psych-pure 0.1.4 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f1767eb97091e21f83527221b9ef4f156d9db40998b9008422b3dc017ff4ada6
-  data.tar.gz: b7e1d3b11886a31fda7fd7aed9937e2c26eaafa6d5e242a65bb89c8a02b00be7
+  metadata.gz: 33f682a9e0bba7ec024cec2e1ca54aa791aa550e066ac708d7b9cf4e765e6d5a
+  data.tar.gz: 5a0b04e41b5700c481658a9604be9f942f41757231f339e8382d53f2a383b431
 SHA512:
-  metadata.gz: 7103d8ceae77bf15b2f21f9462f65fe495bf6e90206749fded264d192beb1caee6ae1a0585703ce914d7268a1e57f498f4cc1b824c21b3c5fe95edfb1dc1e320
-  data.tar.gz: 570ba899c74d5873ba243771ae3ff71605b59f81b1567f5f0712314aa2f8efbbdfab659c4e9bde63609ce2ef5ccce46352624c0f458ef6a8ddd31826edcfa9da
+  metadata.gz: 164cb1c175e35e791e0206d14cfef2233817e90d2a6ebabb8f5b88ba59cad9c6fd946d651a13c7aab2437fa33e2e1a23a8d775597958e8d06d42ea21bcd6a1f5
+  data.tar.gz: dcab2936344a40ddd58495e6a699aebc77c0d5900c53f0692009b5e4b7383e09ae4896c9c76fa2d6b67599da72a313772613ec9440f659558a51d6598f9e4aca

data/CHANGELOG.md

@@ -6,6 +6,15 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/) a
 
 ## [Unreleased]
 
+## [0.3.0] - 2025-12-25
+
+- Support Psych >= 5.3.0 by adding the `parse_symbols` option.
+
+## [0.2.0] - 2025-11-11
+
+- Add `sequence_indent` option to `Psych::Pure.dump` to control whether sequence elements contained within mapping elements are indented.
+- Properly handle mutation methods on loaded objects that mutate in place.
+
 ## [0.1.4] - 2025-11-10
 
 - Fix up comment handling preceding sequence elements.
@@ -37,7 +46,10 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/) a
 
 - 🎉 Initial release. 🎉
 
-[unreleased]: https://github.com/kddnewton/psych-pure/compare/v0.1.3...HEAD
+[unreleased]: https://github.com/kddnewton/psych-pure/compare/v0.3.0...HEAD
+[0.3.0]: https://github.com/kddnewton/psych-pure/compare/v0.2.0...v0.3.0
+[0.2.0]: https://github.com/kddnewton/psych-pure/compare/v0.1.4...v0.2.0
+[0.1.4]: https://github.com/kddnewton/psych-pure/compare/v0.1.3...v0.1.4
 [0.1.3]: https://github.com/kddnewton/psych-pure/compare/v0.1.2...v0.1.3
 [0.1.2]: https://github.com/kddnewton/psych-pure/compare/v0.1.1...v0.1.2
 [0.1.1]: https://github.com/kddnewton/psych-pure/compare/v0.1.0...v0.1.1

data/lib/psych/pure/version.rb

@@ -2,6 +2,6 @@
 
 module Psych
   module Pure
-    VERSION = "0.1.4"
+    VERSION = "0.3.0"
   end
 end

data/lib/psych/pure.rb

@@ -15,6 +15,17 @@ module Psych
     end
   end
 
+  # If Psych is older than 5.3, we need to modify the ScalarScanner to add the
+  # parse_symbols parameter so that we can use it consistently when we
+  # initialize it.
+  if Psych::VERSION < "5.3"
+    ScalarScanner.prepend(Module.new {
+      def initialize(class_loader, strict_integer: false, parse_symbols: true)
+        super(class_loader, strict_integer: strict_integer)
+      end
+    })
+  end
+
   # A YAML parser written in Ruby.
   module Pure
     # An internal exception is an exception that should not have occurred. It is
@@ -74,6 +85,10 @@ module Psych
       def column(offset)
         offset - @line_offsets[line(offset)]
       end
+
+      def point(offset)
+        "line #{line(offset) + 1} column #{column(offset)}"
+      end
     end
 
     # A location represents a range of bytes in the input string.
@@ -201,15 +216,37 @@ module Psych
       # rely on the source formatting, and need to format it ourselves.
       attr_reader :dirty
 
-      def initialize(object, psych_node, dirty = false)
+      def initialize(object, psych_node)
         super(object)
         @psych_node = psych_node
-        @dirty = dirty
+        @dirty = false
       end
 
-      def replace(psych_node)
-        @psych_node = psych_node
-        @dirty = true
+      def initialize_clone(obj, freeze: nil)
+        super
+        @psych_node = obj.psych_node.dup
+      end
+
+      def initialize_dup(obj)
+        super
+        @psych_node = obj.psych_node.dup
+      end
+
+      # Effectively implement the same method_missing as SimpleDelegator, but
+      # additionally track whether or not the object has been mutated.
+      ruby2_keywords def method_missing(name, *args, &block)
+        takes_block = false
+        target = self.__getobj__ { takes_block = true }
+
+        if !takes_block && target_respond_to?(target, name, false)
+          previous = target.dup
+          result = target.__send__(name, *args, &block)
+
+          @dirty = true unless previous.eql?(target)
+          result
+        else
+          super(name, *args, &block)
+        end
       end
     end
 
@@ -623,8 +660,8 @@ module Psych
           defined?(@comments)
         end
 
-        def to_ruby(symbolize_names: false, freeze: false, strict_integer: false, comments: false)
-          Visitors::ToRuby.create(symbolize_names: symbolize_names, freeze: freeze, strict_integer: strict_integer, comments: comments).accept(self)
+        def to_ruby(symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true, comments: false)
+          Visitors::ToRuby.create(symbolize_names: symbolize_names, freeze: freeze, strict_integer: strict_integer, parse_symbols: parse_symbols, comments: comments).accept(self)
         end
       end
 
@@ -710,9 +747,9 @@ module Psych
 
       # Extend the ToRuby singleton to be able to pass the comments option.
       module ToRubySingleton
-        def create(symbolize_names: false, freeze: false, strict_integer: false, comments: false)
+        def create(symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true, comments: false)
           class_loader = ClassLoader.new
-          scanner = ScalarScanner.new(class_loader, strict_integer: strict_integer)
+          scanner = ScalarScanner.new(class_loader, strict_integer: strict_integer, parse_symbols: parse_symbols)
           new(scanner, class_loader, symbolize_names: symbolize_names, freeze: freeze, comments: comments)
         end
       end
@@ -909,6 +946,123 @@ module Psych
     # The parser is responsible for taking a YAML string and converting it into
     # a series of events that can be used by the consumer.
     class Parser
+      # A stack of contexts that the parser is currently within. We use this to
+      # decorate error messages with the context in which they occurred.
+      class Context
+        class BlockMapping
+          attr_reader :pos, :indent
+
+          def initialize(pos, indent)
+            @pos = pos
+            @indent = indent
+          end
+
+          def format(source)
+            "block mapping at #{source.point(pos)}#{indent == -1 ? "" : " (indent=#{indent})"}"
+          end
+        end
+
+        class BlockSequence
+          attr_reader :pos, :indent
+
+          def initialize(pos, indent)
+            @pos = pos
+            @indent = indent
+          end
+
+          def format(source)
+            "block sequence at #{source.point(pos)}#{indent == -1 ? "" : " (indent=#{indent})"}"
+          end
+        end
+
+        class DoubleQuotedScalar
+          attr_reader :pos
+
+          def initialize(pos)
+            @pos = pos
+          end
+
+          def format(source)
+            "double quoted scalar at #{source.point(pos)}"
+          end
+        end
+
+        class FlowMapping
+          attr_reader :pos, :context
+
+          def initialize(pos, context)
+            @pos = pos
+            @context = context
+          end
+
+          def format(source)
+            "flow mapping at #{source.point(pos)} (context=#{context})"
+          end
+        end
+
+        class FlowSequence
+          attr_reader :pos, :context
+
+          def initialize(pos, context)
+            @pos = pos
+            @context = context
+          end
+
+          def format(source)
+            "flow sequence at #{source.point(pos)} (context=#{context})"
+          end
+        end
+
+        private_constant :BlockMapping, :BlockSequence, :DoubleQuotedScalar, :FlowMapping, :FlowSequence
+
+        def initialize
+          @contexts = []
+          @deepest = []
+        end
+
+        def syntax_error(source, filename, pos, message)
+          unless (stack = (@contexts.empty? ? @deepest : @contexts)).empty?
+            pos = stack.last.pos
+            message = "#{message}\nwithin:\n#{stack.map { |element| " #{element.format(source)}\n" }.join}"
+          end
+
+          SyntaxError.new(filename, source.line(pos), source.column(pos), pos, message, nil)
+        end
+
+        def within_block_mapping(pos, indent, &blk)
+          within(BlockMapping.new(pos, indent), &blk)
+        end
+
+        def within_block_sequence(pos, indent, &blk)
+          within(BlockSequence.new(pos, indent), &blk)
+        end
+
+        def within_double_quoted_scalar(pos, &blk)
+          within(DoubleQuotedScalar.new(pos), &blk)
+        end
+
+        def within_flow_mapping(pos, context, &blk)
+          within(FlowMapping.new(pos, context), &blk)
+        end
+
+        def within_flow_sequence(pos, context, &blk)
+          within(FlowSequence.new(pos, context), &blk)
+        end
+
+        private
+
+        def within(object)
+          @contexts.push(object)
+          @deepest = @contexts.dup if @contexts.length > @deepest.length
+
+          begin
+            yield
+          ensure
+            @contexts.pop
+          end
+        end
+      end
+
       # Initialize a new parser with the given source string.
       def initialize(handler)
         # These are used to track the current state of the parser.
@@ -952,6 +1106,11 @@ module Psych
         # This parser can optionally parse comments and attach them to the
         # resulting tree, if the option is passed.
         @comments = nil
+
+        # The context of the parser at any given time, which is used to decorate
+        # error messages to make it easier to find the specific location where
+        # they occurred.
+        @context = Context.new
       end
 
       # Top-level parse function that starts the parsing process.
@@ -985,9 +1144,7 @@ module Psych
 
       # Raise a syntax error with the given message.
       def raise_syntax_error(message)
-        line = @source.line(@scanner.pos)
-        column = @source.column(@scanner.pos)
-        raise SyntaxError.new(@filename, line, column, @scanner.pos, message, nil)
+        raise @context.syntax_error(@source, @filename, @scanner.pos, message)
       end
 
       # ------------------------------------------------------------------------
@@ -1856,30 +2013,32 @@ module Psych
       def parse_c_double_quoted(n, c)
         pos_start = @scanner.pos
 
-        if try { match("\"") && parse_nb_double_text(n, c) && match("\"") }
-          end1 = "(?:\\\\\\r?\\n[ \\t]*)"
-          end2 = "(?:[ \\t]*\\r?\\n[ \\t]*)"
-          hex = "[0-9a-fA-F]"
-          hex2 = "(?:\\\\x(#{hex}{2}))"
-          hex4 = "(?:\\\\u(#{hex}{4}))"
-          hex8 = "(?:\\\\U(#{hex}{8}))"
-
-          value = from(pos_start).byteslice(1...-1)
-          value.gsub!(%r{(?:\r\n|#{end1}|#{end2}+|#{hex2}|#{hex4}|#{hex8}|\\[\\ "/_0abefnrt\tvLNP])}) do |m|
-            case m
-            when /\A(?:#{hex2}|#{hex4}|#{hex8})\z/o
-              m[2..].to_i(16).chr(Encoding::UTF_8)
-            when /\A#{end1}\z/o
-              ""
-            when /\A#{end2}+\z/o
-              m.sub(/#{end2}/, "").gsub(/#{end2}/, "\n").then { |r| r.empty? ? " " : r }
-            else
-              C_DOUBLE_QUOTED_UNESCAPES.fetch(m, m)
+        @context.within_double_quoted_scalar(@scanner.pos) do
+          if try { match("\"") && parse_nb_double_text(n, c) && match("\"") }
+            end1 = "(?:\\\\\\r?\\n[ \\t]*)"
+            end2 = "(?:[ \\t]*\\r?\\n[ \\t]*)"
+            hex = "[0-9a-fA-F]"
+            hex2 = "(?:\\\\x(#{hex}{2}))"
+            hex4 = "(?:\\\\u(#{hex}{4}))"
+            hex8 = "(?:\\\\U(#{hex}{8}))"
+
+            value = from(pos_start).byteslice(1...-1)
+            value.gsub!(%r{(?:\r\n|#{end1}|#{end2}+|#{hex2}|#{hex4}|#{hex8}|\\[\\ "/_0abefnrt\tvLNP])}) do |m|
+              case m
+              when /\A(?:#{hex2}|#{hex4}|#{hex8})\z/o
+                m[2..].to_i(16).chr(Encoding::UTF_8)
+              when /\A#{end1}\z/o
+                ""
+              when /\A#{end2}+\z/o
+                m.sub(/#{end2}/, "").gsub(/#{end2}/, "\n").then { |r| r.empty? ? " " : r }
+              else
+                C_DOUBLE_QUOTED_UNESCAPES.fetch(m, m)
+              end
             end
-          end
 
-          events_push_flush_properties(Scalar.new(Location.new(@source, pos_start, @scanner.pos), from(pos_start), value, Nodes::Scalar::DOUBLE_QUOTED))
-          true
+            events_push_flush_properties(Scalar.new(Location.new(@source, pos_start, @scanner.pos), from(pos_start), value, Nodes::Scalar::DOUBLE_QUOTED))
+            true
+          end
         end
       end
 
@@ -2249,14 +2408,16 @@ module Psych
       def parse_c_flow_sequence(n, c)
         try do
           if match("[")
-            events_push_flush_properties(SequenceStart.new(Location.new(@source, @scanner.pos - 1, @scanner.pos), Nodes::Sequence::FLOW))
+            @context.within_flow_sequence(@scanner.pos, c) do
+              events_push_flush_properties(SequenceStart.new(Location.new(@source, @scanner.pos - 1, @scanner.pos), Nodes::Sequence::FLOW))
 
-            parse_s_separate(n, c)
-            parse_ns_s_flow_seq_entries(n, parse_in_flow(c))
+              parse_s_separate(n, c)
+              parse_ns_s_flow_seq_entries(n, parse_in_flow(c))
 
-            if match("]")
-              events_push_flush_properties(SequenceEnd.new(Location.new(@source, @scanner.pos - 1, @scanner.pos)))
-              true
+              if match("]")
+                events_push_flush_properties(SequenceEnd.new(Location.new(@source, @scanner.pos - 1, @scanner.pos)))
+                true
+              end
             end
           end
         end
@@ -2297,14 +2458,16 @@ module Psych
       def parse_c_flow_mapping(n, c)
         try do
           if match("{")
-            events_push_flush_properties(MappingStart.new(Location.new(@source, @scanner.pos - 1, @scanner.pos), Nodes::Mapping::FLOW))
+            @context.within_flow_mapping(@scanner.pos, c) do
+              events_push_flush_properties(MappingStart.new(Location.new(@source, @scanner.pos - 1, @scanner.pos), Nodes::Mapping::FLOW))
 
-            parse_s_separate(n, c)
-            parse_ns_s_flow_map_entries(n, parse_in_flow(c))
+              parse_s_separate(n, c)
+              parse_ns_s_flow_map_entries(n, parse_in_flow(c))
 
-            if match("}")
-              events_push_flush_properties(MappingEnd.new(Location.new(@source, @scanner.pos - 1, @scanner.pos)))
-              true
+              if match("}")
+                events_push_flush_properties(MappingEnd.new(Location.new(@source, @scanner.pos - 1, @scanner.pos)))
+                true
+              end
             end
           end
         end
@@ -3017,18 +3180,20 @@ module Psych
       def parse_l_block_sequence(n)
         return false if (m = detect_indent(n)) == 0
 
-        events_cache_push
-        events_push_flush_properties(SequenceStart.new(Location.point(@source, @scanner.pos), Nodes::Sequence::BLOCK))
+        @context.within_block_sequence(@scanner.pos, n) do
+          events_cache_push
+          events_push_flush_properties(SequenceStart.new(Location.point(@source, @scanner.pos), Nodes::Sequence::BLOCK))
 
-        if try { plus { try { parse_s_indent(n + m) && parse_c_l_block_seq_entry(n + m) } } }
-          events_cache_flush
-          events_push_flush_properties(SequenceEnd.new(Location.point(@source, @scanner.pos)))
-          true
-        else
-          event = events_cache_pop[0]
-          @anchor = event.anchor
-          @tag = event.tag
-          false
+          if try { plus { try { parse_s_indent(n + m) && parse_c_l_block_seq_entry(n + m) } } }
+            events_cache_flush
+            events_push_flush_properties(SequenceEnd.new(Location.point(@source, @scanner.pos)))
+            true
+          else
+            event = events_cache_pop[0]
+            @anchor = event.anchor
+            @tag = event.tag
+            false
+          end
         end
       end
 
@@ -3093,16 +3258,18 @@ module Psych
       def parse_l_block_mapping(n)
         return false if (m = detect_indent(n)) == 0
 
-        events_cache_push
-        events_push_flush_properties(MappingStart.new(Location.point(@source, @scanner.pos), Nodes::Mapping::BLOCK))
+        @context.within_block_mapping(@scanner.pos, n) do
+          events_cache_push
+          events_push_flush_properties(MappingStart.new(Location.point(@source, @scanner.pos), Nodes::Mapping::BLOCK))
 
-        if try { plus { try { parse_s_indent(n + m) && parse_ns_l_block_map_entry(n + m) } } }
-          events_cache_flush
-          events_push_flush_properties(MappingEnd.new(Location.point(@source, @scanner.pos)))
-          true
-        else
-          events_cache_pop
-          false
+          if try { plus { try { parse_s_indent(n + m) && parse_ns_l_block_map_entry(n + m) } } }
+            events_cache_flush
+            events_push_flush_properties(MappingEnd.new(Location.point(@source, @scanner.pos)))
+            true
+          else
+            events_cache_pop
+            false
+          end
         end
       end
 
@@ -3559,8 +3726,9 @@ module Psych
       # The visitor is responsible for walking the tree and generating the YAML
       # output.
       class Visitor
-        def initialize(q)
+        def initialize(q, sequence_indent: false)
           @q = q
+          @sequence_indent = sequence_indent
         end
 
         # Visit an AliasNode.
@@ -3777,6 +3945,11 @@ module Psych
             elsif inlined || value.anchor || value.tag || value.value.empty?
               @q.text(" ")
               @q.nest(2) { visit(value) }
+            elsif @sequence_indent
+              @q.nest(2) do
+                @q.breakable
+                visit(value)
+              end
             else
               @q.breakable
               visit(value)
@@ -4006,7 +4179,7 @@ module Psych
             q.text(" ")
           end
 
-          node.accept(Visitor.new(q))
+          node.accept(Visitor.new(q, sequence_indent: @options.fetch(:sequence_indent, false)))
           q.breakable
           q.current_group.break
           q.flush
@@ -4193,19 +4366,19 @@ module Psych
       end
     end
 
-    def self.unsafe_load(yaml, filename: nil, fallback: false, symbolize_names: false, freeze: false, strict_integer: false, comments: false)
+    def self.unsafe_load(yaml, filename: nil, fallback: false, symbolize_names: false, freeze: false, strict_integer: false, parse_symbols: true, comments: false)
       result = parse(yaml, filename: filename, comments: comments)
       return fallback unless result
 
-      result.to_ruby(symbolize_names: symbolize_names, freeze: freeze, strict_integer: strict_integer, comments: comments)
+      result.to_ruby(symbolize_names: symbolize_names, freeze: freeze, strict_integer: strict_integer, parse_symbols: parse_symbols, comments: comments)
     end
 
-    def self.safe_load(yaml, permitted_classes: [], permitted_symbols: [], aliases: false, filename: nil, fallback: nil, symbolize_names: false, freeze: false, strict_integer: false, comments: false)
+    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, comments: false)
       result = parse(yaml, filename: filename, comments: comments)
       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)
+      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, comments: comments)
@@ -4216,7 +4389,7 @@ module Psych
       visitor.accept(result)
     end
 
-    def self.load(yaml, permitted_classes: [Symbol], permitted_symbols: [], aliases: false, filename: nil, fallback: nil, symbolize_names: false, freeze: false, strict_integer: false, comments: false)
+    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, comments: false)
       safe_load(
         yaml,
         permitted_classes: permitted_classes,
@@ -4227,6 +4400,7 @@ module Psych
         symbolize_names: symbolize_names,
         freeze: freeze,
         strict_integer: strict_integer,
+        parse_symbols: parse_symbols,
         comments: comments
       )
     end
@@ -4245,6 +4419,21 @@ module Psych
       result
     end
 
+    def self.safe_load_stream(yaml, filename: nil, permitted_classes: [], aliases: false, comments: 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, comments: comments)
+      end
+
+      if block_given?
+        documents.each { |doc| yield doc }
+        nil
+      else
+        documents
+      end
+    end
+
     def self.unsafe_load_file(filename, **kwargs)
       File.open(filename, "r:bom|utf-8") do |f|
         self.unsafe_load(f, filename: filename, **kwargs)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: psych-pure
 version: !ruby/object:Gem::Version
-  version: 0.1.4
+  version: 0.3.0
 platform: ruby
 authors:
 - Kevin Newton
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2025-11-11 00:00:00.000000000 Z
+date: 2025-12-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: psych
@@ -113,7 +113,7 @@ licenses:
 - MIT
 metadata:
   bug_tracker_uri: https://github.com/kddnewton/psych-pure/issues
-  changelog_uri: https://github.com/kddnewton/psych-pure/blob/v0.1.4/CHANGELOG.md
+  changelog_uri: https://github.com/kddnewton/psych-pure/blob/v0.3.0/CHANGELOG.md
   source_code_uri: https://github.com/kddnewton/psych-pure
   rubygems_mfa_required: 'true'
 post_install_message: