message = "(%s): %s at line %d column %d" % [filename, err, line, col] + + @file = file + @line = line + @column = col + @offset = offset + @problem = problem + @context = context + super(message) + end + end +end diff --git a/ext/psych/lib/psych/tree_builder.rb b/ext/psych/lib/psych/tree_builder.rb new file mode 100644 index 0000000000..83115bd721 --- /dev/null +++ b/ext/psych/lib/psych/tree_builder.rb @@ -0,0 +1,137 @@ +# frozen_string_literal: true +require_relative 'handler' + +module Psych + ### + # This class works in conjunction with Psych::Parser to build an in-memory + # parse tree that represents a YAML document. + # + # == Example + # + # parser = Psych::Parser.new Psych::TreeBuilder.new + # parser.parse('--- foo') + # tree = parser.handler.root + # + # See Psych::Handler for documentation on the event methods used in this + # class. + class TreeBuilder < Psych::Handler + # Returns the root node for the built tree + attr_reader :root + + # Create a new TreeBuilder instance + def initialize + @stack = [] + @last = nil + @root = nil + + @start_line = nil + @start_column = nil + @end_line = nil + @end_column = nil + end + + def event_location(start_line, start_column, end_line, end_column) + @start_line = start_line + @start_column = start_column + @end_line = end_line + @end_column = end_column + end + + %w{ + Sequence + Mapping + }.each do |node| + class_eval <<~RUBY, __FILE__, __LINE__ + 1 + def start_#{node.downcase}(anchor, tag, implicit, style) + n = Nodes::#{node}.new(anchor, tag, implicit, style) + set_start_location(n) + @last.children << n + push n + end + + def end_#{node.downcase} + n = pop + set_end_location(n) + n + end + RUBY + end + + ### + # Handles start_document events with +version+, +tag_directives+, + # and +implicit+ styling. + # + # See Psych::Handler#start_document + def start_document version, tag_directives, implicit + n = Nodes::Document.new version, tag_directives, implicit + set_start_location(n) + @last.children << n + push n + end + + ### + # Handles end_document events with +version+, +tag_directives+, + # and +implicit+ styling. + # + # See Psych::Handler#start_document + def end_document implicit_end = !streaming? + @last.implicit_end = implicit_end + n = pop + set_end_location(n) + n + end + + def start_stream encoding + @root = Nodes::Stream.new(encoding) + set_start_location(@root) + push @root + end + + def end_stream + n = pop + set_end_location(n) + n + end + + def scalar value, anchor, tag, plain, quoted, style + s = Nodes::Scalar.new(value,anchor,tag,plain,quoted,style) + set_location(s) + @last.children << s + s + end + + def alias anchor + a = Nodes::Alias.new(anchor) + set_location(a) + @last.children << a + a + end + + private + def push value + @stack.push value + @last = value + end + + def pop + x = @stack.pop + @last = @stack.last + x + end + + def set_location(node) + set_start_location(node) + set_end_location(node) + end + + def set_start_location(node) + node.start_line = @start_line + node.start_column = @start_column + end + + def set_end_location(node) + node.end_line = @end_line + node.end_column = @end_column + end + end +end diff --git a/ext/psych/lib/psych/versions.rb b/ext/psych/lib/psych/versions.rb new file mode 100644 index 0000000000..6c1679bf65 --- /dev/null +++ b/ext/psych/lib/psych/versions.rb @@ -0,0 +1,10 @@ +# frozen_string_literal: true + +module Psych + # The version of Psych you are using + VERSION = '5.4.0' + + if RUBY_ENGINE == 'jruby' + DEFAULT_SNAKEYAML_VERSION = '2.10'.freeze + end +end diff --git a/ext/psych/lib/psych/visitors.rb b/ext/psych/lib/psych/visitors.rb new file mode 100644 index 0000000000..508290d862 --- /dev/null +++ b/ext/psych/lib/psych/visitors.rb @@ -0,0 +1,7 @@ +# frozen_string_literal: true +require_relative 'visitors/visitor' +require_relative 'visitors/to_ruby' +require_relative 'visitors/emitter' +require_relative 'visitors/yaml_tree' +require_relative 'visitors/json_tree' +require_relative 'visitors/depth_first' diff --git a/ext/psych/lib/psych/visitors/depth_first.rb b/ext/psych/lib/psych/visitors/depth_first.rb new file mode 100644 index 0000000000..b4ff9e40e7 --- /dev/null +++ b/ext/psych/lib/psych/visitors/depth_first.rb @@ -0,0 +1,27 @@ +# frozen_string_literal: true +module Psych + module Visitors + class DepthFirst < Psych::Visitors::Visitor + def initialize block + @block = block + end + + private + + def nary o + o.children.each { |x| visit x } + @block.call o + end + alias :visit_Psych_Nodes_Stream :nary + alias :visit_Psych_Nodes_Document :nary + alias :visit_Psych_Nodes_Sequence :nary + alias :visit_Psych_Nodes_Mapping :nary + + def terminal o + @block.call o + end + alias :visit_Psych_Nodes_Scalar :terminal + alias :visit_Psych_Nodes_Alias :terminal + end + end +end diff --git a/ext/psych/lib/psych/visitors/emitter.rb b/ext/psych/lib/psych/visitors/emitter.rb new file mode 100644 index 0000000000..e3b92b7d03 --- /dev/null +++ b/ext/psych/lib/psych/visitors/emitter.rb @@ -0,0 +1,52 @@ +# frozen_string_literal: true +module Psych + module Visitors + class Emitter < Psych::Visitors::Visitor + def initialize io, options = {} + opts = [:indentation, :canonical, :line_width].find_all { |opt| + options.key?(opt) + } + + if opts.empty? + @handler = Psych::Emitter.new io + else + du = Handler::DumperOptions.new + opts.each { |option| du.send :"#{option}=", options[option] } + @handler = Psych::Emitter.new io, du + end + end + + def visit_Psych_Nodes_Stream o + @handler.start_stream o.encoding + o.children.each { |c| accept c } + @handler.end_stream + end + + def visit_Psych_Nodes_Document o + @handler.start_document o.version, o.tag_directives, o.implicit + o.children.each { |c| accept c } + @handler.end_document o.implicit_end + end + + def visit_Psych_Nodes_Scalar o + @handler.scalar o.value, o.anchor, o.tag, o.plain, o.quoted, o.style + end + + def visit_Psych_Nodes_Sequence o + @handler.start_sequence o.anchor, o.tag, o.implicit, o.style + o.children.each { |c| accept c } + @handler.end_sequence + end + + def visit_Psych_Nodes_Mapping o + @handler.start_mapping o.anchor, o.tag, o.implicit, o.style + o.children.each { |c| accept c } + @handler.end_mapping + end + + def visit_Psych_Nodes_Alias o + @handler.alias o.anchor + end + end + end +end diff --git a/ext/psych/lib/psych/visitors/json_tree.rb b/ext/psych/lib/psych/visitors/json_tree.rb new file mode 100644 index 0000000000..979fc100bd --- /dev/null +++ b/ext/psych/lib/psych/visitors/json_tree.rb @@ -0,0 +1,25 @@ +# frozen_string_literal: true +require_relative '../json/ruby_events' + +module Psych + module Visitors + class JSONTree < YAMLTree + include Psych::JSON::RubyEvents + + def self.create options = {} + emitter = Psych::JSON::TreeBuilder.new + class_loader = ClassLoader.new + ss = ScalarScanner.new class_loader + new(emitter, ss, options) + end + + def accept target + if target.respond_to?(:encode_with) + dump_coder target + else + send(@dispatch_cache[target.class], target) + end + end + end + end +end diff --git a/ext/psych/lib/psych/visitors/to_ruby.rb b/ext/psych/lib/psych/visitors/to_ruby.rb new file mode 100644 index 0000000000..475444e589 --- /dev/null +++ b/ext/psych/lib/psych/visitors/to_ruby.rb @@ -0,0 +1,479 @@ +# frozen_string_literal: true +require_relative '../scalar_scanner' +require_relative '../class_loader' +require_relative '../exception' + +unless defined?(Regexp::NOENCODING) + Regexp::NOENCODING = 32 +end + +module Psych + module Visitors + ### + # This class walks a YAML AST, converting each node to Ruby + class ToRuby < Psych::Visitors::Visitor + unless RUBY_VERSION < "3.2" + DATA_INITIALIZE = Data.instance_method(:initialize) + end + + def self.create(symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true) + class_loader = ClassLoader.new + scanner = ScalarScanner.new class_loader, strict_integer: strict_integer, parse_symbols: parse_symbols + new(scanner, class_loader, symbolize_names: symbolize_names, freeze: freeze) + end + + attr_reader :class_loader + + def initialize ss, class_loader, symbolize_names: false, freeze: false + super() + @st = {} + @ss = ss + @load_tags = Psych.load_tags + @domain_types = Psych.domain_types + @class_loader = class_loader + @symbolize_names = symbolize_names + @freeze = freeze + end + + def accept target + result = super + + unless @domain_types.empty? || !target.tag + key = target.tag.sub(/^[!\/]*/, '').sub(/(,\d+)\//, '\1:') + key = "tag:#{key}" unless key.match?(/^(?:tag:|x-private)/) + + if @domain_types.key? key + value, block = @domain_types[key] + result = block.call value, result + end + end + + result = deduplicate(result).freeze if @freeze + result + end + + def deserialize o + if klass = resolve_class(@load_tags[o.tag]) + instance = klass.allocate + + if instance.respond_to?(:init_with) + coder = Psych::Coder.new(o.tag) + coder.scalar = o.value + instance.init_with coder + end + + return instance + end + + return o.value if o.quoted + return @ss.tokenize(o.value) unless o.tag + + case o.tag + when '!binary', 'tag:yaml.org,2002:binary' + o.value.unpack('m').first + when /^!(?:str|ruby\/string)(?::(.*))?$/, 'tag:yaml.org,2002:str' + klass = resolve_class($1) + if klass + klass.allocate.replace o.value + else + o.value + end + when '!ruby/object:BigDecimal' + require 'bigdecimal' unless defined? BigDecimal + class_loader.big_decimal._load o.value + when "!ruby/object:DateTime" + class_loader.date_time + t = @ss.parse_time(o.value) + DateTime.civil(*t.to_a[0, 6].reverse, Rational(t.utc_offset, 86400)) + + (t.subsec/86400) + when '!ruby/encoding' + ::Encoding.find o.value + when "!ruby/object:Complex" + class_loader.complex + Complex(o.value) + when "!ruby/object:Rational" + class_loader.rational + Rational(o.value) + when "!ruby/class", "!ruby/module" + resolve_class o.value + when "tag:yaml.org,2002:float", "!float" + Float(@ss.tokenize(o.value)) + when "!ruby/regexp" + klass = class_loader.regexp + matches = /^\/(?<string>.*)\/(?<options>[mixn]*)$/m.match(o.value) + source = matches[:string].gsub('\/', '/') + options = 0 + lang = nil + matches[:options].each_char do |option| + case option + when 'x' then options |= Regexp::EXTENDED + when 'i' then options |= Regexp::IGNORECASE + when 'm' then options |= Regexp::MULTILINE + when 'n' then options |= Regexp::NOENCODING + else lang = option + end + end + klass.new(*[source, options, lang].compact) + when "!ruby/range" + klass = class_loader.range + args = o.value.split(/([.]{2,3})/, 2).map { |s| + accept Nodes::Scalar.new(s) + } + args.push(args.delete_at(1) == '...') + klass.new(*args) + when /^!ruby\/sym(bol)?:?(.*)?$/ + class_loader.symbolize o.value + else + @ss.tokenize o.value + end + end + private :deserialize + + def visit_Psych_Nodes_Scalar o + register o, deserialize(o) + end + + def visit_Psych_Nodes_Sequence o + if klass = resolve_class(@load_tags[o.tag]) + instance = klass.allocate + + if instance.respond_to?(:init_with) + coder = Psych::Coder.new(o.tag) + coder.seq = o.children.map { |c| accept c } + instance.init_with coder + end + + return instance + end + + case o.tag + when nil + register_empty(o) + when '!omap', 'tag:yaml.org,2002:omap' + map = register(o, Psych::Omap.new) + o.children.each { |a| + map[accept(a.children.first)] = accept a.children.last + } + map + when /^!(?:seq|ruby\/array):(.*)$/ + klass = resolve_class($1) + list = register(o, klass.allocate) + o.children.each { |c| list.push accept c } + list + else + register_empty(o) + end + end + + def visit_Psych_Nodes_Mapping o + if @load_tags[o.tag] + return revive(resolve_class(@load_tags[o.tag]), o) + end + return revive_hash(register(o, {}), o) unless o.tag + + case o.tag + when /^!ruby\/struct:?(.*)?$/ + klass = resolve_class($1) if $1 + + if klass + s = register(o, klass.allocate) + + members = {} + struct_members = s.members.map { |x| class_loader.symbolize x } + o.children.each_slice(2) do |k,v| + member = accept(k) + value = accept(v) + if struct_members.include?(class_loader.symbolize(member)) + s.send("#{member}=", value) + else + members[member.to_s.sub(/^@/, '')] = value + end + end + init_with(s, members, o) + else + klass = class_loader.struct + members = o.children.map { |c| accept c } + h = Hash[*members] + s = klass.new(*h.map { |k,v| + class_loader.symbolize k + }).new(*h.map { |k,v| v }) + register(o, s) + s + end + + when /^!ruby\/data(-with-ivars)?(?::(.*))?$/ + data = register(o, resolve_class($2).allocate) if $2 + members = {} + + if $1 # data-with-ivars + ivars = {} + o.children.each_slice(2) do |type, vars| + case accept(type) + when 'members' + revive_data_members(members, vars) + data ||= allocate_anon_data(o, members) + when 'ivars' + revive_hash(ivars, vars) + end + end + ivars.each do |ivar, v| + data.instance_variable_set ivar, v + end + else + revive_data_members(members, o) + end + data ||= allocate_anon_data(o, members) + DATA_INITIALIZE.bind_call(data, **members) + data.freeze + data + + when /^!ruby\/object:?(.*)?$/ + name = $1 || 'Object' + + if name == 'Complex' + class_loader.complex + h = Hash[*o.children.map { |c| accept c }] + register o, Complex(h['real'], h['image']) + elsif name == 'Rational' + class_loader.rational + h = Hash[*o.children.map { |c| accept c }] + register o, Rational(h['numerator'], h['denominator']) + elsif name == 'Hash' + revive_hash(register(o, {}), o) + else + obj = revive((resolve_class(name) || class_loader.object), o) + obj + end + + when /^!(?:str|ruby\/string)(?::(.*))?$/, 'tag:yaml.org,2002:str' + klass = resolve_class($1) + members = {} + string = nil + + o.children.each_slice(2) do |k,v| + key = accept k + value = accept v + + if key == 'str' + if klass + string = klass.allocate.replace value + else + string = value + end + register(o, string) + else + members[key] = value + end + end + init_with(string, members.map { |k,v| [k.to_s.sub(/^@/, ''),v] }, o) + when /^!ruby\/array:(.*)$/ + klass = resolve_class($1) + list = register(o, klass.allocate) + + members = Hash[o.children.map { |c| accept c }.each_slice(2).to_a] + list.replace members['internal'] + + members['ivars'].each do |ivar, v| + list.instance_variable_set ivar, v + end + list + + when '!ruby/range' + klass = class_loader.range + h = Hash[*o.children.map { |c| accept c }] + register o, klass.new(h['begin'], h['end'], h['excl']) + + when /^!ruby\/exception:?(.*)?$/ + h = Hash[*o.children.map { |c| accept c }] + + e = build_exception((resolve_class($1) || class_loader.exception), + h.delete('message')) + + e.set_backtrace h.delete('backtrace') if h.key? 'backtrace' + init_with(e, h, o) + + when '!set', 'tag:yaml.org,2002:set' + set = class_loader.psych_set.new + @st[o.anchor] = set if o.anchor + o.children.each_slice(2) do |k,v| + set[accept(k)] = accept(v) + end + set + + when /^!ruby\/hash-with-ivars(?::(.*))?$/ + hash = $1 ? resolve_class($1).allocate : {} + register o, hash + o.children.each_slice(2) do |key, value| + case key.value + when 'elements' + revive_hash hash, value + when 'ivars' + value.children.each_slice(2) do |k,v| + hash.instance_variable_set accept(k), accept(v) + end + end + end + hash + + when /^!map:(.*)$/, /^!ruby\/hash:(.*)$/ + revive_hash register(o, resolve_class($1).allocate), o + + when '!omap', 'tag:yaml.org,2002:omap' + map = register(o, class_loader.psych_omap.new) + o.children.each_slice(2) do |l,r| + map[accept(l)] = accept r + end + map + + when /^!ruby\/marshalable:(.*)$/ + name = $1 + klass = resolve_class(name) + obj = register(o, klass.allocate) + + if obj.respond_to?(:init_with) + init_with(obj, revive_hash({}, o), o) + elsif obj.respond_to?(:marshal_load) + marshal_data = o.children.map(&method(:accept)) + obj.marshal_load(marshal_data) + obj + else + raise ArgumentError, "Cannot deserialize #{name}" + end + + else + revive_hash(register(o, {}), o) + end + end + + def visit_Psych_Nodes_Document o + accept o.root + end + + def visit_Psych_Nodes_Stream o + o.children.map { |c| accept c } + end + + def visit_Psych_Nodes_Alias o + @st.fetch(o.anchor) { raise AnchorNotDefined, o.anchor } + end + + private + + def register node, object + @st[node.anchor] = object if node.anchor + object + end + + def register_empty object + list = register(object, []) + object.children.each { |c| list.push accept c } + list + end + + def allocate_anon_data node, members + klass = class_loader.data.define(*members.keys) + register(node, klass.allocate) + end + + def revive_data_members hash, o + o.children.each_slice(2) do |k,v| + name = accept(k) + value = accept(v) + hash[class_loader.symbolize(name)] = value + end + hash + end + + def revive_hash hash, o, tagged= false + o.children.each_slice(2) { |k,v| + key = accept(k) + val = accept(v) + + if key == '<<' && k.tag != "tag:yaml.org,2002:str" + case v + when Nodes::Alias, Nodes::Mapping + begin + hash.merge! val + rescue TypeError + hash[key] = val + end + when Nodes::Sequence + begin + h = {} + val.reverse_each do |value| + h.merge! value + end + hash.merge! h + rescue TypeError + hash[key] = val + end + else + hash[key] = val + end + else + if !tagged && @symbolize_names && key.is_a?(String) + key = key.to_sym + elsif !@freeze + key = deduplicate(key) + end + + hash[key] = val + end + + } + hash + end + + if RUBY_VERSION < '2.7' + def deduplicate key + if key.is_a?(String) + # It is important to untaint the string, otherwise it won't + # be deduplicated into an fstring, but simply frozen. + -(key.untaint) + else + key + end + end + else + def deduplicate key + if key.is_a?(String) + -key + else + key + end + end + end + + def merge_key hash, key, val + end + + def revive klass, node + s = register(node, klass.allocate) + init_with(s, revive_hash({}, node, true), node) + end + + def init_with o, h, node + c = Psych::Coder.new(node.tag) + c.map = h + + if o.respond_to?(:init_with) + o.init_with c + else + h.each { |k,v| o.instance_variable_set(:"@#{k}", v) } + end + o + end + + # Convert +klassname+ to a Class + def resolve_class klassname + class_loader.load klassname + end + end + + class NoAliasRuby < ToRuby + def visit_Psych_Nodes_Alias o + raise AliasesNotEnabled + end + end + end +end diff --git a/ext/psych/lib/psych/visitors/visitor.rb b/ext/psych/lib/psych/visitors/visitor.rb new file mode 100644 index 0000000000..21052aa66f --- /dev/null +++ b/ext/psych/lib/psych/visitors/visitor.rb @@ -0,0 +1,34 @@ +# frozen_string_literal: true +module Psych + module Visitors + class Visitor + def accept target + visit target + end + + private + + # @api private + def self.dispatch_cache + Hash.new do |hash, klass| + hash[klass] = :"visit_#{klass.name.gsub('::', '_')}" + end.compare_by_identity + end + + if defined?(Ractor) + def dispatch + @dispatch_cache ||= (Ractor.current[:Psych_Visitors_Visitor] ||= Visitor.dispatch_cache) + end + else + DISPATCH = dispatch_cache + def dispatch + DISPATCH + end + end + + def visit target + send dispatch[target.class], target + end + end + end +end diff --git a/ext/psych/lib/psych/visitors/yaml_tree.rb b/ext/psych/lib/psych/visitors/yaml_tree.rb new file mode 100644 index 0000000000..b6c86f4c94 --- /dev/null +++ b/ext/psych/lib/psych/visitors/yaml_tree.rb @@ -0,0 +1,626 @@ +# frozen_string_literal: true +require_relative '../tree_builder' +require_relative '../scalar_scanner' +require_relative '../class_loader' + +module Psych + module Visitors + ### + # YAMLTree builds a YAML ast given a Ruby object. For example: + # + # builder = Psych::Visitors::YAMLTree.new + # builder << { :foo => 'bar' } + # builder.tree # => #<Psych::Nodes::Stream .. } + # + class YAMLTree < Psych::Visitors::Visitor + class Registrar # :nodoc: + def initialize + @obj_to_id = {}.compare_by_identity + @obj_to_node = {}.compare_by_identity + @counter = 0 + end + + def register target, node + @obj_to_node[target] = node + end + + def key? target + @obj_to_node.key? target + end + + def id_for target + @obj_to_id[target] ||= (@counter += 1) + end + + def node_for target + @obj_to_node[target] + end + end + + attr_reader :started, :finished + alias :finished? :finished + alias :started? :started + + def self.create options = {}, emitter = nil + emitter ||= TreeBuilder.new + class_loader = ClassLoader.new + ss = ScalarScanner.new class_loader + new(emitter, ss, options) + end + + def initialize emitter, ss, options + super() + @started = false + @finished = false + @emitter = emitter + @st = Registrar.new + @ss = ss + @options = options + @line_width = options[:line_width] + if @line_width && @line_width < 0 + if @line_width == -1 + # Treat -1 as unlimited line-width, same as libyaml does. + @line_width = nil + else + fail(ArgumentError, "Invalid line_width #{@line_width}, must be non-negative or -1 for unlimited.") + end + end + @stringify_names = options[:stringify_names] + @coders = [] + + @dispatch_cache = Hash.new do |h,klass| + method = "visit_#{(klass.name || '').split('::').join('_')}" + + method = respond_to?(method) ? method : h[klass.superclass] + + raise(TypeError, "can't dump #{klass.name}") unless method + + h[klass] = method + end.compare_by_identity + end + + def start encoding = Nodes::Stream::UTF8 + @emitter.start_stream(encoding).tap do + @started = true + end + end + + def finish + @emitter.end_stream.tap do + @finished = true + end + end + + def tree + finish unless finished? + @emitter.root + end + + def push object + start unless started? + version = [] + version = [1,1] if @options[:header] + + case @options[:version] + when Array + version = @options[:version] + when String + version = @options[:version].split('.').map { |x| x.to_i } + else + version = [1,1] + end if @options.key? :version + + @emitter.start_document version, [], false + accept object + @emitter.end_document !@emitter.streaming? + end + alias :<< :push + + def accept target + # return any aliases we find + if @st.key? target + oid = @st.id_for target + node = @st.node_for target + anchor = oid.to_s + node.anchor = anchor + return @emitter.alias anchor + end + + if target.respond_to?(:encode_with) + dump_coder target + else + send(@dispatch_cache[target.class], target) + end + end + + def visit_Psych_Omap o + seq = @emitter.start_sequence(nil, 'tag:yaml.org,2002:omap', false, Nodes::Sequence::BLOCK) + register(o, seq) + + o.each { |k,v| visit_Hash k => v } + @emitter.end_sequence + end + + def visit_Encoding o + tag = "!ruby/encoding" + @emitter.scalar o.name, nil, tag, false, false, Nodes::Scalar::ANY + end + + def visit_Object o + tag = Psych.dump_tags[o.class] + unless tag + klass = o.class == Object ? nil : o.class.name + tag = ['!ruby/object', klass].compact.join(':') + end + + map = @emitter.start_mapping(nil, tag, false, Nodes::Mapping::BLOCK) + register(o, map) + + dump_ivars o + @emitter.end_mapping + end + + alias :visit_Delegator :visit_Object + + def visit_Data o + ivars = o.instance_variables + if ivars.empty? + tag = ['!ruby/data', o.class.name].compact.join(':') + register o, @emitter.start_mapping(nil, tag, false, Nodes::Mapping::BLOCK) + o.members.each do |member| + @emitter.scalar member.to_s, nil, nil, true, false, Nodes::Scalar::ANY + accept o.send member + end + @emitter.end_mapping + + else + tag = ['!ruby/data-with-ivars', o.class.name].compact.join(':') + node = @emitter.start_mapping(nil, tag, false, Psych::Nodes::Mapping::BLOCK) + register(o, node) + + # Dump the members + accept 'members' + @emitter.start_mapping nil, nil, true, Nodes::Mapping::BLOCK + o.members.each do |member| + @emitter.scalar member.to_s, nil, nil, true, false, Nodes::Scalar::ANY + accept o.send member + end + @emitter.end_mapping + + # Dump the ivars + accept 'ivars' + @emitter.start_mapping nil, nil, true, Nodes::Mapping::BLOCK + ivars.each do |ivar| + accept ivar.to_s + accept o.instance_variable_get ivar + end + @emitter.end_mapping + + @emitter.end_mapping + end + end unless RUBY_VERSION < "3.2" + + def visit_Struct o + tag = ['!ruby/struct', o.class.name].compact.join(':') + + register o, @emitter.start_mapping(nil, tag, false, Nodes::Mapping::BLOCK) + o.members.each do |member| + @emitter.scalar member.to_s, nil, nil, true, false, Nodes::Scalar::ANY + accept o[member] + end + + dump_ivars o + + @emitter.end_mapping + end + + def visit_Exception o + dump_exception o, o.message.to_s + end + + def visit_NameError o + dump_exception o, o.message.to_s + end + + def visit_Regexp o + register o, @emitter.scalar(o.inspect, nil, '!ruby/regexp', false, false, Nodes::Scalar::ANY) + end + + def visit_Date o + formatted = format_date o + register o, @emitter.scalar(formatted, nil, nil, true, false, Nodes::Scalar::ANY) + end + + def visit_DateTime o + t = o.italy + formatted = format_time t, t.offset.zero? + tag = '!ruby/object:DateTime' + register o, @emitter.scalar(formatted, nil, tag, false, false, Nodes::Scalar::ANY) + end + + def visit_Time o + formatted = format_time o + register o, @emitter.scalar(formatted, nil, nil, true, false, Nodes::Scalar::ANY) + end + + def visit_Rational o + register o, @emitter.start_mapping(nil, '!ruby/object:Rational', false, Nodes::Mapping::BLOCK) + + [ + 'denominator', o.denominator.to_s, + 'numerator', o.numerator.to_s + ].each do |m| + @emitter.scalar m, nil, nil, true, false, Nodes::Scalar::ANY + end + + @emitter.end_mapping + end + + def visit_Complex o + register o, @emitter.start_mapping(nil, '!ruby/object:Complex', false, Nodes::Mapping::BLOCK) + + ['real', o.real.to_s, 'image', o.imag.to_s].each do |m| + @emitter.scalar m, nil, nil, true, false, Nodes::Scalar::ANY + end + + @emitter.end_mapping + end + + def visit_Integer o + @emitter.scalar o.to_s, nil, nil, true, false, Nodes::Scalar::ANY + end + alias :visit_TrueClass :visit_Integer + alias :visit_FalseClass :visit_Integer + + def visit_Float o + if o.nan? + @emitter.scalar '.nan', nil, nil, true, false, Nodes::Scalar::ANY + elsif o.infinite? + @emitter.scalar((o.infinite? > 0 ? '.inf' : '-.inf'), + nil, nil, true, false, Nodes::Scalar::ANY) + else + @emitter.scalar o.to_s, nil, nil, true, false, Nodes::Scalar::ANY + end + end + + def visit_BigDecimal o + @emitter.scalar o._dump, nil, '!ruby/object:BigDecimal', false, false, Nodes::Scalar::ANY + end + + def visit_String o + plain = true + quote = true + style = Nodes::Scalar::PLAIN + tag = nil + + if binary?(o) + o = [o].pack('m0') + tag = '!binary' # FIXME: change to below when syck is removed + #tag = 'tag:yaml.org,2002:binary' + style = Nodes::Scalar::LITERAL + plain = false + quote = false + elsif o.match?(/\n(?!\Z)/) # match \n except blank line at the end of string + style = Nodes::Scalar::LITERAL + elsif o == '<<' + style = Nodes::Scalar::SINGLE_QUOTED + tag = 'tag:yaml.org,2002:str' + plain = false + quote = false + elsif o == 'y' || o == 'Y' || o == 'n' || o == 'N' + style = Nodes::Scalar::DOUBLE_QUOTED + elsif @line_width && o.length > @line_width + style = Nodes::Scalar::FOLDED + elsif o.match?(/^[^[:word:]][^"]*$/) + style = Nodes::Scalar::DOUBLE_QUOTED + elsif not String === @ss.tokenize(o) or /\A0[0-7]*[89]/.match?(o) + style = Nodes::Scalar::SINGLE_QUOTED + end + + is_primitive = o.class == ::String + ivars = is_primitive ? [] : o.instance_variables + + if ivars.empty? + unless is_primitive + tag = "!ruby/string:#{o.class}" + plain = false + quote = false + end + @emitter.scalar o, nil, tag, plain, quote, style + else + maptag = '!ruby/string'.dup + maptag << ":#{o.class}" unless o.class == ::String + + register o, @emitter.start_mapping(nil, maptag, false, Nodes::Mapping::BLOCK) + @emitter.scalar 'str', nil, nil, true, false, Nodes::Scalar::ANY + @emitter.scalar o, nil, tag, plain, quote, style + + dump_ivars o + + @emitter.end_mapping + end + end + + def visit_Module o + raise TypeError, "can't dump anonymous module: #{o}" unless o.name + register o, @emitter.scalar(o.name, nil, '!ruby/module', false, false, Nodes::Scalar::SINGLE_QUOTED) + end + + def visit_Class o + raise TypeError, "can't dump anonymous class: #{o}" unless o.name + register o, @emitter.scalar(o.name, nil, '!ruby/class', false, false, Nodes::Scalar::SINGLE_QUOTED) + end + + def visit_Range o + register o, @emitter.start_mapping(nil, '!ruby/range', false, Nodes::Mapping::BLOCK) + ['begin', o.begin, 'end', o.end, 'excl', o.exclude_end?].each do |m| + accept m + end + @emitter.end_mapping + end + + def visit_Hash o + if o.class == ::Hash + register(o, @emitter.start_mapping(nil, nil, true, Psych::Nodes::Mapping::BLOCK)) + o.each do |k,v| + accept(@stringify_names && Symbol === k ? k.to_s : k) + accept v + end + @emitter.end_mapping + else + visit_hash_subclass o + end + end + + def visit_Psych_Set o + register(o, @emitter.start_mapping(nil, '!set', false, Psych::Nodes::Mapping::BLOCK)) + + o.each do |k,v| + accept(@stringify_names && Symbol === k ? k.to_s : k) + accept v + end + + @emitter.end_mapping + end + + def visit_Array o + if o.class == ::Array + visit_Enumerator o + else + visit_array_subclass o + end + end + + def visit_Enumerator o + register o, @emitter.start_sequence(nil, nil, true, Nodes::Sequence::BLOCK) + o.each { |c| accept c } + @emitter.end_sequence + end + + def visit_NilClass o + @emitter.scalar('', nil, 'tag:yaml.org,2002:null', true, false, Nodes::Scalar::ANY) + end + + def visit_Symbol o + if o.empty? + @emitter.scalar "", nil, '!ruby/symbol', false, false, Nodes::Scalar::ANY + else + @emitter.scalar ":#{o}", nil, nil, true, false, Nodes::Scalar::ANY + end + end + + def visit_BasicObject o + tag = Psych.dump_tags[o.class] + tag ||= "!ruby/marshalable:#{o.class.name}" + + map = @emitter.start_mapping(nil, tag, false, Nodes::Mapping::BLOCK) + register(o, map) + + o.marshal_dump.each(&method(:accept)) + + @emitter.end_mapping + end + + private + + def binary? string + string.encoding == Encoding::ASCII_8BIT && !string.ascii_only? + end + + def visit_array_subclass o + tag = "!ruby/array:#{o.class}" + ivars = o.instance_variables + if ivars.empty? + node = @emitter.start_sequence(nil, tag, false, Nodes::Sequence::BLOCK) + register o, node + o.each { |c| accept c } + @emitter.end_sequence + else + node = @emitter.start_mapping(nil, tag, false, Nodes::Sequence::BLOCK) + register o, node + + # Dump the internal list + accept 'internal' + @emitter.start_sequence(nil, nil, true, Nodes::Sequence::BLOCK) + o.each { |c| accept c } + @emitter.end_sequence + + # Dump the ivars + accept 'ivars' + @emitter.start_mapping(nil, nil, true, Nodes::Sequence::BLOCK) + ivars.each do |ivar| + accept ivar + accept o.instance_variable_get ivar + end + @emitter.end_mapping + + @emitter.end_mapping + end + end + + def visit_hash_subclass o + ivars = o.instance_variables + if ivars.any? + tag = "!ruby/hash-with-ivars:#{o.class}" + node = @emitter.start_mapping(nil, tag, false, Psych::Nodes::Mapping::BLOCK) + register(o, node) + + # Dump the ivars + accept 'ivars' + @emitter.start_mapping nil, nil, true, Nodes::Mapping::BLOCK + o.instance_variables.each do |ivar| + accept ivar + accept o.instance_variable_get ivar + end + @emitter.end_mapping + + # Dump the elements + accept 'elements' + @emitter.start_mapping nil, nil, true, Nodes::Mapping::BLOCK + o.each do |k,v| + accept k + accept v + end + @emitter.end_mapping + + @emitter.end_mapping + else + tag = "!ruby/hash:#{o.class}" + node = @emitter.start_mapping(nil, tag, false, Psych::Nodes::Mapping::BLOCK) + register(o, node) + o.each do |k,v| + accept k + accept v + end + @emitter.end_mapping + end + end + + def dump_list o + end + + def dump_exception o, msg + tag = ['!ruby/exception', o.class.name].join ':' + + @emitter.start_mapping nil, tag, false, Nodes::Mapping::BLOCK + + if msg + @emitter.scalar 'message', nil, nil, true, false, Nodes::Scalar::ANY + accept msg + end + + @emitter.scalar 'backtrace', nil, nil, true, false, Nodes::Scalar::ANY + accept o.backtrace + + dump_ivars o + + @emitter.end_mapping + end + + def format_time time, utc = time.utc? + if utc + time.strftime("%Y-%m-%d %H:%M:%S.%9N Z") + else + time.strftime("%Y-%m-%d %H:%M:%S.%9N %:z") + end + end + + def format_date date + date.strftime("%Y-%m-%d") + end + + def register target, yaml_obj + @st.register target, yaml_obj + yaml_obj + end + + def dump_coder o + @coders << o + tag = Psych.dump_tags[o.class] + unless tag + klass = o.class == Object ? nil : o.class.name + tag = ['!ruby/object', klass].compact.join(':') + end + + c = Psych::Coder.new(tag) + o.encode_with(c) + emit_coder c, o + end + + def emit_coder c, o + case c.type + when :scalar + @emitter.scalar c.scalar, nil, c.tag, c.tag.nil?, false, c.style + when :seq + @emitter.start_sequence nil, c.tag, c.tag.nil?, c.style + c.seq.each do |thing| + accept thing + end + @emitter.end_sequence + when :map + register o, @emitter.start_mapping(nil, c.tag, c.implicit, c.style) + c.map.each do |k,v| + accept k + accept v + end + @emitter.end_mapping + when :object + accept c.object + end + end + + def dump_ivars target + target.instance_variables.each do |iv| + @emitter.scalar("#{iv.to_s.sub(/^@/, '')}", nil, nil, true, false, Nodes::Scalar::ANY) + accept target.instance_variable_get(iv) + end + end + end + + class RestrictedYAMLTree < YAMLTree + DEFAULT_PERMITTED_CLASSES = { + TrueClass => true, + FalseClass => true, + NilClass => true, + Integer => true, + Float => true, + String => true, + Array => true, + Hash => true, + }.compare_by_identity.freeze + + def initialize emitter, ss, options + super + @permitted_classes = DEFAULT_PERMITTED_CLASSES.dup + Array(options[:permitted_classes]).each do |klass| + @permitted_classes[klass] = true + end + @permitted_symbols = {}.compare_by_identity + Array(options[:permitted_symbols]).each do |symbol| + @permitted_symbols[symbol] = true + end + @aliases = options.fetch(:aliases, false) + end + + def accept target + if !@aliases && @st.key?(target) + raise BadAlias, "Tried to dump an aliased object" + end + + unless Symbol === target || @permitted_classes[target.class] + raise DisallowedClass.new('dump', target.class.name || target.class.inspect) + end + + super + end + + def visit_Symbol sym + unless @permitted_classes[Symbol] || @permitted_symbols[sym] + raise DisallowedClass.new('dump', "Symbol(#{sym.inspect})") + end + + super + end + end + end +end diff --git a/ext/psych/lib/psych/y.rb b/ext/psych/lib/psych/y.rb new file mode 100644 index 0000000000..e857953c04 --- /dev/null +++ b/ext/psych/lib/psych/y.rb @@ -0,0 +1,10 @@ +# frozen_string_literal: true +module Kernel + ### + # An alias for Psych.dump_stream meant to be used with IRB. + def y *objects + puts Psych.dump_stream(*objects) + end + private :y +end + |