birman 0.1.0

data/lib/birman/rules.rb

@@ -0,0 +1,77 @@
+# frozen_string_literal: true
+
+module Birman
+  # A standard library of rule modules. Include the ones you need;
+  # later includes override earlier ones (Comments replaces
+  # Whitespace's `spacing`, for example).
+  #
+  # Value rules (number, string, identifier) are lexemes: in a grammar
+  # that declares `ignore :spacing`, they skip trailing spacing on their own.
+  module Rules
+    # Whitespace as the `spacing` rule, plus a `token` helper for grammars
+    # that don't declare `ignore`.
+    module Whitespace
+      extend RuleSet
+
+      rule(:spacing) { match(/\s*/).quiet }
+
+      # A matcher followed by whitespace. Only needed in grammars that
+      # don't declare `ignore`.
+      def token(m) = Pick.new([Birman.coerce(m), spacing], 0)
+    end
+
+    # Whitespace including comments. Works on its own or after
+    # Whitespace. Override comment_pattern for other comment syntaxes.
+    module Comments
+      extend RuleSet
+
+      rule(:spacing) { (match(/\s+/) | comment).rep.quiet }
+      rule(:comment) { match(comment_pattern).label("comment") }
+
+      def comment_pattern = /#[^\n]*/
+      def token(m) = Pick.new([Birman.coerce(m), spacing], 0)
+    end
+
+    # Integers and floats, as Ruby numbers.
+    module Numbers
+      extend RuleSet
+
+      lexeme(:integer) { match(/\d+/).map(&:to_i).label("integer") }
+      lexeme(:float)   { match(/\d+\.\d+/).map(&:to_f).label("float") }
+      lexeme(:number)  { (float | integer).label("number") }
+    end
+
+    # Double-quoted strings with backslash escapes. The opening quote
+    # commits, so an unterminated string is reported where it breaks.
+    module Strings
+      extend RuleSet
+
+      ESCAPES = { "n" => "\n", "t" => "\t", "r" => "\r", "0" => "\0" }.freeze
+
+      lexeme(:string) do
+        (str('"') > (string_part.rep >> str('"').label("closing quote")))
+          .map { |(_, parts, _)| parts.join }
+          .label("string")
+      end
+
+      rule(:string_part) do
+        (match(/\\./).map { ESCAPES.fetch(_1[1], _1[1]) } |
+          match(/[^"\\\n]+/)).label("string character")
+      end
+    end
+
+    # Identifiers that can't be reserved words. Declare those in the
+    # grammar with `keywords "if", "then", ...`.
+    module Identifiers
+      extend RuleSet
+
+      lexeme(:identifier) { (reserved_word.absent & match(/[A-Za-z_]\w*/)).label("identifier") }
+
+      rule(:reserved_word) do
+        next never if keywords.empty?
+
+        match(/(?:#{keywords.map { Regexp.escape(_1) }.join("|")})\b/)
+      end
+    end
+  end
+end

data/lib/birman/version.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+module Birman
+  VERSION = "0.1.0"
+end

data/lib/birman.rb

@@ -0,0 +1,42 @@
+# frozen_string_literal: true
+
+# A small PEG parser library for Ruby 3.2+.
+#
+#   Greeting = Birman.node(:name)
+#
+#   class MyGrammar
+#     include Birman::Grammar
+#     include Birman::Rules::Whitespace
+#     include Birman::Rules::Identifiers
+#
+#     root :greeting
+#     ignore :spacing
+#     rule(:greeting) { seq("hello", commit, identifier.as(:name)).to(Greeting) }
+#   end
+#
+#   case MyGrammar.parse("hello world")
+#   in Birman::Success(value) then value
+#   in Birman::Failure => f   then warn f.message
+#   end
+module Birman
+  # The character offset of a byte offset into input. Free for ASCII
+  # input (Ruby caches ascii_only?), linear otherwise; used once per
+  # parse for a Failure, and only on demand for a Span.
+  def self.char_offset(input, byte_pos)
+    input.ascii_only? ? byte_pos : input.byteslice(0, byte_pos).length
+  end
+end
+
+require "strscan"
+
+require_relative "birman/version"
+require_relative "birman/result"
+require_relative "birman/nodes"
+require_relative "birman/context"
+require_relative "birman/matchers"
+require_relative "birman/rule_set"
+require_relative "birman/combinators"
+require_relative "birman/grammar"
+require_relative "birman/error_formatter"
+require_relative "birman/ast"
+require_relative "birman/rules"

metadata

@@ -0,0 +1,60 @@
+--- !ruby/object:Gem::Specification
+name: birman
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Tad Thorley
+bindir: exe
+cert_chain: []
+date: 1980-01-02 00:00:00.000000000 Z
+dependencies: []
+description: Module-based grammars, immutable Data matchers and AST nodes with source
+  spans, packrat memoization, commit points, labels, and Success/Failure results.
+email:
+- phaedryx@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- CHANGELOG.md
+- LICENSE.txt
+- README.md
+- lib/birman.rb
+- lib/birman/ast.rb
+- lib/birman/combinators.rb
+- lib/birman/context.rb
+- lib/birman/error_formatter.rb
+- lib/birman/grammar.rb
+- lib/birman/matchers.rb
+- lib/birman/nodes.rb
+- lib/birman/result.rb
+- lib/birman/rule_set.rb
+- lib/birman/rules.rb
+- lib/birman/version.rb
+homepage: https://github.com/phaedryx/birman
+licenses:
+- MIT
+metadata:
+  homepage_uri: https://github.com/phaedryx/birman
+  source_code_uri: https://github.com/phaedryx/birman
+  changelog_uri: https://github.com/phaedryx/birman/blob/main/CHANGELOG.md
+  rubygems_mfa_required: 'true'
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 3.2.0
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 4.0.3
+specification_version: 4
+summary: A small PEG parser library for Ruby.
+test_files: []