parsanol 1.3.28-arm-linux → 1.3.30-arm-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 017e97c2ca6e150818f1662a410ee31ad76fa22a2b0b3f55261ec8ed2bd5e5c5
-  data.tar.gz: 952e25087016352b3633011efb637514b6b375c453418b87b4fe5d87cab264be
+  metadata.gz: b4ac5d42ef4718a350220c0d138e4e976e8b636114eb3d1ac19f60813fb8cbed
+  data.tar.gz: 25829d9ffbf70bd209f518d480ec96949d72f326e437df69794e1339a2ddb0af
 SHA512:
-  metadata.gz: 987de76ae18f3dd0d73da7afde2a02528159384360700f0b579d0656a9c5fbd59824ad7998f1063a0be9020f9d5bbd14989fe6852458cfa70508341c8f72753e
-  data.tar.gz: 508a9ff925772c33baa5c6e3047d3e61864040f976e5be1a3574629d011ab5e52a106a0dd7a34fee887985291f95363f6afe69562fcb81acc4fad389f2b03360
+  metadata.gz: 2f396be06ff1b3bc405badb6f30ca772e13429f2ea4bd686017843a3206c157eef91a6ac618e98e62729f23429596f59ce02d1caff7eacb63bdee2c2affed9cf
+  data.tar.gz: d535746fa5d0466276a3dc3f5502ecda5202ddea2d059e3f1bd5af93347745b3149f0369840bf64de8ae245614290c1350b878da605e03853e5e269bcf95af0f

data/lib/parsanol/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Parsanol
-  VERSION = "1.3.28"
+  VERSION = "1.3.30"
 end

data/lib/parsanol/vm.rb

@@ -43,6 +43,7 @@ module Parsanol
     RUN_RE = 24
     RUN_STR = 25
     SEQ_SIMPLE = 26
+    BYTE_DISPATCH = 27
 
     K_ALT = 0
     K_REP = 1
@@ -247,6 +248,147 @@ module Parsanol
         @ops.flatten!(1)
       end
 
+      # Conservative lead-byte table for a branch: table[b] true iff the
+      # atom can start with byte b. nil = cannot prove a constraint (no
+      # guard emitted). Conservative in the safe direction only: a guard
+      # may allow a branch that then fails normally, never reject one
+      # that could match.
+      def pairwise_disjoint?(tables)
+        seen = Array.new(256, false)
+        tables.each do |t|
+          i = 0
+          while i < 256
+            return false if t[i] && seen[i]
+
+            seen[i] = true if t[i]
+            i += 1
+          end
+        end
+        true
+      end
+
+      def first_byte_table(atom, depth = 0)
+        return nil if depth > 10
+
+        case atom
+        when Parsanol::Atoms::Str
+          b = atom.str.getbyte(0)
+          return nil unless b
+
+          t = Array.new(256, false)
+          t[b] = true
+          t
+        when Parsanol::Atoms::Re
+          byte_table(atom.re)
+        when Parsanol::Atoms::Sequence
+          t = nil
+          nullable_all = true
+          atom.parslets.each do |child|
+            ct = first_byte_table(child, depth + 1)
+            return nil if ct.nil?
+
+            if t
+              i = 0
+              while i < 256
+                t[i] ||= ct[i]
+                i += 1
+              end
+            else
+              t = ct
+            end
+            unless nullable?(child, depth + 1)
+              nullable_all = false
+              break
+            end
+          end
+          nullable_all ? nil : t
+        when Parsanol::Atoms::Alternative
+          t = nil
+          atom.alternatives.each do |branch|
+            bt = first_byte_table(branch, depth + 1)
+            return nil if bt.nil?
+
+            if t
+              i = 0
+              while i < 256
+                t[i] ||= bt[i]
+                i += 1
+              end
+            else
+              t = bt
+            end
+          end
+          t
+        when Parsanol::Atoms::Named
+          first_byte_table(atom.parslet, depth + 1)
+        when Parsanol::Atoms::Ignored
+          first_byte_table(atom.wrapped_atom, depth + 1)
+        when Parsanol::Atoms::Lookahead
+          # Positive: body's lead bytes. Negative: body's lead bytes are
+          # a safe superset (a failing !-check just fails the branch
+          # normally).
+          first_byte_table(atom.bound_parslet, depth + 1)
+        when Parsanol::Atoms::Repetition
+          # min 0 can match empty — no byte constraint exists.
+          return nil if atom.min.zero?
+
+          first_byte_table(atom.parslet, depth + 1)
+        when Parsanol::Atoms::Entity
+          inner = begin
+            atom.parslet
+          rescue StandardError
+            nil
+          end
+          return nil if inner.nil?
+
+          first_byte_table(inner, depth + 1)
+        when Parsanol::Atoms::Scope
+          inner = begin
+            atom.block.call
+          rescue StandardError
+            nil
+          end
+          return nil if inner.nil?
+
+          first_byte_table(inner, depth + 1)
+        end
+      end
+
+      # Wrapper branches (Entity/Scope) share body shapes by design.
+      # rubocop:disable-next Lint/DuplicateBranch
+      def nullable?(atom, depth = 0)
+        return false if depth > 10
+
+        case atom
+        when Parsanol::Atoms::Str then atom.str.bytesize.zero?
+        when Parsanol::Atoms::Re then false
+        when Parsanol::Atoms::Sequence
+          atom.parslets.all? { |c| nullable?(c, depth + 1) }
+        when Parsanol::Atoms::Alternative
+          atom.alternatives.any? { |c| nullable?(c, depth + 1) }
+        when Parsanol::Atoms::Repetition then atom.min.zero?
+        when Parsanol::Atoms::Named then nullable?(atom.parslet, depth + 1)
+        when Parsanol::Atoms::Ignored then nullable?(atom.wrapped_atom, depth + 1)
+        when Parsanol::Atoms::Lookahead then true
+        when Parsanol::Atoms::Entity, Parsanol::Atoms::Scope
+          nullable_wrapped?( # rubocop:disable Lint/DuplicateBranch -- Entity/Scope differ only in accessor
+            -> { atom.is_a?(Parsanol::Atoms::Entity) ? atom.parslet : atom.block.call }, depth
+          )
+        else false
+        end
+      end
+
+      # Entity and Scope wrap lazily-resolvable inners through different
+      # accessors; the nullability logic is shared.
+      def nullable_wrapped?(resolver, depth)
+        inner = begin
+          resolver.call
+        rescue StandardError
+          nil
+        end
+        !inner.nil? && nullable?(inner, depth + 1)
+      end
+
       def compile_atom(atom, consume_all)
         case atom
         when Parsanol::Atoms::Str
@@ -282,6 +424,37 @@ module Parsanol
           count = alts.size
           return nil if count.zero?
 
+          # First-set discrimination: when every branch has a provable
+          # lead-byte table and the tables are pairwise disjoint, at most
+          # one branch can match any input — dispatch directly to the
+          # viable branch (or fail) in one lookup. Shared lead bytes
+          # fall back to the plain CHOICE machinery untouched.
+          tables = alts.map { |a| first_byte_table(a) }
+          if tables.all? && pairwise_disjoint?(tables)
+            dispatch_idx = emit(BYTE_DISPATCH, nil, nil, nil)
+            starts = []
+            idx = 0
+            while idx < count
+              starts << flat(@ops.size)
+              return nil unless compile_atom(alts[idx], consume_all)
+
+              idx += 1
+            end
+            table = Array.new(256, -1)
+            tables.each_with_index do |t, bi|
+              pc_i = starts[bi]
+              bi2 = bi
+              j = 0
+              while j < 256
+                table[j] = pc_i if t[j]
+                j += 1
+              end
+              bi2
+            end
+            patch(dispatch_idx, 1, table)
+            return self
+          end
+
           jumps = []
           idx = 0
           while idx < count
@@ -793,6 +966,11 @@ module Parsanol
               frames.slice!(cbase..)
               pc += 4
             end
+          when BYTE_DISPATCH
+            table = ops[pc + 1]
+            b = pos < n ? bytes[pos] : -1
+            target = b == -1 ? -1 : table[b]
+            pc = target >= 0 ? target : FAIL
           when SEQ_SIMPLE
             kids = ops[pc + 1]
             kn = kids.size

data/parsanol.gemspec

@@ -46,22 +46,4 @@ Gem::Specification.new do |spec|
 
   # Required for Rust extension
   spec.add_dependency "rb_sys", "~> 0.9.39"
-
-  spec.add_development_dependency "rake", "~> 13.0"
-  spec.add_development_dependency "rake-compiler", "~> 1.2.0"
-  spec.add_development_dependency "rdoc", "~> 6.0"
-  spec.add_development_dependency "rspec", "~> 3.0"
-
-  # For code style checking
-  spec.add_development_dependency "rubocop", "~> 1.0"
-
-  # For Parslet compatibility verification
-  spec.add_development_dependency "parslet", "~> 2.0.0"
-
-  # For benchmarking
-  spec.add_development_dependency "benchmark-ips", "~> 2.0"
-
-  # For type checking
-  spec.add_development_dependency "rbs", "~> 3.0"
-  spec.add_development_dependency "steep", "~> 1.0"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: parsanol
 version: !ruby/object:Gem::Version
-  version: 1.3.28
+  version: 1.3.30
 platform: arm-linux
 authors:
 - Ribose Inc.
@@ -9,133 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 date: 2026-09-17 00:00:00.000000000 Z
-dependencies:
-- !ruby/object:Gem::Dependency
-  name: rake
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '13.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '13.0'
-- !ruby/object:Gem::Dependency
-  name: rake-compiler
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 1.2.0
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 1.2.0
-- !ruby/object:Gem::Dependency
-  name: rdoc
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '6.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '6.0'
-- !ruby/object:Gem::Dependency
-  name: rspec
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.0'
-- !ruby/object:Gem::Dependency
-  name: rubocop
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.0'
-- !ruby/object:Gem::Dependency
-  name: parslet
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 2.0.0
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 2.0.0
-- !ruby/object:Gem::Dependency
-  name: benchmark-ips
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '2.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '2.0'
-- !ruby/object:Gem::Dependency
-  name: rbs
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.0'
-- !ruby/object:Gem::Dependency
-  name: steep
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.0'
+dependencies: []
 description: A small Ruby library for constructing parsers in the PEG (Parsing Expression
   Grammar) fashion. Parsanol provides Parslet-compatible API with additional features
   including static frozen parsers and dynamic parsers, with optional Rust native extension