dolos 0.1.2 → 0.2.0

data/{lib/example.rb → examples/letter.rb}

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 require_relative 'dolos'
-require_relative 'dolos_common_parsers/common_parsers'
+require_relative 'dolos_common_parsers/arsers/common_parsers'
 
 include Dolos
 
@@ -27,12 +27,12 @@ alpha_with_lt = char_in("ąčęėįšųūžĄČĘĖĮŠŲŪŽ") | alpha
 
 # Capture all letters in a row and join them,
 # because they are captured as elements of array by each alpha_with_lt parser.
-first_name = alpha_with_lt.rep.capture!.map(&:join)
-last_name = alpha_with_lt.rep.capture!.map(&:join)
+first_name = alpha_with_lt.rep.map(&:join).capture!
+last_name = alpha_with_lt.rep.map(&:join).capture!
 
 # Combine first line parsers
 # Consume zero or more whitespace, after that honorific must follow and so on
-name_line = ws.rep0 >> honorific >> first_name >> ws >> last_name >> eol
+name_line = ws.rep0 & honorific & first_name & ws & last_name & eol
 
 # Next line is company info
 # We could choose to accept UAB and AB or just AB and etc.
@@ -42,9 +42,9 @@ quote_open = c("„")
 quote_close = c("“")
 
 # Consume LT alphabet with whitespace
-company_name = (alpha_with_lt | ws).rep.capture!.map(&:join)
-company_info = company_type >> ws.rep0 >> quote_open >> company_name >> quote_close
-second_line = ws.rep0 >> company_info >> eol
+company_name = (alpha_with_lt | ws).rep.map(&:join).capture!
+company_info = company_type & ws.rep0 & quote_open & company_name & quote_close
+second_line = ws.rep0 & company_info & eol
 
 # Address line
 # 'char_while' will consume characters while passed predicate is true
@@ -52,18 +52,18 @@ second_line = ws.rep0 >> company_info >> eol
 # After that result is captured and mapped to hash
 # Mapping to hash so at the end its easy to tell tuples apart
 # Also while mapping, doing some cleaning with '.strip'
-street_name = char_while(->(char) { !char.match(/\d/) }).capture!.map(&:first).map { |s| { street: s.strip } }
-building = digits.capture!.map(&:first).map { |s| { building: s.strip } }
-address_line = ws.rep0 >> street_name >> building >> eol
+street_name = char_while(->(char) { !char.match(/\d/) }).map { |s| { street: s.strip } }.capture!
+building = digits.map { |s| { building: s.strip } }.capture!
+address_line = ws.rep0 & street_name & building & eol
 
 # City line
 # All digits can be matched here or 'digits.rep(5)' could be used. Also joining with map.
-postcode = digits.capture!.map(&:join).map { |s| { postcode: s.strip } }
-city = alpha_with_lt.rep.capture!.map(&:join).map { |s| { city: s.strip } }
-city_line = ws.rep0 >> postcode >> ws >> city >> eol
+postcode = digits.map { |s| { postcode: s.strip } }.capture!
+city = alpha_with_lt.rep.map(&:join).map { |s| { city: s.strip } }.capture!
+city_line = ws.rep0 & postcode & ws & city & eol
 
 # Full letter parser which is combined from all previous parsers. All previous parsers can be ran separately.
-letter_parser = name_line >> second_line >> address_line >> city_line
+letter_parser = name_line & second_line & address_line & city_line
 result = letter_parser.run(letter)
 
 pp result.captures

data/lib/dolos/parser_state.rb

@@ -3,10 +3,12 @@
 module Dolos
   class ParserState
     attr_reader :input
+    attr_accessor :last_success_position
 
     def initialize(input)
       @input = StringIOWrapper.new(input)
+      @last_success_position = 0
     end
   end
-
 end
+

data/lib/dolos/parsers.rb

@@ -10,14 +10,17 @@ module Dolos
           Success.new(utf8_str, str.bytesize)
         else
           advanced = state.input.offset
+          got_error = state.input.io.string.byteslice(state.input.backup, advanced)
           state.input.rollback
           Failure.new(
-            "Expected #{str.inspect} but got #{state.input.io.string.inspect}",
-            advanced
+            "Expected #{str.inspect} but got #{got_error.inspect}",
+            advanced,
+            state
           )
         end
       end
     end
+
     alias_method :c, :string
 
     def regex(pattern)
@@ -30,13 +33,13 @@ module Dolos
           state.input.rollback
           Failure.new(
             "Expected pattern #{pattern.inspect} but got #{state.input.io.string.inspect}",
-            advanced
+            advanced,
+            state
           )
         end
       end
     end
 
-
     def any_char
       Parser.new do |state|
         state.input.mark_offset
@@ -48,7 +51,11 @@ module Dolos
         else
           advanced = state.input.offset
           state.input.rollback
-          Failure.new('Expected any character but got end of input', advanced)
+          Failure.new(
+            'Expected any character but got end of input',
+            advanced,
+            state
+          )
         end
       end
     end
@@ -71,7 +78,8 @@ module Dolos
           state.input.rollback
           Failure.new(
             "Expected one of #{characters_array.inspect} but got #{char.inspect}",
-            advanced
+            advanced,
+            state
           )
         end
       end
@@ -92,12 +100,37 @@ module Dolos
 
         if buffer.empty?
           advanced = state.input.offset
-          Failure.new("Predicate never returned true", advanced)
+          Failure.new(
+            "Predicate never returned true",
+            advanced,
+            state
+          )
         else
           Success.new(buffer, 0)
         end
       end
     end
 
+    # Unstable API
+    def recursive(&block)
+      recursive_parser = nil
+
+      placeholder = Parser.new do |state|
+        raise "Recursive parser accessed before it was initialized!" if recursive_parser.nil?
+
+        recursive_parser.call.run_with_state(state).tap do |result|
+          if result.failure?
+            error_msg = "Error in recursive structure around position #{state.input.offset}: #{result.message}"
+            Failure.new(error_msg, state.input.offset, state)
+          end
+        end
+      end
+
+      recursive_parser = -> { block.call(placeholder) }
+      placeholder
+    end
+
+
+
   end
 end

data/lib/dolos/result.rb

@@ -10,20 +10,21 @@ module Dolos
     def initialize(value, length, captures = [])
       @value = value
       @length = length
-      # @captures = captures || value
       @captures = captures
     end
 
-    def capture!
-      if value.is_a?(Array)
-        value.each do |v|
-          captures << v
-        end
+    # can be some named capture, :street, {:street => capture }
+    # or an array, [], [capture]
+    def capture!(wrap_in = nil)
+      mapped_value = self.value  # use the transformed value here
+
+      if wrap_in.is_a?(Array)
+        save_capture([mapped_value])
+      elsif wrap_in.is_a?(Symbol)
+        save_capture({ wrap_in => mapped_value })
       else
-        captures << value
+        save_capture(mapped_value)
       end
-
-      Success.new(value, length, captures)
     end
 
     def inspect
@@ -37,21 +38,54 @@ module Dolos
     def failure?
       false
     end
+
+    private
+
+    def save_capture(val)
+      if val.is_a?(Array)
+        val.each do |v|
+          captures << v
+        end
+      else
+        captures << val
+      end
+
+      Success.new(val, length, captures)
+    end
   end
 
   class Failure < Result
-    attr_reader :message, :committed
+    attr_reader :message, :error_position, :state
 
-    def initialize(message, committed)
+    def initialize(message, error_position, state)
       @message = message
-      @committed = committed
+      @error_position = error_position
+      @state = state
     end
 
     def inspect
+      pretty_print
+    end
+
+    def pretty_print
+      input_string = state.input.io.string
+
+      pointer = "^" # This will point to the error position
+
+      context_range = 10 # Chars before and after the error to display
+
+      start_index = [error_position - context_range, 0].max
+      end_index = [error_position + context_range, input_string.length].max
+
+      substring = input_string[start_index..end_index]
+
+      padding = error_position - start_index
+
       [
-        "Failure",
-        "message: #{message}",
-        "committed: #{committed}"
+        "Failure: #{message}",
+        substring,
+        "#{' ' * padding}#{pointer}",
+        "Error Position: #{error_position}, Last Success Position: #{state.last_success_position}"
       ].join("\n")
     end
 

data/lib/dolos/string_io_wrapper.rb

@@ -22,7 +22,6 @@ module Dolos
 
     def matches?(utf8_str)
       read = io.read(utf8_str.bytesize)
-      io.seek(offset)
 
       if read.nil?
         false

data/lib/dolos/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Dolos
-  VERSION = "0.1.2"
+  VERSION = "0.2.0"
 end

data/lib/dolos.rb

@@ -22,21 +22,26 @@ module Dolos
     end
 
     def run_with_state(state)
-      parser_proc.call(state)
+      result = parser_proc.call(state)
+      if result.success?
+        state.last_success_position = state.input.offset
+      end
+      result
     end
 
-    def capture!
+    def capture!(wrap_in = nil)
       Parser.new do |state|
         result = run_with_state(state)
         if result.success?
-          result.capture!
+          result.capture!(wrap_in)
         else
           result
         end
       end
     end
 
-    def map(&block)
+    # Will call block on captures
+    def map_captures(&block)
       Parser.new do |state|
         result = run_with_state(state)
         if result.success?
@@ -47,7 +52,8 @@ module Dolos
       end
     end
 
-    def map_value(&block)
+    # Will call block on tuple of value
+    def map(&block)
       Parser.new do |state|
         result = run_with_state(state)
         if result.success?
@@ -58,7 +64,7 @@ module Dolos
       end
     end
 
-    def flat_map(&block)
+    def combine(&block)
       Parser.new do |state|
         result = run_with_state(state)
         if result.success?
@@ -73,22 +79,44 @@ module Dolos
     end
 
     def flatten
-      map do |captures|
+      map_captures do |captures|
         captures.flatten
       end
     end
 
     def product(other_parser)
-      flat_map do |value1, capture1|
-        other_parser.map_value do |value2|
+      combine do |value1, capture1|
+        other_parser.map do |value2|
           [value1, value2]
-        end.map do |capture2|
+        end.map_captures do |capture2|
+          [capture1, capture2].flatten
+        end
+      end
+    end
+    alias_method :&, :product
+
+    def product_l(other_parser)
+      combine do |value1, capture1|
+        other_parser.map do |_|
+          value1
+        end.map_captures do |capture2|
           [capture1, capture2].flatten
         end
       end
     end
 
-    alias_method :>>, :product
+    def product_r(other_parser)
+      combine do |_, capture1|
+        other_parser.map do |value2|
+          value2
+        end.map_captures do |capture2|
+          [capture1, capture2].flatten
+        end
+      end
+    end
+
+    alias_method :<<, :product_l
+    alias_method :>>, :product_r
 
     def choice(other_parser)
       Parser.new do |state|
@@ -107,25 +135,40 @@ module Dolos
     # rep(n = 2)   # exactly 2
     # repeat(n_min: 2, n_max: 4) # 2 to 4
     # repeat(n_min: 2) # 2 or more
-    def repeat(n_min:, n_max: Float::INFINITY)
+    def repeat(n_min:, n_max: Float::INFINITY, separator: nil)
       Parser.new do |state|
         values = []
         captures = []
         count = 0
+        state.input.mark_offset
 
-        while count < n_max
+        loop do
           result = run_with_state(state.dup)
 
-          break if result.failure?
+          if result.failure? || count >= n_max
+            break
+          end
 
           values << result.value
           captures.concat(result.captures)
           state.input.advance(result.length)
           count += 1
+
+          if separator && count < n_max
+            sep_result = separator.run_with_state(state.dup)
+            break if sep_result.failure?
+
+            state.input.advance(sep_result.length)
+          end
         end
 
         if count < n_min
-          Failure.new("Expected parser to match at least #{n_min} times but matched only #{count} times", false)
+          error_pos = state.input.offset
+          Failure.new(
+            "Expected parser to match at least #{n_min} times but matched only #{count} times",
+            error_pos,
+            state
+          )
         else
           Success.new(values, 0, captures)
         end
@@ -158,5 +201,22 @@ module Dolos
     end
     alias_method :opt, :optional
 
+    # Unstable API
+    # Used to declare lazy parser to avoid infinite loops in recursive parsers
+    def lazy
+      parser_memo = nil
+
+      Parser.new do |state|
+        parser_memo ||= self
+        parser_memo.run_with_state(state)
+      end
+    end
+
+    private
+
+    def combine_and_discard_empty(*arrays)
+      arrays.compact.reject { |arr| arr.is_a?(Array) && arr.empty? }
+    end
+
   end
 end

data/lib/dolos_common_parsers/common_parsers.rb

@@ -10,9 +10,12 @@ module Dolos
       regex(/\n|\r\n|\r/)
     end
 
-    # Capture as String and convert to integer
     def digit
-      regex(/\d/).capture!.map { |capt| capt.map(&:to_i) }
+      regex(/\d/)
+    end
+
+    def int
+      digit.map(&:to_i)
     end
 
     # Capture as string

data/sig/dolos/parser_state.rbs

@@ -1,6 +1,7 @@
 module Dolos
   class ParserState
     attr_reader input: Dolos::StringIOWrapper
+    attr_accessor last_success_position: Integer
 
     def initialize: (String) -> void
   end

data/sig/dolos/result.rbs

@@ -16,6 +16,7 @@ module Dolos
 
   class Failure < Result[bot]
     attr_reader committed: bool
+    attr_reader error_position: Integer
     attr_reader message: String
 
     def captures: -> []

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: dolos
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: 0.2.0
 platform: ruby
 authors:
 - benetis
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-08-16 00:00:00.000000000 Z
+date: 2023-08-19 00:00:00.000000000 Z
 dependencies: []
 description: Parser combinators library for Ruby. In active development, not stable
   yet.
@@ -23,7 +23,10 @@ files:
 - LICENSE.txt
 - README.md
 - Rakefile
+- benchmarks/json/json.rb
+- benchmarks/json/nested_json.json
 - docs/dolos_stable_diff.png
+- examples/letter.rb
 - lib/dolos.rb
 - lib/dolos/parser_state.rb
 - lib/dolos/parsers.rb
@@ -31,7 +34,6 @@ files:
 - lib/dolos/string_io_wrapper.rb
 - lib/dolos/version.rb
 - lib/dolos_common_parsers/common_parsers.rb
-- lib/example.rb
 - sig/dolos.rbs
 - sig/dolos/common_parsers.rbs
 - sig/dolos/parser.rbs