v5.1.0

This takes the place of #line directives

.github/workflows/run-tests.yml

@@ -17,7 +17,7 @@ jobs:
     steps:
     - name: Install dependencies (Linux)
       if: runner.os == 'Linux'
-      run: sudo apt-get update && sudo apt-get install -y gcc gdc ldc
+      run: sudo apt-get update && sudo apt-get install -y gcc gdc ldc valgrind
 
     - name: Install dependencies (macOS)
       if: runner.os == 'macOS'
@@ -31,6 +31,9 @@ jobs:
       with:
         ruby-version: ${{ matrix.ruby-version }}
 
+    - name: Set up Rust
+      uses: dtolnay/rust-toolchain@stable
+
     - name: Install dependencies
       run: bundle install
 

CHANGELOG.md

@@ -1,3 +1,127 @@
+## v5.1.0
+
+### New Features
+
+- Add a `node_id()` accessor to the C++ and D tree node handle types, for node
+  identity comparison. This matches the existing `p_node_id()` macro (C) and
+  `node_id()` method (Rust).
+
+## v5.0.0
+
+### New Features
+
+- Add Rust target language output.
+- Add Rust language detection in propane.vim.
+
+### API Changes
+
+- The matched text argument passed to lexer user code blocks is now named
+  `match_text` instead of `match`, since `match` is a keyword in Rust. The
+  `match_length` argument (C and C++) is unchanged.
+- Tree generation mode now stores all tree nodes in a compact arena owned by
+  the parser context (a flat node array plus a shared child-link array).
+  This replaces the previous design of one heap allocation per node with
+  layout-punned typed structs.
+- Tree nodes are now referenced by lightweight handles rather than pointers.
+  `p_result()` and the field accessors now return handle values in tree
+  generation mode.
+- The whole tree is freed together with the context by `p_context_delete()`.
+  The `p_tree_delete()` / `p_tree_delete_XXX()` functions have been removed;
+  tree node handles are only valid while the context is alive.
+- Tree node field access changed per target language:
+  - C: per-field accessor functions (e.g. `p_Start_pItems(node)`) plus tree
+    walk macros (e.g. `p_tree_walk_Start(node, pItems, pItem, pToken1, token)`),
+    and generic accessors `p_node_valid()`, `p_node_position()`,
+    `p_node_end_position()`, `p_node_n_fields()`, `p_node_data()`, `p_node_id()`.
+  - C++: handle methods called with `()` (e.g. `node.pItems().pToken1().token()`),
+    plus the same C-style functions/macros for convenience.
+  - D: `@property` accessors preserving the previous field-access syntax
+    (e.g. `node.pItems.pToken1.token`); null checks use `.valid` instead of
+    `is null`.
+- Tree-mode parser rule user code: `$$` and `$1` etc. now yield node handles.
+  Reference child fields through the target-language accessors described above
+  rather than through struct pointer members.
+
+### Improvements
+
+- Improve D language detection in propane.vim
+- Speed up specs
+
+## v4.8.1
+
+### Fixes
+
+- Fix tree node struct type forward-declarations for C/C++
+
+## v4.8.0
+
+### New Features
+
+- Add `p_parse_inner_XXX()` APIs that accept a caller-provided set of follow
+  tokens. These behave the same as `p_parse_XXX()` by parsing starting at the
+  given start rule, but instead of expecting the rest of the input to match
+  the start rule they allow specifying a set of tokens that may follow the
+  start rule.
+- Add `p_set_position()` API to set the current text position stored in the
+  context. Useful for setting the initial text position to something other
+  than `(1, 1)` for a nested parse operation.
+- Add `p_input_index()` API to get the current input text byte offset.
+- Add `p_set_input_index()` API to set the current input text byte offset.
+  Useful together with `p_set_position()` to rewind the input part-way through
+  a parse in order to re-read an earlier section of the input.
+
+## v4.7.0
+
+### New Features
+
+- Support parser rule user code blocks in tree generation mode.
+
+### Fixes
+
+- propane.vim: do not highlight rule components as propane keywords
+
+## v4.6.0
+
+### New Features
+
+- Add lexer user code API to access matched input text positions
+- Track rule component text positions and add parser user code API to access
+
+### Fixes
+
+- Fixed a few user guide and source comments related to text input positions
+
+## v4.5.0
+
+### New Features
+
+- Add `noline` grammar statement to skip emitting `#line` directives
+- Attempt to autodetect target language (D/C++) in extra/vim/syntax/propane.vim
+
+### Fixes
+
+- Fix #line reset directives
+- Update keyword list in extra/vim/syntax/propane.vim
+- Fix propane.vim keyword detection
+
+## v4.4.0
+
+### New Features
+
+- Add p_value_get() / p_value_get_XXX() accessors
+
+## v4.3.0
+
+### New Features
+
+- Use #line for user code blocks to report input grammar position for errors.
+
+## v4.2.0
+
+### New Features
+
+- Add support for a custom lex function.
+
 ## v4.1.0
 
 ### New Features

Gemfile.lock

@@ -5,11 +5,11 @@ GEM
     date (3.5.1)
     diff-lcs (1.6.2)
     docile (1.4.1)
-    erb (6.0.1)
-    psych (5.3.1)
+    erb (6.0.4)
+    psych (5.4.0)
       date
       stringio
-    rake (13.3.1)
+    rake (13.4.2)
     rdoc (7.2.0)
       erb
       psych (>= 4.0.0)
@@ -24,7 +24,7 @@ GEM
     rspec-expectations (3.13.5)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
-    rspec-mocks (3.13.7)
+    rspec-mocks (3.13.8)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
     rspec-support (3.13.7)
@@ -51,4 +51,4 @@ DEPENDENCIES
   syntax
 
 BUNDLED WITH
-   2.3.7
+  4.0.14

LICENSE.txt

@@ -1,6 +1,6 @@
 The MIT License (MIT)
 
-Copyright (c) 2010-2024 Josh Holtrop
+Copyright (c) 2010-2026 Josh Holtrop
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

README.md

@@ -6,7 +6,7 @@ Propane is a LALR Parser Generator (LPG) which:
   * generates a built-in lexer to tokenize input
   * supports UTF-8 lexer inputs
   * generates a table-driven shift/reduce parser to parse input in linear time
-  * targets C, C++, or D language outputs
+  * targets C, C++, D, or Rust language outputs
   * optionally supports automatic full parse tree generation
   * supports starting parsing from multiple start rules
   * tracks input text start and end positions for all matched tokens/rules
@@ -69,7 +69,7 @@ token times /\*/;
 token power /\*\*/;
 token integer /\d+/ <<
   ulong v;
-  foreach (c; match)
+  foreach (c; match_text)
   {
     v *= 10;
     v += (c - '0');

Rakefile.rb

@@ -1,3 +1,4 @@
+require "fileutils"
 require "rake/clean"
 require "rspec/core/rake_task"
 require "simplecov"
@@ -11,7 +12,10 @@ end
 
 RSpec::Core::RakeTask.new(:spec, :example_pattern) do |task, args|
   if args.example_pattern
+    ENV["partial_specs"] = "1"
     task.rspec_opts = %W[-e "#{args.example_pattern}" -f documentation]
+  else
+    FileUtils.rm_rf("coverage")
   end
 end
 task :spec do |task, args|
@@ -19,7 +23,7 @@ task :spec do |task, args|
     original_stdout = $stdout
     sio = StringIO.new
     $stdout = sio
-    SimpleCov.collate Dir["coverage/.resultset.json"]
+    SimpleCov.collate Dir["coverage/parts/*/.resultset.json"]
     $stdout = original_stdout
     sio.string.lines.each do |line|
       $stdout.write(line) unless line =~ /Coverage report generated for/
@@ -27,6 +31,15 @@ task :spec do |task, args|
   end
 end
 
+task :valgrind do
+  begin
+    ENV["spec-valgrind"] = "1"
+    Rake::Task[:spec].execute
+  ensure
+    ENV.delete("spec-valgrind")
+  end
+end
+
 # dspec task is useful to test the distributable release script, but is not
 # useful for coverage information.
 desc "Dist Specs"
@@ -43,4 +56,4 @@ task :user_guide do
   system("ruby", "-Ilib", "rb/gen_user_guide.rb")
 end
 
-task :all => [:spec, :dspec, :user_guide]
+task :all => [:valgrind, :dspec, :user_guide]

UPGRADING.md

@@ -1,3 +1,92 @@
+## v5.0.0
+
+The generated API for tree generation mode (`tree;`) has been changed
+significantly for this version.
+Aside from the lexer user code block matched text rename described below, the
+lexer/parser value APIs for non-tree grammars are unchanged.
+
+### Lexer user code block matched text
+
+The matched text argument passed to lexer user code blocks has been renamed
+from `match` to `match_text` for all target languages.
+
+- C, C++, and D: rename references to `match` in lexer user code blocks to
+  `match_text` (for example `$$ = match[0];` becomes `$$ = match_text[0];`).
+
+The `match_length` argument (C, C++) is unchanged.
+
+### Tree memory management
+
+- Remove all calls to `p_tree_delete()` / `p_tree_delete_XXX()`. Tree nodes now
+  live in the parser context and are freed by `p_context_delete()`.
+- Tree node handles (returned by `p_result()` and the field accessors) are only
+  valid while the context is alive. Do not use them after `p_context_delete()`.
+
+### Tree node field access
+
+Tree nodes are now referenced by handle values instead of pointers, and field
+access differs per target language:
+
+- C: replace `node->field` with the accessor function `p_TYPE_field(node)`, or
+  use the tree walk macro `p_tree_walk_TYPE(node, field1, field2, ...)`. Replace
+  `x != NULL` / `x == NULL` node checks with `p_node_valid(x)` /
+  `!p_node_valid(x)`. Read positions with `p_node_position(node)` /
+  `p_node_end_position(node)`, token payload with `p_TYPE_token(node)` /
+  `p_TYPE_pvalue(node)` or `p_node_data(node)->field`, and compare node identity
+  with `p_node_id(a) == p_node_id(b)`.
+- C++: replace `node->field` with the handle method `node.field()`. Use
+  `node.valid()`, `node.position()`, `node.token()`, `node.pvalue()`, and
+  `node.data()->field` for user token fields. (The C-style functions and macros
+  above are also available.)
+- D: replace pointer declarations (`Start * s`) with value handles (`Start s`)
+  and replace `x !is null` / `x is null` with `x.valid` / `!x.valid`. Field
+  access syntax (`node.field.field`) is otherwise unchanged.
+
+### Tree-mode parser rule user code
+
+In tree generation mode `$$` and `$1`, `$2`, ... now expand to node handles.
+Reference child fields through the target-language accessors above (for example
+`$$->pA->pToken1->pvalue` becomes `p_tree_walk_Start($$, pA, pToken1, pvalue)`
+in C, `$$.pA().pToken1().pvalue()` in C++, and `$$.pA.pToken1.pvalue` in D).
+
+### Pointers into tree node storage
+
+Tree nodes previously each had their own allocation, so a pointer to a node
+stayed valid for the life of the tree. They are now held in a single array
+which is reallocated as it grows, so a pointer or reference into that array may
+be invalidated whenever a new node is created.
+
+New nodes are created while parsing, so this matters for a pointer taken in a
+tree-mode parser rule user code block, which runs before the parse has
+finished. Keep the node handle instead, which stores a node ID rather than an
+address and stays valid, and obtain the pointer from it when it is needed.
+
+For example, replace a saved pointer:
+
+```
+context_user_fields <<
+    p_node_data_t * saved;
+>>
+Items -> Items a << ${context.saved} = p_node_data($$); >>
+```
+
+with a saved handle:
+
+```
+context_user_fields <<
+    Items saved_node;
+>>
+Items -> Items a << ${context.saved_node} = $$; >>
+```
+
+```
+p_node_data_t * data = p_node_data(context->saved_node);
+```
+
+Once parsing has finished, no further nodes are created, so a pointer obtained
+after `p_parse()` returns stays valid until the context is deleted, as long as
+no further parsing is performed with the same context.
+
 ## v4.0.0
 
 ### API Changes

assets/parser.c.erb

@@ -68,6 +68,18 @@ const char * <%= @grammar.prefix %>token_names[] = {
     context->text_position.row = 1u;
     context->text_position.col = 1u;
     context->mode = <%= @lexer.mode_id("default") %>;
+<% if @grammar.tree %>
+
+    /* Reserve node ID 0 as the null tree node. */
+<%   if @cpp %>
+    context-><%= @grammar.prefix %>tree_nodes.resize(1);
+<%   else %>
+    context-><%= @grammar.prefix %>tree_nodes_capacity = 16u;
+    context-><%= @grammar.prefix %>tree_nodes = (<%= @grammar.prefix %>node_data_t *)malloc(16u * sizeof(<%= @grammar.prefix %>node_data_t));
+    memset(&context-><%= @grammar.prefix %>tree_nodes[0], 0, sizeof(<%= @grammar.prefix %>node_data_t));
+    context-><%= @grammar.prefix %>tree_nodes_length = 1u;
+<%   end %>
+<% end %>
 
     return context;
 }
@@ -84,9 +96,27 @@ const char * <%= @grammar.prefix %>token_names[] = {
  */
 void <%= @grammar.prefix %>context_delete(<%= @grammar.prefix %>context_t * context)
 {
+<% if @grammar.tree && @grammar.free_token_node != "" %>
+<%   if @cpp %>
+    for (size_t i = 0u; i < context-><%= @grammar.prefix %>tree_nodes.size(); i++)
+<%   else %>
+    for (size_t i = 0u; i < context-><%= @grammar.prefix %>tree_nodes_length; i++)
+<%   end %>
+    {
+        if (context-><%= @grammar.prefix %>tree_nodes[i].is_token)
+        {
+            <%= @grammar.prefix %>node_data_t * token_tree_node = &context-><%= @grammar.prefix %>tree_nodes[i];
+<%= expand_code(@grammar.free_token_node, false, nil, nil) %>
+        }
+    }
+<% end %>
 <% if @cpp %>
     delete context;
 <% else %>
+<%   if @grammar.tree %>
+    free(context-><%= @grammar.prefix %>tree_nodes);
+    free(context-><%= @grammar.prefix %>tree_children);
+<%   end %>
     free(context);
 <% end %>
 }
@@ -289,7 +319,7 @@ static lexer_mode_t lexer_mode_table[] = {
  *   Lexer/parser context structure.
  * @param code_id
  *   The ID of the user code block to execute.
- * @param match
+ * @param match_text
  *   Matched text for this pattern.
  * @param match_length
  *   Matched text length.
@@ -300,7 +330,7 @@ static lexer_mode_t lexer_mode_table[] = {
  *   not explicitly return a token.
  */
 static <%= @grammar.prefix %>token_t lexer_user_code(<%= @grammar.prefix %>context_t * context,
-    lexer_user_code_id_t code_id, uint8_t const * match,
+    lexer_user_code_id_t code_id, uint8_t const * match_text,
     size_t match_length, <%= @grammar.prefix %>token_info_t * out_token_info)
 {
     switch (code_id)
@@ -486,11 +516,27 @@ static size_t attempt_lex_token(<%= @grammar.prefix %>context_t * context, <%= @
     case P_SUCCESS:
         {
         <%= @grammar.prefix %>token_t token_to_accept = match_info.accepting_state->token;
+        /* Calculate the token length and start/end positions before invoking
+         * the lexer user code so that the user code can access them. The
+         * context input text position tracking is not updated until after the
+         * user code has run so that it is left unchanged if the user code
+         * requests to terminate the lexer. */
+        token_info.length = match_info.length;
+        if (match_info.end_delta_position.row != 0u)
+        {
+            token_info.end_position.row = token_info.position.row + match_info.end_delta_position.row;
+            token_info.end_position.col = match_info.end_delta_position.col;
+        }
+        else
+        {
+            token_info.end_position.row = token_info.position.row;
+            token_info.end_position.col = token_info.position.col + match_info.end_delta_position.col;
+        }
         if (match_info.accepting_state->code_id != INVALID_USER_CODE_ID)
         {
-            uint8_t const * match = &context->input[context->input_index];
+            uint8_t const * match_text = &context->input[context->input_index];
             <%= @grammar.prefix %>token_t user_code_token = lexer_user_code(context,
-                match_info.accepting_state->code_id, match, match_info.length, &token_info);
+                match_info.accepting_state->code_id, match_text, match_info.length, &token_info);
             /* A TERMINATE_TOKEN_ID return code from lexer_user_code() means
              * that the user code is requesting to terminate the lexer. */
             if (user_code_token == TERMINATE_TOKEN_ID)
@@ -524,17 +570,6 @@ static size_t attempt_lex_token(<%= @grammar.prefix %>context_t * context, <%= @
             return P_DROP;
         }
         token_info.token = token_to_accept;
-        token_info.length = match_info.length;
-        if (match_info.end_delta_position.row != 0u)
-        {
-            token_info.end_position.row = token_info.position.row + match_info.end_delta_position.row;
-            token_info.end_position.col = match_info.end_delta_position.col;
-        }
-        else
-        {
-            token_info.end_position.row = token_info.position.row;
-            token_info.end_position.col = token_info.position.col + match_info.end_delta_position.col;
-        }
         *out_token_info = token_info;
         }
         return P_SUCCESS;
@@ -707,26 +742,16 @@ typedef struct
     /** Parser state ID. */
     size_t state_id;
 
-    /** Parser value from this state. */
-    <%= @grammar.prefix %>value_t pvalue;
-
 <% if @grammar.tree %>
-    /** tree node. */
-    void * tree_node;
-<% end %>
-} state_value_t;
-
-<% if @grammar.tree %>
-/** Common tree node structure. */
-typedef struct TreeNode_s
-{
+    /** Tree node ID. */
+    <%= @grammar.prefix %>node_id_t node_id;
+<% else %>
     <%= @grammar.prefix %>position_t position;
     <%= @grammar.prefix %>position_t end_position;
-    uint16_t n_fields;
-    uint8_t is_token;
-    struct TreeNode_s * fields[];
-} TreeNode;
+    /** Parser value from this state. */
+    <%= @grammar.prefix %>value_t pvalue;
 <% end %>
+} state_value_t;
 
 /** Parser shift table. */
 static const shift_t parser_shift_table[] = {
@@ -865,7 +890,131 @@ static void state_values_stack_free(state_values_stack_t * stack)
      free(stack->entries);
 }
 
+<% if @grammar.tree %>
+/* Tree arena helpers. */
+
+/**
+ * Allocate a new (zeroed) tree node in the context arena.
+ *
+ * @return The new node ID.
+ */
+static <%= @grammar.prefix %>node_id_t tree_new_node(<%= @grammar.prefix %>context_t * context)
+{
+<% if @cpp %>
+    <%= @grammar.prefix %>node_id_t id = (<%= @grammar.prefix %>node_id_t)context-><%= @grammar.prefix %>tree_nodes.size();
+    context-><%= @grammar.prefix %>tree_nodes.emplace_back();
+    return id;
+<% else %>
+    if (context-><%= @grammar.prefix %>tree_nodes_length >= context-><%= @grammar.prefix %>tree_nodes_capacity)
+    {
+        size_t new_capacity = context-><%= @grammar.prefix %>tree_nodes_capacity * 2u;
+        <%= @grammar.prefix %>node_data_t * new_nodes = (<%= @grammar.prefix %>node_data_t *)malloc(new_capacity * sizeof(<%= @grammar.prefix %>node_data_t));
+        memcpy(new_nodes, context-><%= @grammar.prefix %>tree_nodes, context-><%= @grammar.prefix %>tree_nodes_length * sizeof(<%= @grammar.prefix %>node_data_t));
+        free(context-><%= @grammar.prefix %>tree_nodes);
+        context-><%= @grammar.prefix %>tree_nodes = new_nodes;
+        context-><%= @grammar.prefix %>tree_nodes_capacity = new_capacity;
+    }
+    <%= @grammar.prefix %>node_id_t id = (<%= @grammar.prefix %>node_id_t)context-><%= @grammar.prefix %>tree_nodes_length;
+    memset(&context-><%= @grammar.prefix %>tree_nodes[id], 0, sizeof(<%= @grammar.prefix %>node_data_t));
+    context-><%= @grammar.prefix %>tree_nodes_length += 1u;
+    return id;
+<% end %>
+}
+
+/**
+ * Reserve n contiguous (zeroed) child slots in the shared children array.
+ *
+ * @return The offset of the first reserved slot.
+ */
+static <%= @grammar.prefix %>node_id_t tree_reserve_children(<%= @grammar.prefix %>context_t * context, size_t n)
+{
+<% if @cpp %>
+    <%= @grammar.prefix %>node_id_t offset = (<%= @grammar.prefix %>node_id_t)context-><%= @grammar.prefix %>tree_children.size();
+    context-><%= @grammar.prefix %>tree_children.resize(context-><%= @grammar.prefix %>tree_children.size() + n);
+    return offset;
+<% else %>
+    size_t offset = context-><%= @grammar.prefix %>tree_children_length;
+    size_t needed = offset + n;
+    if (needed > context-><%= @grammar.prefix %>tree_children_capacity)
+    {
+        size_t new_capacity = context-><%= @grammar.prefix %>tree_children_capacity ? context-><%= @grammar.prefix %>tree_children_capacity : 1u;
+        while (new_capacity < needed)
+        {
+            new_capacity *= 2u;
+        }
+        <%= @grammar.prefix %>node_id_t * new_children = (<%= @grammar.prefix %>node_id_t *)malloc(new_capacity * sizeof(<%= @grammar.prefix %>node_id_t));
+        if (context-><%= @grammar.prefix %>tree_children != NULL)
+        {
+            memcpy(new_children, context-><%= @grammar.prefix %>tree_children, context-><%= @grammar.prefix %>tree_children_length * sizeof(<%= @grammar.prefix %>node_id_t));
+            free(context-><%= @grammar.prefix %>tree_children);
+        }
+        context-><%= @grammar.prefix %>tree_children = new_children;
+        context-><%= @grammar.prefix %>tree_children_capacity = new_capacity;
+    }
+    memset(&context-><%= @grammar.prefix %>tree_children[offset], 0, n * sizeof(<%= @grammar.prefix %>node_id_t));
+    context-><%= @grammar.prefix %>tree_children_length = needed;
+    return (<%= @grammar.prefix %>node_id_t)offset;
+<% end %>
+}
+
+/* Tree node field accessor functions. */
+<%= c_tree_accessor_defs %>
+<% end %>
+
 <% unless @grammar.tree %>
+/**
+ * Get the rule position (start or end) for the currently matched rule.
+ */
+static <%= @grammar.prefix %>position_t get_rule_position(state_values_stack_t * statevalues, size_t i, size_t n_states, bool get_end)
+{
+    if (n_states > 0u)
+    {
+        if (i == 0u)
+        {
+            if (get_end)
+            {
+                int stack_index = -1;
+                for (size_t j = 0u; j < n_states; j++)
+                {
+                    state_value_t * sv = state_values_stack_index(statevalues, stack_index - (int)j);
+                    if (<%= @grammar.prefix %>position_valid(sv->end_position))
+                    {
+                        return sv->end_position;
+                    }
+                }
+            }
+            else
+            {
+                int stack_index = -(int)n_states;
+                for (size_t j = 0u; j < n_states; j++)
+                {
+                    state_value_t * sv = state_values_stack_index(statevalues, stack_index + (int)j);
+                    if (<%= @grammar.prefix %>position_valid(sv->position))
+                    {
+                        return sv->position;
+                    }
+                }
+            }
+        }
+        else
+        {
+            if (get_end)
+            {
+                return state_values_stack_index(statevalues, -1 - (int)n_states + (int)i)->end_position;
+            }
+            else
+            {
+                return state_values_stack_index(statevalues, -1 - (int)n_states + (int)i)->position;
+            }
+        }
+    }
+    <%= @grammar.prefix %>position_t empty_pos;
+    memset(&empty_pos, 0, sizeof(empty_pos));
+    return empty_pos;
+}
+<% end %>
+
+<% if !@grammar.tree || @grammar.parser_user_code_used? %>
 /**
  * Execute user code associated with a parser rule.
  *
@@ -876,7 +1025,7 @@ static void state_values_stack_free(state_values_stack_t * stack)
  * @retval P_USER_TERMINATED
  *   User requested to terminate parsing.
  */
-static size_t parser_user_code(<%= @grammar.prefix %>value_t * _pvalue, uint32_t rule, state_values_stack_t * statevalues, uint32_t n_states, <%= @grammar.prefix %>context_t * context)
+static size_t parser_user_code(<%= @grammar.tree ? "#{@grammar.prefix}node_id_t _node_id" : "#{@grammar.prefix}value_t * _pvalue" %>, uint32_t rule, state_values_stack_t * statevalues, uint32_t n_states, <%= @grammar.prefix %>context_t * context)
 {
     switch (rule)
     {
@@ -926,7 +1075,7 @@ static size_t check_shift(size_t state_id, size_t symbol_id)
  * @param token
  *   Incoming token.
  *
- * @return State to reduce to, or INVALID_ID if none.
+ * @return Reduce table index to reduce with, or INVALID_ID if none.
  */
 static size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token)
 {
@@ -948,8 +1097,17 @@ static size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token)
  *
  * @param context
  *   Lexer/parser context structure.
- * @start_state_id
+ * @param start_state_id
  *   ID of the state in which to start.
+ * @param start_rule_set_id
+ *   Rule set ID for the requested start rule. Only used when
+ *   @p follow_tokens is non-NULL, to gate follow-token shift success.
+ * @param follow_tokens
+ *   Optional array of caller-provided follow tokens (tokens expected to
+ *   appear immediately after the start rule in some outer context). Used to
+ *   drive the "parse inner" retry logic. May be NULL for a standard parse.
+ * @param n_follow_tokens
+ *   Number of entries in @p follow_tokens.
  *
  * @retval P_SUCCESS
  *   The parser successfully matched the input text. The parse result value
@@ -962,15 +1120,20 @@ static size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token)
  * @reval P_UNEXPECTED_INPUT
  *   Input text does not match any lexer pattern.
  */
-static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start_state_id)
+static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start_state_id,
+    size_t start_rule_set_id,
+    <%= @grammar.prefix %>token_t const * follow_tokens, size_t n_follow_tokens)
 {
     <%= @grammar.prefix %>token_info_t token_info;
     <%= @grammar.prefix %>token_t token = INVALID_TOKEN_ID;
     state_values_stack_t statevalues;
     size_t reduced_rule_set = INVALID_ID;
+    size_t last_shifted_rule_set_id = INVALID_ID;
 <% if @grammar.tree %>
-    void * reduced_parser_node;
+    <%= @grammar.prefix %>node_id_t reduced_parser_node;
 <% else %>
+    <%= @grammar.prefix %>position_t reduced_position;
+    <%= @grammar.prefix %>position_t reduced_end_position;
     <%= @grammar.prefix %>value_t reduced_parser_value;
 <% end %>
     state_values_stack_init(&statevalues);
@@ -981,7 +1144,7 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
     {
         if (token == INVALID_TOKEN_ID)
         {
-            size_t lexer_result = <%= @grammar.prefix %>lex(context, &token_info);
+            size_t lexer_result = <%= lex_fn %>(context, &token_info);
             if (lexer_result != P_SUCCESS)
             {
                 result = lexer_result;
@@ -989,6 +1152,18 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
             }
             token = token_info.token;
         }
+        /* For a "parse inner" operation, determine once per iteration whether
+         * the current token is a member of the caller-provided follow token
+         * set. Used by both the shift-side and reduce-side retries below. */
+        bool token_is_follow = false;
+        for (size_t i = 0u; i < n_follow_tokens; i++)
+        {
+            if (token == follow_tokens[i])
+            {
+                token_is_follow = true;
+                break;
+            }
+        }
         size_t shift_state = INVALID_ID;
         if (reduced_rule_set != INVALID_ID)
         {
@@ -1001,28 +1176,58 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
             {
                 /* Successful parse. */
 <% if @grammar.tree %>
-                context->parse_result = state_values_stack_index(&statevalues, -1)->tree_node;
+                context->parse_result = state_values_stack_index(&statevalues, -1)->node_id;
 <% else %>
                 context->parse_result = state_values_stack_index(&statevalues, -1)->pvalue;
 <% end %>
                 result = P_SUCCESS;
                 break;
             }
+            if ((shift_state == INVALID_ID) && token_is_follow)
+            {
+                /* For a "parse inner" operation, if the incoming token is one
+                 * of the caller's follow tokens, retry the shift as
+                 * TOKEN___EOF. Only consider the parse complete if the reduced
+                 * start rule is the only thing on the parse stack (i.e. the
+                 * initial state plus a single shifted start rule set entry). */
+                size_t retry_shift_state = check_shift(state_values_stack_index(&statevalues, -1)->state_id, TOKEN___EOF);
+                if ((retry_shift_state != INVALID_ID) &&
+                    (statevalues.length == 2u) &&
+                    (last_shifted_rule_set_id == start_rule_set_id))
+                {
+                    /* Successful parse via follow token. Rewind the input
+                     * position so that the follow token is not consumed from
+                     * the input stream and remains available for a subsequent
+                     * call to <%= @grammar.prefix %>lex() or a
+                     * <%= @grammar.prefix %>parse*() function. */
+                    context->input_index -= token_info.length;
+                    context->text_position = token_info.position;
+<% if @grammar.tree %>
+                    context->parse_result = state_values_stack_index(&statevalues, -1)->node_id;
+<% else %>
+                    context->parse_result = state_values_stack_index(&statevalues, -1)->pvalue;
+<% end %>
+                    result = P_SUCCESS;
+                    break;
+                }
+            }
         }
         if (shift_state != INVALID_ID)
         {
-            /* We have something to shift. */
+            /* We have something to shift. Track the last shifted rule set ID
+             * (INVALID_ID if we just shifted a token) so the follow-token
+             * shift retry can gate success on the reduced start rule being the
+             * only thing on top of the initial state. */
+            last_shifted_rule_set_id = reduced_rule_set;
             state_values_stack_push(&statevalues);
-            state_values_stack_index(&statevalues, -1)->state_id = shift_state;
+            state_value_t * new_state_info = state_values_stack_index(&statevalues, -1);
+            new_state_info->state_id = shift_state;
             if (reduced_rule_set == INVALID_ID)
             {
                 /* We shifted a token, mark it consumed. */
 <% if @grammar.tree %>
-<%   if @cpp %>
-                <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %> * token_tree_node = new <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>();
-<%   else %>
-                <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %> * token_tree_node = (<%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %> *)malloc(sizeof(<%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>));
-<%   end %>
+                <%= @grammar.prefix %>node_id_t token_node_id = tree_new_node(context);
+                <%= @grammar.prefix %>node_data_t * token_tree_node = &context-><%= @grammar.prefix %>tree_nodes[token_node_id];
                 token_tree_node->position = token_info.position;
                 token_tree_node->end_position = token_info.end_position;
                 token_tree_node->n_fields = 0u;
@@ -1030,9 +1235,11 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
                 token_tree_node->token = token;
                 token_tree_node->pvalue = token_info.pvalue;
 <%= expand_code(@grammar.on_token_node, false, nil, nil) %>
-                state_values_stack_index(&statevalues, -1)->tree_node = token_tree_node;
+                new_state_info->node_id = token_node_id;
 <% else %>
-                state_values_stack_index(&statevalues, -1)->pvalue = token_info.pvalue;
+                new_state_info->position = token_info.position;
+                new_state_info->end_position = token_info.end_position;
+                new_state_info->pvalue = token_info.pvalue;
 <% end %>
                 token = INVALID_TOKEN_ID;
             }
@@ -1040,9 +1247,11 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
             {
                 /* We shifted a RuleSet. */
 <% if @grammar.tree %>
-                state_values_stack_index(&statevalues, -1)->tree_node = reduced_parser_node;
+                new_state_info->node_id = reduced_parser_node;
 <% else %>
-                state_values_stack_index(&statevalues, -1)->pvalue = reduced_parser_value;
+                new_state_info->pvalue = reduced_parser_value;
+                new_state_info->position = reduced_position;
+                new_state_info->end_position = reduced_end_position;
                 <%= @grammar.prefix %>value_t new_parse_result;
                 memset(&new_parse_result, 0, sizeof(new_parse_result));
                 reduced_parser_value = new_parse_result;
@@ -1053,57 +1262,77 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
         }
 
         size_t reduce_index = check_reduce(state_values_stack_index(&statevalues, -1)->state_id, token);
+        if ((reduce_index == INVALID_ID) && token_is_follow)
+        {
+            /* For a "parse inner" operation, if the incoming token is one of
+             * the caller's follow tokens, retry the reduce lookup as
+             * TOKEN___EOF. Whatever reduce_index results (if any) is used
+             * regardless of which rule set it reduces to; this allows chains
+             * of reductions leading up to the start rule. */
+            reduce_index = check_reduce(state_values_stack_index(&statevalues, -1)->state_id, TOKEN___EOF);
+        }
         if (reduce_index != INVALID_ID)
         {
             /* We have something to reduce. */
 <% if @grammar.tree %>
             if (parser_reduce_table[reduce_index].propagate_optional_target)
             {
-                reduced_parser_node = state_values_stack_index(&statevalues, -1)->tree_node;
+                reduced_parser_node = state_values_stack_index(&statevalues, -1)->node_id;
             }
             else if (parser_reduce_table[reduce_index].n_states > 0)
             {
-                size_t n_fields = parser_reduce_table[reduce_index].rule_set_node_field_array_size;
-                size_t bytes = sizeof(TreeNode) + n_fields * sizeof(void *);
-                TreeNode * node = (TreeNode *)malloc(bytes);
-                memset(node, 0, bytes);
-                node->position = INVALID_POSITION;
-                node->end_position = INVALID_POSITION;
-                node->n_fields = n_fields;
+                uint16_t n_fields = parser_reduce_table[reduce_index].rule_set_node_field_array_size;
+                /* Reserve child slots. New slots are zero-initialized
+                 * (null node ID) so absent optional children remain null. */
+                <%= @grammar.prefix %>node_id_t child_offset = tree_reserve_children(context, n_fields);
                 if (parser_reduce_table[reduce_index].rule_set_node_field_index_map == NULL)
                 {
                     for (size_t i = 0; i < parser_reduce_table[reduce_index].n_states; i++)
                     {
-                        node->fields[i] = (TreeNode *)state_values_stack_index(&statevalues, -(int)parser_reduce_table[reduce_index].n_states + (int)i)->tree_node;
+                        context-><%= @grammar.prefix %>tree_children[child_offset + i] = state_values_stack_index(&statevalues, -(int)parser_reduce_table[reduce_index].n_states + (int)i)->node_id;
                     }
                 }
                 else
                 {
                     for (size_t i = 0; i < parser_reduce_table[reduce_index].n_states; i++)
                     {
-                        node->fields[parser_reduce_table[reduce_index].rule_set_node_field_index_map[i]] = (TreeNode *)state_values_stack_index(&statevalues, -(int)parser_reduce_table[reduce_index].n_states + (int)i)->tree_node;
+                        context-><%= @grammar.prefix %>tree_children[child_offset + parser_reduce_table[reduce_index].rule_set_node_field_index_map[i]] = state_values_stack_index(&statevalues, -(int)parser_reduce_table[reduce_index].n_states + (int)i)->node_id;
                     }
                 }
+                <%= @grammar.prefix %>node_id_t node_id = tree_new_node(context);
+                <%= @grammar.prefix %>node_data_t * node = &context-><%= @grammar.prefix %>tree_nodes[node_id];
+                node->position = INVALID_POSITION;
+                node->end_position = INVALID_POSITION;
+                node->child_offset = child_offset;
+                node->n_fields = n_fields;
+                node->is_token = 0u;
                 bool position_found = false;
-                for (size_t i = 0; i < n_fields; i++)
+                for (uint16_t i = 0; i < n_fields; i++)
                 {
-                    TreeNode * child = node->fields[i];
-                    if ((child != NULL) && <%= @grammar.prefix %>position_valid(child->position))
+                    <%= @grammar.prefix %>node_id_t child_id = context-><%= @grammar.prefix %>tree_children[child_offset + i];
+                    if ((child_id != 0u) && <%= @grammar.prefix %>position_valid(context-><%= @grammar.prefix %>tree_nodes[child_id].position))
                     {
                         if (!position_found)
                         {
-                            node->position = child->position;
+                            node->position = context-><%= @grammar.prefix %>tree_nodes[child_id].position;
                             position_found = true;
                         }
-                        node->end_position = child->end_position;
+                        node->end_position = context-><%= @grammar.prefix %>tree_nodes[child_id].end_position;
                     }
                 }
-                reduced_parser_node = node;
+                reduced_parser_node = node_id;
             }
             else
             {
-                reduced_parser_node = NULL;
+                reduced_parser_node = 0u;
             }
+<%   if @grammar.parser_user_code_used? %>
+            if (parser_user_code(reduced_parser_node, parser_reduce_table[reduce_index].rule, &statevalues, parser_reduce_table[reduce_index].n_states, context) == P_USER_TERMINATED)
+            {
+                state_values_stack_free(&statevalues);
+                return P_USER_TERMINATED;
+            }
+<%   end %>
 <% else %>
             <%= @grammar.prefix %>value_t reduced_parser_value2;
             memset(&reduced_parser_value2, 0, sizeof(reduced_parser_value2));
@@ -1113,6 +1342,16 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
                 return P_USER_TERMINATED;
             }
             reduced_parser_value = reduced_parser_value2;
+            if (parser_reduce_table[reduce_index].n_states > 0u)
+            {
+                reduced_position = get_rule_position(&statevalues, 0u, parser_reduce_table[reduce_index].n_states, false);
+                reduced_end_position = get_rule_position(&statevalues, 0u, parser_reduce_table[reduce_index].n_states, true);
+            }
+            else
+            {
+                memset(&reduced_position, 0, sizeof(reduced_position));
+                memset(&reduced_end_position, 0, sizeof(reduced_end_position));
+            }
 <% end %>
             reduced_rule_set = parser_reduce_table[reduce_index].rule_set;
             state_values_stack_pop(&statevalues, parser_reduce_table[reduce_index].n_states);
@@ -1135,14 +1374,20 @@ static size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start
 
 size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context)
 {
-    return parse_from(context, 0u);
+    return parse_from(context, 0u, <%= @parser.rule_sets[@grammar.start_rules[0]].id %>u, NULL, 0u);
 }
 
 <% @grammar.start_rules.each_with_index do |start_rule, i| %>
 
 size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix %>context_t * context)
 {
-    return parse_from(context, <%= i %>u);
+    return parse_from(context, <%= i %>u, <%= @parser.rule_sets[start_rule].id %>u, NULL, 0u);
+}
+
+size_t <%= @grammar.prefix %>parse_inner_<%= start_rule %>(<%= @grammar.prefix %>context_t * context,
+    <%= @grammar.prefix %>token_t const * follow_tokens, size_t n_follow_tokens)
+{
+    return parse_from(context, <%= i %>u, <%= @parser.rule_sets[start_rule].id %>u, follow_tokens, n_follow_tokens);
 }
 <% end %>
 
@@ -1155,14 +1400,14 @@ size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix %>conte
  * @return Parse result value.
  */
 <% if @grammar.tree %>
-<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> * <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context)
+<%= h_type(@grammar.start_rules[0]) %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context)
 {
-    return (<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> *) context->parse_result;
+    return <%= tree_handle(h_type(@grammar.start_rules[0]), "context->parse_result") %>;
 }
 <%   @grammar.start_rules.each_with_index do |start_rule, i| %>
-<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> * <%= @grammar.prefix %>result_<%= start_rule %>(<%= @grammar.prefix %>context_t * context)
+<%= h_type(start_rule) %> <%= @grammar.prefix %>result_<%= start_rule %>(<%= @grammar.prefix %>context_t * context)
 {
-    return (<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> *) context->parse_result;
+    return <%= tree_handle(h_type(start_rule), "context->parse_result") %>;
 }
 <%   end %>
 <% else %>
@@ -1191,6 +1436,58 @@ size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix %>conte
     return context->text_position;
 }
 
+/**
+ * Set the current text input position.
+ *
+ * This can be used to set the initial text position to something other than
+ * (1, 1) for a nested parse operation so that error positions reported by
+ * subsequent lexer/parser calls are relative to a larger enclosing document.
+ *
+ * @param context
+ *   Lexer/parser context structure.
+ * @param position
+ *   Text position to set.
+ */
+void <%= @grammar.prefix %>set_position(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>position_t position)
+{
+    context->text_position = position;
+}
+
+/**
+ * Get the current input text byte offset.
+ *
+ * @param context
+ *   Lexer/parser context structure.
+ *
+ * @return Current input text byte offset (measured from the start of the
+ *   input text passed to <%= @grammar.prefix %>context_new()).
+ */
+size_t <%= @grammar.prefix %>input_index(<%= @grammar.prefix %>context_t * context)
+{
+    return context->input_index;
+}
+
+/**
+ * Set the current input text byte offset.
+ *
+ * This moves the lexer's read cursor to the given byte offset (measured from
+ * the start of the input text passed to <%= @grammar.prefix %>context_new()).
+ * It can be used together with <%= @grammar.prefix %>set_position() to rewind
+ * the input part-way through a parse in order to re-read an earlier section of
+ * the input. The byte offset is not validated; the caller is responsible for
+ * providing an offset within the bounds of the input text. A value previously
+ * returned by <%= @grammar.prefix %>input_index() is a suitable argument.
+ *
+ * @param context
+ *   Lexer/parser context structure.
+ * @param input_index
+ *   Input text byte offset to set.
+ */
+void <%= @grammar.prefix %>set_input_index(<%= @grammar.prefix %>context_t * context, size_t input_index)
+{
+    context->input_index = input_index;
+}
+
 /**
  * Get the user terminate code.
  *
@@ -1213,48 +1510,3 @@ size_t <%= @grammar.prefix %>user_terminate_code(<%= @grammar.prefix %>context_t
 {
     return context->token;
 }
-<% if @grammar.tree %>
-
-static void tree_delete(TreeNode * node)
-{
-    if (node->is_token)
-    {
-        <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %> * token_tree_node = (<%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %> *)node;
-<%= expand_code(@grammar.free_token_node, false, nil, nil) %>
-<%   if @cpp %>
-        delete token_tree_node;
-<%   else %>
-        free(token_tree_node);
-<%   end %>
-    }
-    else if (node->n_fields > 0u)
-    {
-        for (size_t i = 0u; i < node->n_fields; i++)
-        {
-            if (node->fields[i] != NULL)
-            {
-                tree_delete(node->fields[i]);
-            }
-        }
-        free(node);
-    }
-}
-
-/**
- * Free all tree node memory.
- */
-void <%= @grammar.prefix %>tree_delete(<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> * tree)
-{
-    tree_delete((TreeNode *)tree);
-}
-<%   @grammar.start_rules.each_with_index do |start_rule, i| %>
-
-/**
- * Free all tree node memory.
- */
-void <%= @grammar.prefix %>tree_delete_<%= start_rule %>(<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> * tree)
-{
-    tree_delete((TreeNode *)tree);
-}
-<%   end %>
-<% end %>

assets/parser.d.erb

@@ -59,10 +59,10 @@ public alias <%= @grammar.prefix %>code_point_t = uint;
  */
 public struct <%= @grammar.prefix %>position_t
 {
-    /** Input text row (0-based). */
+    /** Input text row (1-based). */
     uint row;
 
-    /** Input text column (0-based). */
+    /** Input text column (1-based). */
     uint col;
 
     /** Invalid position value. */
@@ -86,25 +86,42 @@ public union <%= @grammar.prefix %>value_t
     <%= typestring %> v_<%= name %>;
 <%   end %>
 }
+
+/** Parser value constructor(s). */
+<%   @grammar.ptypes.each do |name, typestring| %>
+public <%= @grammar.prefix %>value_t <%= @grammar.prefix %>value<%= name == "default" ? "" : "_#{name}" %>(T)(T v)
+{
+    return <%= @grammar.prefix %>value_t(v_<%= name %>: v);
+}
+<%   end %>
+
+/** Parser value accessor(s). */
+<%   @grammar.ptypes.each do |name, typestring| %>
+public <%= typestring %> <%= @grammar.prefix %>value_get<%= name == "default" ? "" : "_#{name}" %>(<%= @grammar.prefix %>value_t * pvalue)
+{
+    return pvalue.v_<%= name %>;
+}
+<%   end %>
 <% end %>
 
 <% if @grammar.tree %>
-/** Common tree node structure. */
-private struct TreeNode
-{
-    <%= @grammar.prefix %>position_t position;
-    <%= @grammar.prefix %>position_t end_position;
-    ushort n_fields;
-    bool is_token;
-    void *[0] fields;
-}
+/** Tree node ID type (index into the context node arena). ID 0 is null. */
+public alias <%= @grammar.prefix %>node_id_t = uint;
 
-/** Tree node types. @{ */
-public struct <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>
+/**
+ * Tree node record.
+ *
+ * All tree nodes are stored contiguously in the context node arena. Child
+ * links are stored in a shared children array: a node's children
+ * occupy children[child_offset .. child_offset + n_fields]. Token payload
+ * fields (token, pvalue, and any user fields) are only meaningful when
+ * is_token is true.
+ */
+private struct <%= @grammar.prefix %>node_data_t
 {
-    /* TreeNode fields must be present in the same order here. */
     <%= @grammar.prefix %>position_t position;
     <%= @grammar.prefix %>position_t end_position;
+    <%= @grammar.prefix %>node_id_t child_offset;
     ushort n_fields;
     bool is_token;
     <%= @grammar.prefix %>token_t token;
@@ -112,22 +129,96 @@ public struct <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>
 <%= @grammar.token_user_fields %>
 }
 
-<%   @parser.rule_sets.each do |name, rule_set| %>
-<%     next if name.start_with?("$") %>
-<%     next if rule_set.optional? %>
-public struct <%= @grammar.tree_prefix %><%= name %><%= @grammar.tree_suffix %>
+/** Tree node handle types. @{ */
+
+/** Token tree node handle. */
+public struct <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>
 {
-    <%= @grammar.prefix %>position_t position;
-    <%= @grammar.prefix %>position_t end_position;
-    ushort n_fields;
-    bool is_token;
-<%     rule_set.tree_fields.each do |fields| %>
-    union
+    private <%= @grammar.prefix %>context_t * __context;
+    private <%= @grammar.prefix %>node_id_t __id;
+
+    this(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>node_id_t id)
     {
-<%       fields.each do |field_name, type| %>
-        <%= type %> * <%= field_name %>;
-<%       end %>
+        this.__context = context;
+        this.__id = id;
     }
+
+    /** Return whether this handle refers to a valid (non-null) node. */
+    @property bool valid()
+    {
+        return __id != 0u;
+    }
+
+    /** Return the node ID (for identity comparison). */
+    @property <%= @grammar.prefix %>node_id_t node_id()
+    {
+        return __id;
+    }
+
+    /** Access the underlying node record (token, pvalue, and user fields). */
+    @property ref <%= @grammar.prefix %>node_data_t __node()
+    {
+        return __context.<%= @grammar.prefix %>tree_nodes[__id];
+    }
+
+    alias __node this;
+}
+
+<%   tree_node_rule_sets.each do |rule_set| %>
+/** <%= rule_set.name %> tree node handle. */
+public struct <%= @grammar.tree_prefix %><%= rule_set.name %><%= @grammar.tree_suffix %>
+{
+    private <%= @grammar.prefix %>context_t * __context;
+    private <%= @grammar.prefix %>node_id_t __id;
+
+    this(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>node_id_t id)
+    {
+        this.__context = context;
+        this.__id = id;
+    }
+
+    /** Return whether this handle refers to a valid (non-null) node. */
+    @property bool valid()
+    {
+        return __id != 0u;
+    }
+
+    /** Return the node ID (for identity comparison). */
+    @property <%= @grammar.prefix %>node_id_t node_id()
+    {
+        return __id;
+    }
+
+    /** Text position of the first code point spanned by this node. */
+    @property <%= @grammar.prefix %>position_t position()
+    {
+        return __context.<%= @grammar.prefix %>tree_nodes[__id].position;
+    }
+
+    /** Text position of the last code point spanned by this node. */
+    @property <%= @grammar.prefix %>position_t end_position()
+    {
+        return __context.<%= @grammar.prefix %>tree_nodes[__id].end_position;
+    }
+
+    /** Number of child fields in this node. */
+    @property ushort n_fields()
+    {
+        return __id ? __context.<%= @grammar.prefix %>tree_nodes[__id].n_fields : cast(ushort)0u;
+    }
+<%     rule_set.tree_fields.each_with_index do |fields, i| %>
+<%       fields.each do |field_name, type| %>
+
+    /** Access the <%= field_name %> child node. */
+    @property <%= type %> <%= field_name %>()
+    {
+        if (__id == 0u)
+        {
+            return <%= type %>(__context, 0u);
+        }
+        return <%= type %>(__context, __context.<%= @grammar.prefix %>tree_children[__context.<%= @grammar.prefix %>tree_nodes[__id].child_offset + <%= i %>u]);
+    }
+<%       end %>
 <%     end %>
 }
 
@@ -180,7 +271,13 @@ public struct <%= @grammar.prefix %>context_t
 
     /** Parse result value. */
 <% if @grammar.tree %>
-    void * parse_result;
+    <%= @grammar.prefix %>node_id_t parse_result;
+
+    /** Tree node arena. Node ID 0 is reserved as the null node. */
+    <%= @grammar.prefix %>node_data_t[] <%= @grammar.prefix %>tree_nodes;
+
+    /** Shared tree child links. */
+    <%= @grammar.prefix %>node_id_t[] <%= @grammar.prefix %>tree_children;
 <% else %>
     <%= @grammar.prefix %>value_t parse_result;
 <% end %>
@@ -237,8 +334,6 @@ private enum size_t INVALID_ID = cast(size_t)-1;
  *
  * @param input
  *   Text input.
- * @param input_length
- *   Text input length.
  *
  * @return Context structure for lexer/parser.
  */
@@ -252,6 +347,11 @@ private enum size_t INVALID_ID = cast(size_t)-1;
     context.text_position.row = 1u;
     context.text_position.col = 1u;
     context.mode = <%= @lexer.mode_id("default") %>;
+<% if @grammar.tree %>
+
+    /* Reserve node ID 0 as the null tree node. */
+    context.<%= @grammar.prefix %>tree_nodes = new <%= @grammar.prefix %>node_data_t[](1);
+<% end %>
 
     return context;
 }
@@ -264,6 +364,16 @@ private enum size_t INVALID_ID = cast(size_t)-1;
  */
 void <%= @grammar.prefix %>context_delete(<%= @grammar.prefix %>context_t * context)
 {
+<% if @grammar.tree && @grammar.free_token_node != "" %>
+    foreach (ref node; context.<%= @grammar.prefix %>tree_nodes)
+    {
+        if (node.is_token)
+        {
+            <%= @grammar.prefix %>node_data_t * token_tree_node = &node;
+<%= expand_code(@grammar.free_token_node, false, nil, nil) %>
+        }
+    }
+<% end %>
 }
 
 /**************************************************************************
@@ -462,7 +572,7 @@ private immutable lexer_mode_t[] lexer_mode_table = [
  *   Lexer/parser context structure.
  * @param code_id
  *   The ID of the user code block to execute.
- * @param match
+ * @param match_text
  *   Matched text for this pattern.
  * @param out_token_info
  *   Lexer token info in progress.
@@ -471,7 +581,7 @@ private immutable lexer_mode_t[] lexer_mode_table = [
  *   not explicitly return a token.
  */
 private <%= @grammar.prefix %>token_t lexer_user_code(<%= @grammar.prefix %>context_t * context,
-    lexer_user_code_id_t code_id, string match,
+    lexer_user_code_id_t code_id, string match_text,
     <%= @grammar.prefix %>token_info_t * out_token_info)
 {
     switch (code_id)
@@ -649,11 +759,27 @@ private size_t attempt_lex_token(<%= @grammar.prefix %>context_t * context, <%=
     {
     case P_SUCCESS:
         <%= @grammar.prefix %>token_t token_to_accept = match_info.accepting_state.token;
+        /* Calculate the token length and start/end positions before invoking
+         * the lexer user code so that the user code can access them. The
+         * context input text position tracking is not updated until after the
+         * user code has run so that it is left unchanged if the user code
+         * requests to terminate the lexer. */
+        token_info.length = match_info.length;
+        if (match_info.end_delta_position.row != 0u)
+        {
+            token_info.end_position.row = token_info.position.row + match_info.end_delta_position.row;
+            token_info.end_position.col = match_info.end_delta_position.col;
+        }
+        else
+        {
+            token_info.end_position.row = token_info.position.row;
+            token_info.end_position.col = token_info.position.col + match_info.end_delta_position.col;
+        }
         if (match_info.accepting_state.code_id != INVALID_USER_CODE_ID)
         {
-            string match = context.input[context.input_index..(context.input_index + match_info.length)];
+            string match_text = context.input[context.input_index..(context.input_index + match_info.length)];
             <%= @grammar.prefix %>token_t user_code_token = lexer_user_code(context,
-                match_info.accepting_state.code_id, match, &token_info);
+                match_info.accepting_state.code_id, match_text, &token_info);
             /* A TERMINATE_TOKEN_ID return code from lexer_user_code() means
              * that the user code is requesting to terminate the lexer. */
             if (user_code_token == TERMINATE_TOKEN_ID)
@@ -687,17 +813,6 @@ private size_t attempt_lex_token(<%= @grammar.prefix %>context_t * context, <%=
             return P_DROP;
         }
         token_info.token = token_to_accept;
-        token_info.length = match_info.length;
-        if (match_info.end_delta_position.row != 0u)
-        {
-            token_info.end_position.row = token_info.position.row + match_info.end_delta_position.row;
-            token_info.end_position.col = match_info.end_delta_position.col;
-        }
-        else
-        {
-            token_info.end_position.row = token_info.position.row;
-            token_info.end_position.col = token_info.position.col + match_info.end_delta_position.col;
-        }
         *out_token_info = token_info;
         return P_SUCCESS;
 
@@ -866,12 +981,14 @@ private struct state_value_t
     /** Parser state ID. */
     size_t state_id;
 
+<% if @grammar.tree %>
+    /** Tree node ID. */
+    <%= @grammar.prefix %>node_id_t node_id;
+<% else %>
+    <%= @grammar.prefix %>position_t position;
+    <%= @grammar.prefix %>position_t end_position;
     /** Parser value from this state. */
     <%= @grammar.prefix %>value_t pvalue;
-
-<% if @grammar.tree %>
-    /** Tree node. */
-    void * tree_node;
 <% end %>
 
     this(size_t state_id)
@@ -925,6 +1042,55 @@ private immutable parser_state_t[] parser_state_table = [
 ];
 
 <% unless @grammar.tree %>
+/**
+ * Get the rule position (start or end) for the currently matched rule.
+ */
+private <%= @grammar.prefix %>position_t get_rule_position(state_value_t[] statevalues, size_t i, size_t n_states, bool get_end)
+{
+    if (n_states > 0u)
+    {
+        if (i == 0u)
+        {
+            if (get_end)
+            {
+                for (size_t j = 0u; j < n_states; j++)
+                {
+                    state_value_t * sv = &statevalues[$-1-j];
+                    if (sv.end_position.valid)
+                    {
+                        return sv.end_position;
+                    }
+                }
+            }
+            else
+            {
+                for (size_t j = 0u; j < n_states; j++)
+                {
+                    state_value_t * sv = &statevalues[$-n_states+j];
+                    if (sv.position.valid)
+                    {
+                        return sv.position;
+                    }
+                }
+            }
+        }
+        else
+        {
+            if (get_end)
+            {
+                return statevalues[$-1-n_states+i].end_position;
+            }
+            else
+            {
+                return statevalues[$-1-n_states+i].position;
+            }
+        }
+    }
+    return <%= @grammar.prefix %>position_t.INVALID;
+}
+<% end %>
+
+<% if !@grammar.tree || @grammar.parser_user_code_used? %>
 /**
  * Execute user code associated with a parser rule.
  *
@@ -935,7 +1101,7 @@ private immutable parser_state_t[] parser_state_table = [
  * @retval P_USER_TERMINATED
  *   User requested to terminate parsing.
  */
-private size_t parser_user_code(<%= @grammar.prefix %>value_t * _pvalue, uint rule, state_value_t[] statevalues, uint n_states, <%= @grammar.prefix %>context_t * context)
+private size_t parser_user_code(<%= @grammar.tree ? "#{@grammar.prefix}node_id_t _node_id" : "#{@grammar.prefix}value_t * _pvalue" %>, uint rule, state_value_t[] statevalues, uint n_states, <%= @grammar.prefix %>context_t * context)
 {
     switch (rule)
     {
@@ -985,7 +1151,7 @@ private size_t check_shift(size_t state_id, size_t symbol_id)
  * @param token
  *   Incoming token.
  *
- * @return State to reduce to, or INVALID_ID if none.
+ * @return Reduce table index to reduce with, or INVALID_ID if none.
  */
 private size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token)
 {
@@ -1007,8 +1173,16 @@ private size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token
  *
  * @param context
  *   Lexer/parser context structure.
- * @start_state_id
+ * @param start_state_id
  *   ID of the state in which to start.
+ * @param start_rule_set_id
+ *   Rule set ID for the requested start rule. Only used when
+ *   @p follow_tokens is non-empty, to gate follow-token shift success.
+ * @param follow_tokens
+ *   Optional slice of caller-provided follow tokens (tokens expected to
+ *   appear immediately after the start rule in some outer context). Used to
+ *   drive the "parse inner" retry logic. May be null/empty for a standard
+ *   parse.
  *
  * @retval P_SUCCESS
  *   The parser successfully matched the input text. The parse result value
@@ -1021,29 +1195,46 @@ private size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token
  * @reval P_UNEXPECTED_INPUT
  *   Input text does not match any lexer pattern.
  */
-private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start_state_id)
+private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t start_state_id,
+    size_t start_rule_set_id,
+    const(<%= @grammar.prefix %>token_t)[] follow_tokens)
 {
     <%= @grammar.prefix %>token_info_t token_info;
     <%= @grammar.prefix %>token_t token = INVALID_TOKEN_ID;
     state_value_t[] statevalues = new state_value_t[](1);
     statevalues[0].state_id = start_state_id;
     size_t reduced_rule_set = INVALID_ID;
+    size_t last_shifted_rule_set_id = INVALID_ID;
 <% if @grammar.tree %>
-    void * reduced_parser_node;
+    <%= @grammar.prefix %>node_id_t reduced_parser_node;
 <% else %>
+    <%= @grammar.prefix %>position_t reduced_position;
+    <%= @grammar.prefix %>position_t reduced_end_position;
     <%= @grammar.prefix %>value_t reduced_parser_value;
 <% end %>
     for (;;)
     {
         if (token == INVALID_TOKEN_ID)
         {
-            size_t lexer_result = <%= @grammar.prefix %>lex(context, &token_info);
+            size_t lexer_result = <%= lex_fn %>(context, &token_info);
             if (lexer_result != P_SUCCESS)
             {
                 return lexer_result;
             }
             token = token_info.token;
         }
+        /* For a "parse inner" operation, determine once per iteration whether
+         * the current token is a member of the caller-provided follow token
+         * set. Used by both the shift-side and reduce-side retries below. */
+        bool token_is_follow = false;
+        foreach (eof_token; follow_tokens)
+        {
+            if (token == eof_token)
+            {
+                token_is_follow = true;
+                break;
+            }
+        }
         size_t shift_state = INVALID_ID;
         if (reduced_rule_set != INVALID_ID)
         {
@@ -1056,25 +1247,66 @@ private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t star
             {
                 /* Successful parse. */
 <% if @grammar.tree %>
-                context.parse_result = statevalues[$-1].tree_node;
+                context.parse_result = statevalues[$-1].node_id;
 <% else %>
                 context.parse_result = statevalues[$-1].pvalue;
 <% end %>
                 return P_SUCCESS;
             }
+            if ((shift_state == INVALID_ID) && token_is_follow)
+            {
+                /* For a "parse inner" operation, if the incoming token is one
+                 * of the caller's follow tokens, retry the shift as
+                 * TOKEN___EOF. Only consider the parse complete if the reduced
+                 * start rule is the only thing on the parse stack (i.e. the
+                 * initial state plus a single shifted start rule set entry). */
+                size_t retry_shift_state = check_shift(statevalues[$-1].state_id, TOKEN___EOF);
+                if ((retry_shift_state != INVALID_ID) &&
+                    (statevalues.length == 2u) &&
+                    (last_shifted_rule_set_id == start_rule_set_id))
+                {
+                    /* Successful parse via follow token. Rewind the input
+                     * position so that the follow token is not consumed from
+                     * the input stream and remains available for a subsequent
+                     * call to <%= @grammar.prefix %>lex() or a
+                     * <%= @grammar.prefix %>parse*() function. */
+                    context.input_index -= token_info.length;
+                    context.text_position = token_info.position;
+<% if @grammar.tree %>
+                    context.parse_result = statevalues[$-1].node_id;
+<% else %>
+                    context.parse_result = statevalues[$-1].pvalue;
+<% end %>
+                    return P_SUCCESS;
+                }
+            }
         }
         if (shift_state != INVALID_ID)
         {
-            /* We have something to shift. */
+            /* We have something to shift. Track the last shifted rule set ID
+             * (INVALID_ID if we just shifted a token) so the follow-token
+             * shift retry can gate success on the reduced start rule being the
+             * only thing on top of the initial state. */
+            last_shifted_rule_set_id = reduced_rule_set;
             statevalues ~= state_value_t(shift_state);
             if (reduced_rule_set == INVALID_ID)
             {
                 /* We shifted a token, mark it consumed. */
 <% if @grammar.tree %>
-                <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %> * token_tree_node = new <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>(token_info.position, token_info.end_position, 0u, true, token, token_info.pvalue);
+                <%= @grammar.prefix %>node_id_t token_node_id = cast(<%= @grammar.prefix %>node_id_t)context.<%= @grammar.prefix %>tree_nodes.length;
+                context.<%= @grammar.prefix %>tree_nodes ~= <%= @grammar.prefix %>node_data_t.init;
+                <%= @grammar.prefix %>node_data_t * token_tree_node = &context.<%= @grammar.prefix %>tree_nodes[token_node_id];
+                token_tree_node.position = token_info.position;
+                token_tree_node.end_position = token_info.end_position;
+                token_tree_node.n_fields = 0u;
+                token_tree_node.is_token = true;
+                token_tree_node.token = token;
+                token_tree_node.pvalue = token_info.pvalue;
 <%= expand_code(@grammar.on_token_node, false, nil, nil) %>
-                statevalues[$-1].tree_node = token_tree_node;
+                statevalues[$-1].node_id = token_node_id;
 <% else %>
+                statevalues[$-1].position = token_info.position;
+                statevalues[$-1].end_position = token_info.end_position;
                 statevalues[$-1].pvalue = token_info.pvalue;
 <% end %>
                 token = INVALID_TOKEN_ID;
@@ -1083,9 +1315,11 @@ private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t star
             {
                 /* We shifted a RuleSet. */
 <% if @grammar.tree %>
-                statevalues[$-1].tree_node = reduced_parser_node;
+                statevalues[$-1].node_id = reduced_parser_node;
 <% else %>
                 statevalues[$-1].pvalue = reduced_parser_value;
+                statevalues[$-1].position = reduced_position;
+                statevalues[$-1].end_position = reduced_end_position;
                 <%= @grammar.prefix %>value_t new_parse_result;
                 reduced_parser_value = new_parse_result;
 <% end %>
@@ -1095,62 +1329,78 @@ private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t star
         }
 
         size_t reduce_index = check_reduce(statevalues[$-1].state_id, token);
+        if ((reduce_index == INVALID_ID) && token_is_follow)
+        {
+            /* For a "parse inner" operation, if the incoming token is one of
+             * the caller's follow tokens, retry the reduce lookup as
+             * TOKEN___EOF. Whatever reduce_index results (if any) is used
+             * regardless of which rule set it reduces to; this allows chains
+             * of reductions leading up to the start rule. */
+            reduce_index = check_reduce(statevalues[$-1].state_id, TOKEN___EOF);
+        }
         if (reduce_index != INVALID_ID)
         {
             /* We have something to reduce. */
 <% if @grammar.tree %>
             if (parser_reduce_table[reduce_index].propagate_optional_target)
             {
-                reduced_parser_node = statevalues[$ - 1].tree_node;
+                reduced_parser_node = statevalues[$ - 1].node_id;
             }
             else if (parser_reduce_table[reduce_index].n_states > 0)
             {
-                size_t n_fields = parser_reduce_table[reduce_index].rule_set_node_field_array_size;
-                size_t node_size = TreeNode.sizeof + n_fields * (void *).sizeof;
-                TreeNode * node = cast(TreeNode *)malloc(node_size);
-                GC.addRange(node, node_size);
-                node.position = <%= @grammar.prefix %>position_t.INVALID;
-                node.end_position = <%= @grammar.prefix %>position_t.INVALID;
-                node.n_fields = cast(ushort)n_fields;
-                node.is_token = false;
-                foreach (i; 0..n_fields)
-                {
-                    node.fields[i] = null;
-                }
+                ushort n_fields = parser_reduce_table[reduce_index].rule_set_node_field_array_size;
+                /* Reserve child slots. New slots are zero-initialized
+                 * (null node ID) so absent optional children remain null. */
+                <%= @grammar.prefix %>node_id_t child_offset = cast(<%= @grammar.prefix %>node_id_t)context.<%= @grammar.prefix %>tree_children.length;
+                context.<%= @grammar.prefix %>tree_children.length += n_fields;
                 if (parser_reduce_table[reduce_index].rule_set_node_field_index_map is null)
                 {
                     foreach (i; 0..parser_reduce_table[reduce_index].n_states)
                     {
-                        node.fields[i] = statevalues[$ - parser_reduce_table[reduce_index].n_states + i].tree_node;
+                        context.<%= @grammar.prefix %>tree_children[child_offset + i] = statevalues[$ - parser_reduce_table[reduce_index].n_states + i].node_id;
                     }
                 }
                 else
                 {
                     foreach (i; 0..parser_reduce_table[reduce_index].n_states)
                     {
-                        node.fields[parser_reduce_table[reduce_index].rule_set_node_field_index_map[i]] = statevalues[$ - parser_reduce_table[reduce_index].n_states + i].tree_node;
+                        context.<%= @grammar.prefix %>tree_children[child_offset + parser_reduce_table[reduce_index].rule_set_node_field_index_map[i]] = statevalues[$ - parser_reduce_table[reduce_index].n_states + i].node_id;
                     }
                 }
+                <%= @grammar.prefix %>node_id_t node_id = cast(<%= @grammar.prefix %>node_id_t)context.<%= @grammar.prefix %>tree_nodes.length;
+                context.<%= @grammar.prefix %>tree_nodes ~= <%= @grammar.prefix %>node_data_t.init;
+                <%= @grammar.prefix %>node_data_t * node = &context.<%= @grammar.prefix %>tree_nodes[node_id];
+                node.position = <%= @grammar.prefix %>position_t.INVALID;
+                node.end_position = <%= @grammar.prefix %>position_t.INVALID;
+                node.child_offset = child_offset;
+                node.n_fields = n_fields;
+                node.is_token = false;
                 bool position_found = false;
                 foreach (i; 0..n_fields)
                 {
-                    TreeNode * child = cast(TreeNode *)node.fields[i];
-                    if (child && child.position.valid)
+                    <%= @grammar.prefix %>node_id_t child_id = context.<%= @grammar.prefix %>tree_children[child_offset + i];
+                    if (child_id != 0u && context.<%= @grammar.prefix %>tree_nodes[child_id].position.valid)
                     {
                         if (!position_found)
                         {
-                            node.position = child.position;
+                            node.position = context.<%= @grammar.prefix %>tree_nodes[child_id].position;
                             position_found = true;
                         }
-                        node.end_position = child.end_position;
+                        node.end_position = context.<%= @grammar.prefix %>tree_nodes[child_id].end_position;
                     }
                 }
-                reduced_parser_node = node;
+                reduced_parser_node = node_id;
             }
             else
             {
-                reduced_parser_node = null;
+                reduced_parser_node = 0u;
             }
+<%   if @grammar.parser_user_code_used? %>
+            if (parser_user_code(reduced_parser_node, parser_reduce_table[reduce_index].rule, statevalues, parser_reduce_table[reduce_index].n_states, context) == P_USER_TERMINATED)
+            {
+                return P_USER_TERMINATED;
+            }
+<%   end %>
 <% else %>
             <%= @grammar.prefix %>value_t reduced_parser_value2;
             if (parser_user_code(&reduced_parser_value2, parser_reduce_table[reduce_index].rule, statevalues, parser_reduce_table[reduce_index].n_states, context) == P_USER_TERMINATED)
@@ -1158,6 +1408,16 @@ private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t star
                 return P_USER_TERMINATED;
             }
             reduced_parser_value = reduced_parser_value2;
+            if (parser_reduce_table[reduce_index].n_states > 0u)
+            {
+                reduced_position = get_rule_position(statevalues, 0u, parser_reduce_table[reduce_index].n_states, false);
+                reduced_end_position = get_rule_position(statevalues, 0u, parser_reduce_table[reduce_index].n_states, true);
+            }
+            else
+            {
+                reduced_position = <%= @grammar.prefix %>position_t.INVALID;
+                reduced_end_position = <%= @grammar.prefix %>position_t.INVALID;
+            }
 <% end %>
             reduced_rule_set = parser_reduce_table[reduce_index].rule_set;
             statevalues.length -= parser_reduce_table[reduce_index].n_states;
@@ -1177,14 +1437,20 @@ private size_t parse_from(<%= @grammar.prefix %>context_t * context, size_t star
 
 public size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context)
 {
-    return parse_from(context, 0u);
+    return parse_from(context, 0u, <%= @parser.rule_sets[@grammar.start_rules[0]].id %>u, null);
 }
 
 <% @grammar.start_rules.each_with_index do |start_rule, i| %>
 
 public size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix %>context_t * context)
 {
-    return parse_from(context, <%= i %>u);
+    return parse_from(context, <%= i %>u, <%= @parser.rule_sets[start_rule].id %>u, null);
+}
+
+public size_t <%= @grammar.prefix %>parse_inner_<%= start_rule %>(<%= @grammar.prefix %>context_t * context,
+    const(<%= @grammar.prefix %>token_t)[] follow_tokens)
+{
+    return parse_from(context, <%= i %>u, <%= @parser.rule_sets[start_rule].id %>u, follow_tokens);
 }
 <% end %>
 
@@ -1197,14 +1463,14 @@ public size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix
  * @return Parse result value.
  */
 <% if @grammar.tree %>
-public <%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> * <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context)
+public <%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context)
 {
-    return cast(<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> *)context.parse_result;
+    return <%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %>(context, context.parse_result);
 }
 <%   @grammar.start_rules.each_with_index do |start_rule, i| %>
-public <%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> * <%= @grammar.prefix %>result_<%= start_rule %>(<%= @grammar.prefix %>context_t * context)
+public <%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> <%= @grammar.prefix %>result_<%= start_rule %>(<%= @grammar.prefix %>context_t * context)
 {
-    return cast(<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> *)context.parse_result;
+    return <%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %>(context, context.parse_result);
 }
 <%   end %>
 <% else %>
@@ -1220,35 +1486,6 @@ public <%= start_rule_type(i)[1] %> <%= @grammar.prefix %>result_<%= start_rule
 <%   end %>
 <% end %>
 
-<% if @grammar.tree %>
-private void tree_delete(TreeNode * node)
-{
-    if (!node.is_token)
-    {
-        for (size_t i = 0u; i < node.n_fields; i++)
-        {
-            if (node.fields[i])
-            {
-                tree_delete(cast(TreeNode *)node.fields[i]);
-            }
-        }
-        GC.removeRange(node);
-        free(node);
-    }
-}
-
-void <%= @grammar.prefix %>tree_delete(<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> * tree)
-{
-    tree_delete(cast(TreeNode *)tree);
-}
-<%   @grammar.start_rules.each_with_index do |start_rule, i| %>
-void <%= @grammar.prefix %>tree_delete_<%= start_rule %>(<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> * tree)
-{
-    tree_delete(cast(TreeNode *)tree);
-}
-<%   end %>
-<% end %>
-
 /**
  * Get the current text input position.
  *
@@ -1262,6 +1499,58 @@ public <%= @grammar.prefix %>position_t <%= @grammar.prefix %>position(<%= @gram
     return context.text_position;
 }
 
+/**
+ * Set the current text input position.
+ *
+ * This can be used to set the initial text position to something other than
+ * (1, 1) for a nested parse operation so that error positions reported by
+ * subsequent lexer/parser calls are relative to a larger enclosing document.
+ *
+ * @param context
+ *   Lexer/parser context structure.
+ * @param position
+ *   Text position to set.
+ */
+public void <%= @grammar.prefix %>set_position(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>position_t position)
+{
+    context.text_position = position;
+}
+
+/**
+ * Get the current input text byte offset.
+ *
+ * @param context
+ *   Lexer/parser context structure.
+ *
+ * @return Current input text byte offset (measured from the start of the
+ *   input text passed to <%= @grammar.prefix %>context_new()).
+ */
+public size_t <%= @grammar.prefix %>input_index(<%= @grammar.prefix %>context_t * context)
+{
+    return context.input_index;
+}
+
+/**
+ * Set the current input text byte offset.
+ *
+ * This moves the lexer's read cursor to the given byte offset (measured from
+ * the start of the input text passed to <%= @grammar.prefix %>context_new()).
+ * It can be used together with <%= @grammar.prefix %>set_position() to rewind
+ * the input part-way through a parse in order to re-read an earlier section of
+ * the input. The byte offset is not validated; the caller is responsible for
+ * providing an offset within the bounds of the input text. A value previously
+ * returned by <%= @grammar.prefix %>input_index() is a suitable argument.
+ *
+ * @param context
+ *   Lexer/parser context structure.
+ * @param input_index
+ *   Input text byte offset to set.
+ */
+public void <%= @grammar.prefix %>set_input_index(<%= @grammar.prefix %>context_t * context, size_t input_index)
+{
+    context.input_index = input_index;
+}
+
 /**
  * Get the user terminate code.
  *

assets/parser.h.erb

@@ -8,6 +8,9 @@
 
 #include <stdint.h>
 #include <stddef.h>
+<% if @cpp %>
+#include <vector>
+<% end %>
 
 /**************************************************************************
  * Public types
@@ -45,10 +48,10 @@ typedef uint32_t <%= @grammar.prefix %>code_point_t;
  */
 typedef struct
 {
-    /** Input text row (0-based). */
+    /** Input text row (1-based). */
     uint32_t row;
 
-    /** Input text column (0-based). */
+    /** Input text column (1-based). */
     uint32_t col;
 } <%= @grammar.prefix %>position_t;
 
@@ -69,50 +72,48 @@ typedef union
     <%= typestring %> v_<%= name %>;
 <%   end %>
 } <%= @grammar.prefix %>value_t;
+
+/** Parser value constructor(s). */
+<%   @grammar.ptypes.each do |name, typestring| %>
+static inline <%= @grammar.prefix %>value_t <%= @grammar.prefix %>value<%= name == "default" ? "" : "_#{name}" %>(<%= typestring %> v)
+{
+    return (<%= @grammar.prefix %>value_t){.v_<%= name %> = v};
+}
+<%   end %>
+
+/** Parser value accessor(s). */
+<%   @grammar.ptypes.each do |name, typestring| %>
+static inline <%= typestring %> <%= @grammar.prefix %>value_get<%= name == "default" ? "" : "_#{name}" %>(<%= @grammar.prefix %>value_t const * pvalue)
+{
+    return pvalue->v_<%= name %>;
+}
+<%   end %>
 <% end %>
 
 <% if @grammar.tree %>
-/** Tree node types. @{ */
-typedef struct <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>
+/** Tree node ID type (index into the context node arena). ID 0 is null. */
+typedef uint32_t <%= @grammar.prefix %>node_id_t;
+
+/**
+ * Tree node record.
+ *
+ * All tree nodes are stored contiguously in the context node arena. Child
+ * links are stored in a shared children array: a node's children
+ * occupy children[child_offset .. child_offset + n_fields]. Token payload
+ * fields (token, pvalue, and any user fields) are only meaningful when
+ * is_token is nonzero.
+ */
+typedef struct
 {
-    <% # TreeNode fields must be present in the same order here. # %>
     <%= @grammar.prefix %>position_t position;
     <%= @grammar.prefix %>position_t end_position;
+    <%= @grammar.prefix %>node_id_t child_offset;
     uint16_t n_fields;
     uint8_t is_token;
-<%= @grammar.token_user_fields %>
     <%= @grammar.prefix %>token_t token;
     <%= @grammar.prefix %>value_t pvalue;
-} <%= @grammar.tree_prefix %>Token<%= @grammar.tree_suffix %>;
-
-<%   @parser.rule_sets.each do |name, rule_set| %>
-<%     next if name.start_with?("$") %>
-<%     next if rule_set.optional? %>
-struct <%= name %>;
-<%   end %>
-
-<%   @parser.rule_sets.each do |name, rule_set| %>
-<%     next if name.start_with?("$") %>
-<%     next if rule_set.optional? %>
-typedef struct <%= @grammar.tree_prefix %><%= name %><%= @grammar.tree_suffix %>
-{
-    <% # TreeNode fields must be present in the same order here. # %>
-    <%= @grammar.prefix %>position_t position;
-    <%= @grammar.prefix %>position_t end_position;
-    uint16_t n_fields;
-    uint8_t is_token;
-<%     rule_set.tree_fields.each do |fields| %>
-    union
-    {
-<%       fields.each do |field_name, type| %>
-        struct <%= type %> * <%= field_name %>;
-<%       end %>
-    };
-<%     end %>
-} <%= @grammar.tree_prefix %><%= name %><%= @grammar.tree_suffix %>;
-
-<%   end %>
-/** @} */
+<%= @grammar.token_user_fields %>
+} <%= @grammar.prefix %>node_data_t;
 <% end %>
 
 /** Lexed token information. */
@@ -134,13 +135,19 @@ typedef struct
     <%= @grammar.prefix %>value_t pvalue;
 } <%= @grammar.prefix %>token_info_t;
 
+typedef struct <%= @grammar.prefix %>context_s <%= @grammar.prefix %>context_t;
+
+<% if @grammar.tree %>
+<%= c_tree_handle_types_header %>
+
+<% end %>
 /**
  * Lexer and parser context.
  *
  * The user must allocate an instance of this structure and pass it to any
  * public API function.
  */
-typedef struct
+struct <%= @grammar.prefix %>context_s
 {
     /* Lexer context data. */
 
@@ -163,7 +170,25 @@ typedef struct
 
     /** Parse result value. */
 <% if @grammar.tree %>
-    void * parse_result;
+    <%= @grammar.prefix %>node_id_t parse_result;
+
+<%   if @cpp %>
+    /** Tree node arena. Node ID 0 is reserved as the null node. */
+    std::vector<<%= @grammar.prefix %>node_data_t> <%= @grammar.prefix %>tree_nodes;
+
+    /** Shared tree child links. */
+    std::vector<<%= @grammar.prefix %>node_id_t> <%= @grammar.prefix %>tree_children;
+<%   else %>
+    /** Tree node arena. Node ID 0 is reserved as the null node. */
+    <%= @grammar.prefix %>node_data_t * <%= @grammar.prefix %>tree_nodes;
+    size_t <%= @grammar.prefix %>tree_nodes_length;
+    size_t <%= @grammar.prefix %>tree_nodes_capacity;
+
+    /** Shared tree child links. */
+    <%= @grammar.prefix %>node_id_t * <%= @grammar.prefix %>tree_children;
+    size_t <%= @grammar.prefix %>tree_children_length;
+    size_t <%= @grammar.prefix %>tree_children_capacity;
+<%   end %>
 <% else %>
     <%= @grammar.prefix %>value_t parse_result;
 <% end %>
@@ -175,7 +200,11 @@ typedef struct
     size_t user_terminate_code;
 
 <%= @grammar.context_user_fields %>
-} <%= @grammar.prefix %>context_t;
+};
+
+<% if @grammar.tree %>
+<%= c_tree_types_header %>
+<% end %>
 
 /**************************************************************************
  * Public data
@@ -196,12 +225,14 @@ size_t <%= @grammar.prefix %>lex(<%= @grammar.prefix %>context_t * context, <%=
 size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context);
 <%   @grammar.start_rules.each_with_index do |start_rule, i| %>
 size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix %>context_t * context);
+size_t <%= @grammar.prefix %>parse_inner_<%= start_rule %>(<%= @grammar.prefix %>context_t * context,
+        <%= @grammar.prefix %>token_t const * follow_tokens, size_t n_follow_tokens);
 <%   end %>
 
 <% if @grammar.tree %>
-<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> * <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context);
+<%= h_type(@grammar.start_rules[0]) %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context);
 <%   @grammar.start_rules.each_with_index do |start_rule, i| %>
-<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> * <%= @grammar.prefix %>result_<%= start_rule %>(<%= @grammar.prefix %>context_t * context);
+<%= h_type(start_rule) %> <%= @grammar.prefix %>result_<%= start_rule %>(<%= @grammar.prefix %>context_t * context);
 <%   end %>
 <% else %>
 <%= start_rule_type[1] %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context);
@@ -210,15 +241,14 @@ size_t <%= @grammar.prefix %>parse_<%= start_rule %>(<%= @grammar.prefix %>conte
 <%   end %>
 <% end %>
 
-<% if @grammar.tree %>
-void <%= @grammar.prefix %>tree_delete(<%= @grammar.tree_prefix %><%= @grammar.start_rules[0] %><%= @grammar.tree_suffix %> * tree);
-<%   @grammar.start_rules.each_with_index do |start_rule, i| %>
-void <%= @grammar.prefix %>tree_delete_<%= start_rule %>(<%= @grammar.tree_prefix %><%= start_rule %><%= @grammar.tree_suffix %> * tree);
-<%   end %>
-<% end %>
-
 <%= @grammar.prefix %>position_t <%= @grammar.prefix %>position(<%= @grammar.prefix %>context_t * context);
 
+void <%= @grammar.prefix %>set_position(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>position_t position);
+
+size_t <%= @grammar.prefix %>input_index(<%= @grammar.prefix %>context_t * context);
+
+void <%= @grammar.prefix %>set_input_index(<%= @grammar.prefix %>context_t * context, size_t input_index);
+
 size_t <%= @grammar.prefix %>user_terminate_code(<%= @grammar.prefix %>context_t * context);
 
 <%= @grammar.prefix %>token_t <%= @grammar.prefix %>token(<%= @grammar.prefix %>context_t * context);

extra/vim/syntax/propane.vim

@@ -7,8 +7,46 @@ if exists("b:current_syntax")
   finish
 endif
 
+" Guess the language of the user code blocks from their contents so that the
+" matching syntax file can be included below. b:propane_subtype may also be set
+" before this file is sourced to select the language explicitly.
 if !exists("b:propane_subtype")
-  let b:propane_subtype = "d"
+  " Rust markers. Each keyword requires the syntax that follows it in Rust so
+  " that a plain identifier of the same name in another language does not match
+  " (`int fn = 3;' in C, for example). Type names are only accepted within a
+  " `ptype' statement for the same reason.
+  let s:rust = '\<let\s\+\%(mut\s\+\)\?\w'
+  let s:rust .= '\|\<fn\s\+\w\+\s*('
+  let s:rust .= '\|&mut\>\|\<pub\s\+\w\|\<impl\s\+\w'
+  let s:rust .= '\|#\[\|\<use\s\+\%(std\|core\)::'
+  let s:rust .= '\|\<ptype\>[^;]*\<\%(isize\|usize\|i8\|i16\|i32\|i64\|i128'
+  let s:rust .= '\|u8\|u16\|u32\|u64\|u128\|f32\|f64\|String\)\>'
+  " D markers. These are spellings that have no valid C, C++, or Rust
+  " equivalent, so `import' is deliberately not among them: it is a D keyword
+  " but is also a C++20 module declaration.
+  let s:d = '\<foreach\%(_reverse\)\?\s*([^)]*;'
+  let s:d .= '\|\~=\|\<static\s\+if\s*(\|\<version\s*(\s*\w\+\s*)'
+  let s:d .= '\|\<scope\s*(\s*\%(exit\|failure\|success\)\s*)'
+  let s:d .= '\|\<\%(unittest\|mixin\|immutable\|__gshared\|invariant\)\>'
+  let s:d .= '\|\<alias\s\+\w\+\s*=\|\<enum\s\+\w\+\s*='
+  let s:d .= '\|@\%(property\|safe\|trusted\|system\|nogc\|disable\)\>'
+  let s:d .= '\|\<is\s\+null\>\|\<cast\s*(\s*\w\+\s*)'
+  let s:d .= '\|\<write\%(ln\|fln\|f\)\s*('
+  let s:d .= '\|\<\%(dchar\|dstring\|wstring\|cent\|ucent\)\>'
+  " A module import on its own is ambiguous between D and C++20, so only take
+  " it as D when nothing else in the file looks like C++.
+  let s:import = '\<import\s\+[A-Za-z_][A-Za-z0-9_.]*\s*;'
+  let s:cpp = '::\|\<template\s*<\|\<namespace\>\|\<nullptr\>\|#include\s*[<"]'
+  if search(s:rust, 'nw') > 0
+    let b:propane_subtype = "rust"
+  elseif search(s:d, 'nw') > 0
+    let b:propane_subtype = "d"
+  elseif search(s:import, 'nw') > 0 && search(s:cpp, 'nw') == 0
+    let b:propane_subtype = "d"
+  else
+    let b:propane_subtype = "cpp"
+  endif
+  unlet s:rust s:d s:import s:cpp
 endif
 
 exe "syn include @propaneTarget syntax/".b:propane_subtype.".vim"
@@ -16,18 +54,32 @@ exe "syn include @propaneTarget syntax/".b:propane_subtype.".vim"
 syn region propaneTarget matchgroup=propaneDelimiter start="<<" end=">>$" contains=@propaneTarget keepend
 
 syn match propaneComment "#.*"
-syn match propaneOperator "->"
 syn match propaneFieldAlias ":[a-zA-Z0-9_]\+" contains=propaneFieldOperator
 syn match propaneFieldOperator ":" contained
 syn match propaneOperator "?"
-syn keyword propaneKeyword drop free_token_node free_token_user_fields module prefix ptype start token token_user_fields tokenid tree tree_prefix tree_suffix
+" The right-hand side of a rule (after '->' up to '<<' or ';') lists symbol
+" names that may coincide with propane keywords (e.g. 'start', 'token',
+" 'tree'). Wrap it in a region that excludes keyword matches so those names
+" are not highlighted as keywords. The '<<' is left unconsumed so the
+" propaneTarget region can still match it.
+syn region propaneRuleRhs matchgroup=propaneOperator start="->" end="\ze<<" end=";" contains=propaneFieldAlias,propaneRuleOperator,propaneComment keepend
+syn match propaneRuleOperator "?" contained
+" Keywords that introduce a user-defined name. The name is consumed by
+" propaneName via nextgroup so a name matching a keyword (e.g. 'token start')
+" is not highlighted as a keyword. These must be a match (not syn keyword)
+" because a syn keyword always wins over a contained nextgroup match.
+syn match propaneNameDecl "\<\%(tokenid\|token\|lex_fn\|module\|start\|tree_prefix\|tree_suffix\)\>" nextgroup=propaneName skipwhite
+syn match propaneName "\<\h\w*\>" contained
+syn match propaneKeyword "\<\%(context_user_fields\|drop\|free_token_node\|noline\|on_token_node\|prefix\|ptype\|token_user_fields\|tree\)\>"
 
 syn region propaneRegex start="/" end="/" skip="\v\\\\|\\/"
 
 hi def link propaneComment Comment
 hi def link propaneKeyword Keyword
+hi def link propaneNameDecl Keyword
 hi def link propaneRegex String
 hi def link propaneOperator Operator
+hi def link propaneRuleOperator Operator
 hi def link propaneFieldOperator Operator
 hi def link propaneDelimiter Delimiter
 hi def link propaneFieldAlias Identifier

lib/propane.rb

@@ -33,7 +33,7 @@ class Propane
 
     def run(input_file, output_file, log_file, options)
       begin
-        grammar = Grammar.new(File.read(input_file))
+        grammar = Grammar.new(File.read(input_file), input_file)
         generator = Generator.new(grammar, output_file, log_file, options)
         generator.generate
       rescue Error => e

lib/propane/generator.rb

@@ -18,6 +18,8 @@ class Propane
         elsif output_file =~ %r{\.(cc|cpp|cxx)$}
           @cpp = true
           "c"
+        elsif output_file.end_with?(".rs")
+          "rust"
         else
           raise Error.new("Could not determine target language from output file name (#{output_file})")
         end
@@ -31,14 +33,37 @@ class Propane
         extensions += %w[h]
       end
       extensions.each do |extension|
-        template = Assets.get("parser.#{extension || @language}.erb")
+        template_language = @language == "rust" ? "rs" : @language
+        template = Assets.get("parser.#{extension || template_language}.erb")
         if extension
           output_file = @output_file.sub(%r{\.[a-z]+$}, ".#{extension}")
         else
           output_file = @output_file
         end
         erb = ERB.new(template, trim_mode: "<>")
-        result = erb.result(binding.clone)
+        # Rust has no #line directive support. For a Rust target the directives
+        # that the grammar embeds around user code blocks are replaced with
+        # comments naming the grammar file and line number the code came from,
+        # so that the origin of a section of user code can still be found by
+        # reading up from a compiler diagnostic pointing into the generated
+        # module.
+        user_code_origin = nil
+        result = erb.result(binding.clone).lines.each_with_index.map do |line, i|
+          if @language == "rust"
+            if md = line.match(/^#line (\d+) "([^"]*)"/)
+              user_code_origin = "#{md[2]} line #{md[1]}"
+              line.sub(/^#line \d+ "[^"]*"/, %[/* Begin user code from #{user_code_origin}. */])
+            elsif line == "#linereset\n"
+              %[/* End user code from #{user_code_origin}. */\n]
+            else
+              line
+            end
+          elsif line == "#linereset\n"
+            %[#line #{i + 2} "#{output_file}"\n]
+          else
+            line
+          end
+        end.join
         File.open(output_file, "wb") do |fh|
           fh.write(result)
         end
@@ -179,7 +204,8 @@ class Propane
                 end
               end
               @grammar.rules << Rule.new(component, [], nil, ptypename, rule.line_number)
-              @grammar.rules << Rule.new(component, [c], "$$ = $1;\n", ptypename, rule.line_number)
+              optcode = @grammar.tree ? nil : "$$ = $1;\n"
+              @grammar.rules << Rule.new(component, [c], optcode, ptypename, rule.line_number)
               optional_rules_added << component
             end
           end
@@ -268,6 +294,8 @@ class Propane
           "context->user_terminate_code = (#{user_terminate_code}); return #{retval};"
         when "d"
           "context.user_terminate_code = (#{user_terminate_code}); return #{retval};"
+        when "rust"
+          "context.user_terminate_code = (#{user_terminate_code}); return #{retval};"
         end
       end
       code = code.gsub(/\$\{context\.(\w+)\}/) do |match|
@@ -277,6 +305,8 @@ class Propane
           "context->#{fieldname}"
         when "d"
           "context.#{fieldname}"
+        when "rust"
+          "context.#{fieldname}"
         end
       end
       code = code.gsub(/\$\{token\.(\w+)\}/) do |match|
@@ -286,35 +316,52 @@ class Propane
           "token_tree_node->#{fieldname}"
         when "d"
           "token_tree_node.#{fieldname}"
+        when "rust"
+          "token_tree_node.#{fieldname}"
         end
       end
       if parser
         code = code.gsub(/\$\$/) do |match|
-          case @language
-          when "c"
-            "_pvalue->v_#{rule.ptypename}"
-          when "d"
-            "_pvalue.v_#{rule.ptypename}"
+          if @grammar.tree
+            typename = "#{@grammar.tree_prefix}#{rule.name}#{@grammar.tree_suffix}"
+            case @language
+            when "c"
+              tree_handle(typename, "_node_id")
+            when "d"
+              tree_handle(typename, "_node_id")
+            when "rust"
+              tree_handle(typename, "_node_id")
+            end
+          else
+            case @language
+            when "c"
+              "_pvalue->v_#{rule.ptypename}"
+            when "d"
+              "_pvalue.v_#{rule.ptypename}"
+            when "rust"
+              "(*_pvalue.v_#{rule.ptypename}_mut())"
+            end
           end
         end
         code = code.gsub(/\$(\d+)/) do |match|
+          parser_component_reference(rule, $1.to_i)
+        end
+        code = code.gsub(/\$\{(\$|\d+)\.position\}/) do |match|
           index = $1.to_i
-          case @language
-          when "c"
-            "state_values_stack_index(statevalues, -1 - (int)n_states + #{index})->pvalue.v_#{rule.components[index - 1].ptypename}"
-          when "d"
-            "statevalues[$-1-n_states+#{index}].pvalue.v_#{rule.components[index - 1].ptypename}"
-          end
+          "get_rule_position(statevalues, #{index}, n_states, false)"
+        end
+        code = code.gsub(/\$\{(\$|\d+)\.end_position\}/) do |match|
+          index = $1.to_i
+          "get_rule_position(statevalues, #{index}, n_states, true)"
         end
         code = code.gsub(/\$\{(\w+)\}/) do |match|
           aliasname = $1
           if index = rule.aliases[aliasname]
-            case @language
-            when "c"
-              "state_values_stack_index(statevalues, -(int)n_states + #{index})->pvalue.v_#{rule.components[index].ptypename}"
-            when "d"
-              "statevalues[$-n_states+#{index}].pvalue.v_#{rule.components[index].ptypename}"
-            end
+            # Field aliases are just a named reference to a positional rule
+            # component, so reuse the same expansion as `$1', `$2', etc. Note
+            # that rule.aliases stores a 0-based component index, so add 1 to
+            # convert it to the 1-based index used for positional references.
+            parser_component_reference(rule, index + 1)
           else
             raise Error.new("Field alias '#{aliasname}' not found")
           end
@@ -327,6 +374,8 @@ class Propane
               "out_token_info->pvalue"
             when "d"
               "out_token_info.pvalue"
+            when "rust"
+              "out_token_info.pvalue"
             end
           else
             case @language
@@ -334,9 +383,31 @@ class Propane
               "out_token_info->pvalue.v_#{pattern.ptypename}"
             when "d"
               "out_token_info.pvalue.v_#{pattern.ptypename}"
+            when "rust"
+              "(*out_token_info.pvalue.v_#{pattern.ptypename}_mut())"
             end
           end
         end
+        code = code.gsub(/\$\{position\}/) do |match|
+          case @language
+          when "c"
+            "out_token_info->position"
+          when "d"
+            "out_token_info.position"
+          when "rust"
+            "out_token_info.position"
+          end
+        end
+        code = code.gsub(/\$\{end_position\}/) do |match|
+          case @language
+          when "c"
+            "out_token_info->end_position"
+          when "d"
+            "out_token_info.end_position"
+          when "rust"
+            "out_token_info.end_position"
+          end
+        end
         code = code.gsub(/\$mode\(([a-zA-Z_][a-zA-Z_0-9]*)\)/) do |match|
           mode_name = $1
           mode_id = @lexer.mode_id(mode_name)
@@ -348,12 +419,411 @@ class Propane
             "context->mode = #{mode_id}u"
           when "d"
             "context.mode = #{mode_id}u"
+          when "rust"
+            "context.mode = #{mode_id}"
           end
         end
       end
       code
     end
 
+    # Expand a positional reference to a parser rule component.
+    #
+    # This is used to expand `$1', `$2', etc. as well as field aliases (which
+    # are just named references to a positional rule component).
+    #
+    # @param rule [Rule]
+    #   The Rule containing the user code.
+    # @param index [Integer]
+    #   1-based index of the rule component to reference.
+    #
+    # @return [String]
+    #   Expanded rule component reference.
+    def parser_component_reference(rule, index)
+      component = rule.components[index - 1]
+      if @grammar.tree
+        # In tree mode a component reference yields a handle to that
+        # component's tree node. An optional component propagates its target
+        # node (or null), so use the optional target's node type.
+        if component.is_a?(RuleSet) && component.optional?
+          component = component.option_target
+        end
+        node_name = component.is_a?(Token) ? "Token" : component.name
+        typename = "#{@grammar.tree_prefix}#{node_name}#{@grammar.tree_suffix}"
+        case @language
+        when "c"
+          tree_handle(typename, "state_values_stack_index(statevalues, -1 - (int)n_states + #{index})->node_id")
+        when "d"
+          tree_handle(typename, "statevalues[$-1-n_states+#{index}].node_id")
+        when "rust"
+          tree_handle(typename, "statevalues[statevalues.len() - 1 - n_states + #{index}].node_id")
+        end
+      else
+        case @language
+        when "c"
+          "state_values_stack_index(statevalues, -1 - (int)n_states + #{index})->pvalue.v_#{component.ptypename}"
+        when "d"
+          "statevalues[$-1-n_states+#{index}].pvalue.v_#{component.ptypename}"
+        when "rust"
+          "statevalues[statevalues.len() - 1 - n_states + #{index}].pvalue.get_v_#{component.ptypename}()"
+        end
+      end
+    end
+
+    # Construct a tree node handle expression for the target language.
+    #
+    # A handle is a small value pairing the parser context with a node ID
+    # (an index into the context's node arena). All handle types share this
+    # layout; the distinct types exist for documentation and, in C, to drive
+    # the tree walk macro's type threading.
+    #
+    # @param typename [String]
+    #   Handle type name.
+    # @param id_expr [String]
+    #   Expression yielding the node ID.
+    # @param parenthesize [Boolean]
+    #   Whether to parenthesize the expression. Parentheses are required where
+    #   the expression is substituted into a user code block, since the
+    #   expression could be followed there by a field access or appear in a
+    #   position where a bare Rust struct literal is not accepted. They are
+    #   unnecessary where the expression stands alone, and Rust warns about
+    #   them there, so this can be disabled for those uses.
+    #
+    # @return [String]
+    #   Handle constructor expression.
+    def tree_handle(typename, id_expr, parenthesize = true)
+      if @cpp
+        "(#{typename}{context, #{id_expr}})"
+      elsif @language == "c"
+        "((#{typename}){context, #{id_expr}})"
+      elsif @language == "rust"
+        expr = "#{typename} { context, id: #{id_expr} }"
+        parenthesize ? "(#{expr})" : expr
+      else
+        "#{typename}(context, #{id_expr})"
+      end
+    end
+
+    # Get the list of non-optional, non-internal rule sets that get a tree node
+    # handle type generated for them.
+    #
+    # @return [Array<Propane::RuleSet>]
+    #   Rule sets with generated tree node handle types.
+    def tree_node_rule_sets
+      @parser.rule_sets.reject do |name, rule_set|
+        name.start_with?("$") || rule_set.optional?
+      end.map {|name, rule_set| rule_set}
+    end
+
+    # Maximum number of chained fields supported by a single C tree walk macro
+    # invocation. Deeper navigation can be expressed by nesting walk calls.
+    C_TREE_WALK_MAX = 16
+
+    # Get the tree node handle type name for a node name.
+    #
+    # @param name [String]
+    #   Rule set name, or "Token".
+    #
+    # @return [String]
+    #   Handle type name.
+    def h_type(name)
+      "#{@grammar.tree_prefix}#{name}#{@grammar.tree_suffix}"
+    end
+
+    # Get the list of all tree node handle type names (Token plus rule sets).
+    #
+    # @return [Array<String>]
+    #   Handle type names.
+    def tree_handle_types
+      [h_type("Token")] + tree_node_rule_sets.map {|rs| h_type(rs.name)}
+    end
+
+    # Enumerate the navigation fields of a rule set's tree node.
+    #
+    # @yield [rtype, field_name, child_type, slot]
+    #   Handle type name, field accessor name, child handle type, and child
+    #   slot index.
+    def each_tree_field(rule_set)
+      rtype = h_type(rule_set.name)
+      rule_set.tree_fields.each_with_index do |fields, slot|
+        fields.each do |field_name, child_type|
+          yield rtype, field_name, child_type, slot
+        end
+      end
+    end
+
+    # Generate the tree node handle type declarations for the header.
+    #
+    # These are emitted before the context structure definition so that a
+    # context_user_fields block can declare a field of a handle type.
+    #
+    # @return [String]
+    #   Handle type declarations.
+    def c_tree_handle_types_header
+      @cpp ? cpp_tree_handle_types_header : c_only_tree_handle_types_header
+    end
+
+    # Generate the remainder of the tree node section for the header.
+    #
+    # This is emitted after the context structure definition since it
+    # dereferences the context and so requires the complete type.
+    #
+    # @return [String]
+    #   Accessors, macros, and out-of-line handle method definitions.
+    def c_tree_types_header
+      @cpp ? cpp_tree_types_header : c_only_tree_types_header
+    end
+
+    # Generate the C (non-C++) tree node handle type section for the header.
+    def c_only_tree_handle_types_header
+      p = @grammar.prefix
+      out = []
+      out << "/** Tree node handle types. @{ */"
+      tree_handle_types.each do |t|
+        out << "typedef struct { #{p}context_t * __context; #{p}node_id_t __id; } #{t};"
+      end
+      out << ""
+      out.join("\n")
+    end
+
+    def c_only_tree_types_header
+      out = []
+      out << c_common_accessors_header
+      out << "/** @} */"
+      out.join("\n")
+    end
+
+    # Generate the C-style (function + macro) tree node accessors shared by the
+    # C and C++ headers. In C++ these are provided in addition to the handle
+    # methods so that C-style code (and the tree walk macros) also works.
+    def c_common_accessors_header
+      p = @grammar.prefix
+      out = []
+      out << "/** Generic tree node accessors (usable on any handle type). */"
+      out << "#define #{p}node_valid(h) ((h).__id != 0u)"
+      out << "#define #{p}node_id(h) ((h).__id)"
+      out << "#define #{p}node_data(h) (&(h).__context->#{p}tree_nodes[(h).__id])"
+      out << "#define #{p}node_position(h) ((h).__context->#{p}tree_nodes[(h).__id].position)"
+      out << "#define #{p}node_end_position(h) ((h).__context->#{p}tree_nodes[(h).__id].end_position)"
+      out << "#define #{p}node_n_fields(h) ((h).__id ? (h).__context->#{p}tree_nodes[(h).__id].n_fields : (uint16_t)0u)"
+      out << ""
+      out << "/** Tree node field accessor functions. */"
+      out << "#{p}token_t #{p}#{h_type("Token")}_token(#{h_type("Token")} node);"
+      out << "#{p}value_t #{p}#{h_type("Token")}_pvalue(#{h_type("Token")} node);"
+      tree_node_rule_sets.each do |rule_set|
+        each_tree_field(rule_set) do |rtype, field_name, child_type, slot|
+          out << "#{child_type} #{p}#{rtype}_#{field_name}(#{rtype} node);"
+        end
+      end
+      out << ""
+      out << c_tree_walk_macros
+      out.join("\n")
+    end
+
+    # Generate the C tree walk macro machinery.
+    def c_tree_walk_macros
+      p = @grammar.prefix
+      max = C_TREE_WALK_MAX
+      out = []
+      out << "/* Tree walk macros: p_tree_walk_<Type>(handle, field, ...). */"
+      out << "#define #{p}CAT_(a, b) a##b"
+      out << "#define #{p}CAT(a, b) #{p}CAT_(a, b)"
+      out << "#define #{p}TA(t, f) #{p}CAT(#{p}CAT(#{p}CAT(#{p}TYPEAFTER_, t), _), f)"
+      out << "#define #{p}ACC(t, f) #{p}CAT(#{p}CAT(#{p}CAT(#{p}, t), _), f)"
+      argn = (1..max).map {|i| "_#{i}"}.join(", ")
+      rseq = (0..max).to_a.reverse.join(", ")
+      out << "#define #{p}ARG_N(#{argn}, N, ...) N"
+      out << "#define #{p}NARG(...) #{p}ARG_N(__VA_ARGS__, #{rseq})"
+      (1..max).each do |n|
+        fparams = (1..n).map {|k| "f#{k}"}.join(", ")
+        call = "h"
+        (1..n).each do |k|
+          texpr = "R"
+          (1...k).each {|j| texpr = "#{p}TA(#{texpr}, f#{j})"}
+          call = "#{p}ACC(#{texpr}, f#{k})(#{call})"
+        end
+        out << "#define #{p}tree_walk_#{n}(R, h, #{fparams}) #{call}"
+      end
+      out << "#define #{p}tree_walk_dispatch(R, h, ...) #{p}CAT(#{p}tree_walk_, #{p}NARG(__VA_ARGS__))(R, h, __VA_ARGS__)"
+      # Type transition map (navigation fields only).
+      tree_node_rule_sets.each do |rule_set|
+        each_tree_field(rule_set) do |rtype, field_name, child_type, slot|
+          out << "#define #{p}TYPEAFTER_#{rtype}_#{field_name} #{child_type}"
+        end
+      end
+      # Per-handle-type walk entry points.
+      tree_handle_types.each do |t|
+        out << "#define #{p}tree_walk_#{t}(...) #{p}tree_walk_dispatch(#{t}, __VA_ARGS__)"
+      end
+      out.join("\n")
+    end
+
+    # Generate the C tree node accessor function definitions for the source.
+    #
+    # @return [String]
+    #   Accessor function definitions.
+    def c_tree_accessor_defs
+      p = @grammar.prefix
+      tt = h_type("Token")
+      out = []
+      out << "#{p}token_t #{p}#{tt}_token(#{tt} node)"
+      out << "{"
+      out << "    return node.__context->#{p}tree_nodes[node.__id].token;"
+      out << "}"
+      out << ""
+      out << "#{p}value_t #{p}#{tt}_pvalue(#{tt} node)"
+      out << "{"
+      out << "    return node.__context->#{p}tree_nodes[node.__id].pvalue;"
+      out << "}"
+      tree_node_rule_sets.each do |rule_set|
+        each_tree_field(rule_set) do |rtype, field_name, child_type, slot|
+          out << ""
+          out << "#{child_type} #{p}#{rtype}_#{field_name}(#{rtype} node)"
+          out << "{"
+          out << "    #{child_type} result;"
+          out << "    result.__context = node.__context;"
+          out << "    if (node.__id == 0u)"
+          out << "    {"
+          out << "        result.__id = 0u;"
+          out << "        return result;"
+          out << "    }"
+          out << "    result.__id = node.__context->#{p}tree_children[node.__context->#{p}tree_nodes[node.__id].child_offset + #{slot}u];"
+          out << "    return result;"
+          out << "}"
+        end
+      end
+      out.join("\n")
+    end
+
+    # Generate the C++ tree node handle class declarations for the header.
+    # Only valid() and node_id() are defined inline; every other method
+    # dereferences the context, which is still an incomplete type here, so
+    # those are declared and defined out of line once the context is
+    # complete.
+    def cpp_tree_handle_types_header
+      p = @grammar.prefix
+      out = []
+      out << "/** Tree node handle types. @{ */"
+      tree_handle_types.each {|t| out << "struct #{t};"}
+      out << ""
+      tt = h_type("Token")
+      out << "struct #{tt}"
+      out << "{"
+      out << "    #{p}context_t * __context;"
+      out << "    #{p}node_id_t __id;"
+      out << "    bool valid() const { return __id != 0u; }"
+      out << "    #{p}node_id_t node_id() const { return __id; }"
+      out << "    #{p}node_data_t * data() const;"
+      out << "    #{p}position_t position() const;"
+      out << "    #{p}position_t end_position() const;"
+      out << "    uint16_t n_fields() const;"
+      out << "    #{p}token_t token() const;"
+      out << "    #{p}value_t pvalue() const;"
+      out << "};"
+      out << ""
+      tree_node_rule_sets.each do |rule_set|
+        rtype = h_type(rule_set.name)
+        out << "struct #{rtype}"
+        out << "{"
+        out << "    #{p}context_t * __context;"
+        out << "    #{p}node_id_t __id;"
+        out << "    bool valid() const { return __id != 0u; }"
+        out << "    #{p}node_id_t node_id() const { return __id; }"
+        out << "    #{p}node_data_t * data() const;"
+        out << "    #{p}position_t position() const;"
+        out << "    #{p}position_t end_position() const;"
+        out << "    uint16_t n_fields() const;"
+        each_tree_field(rule_set) do |rt, field_name, child_type, slot|
+          out << "    #{child_type} #{field_name}() const;"
+        end
+        out << "};"
+        out << ""
+      end
+      out.join("\n")
+    end
+
+    # Generate the out-of-line C++ handle method definitions plus the C-style
+    # accessors. Emitted after the context structure definition.
+    def cpp_tree_types_header
+      p = @grammar.prefix
+      out = []
+      # Common node methods, now that the context type is complete.
+      ([h_type("Token")] + tree_node_rule_sets.map {|rs| h_type(rs.name)}).each do |ht|
+        out << "inline #{p}node_data_t * #{ht}::data() const { return &__context->#{p}tree_nodes[__id]; }"
+        out << "inline #{p}position_t #{ht}::position() const { return __context->#{p}tree_nodes[__id].position; }"
+        out << "inline #{p}position_t #{ht}::end_position() const { return __context->#{p}tree_nodes[__id].end_position; }"
+        out << "inline uint16_t #{ht}::n_fields() const { return __id ? __context->#{p}tree_nodes[__id].n_fields : (uint16_t)0u; }"
+      end
+      tt = h_type("Token")
+      out << "inline #{p}token_t #{tt}::token() const { return __context->#{p}tree_nodes[__id].token; }"
+      out << "inline #{p}value_t #{tt}::pvalue() const { return __context->#{p}tree_nodes[__id].pvalue; }"
+      out << ""
+      # Out-of-line navigation method bodies (all handle types now complete).
+      tree_node_rule_sets.each do |rule_set|
+        rtype = h_type(rule_set.name)
+        each_tree_field(rule_set) do |rt, field_name, child_type, slot|
+          out << "inline #{child_type} #{rtype}::#{field_name}() const"
+          out << "{"
+          out << "    if (__id == 0u)"
+          out << "    {"
+          out << "        return #{child_type}{__context, 0u};"
+          out << "    }"
+          out << "    return #{child_type}{__context, __context->#{p}tree_children[__context->#{p}tree_nodes[__id].child_offset + #{slot}u]};"
+          out << "}"
+        end
+      end
+      out << ""
+      out << "/*"
+      out << " * C-style function and macro accessors, provided in addition to the handle"
+      out << " * methods above so that C-style code and the tree walk macros also work."
+      out << " */"
+      out << c_common_accessors_header
+      out << "/** @} */"
+      out.join("\n")
+    end
+
+    # Rust keywords that must be escaped as raw identifiers when used as a
+    # generated identifier (e.g. a field alias named `type`).
+    RUST_KEYWORDS = %w[
+      as break const continue dyn else enum extern false fn for if impl in let
+      loop match mod move mut pub ref return static struct trait true type
+      unsafe use where while async await abstract become box do final macro
+      override priv typeof unsized virtual yield try gen
+    ]
+
+    # Escape a name as a Rust raw identifier if it is a reserved keyword.
+    #
+    # @param name [String]
+    #   Identifier name.
+    #
+    # @return [String]
+    #   Name, escaped as a raw identifier if necessary.
+    def rust_ident(name)
+      RUST_KEYWORDS.include?(name) ? "r##{name}" : name
+    end
+
+    # Map a ptype type string to a valid Rust type.
+    #
+    # The default ptype is a C "void *"; for Rust with no declared ptype we use
+    # the unit type instead.
+    #
+    # @param typestring [String]
+    #   ptype type string.
+    #
+    # @return [String]
+    #   Rust type string.
+    def rust_ptype(typestring)
+      typestring == "void *" ? "()" : typestring
+    end
+
+    # Get the lex function to use.
+    #
+    # @return [String]
+    #   Lex function to use.
+    def lex_fn
+      @grammar.lex_fn || "#{@grammar.prefix}lex"
+    end
+
     # Get the parser value type for the start rule.
     #
     # @return [Array<String>]
@@ -379,6 +849,8 @@ class Propane
           "uint8_t"
         when "d"
           "ubyte"
+        when "rust"
+          "u8"
         end
       elsif max <= 0xFFFF
         case @language
@@ -386,11 +858,15 @@ class Propane
           "uint16_t"
         when "d"
           "ushort"
+        when "rust"
+          "u16"
         end
       else
         case @language
         when "c"
           "uint32_t"
+        when "rust"
+          "u32"
         else
           "uint"
         end

lib/propane/grammar.rb

@@ -6,6 +6,7 @@ class Propane
     IDENTIFIER_REGEX = /(?:[a-zA-Z]|_[a-zA-Z0-9])[a-zA-Z_0-9]*/
 
     attr_reader :context_user_fields
+    attr_reader :lex_fn
     attr_reader :tree
     attr_reader :tree_prefix
     attr_reader :tree_suffix
@@ -21,7 +22,8 @@ class Propane
     attr_reader :on_token_node
     attr_reader :token_user_fields
 
-    def initialize(input)
+    def initialize(input, filename)
+      @filename = filename
       @patterns = []
       @start_rules = []
       @tokens = []
@@ -56,6 +58,10 @@ class Propane
       @tokens.size + 1
     end
 
+    def parser_user_code_used?
+      @rules.any? {|r| r.code}
+    end
+
     private
 
     def parse_grammar!
@@ -69,6 +75,7 @@ class Propane
       elsif parse_comment_line!
       elsif @modeline.nil? && parse_mode_label!
       elsif parse_context_user_fields_statement!
+      elsif parse_lex_fn!
       elsif parse_tree_statement!
       elsif parse_tree_prefix_statement!
       elsif parse_tree_suffix_statement!
@@ -85,6 +92,7 @@ class Propane
       elsif parse_rule_statement!
       elsif parse_code_block_statement!
       elsif parse_prefix_statement!
+      elsif parse_noline_statement!
       else
         if @input.size > 25
           @input = @input.slice(0..20) + "..."
@@ -117,6 +125,12 @@ class Propane
       end
     end
 
+    def parse_lex_fn!
+      if md = consume!(/lex_fn\b\s*(\w+)\s*;/)
+        @lex_fn = md[1]
+      end
+    end
+
     def parse_tree_statement!
       if consume!(/tree\s*;/)
         @tree = true
@@ -253,12 +267,8 @@ class Propane
         end
         md = consume!(/((?:#{IDENTIFIER_REGEX}\??(?::#{IDENTIFIER_REGEX})?\s*)*)\s*/, "expected rule component list")
         components = md[1].strip.split(/\s+/)
-        if @tree
-          consume!(/;/, "expected `;'")
-        else
-          unless code = parse_code_block!
-            consume!(/;/, "expected `;' or code block")
-          end
+        unless code = parse_code_block!
+          consume!(/;/, "expected `;' or code block")
         end
         @rules << Rule.new(rule_name, components, code, ptypename, @line_number)
         @modeline = nil
@@ -296,8 +306,14 @@ class Propane
     def parse_code_block_statement!
       if md = consume!(/<<([a-z]*)(.*?)>>\n/m)
         name, code = md[1..2]
-        code.sub!(/\A\n/, "")
-        code += "\n" unless code.end_with?("\n")
+        code = code.chomp
+        unless @noline
+          if code.start_with?("\n")
+            code = %[#line #{@line_number + 1} "#{@filename}"#{code}\n#linereset\n]
+          else
+            code = %[#line #{@line_number} "#{@filename}"\n#{code}\n#linereset\n]
+          end
+        end
         if @code_blocks[name]
           @code_blocks[name] += code
         else
@@ -315,6 +331,13 @@ class Propane
       end
     end
 
+    def parse_noline_statement!
+      if md = consume!(/noline\s*;/)
+        @noline = true
+        true
+      end
+    end
+
     def parse_pattern!
       if md = consume!(%r{/})
         pattern = ""
@@ -338,9 +361,14 @@ class Propane
 
     def parse_code_block!
       if md = consume!(/<<(.*?)>>\n/m)
-        code = md[1]
-        code.sub!(/\A\n/, "")
-        code += "\n" unless code.end_with?("\n")
+        code = md[1].chomp
+        unless @noline
+          if code.start_with?("\n")
+            code = %[#line #{@line_number + 1} "#{@filename}"#{code}\n#linereset\n]
+          else
+            code = %[#line #{@line_number} "#{@filename}"\n#{code}\n#linereset\n]
+          end
+        end
         code
       end
     end

lib/propane/version.rb

@@ -1,3 +1,3 @@
 class Propane
-  VERSION = "4.1.0"
+  VERSION = "5.1.0"
 end

spec/json_parser.c.propane

@@ -21,46 +21,46 @@ token number /-?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][-+]?[0-9]+)?/ <<
     double n = 0.0;
     bool negative = false;
     size_t i = 0u;
-    if (match[i] == '-')
+    if (match_text[i] == '-')
     {
         negative = true;
         i++;
     }
-    while ('0' <= match[i] && match[i] <= '9')
+    while ('0' <= match_text[i] && match_text[i] <= '9')
     {
         n *= 10.0;
-        n += (match[i] - '0');
+        n += (match_text[i] - '0');
         i++;
     }
-    if (match[i] == '.')
+    if (match_text[i] == '.')
     {
         i++;
         double mult = 0.1;
-        while ('0' <= match[i] && match[i] <= '9')
+        while ('0' <= match_text[i] && match_text[i] <= '9')
         {
-            n += mult * (match[i] - '0');
+            n += mult * (match_text[i] - '0');
             mult /= 10.0;
             i++;
         }
     }
-    if (match[i] == 'e' || match[i] == 'E')
+    if (match_text[i] == 'e' || match_text[i] == 'E')
     {
         bool exp_negative = false;
         i++;
-        if (match[i] == '-')
+        if (match_text[i] == '-')
         {
             exp_negative = true;
             i++;
         }
-        else if (match[i] == '+')
+        else if (match_text[i] == '+')
         {
             i++;
         }
         long exp = 0.0;
-        while ('0' <= match[i] && match[i] <= '9')
+        while ('0' <= match_text[i] && match_text[i] <= '9')
         {
             exp *= 10;
-            exp += (match[i] - '0');
+            exp += (match_text[i] - '0');
             i++;
         }
         if (exp_negative)
@@ -120,11 +120,11 @@ string: /\\t/ <<
 >>
 string: /\\u[0-9a-fA-F]{4}/ <<
   /* Not actually going to encode the code point for this example... */
-  char s[] = {'{', (char)match[2], (char)match[3], (char)match[4], (char)match[5], '}', 0};
+  char s[] = {'{', (char)match_text[2], (char)match_text[3], (char)match_text[4], (char)match_text[5], '}', 0};
   str_append(&string_value, s);
 >>
 string: /[^\\]/ <<
-  char s[] = {(char)match[0], 0};
+  char s[] = {(char)match_text[0], 0};
   str_append(&string_value, s);
 >>
 Start -> Value <<

spec/json_parser.d.propane

@@ -20,46 +20,46 @@ token number /-?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][-+]?[0-9]+)?/ <<
     double n;
     bool negative;
     size_t i = 0u;
-    if (match[i] == '-')
+    if (match_text[i] == '-')
     {
         negative = true;
         i++;
     }
-    while ('0' <= match[i] && match[i] <= '9')
+    while ('0' <= match_text[i] && match_text[i] <= '9')
     {
         n *= 10.0;
-        n += (match[i] - '0');
+        n += (match_text[i] - '0');
         i++;
     }
-    if (match[i] == '.')
+    if (match_text[i] == '.')
     {
         i++;
         double mult = 0.1;
-        while ('0' <= match[i] && match[i] <= '9')
+        while ('0' <= match_text[i] && match_text[i] <= '9')
         {
-            n += mult * (match[i] - '0');
+            n += mult * (match_text[i] - '0');
             mult /= 10.0;
             i++;
         }
     }
-    if (match[i] == 'e' || match[i] == 'E')
+    if (match_text[i] == 'e' || match_text[i] == 'E')
     {
         bool exp_negative;
         i++;
-        if (match[i] == '-')
+        if (match_text[i] == '-')
         {
             exp_negative = true;
             i++;
         }
-        else if (match[i] == '+')
+        else if (match_text[i] == '+')
         {
             i++;
         }
         long exp;
-        while ('0' <= match[i] && match[i] <= '9')
+        while ('0' <= match_text[i] && match_text[i] <= '9')
         {
             exp *= 10;
-            exp += (match[i] - '0');
+            exp += (match_text[i] - '0');
             i++;
         }
         if (exp_negative)
@@ -117,10 +117,10 @@ string: /\\t/ <<
 >>
 string: /\\u[0-9a-fA-F]{4}/ <<
   /* Not actually going to encode the code point for this example... */
-  string_value ~= "{" ~ match[2..6] ~ "}";
+  string_value ~= "{" ~ match_text[2..6] ~ "}";
 >>
 string: /[^\\]/ <<
-  string_value ~= match;
+  string_value ~= match_text;
 >>
 Start -> Value <<
   $$ = $1;

spec/json_parser.rust.propane

@@ -0,0 +1,176 @@
+<<
+pub const JSON_OBJECT: usize = 0;
+pub const JSON_ARRAY: usize = 1;
+pub const JSON_NUMBER: usize = 2;
+pub const JSON_STRING: usize = 3;
+pub const JSON_TRUE: usize = 4;
+pub const JSON_FALSE: usize = 5;
+pub const JSON_NULL: usize = 6;
+
+#[derive(Clone, Default)]
+pub enum JSONValue {
+    #[default]
+    Null,
+    Object(Vec<(String, JSONValue)>),
+    Array(Vec<JSONValue>),
+    Number(f64),
+    StringVal(String),
+    True,
+    False,
+}
+
+impl JSONValue {
+    pub fn id(&self) -> usize {
+        match self {
+            JSONValue::Object(_) => JSON_OBJECT,
+            JSONValue::Array(_) => JSON_ARRAY,
+            JSONValue::Number(_) => JSON_NUMBER,
+            JSONValue::StringVal(_) => JSON_STRING,
+            JSONValue::True => JSON_TRUE,
+            JSONValue::False => JSON_FALSE,
+            JSONValue::Null => JSON_NULL,
+        }
+    }
+    pub fn number(&self) -> f64 {
+        if let JSONValue::Number(n) = self { *n } else { 0.0 }
+    }
+    pub fn string(&self) -> &str {
+        if let JSONValue::StringVal(s) = self { s.as_str() } else { "" }
+    }
+    pub fn object_len(&self) -> usize {
+        if let JSONValue::Object(e) = self { e.len() } else { 0 }
+    }
+    pub fn array_len(&self) -> usize {
+        if let JSONValue::Array(e) = self { e.len() } else { 0 }
+    }
+}
+>>
+
+context_user_fields <<
+    pub string_value: String,
+>>
+
+ptype JSONValue;
+
+drop /\s+/;
+token lbrace /\{/;
+token rbrace /\}/;
+token lbracket /\[/;
+token rbracket /\]/;
+token comma /,/;
+token colon /:/;
+token number /-?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][-+]?[0-9]+)?/ <<
+    let n: f64 = std::str::from_utf8(match_text).unwrap().parse().unwrap();
+    $$ = JSONValue::Number(n);
+>>
+token true <<
+  $$ = JSONValue::True;
+>>
+token false <<
+  $$ = JSONValue::False;
+>>
+token null <<
+  $$ = JSONValue::Null;
+>>
+/"/ <<
+  $mode(string);
+  ${context.string_value} = String::new();
+>>
+string: token string /"/ <<
+  $$ = JSONValue::StringVal(std::mem::take(&mut ${context.string_value}));
+  $mode(default);
+>>
+string: /\\"/ <<
+  ${context.string_value}.push('"');
+>>
+string: /\\\\/ <<
+  ${context.string_value}.push('\\');
+>>
+string: /\\\// <<
+  ${context.string_value}.push('/');
+>>
+string: /\\b/ <<
+  ${context.string_value}.push('\u{0008}');
+>>
+string: /\\f/ <<
+  ${context.string_value}.push('\u{000C}');
+>>
+string: /\\n/ <<
+  ${context.string_value}.push('\n');
+>>
+string: /\\r/ <<
+  ${context.string_value}.push('\r');
+>>
+string: /\\t/ <<
+  ${context.string_value}.push('\t');
+>>
+string: /\\u[0-9a-fA-F]{4}/ <<
+  /* Not actually going to encode the code point for this example... */
+  let s: String = ['{', match_text[2] as char, match_text[3] as char, match_text[4] as char, match_text[5] as char, '}'].iter().collect();
+  ${context.string_value}.push_str(&s);
+>>
+string: /[^\\]/ <<
+  ${context.string_value}.push(match_text[0] as char);
+>>
+Start -> Value <<
+  $$ = $1;
+>>
+Value -> string <<
+  $$ = $1;
+>>
+Value -> number <<
+  $$ = $1;
+>>
+Value -> Object <<
+  $$ = $1;
+>>
+Value -> Array <<
+  $$ = $1;
+>>
+Value -> true <<
+  $$ = $1;
+>>
+Value -> false <<
+  $$ = $1;
+>>
+Value -> null <<
+  $$ = $1;
+>>
+Object -> lbrace rbrace <<
+  $$ = JSONValue::Object(Vec::new());
+>>
+Object -> lbrace KeyValues rbrace <<
+  $$ = $2;
+>>
+KeyValues -> KeyValue <<
+  $$ = $1;
+>>
+KeyValues -> KeyValues comma KeyValue <<
+  let mut obj = $1;
+  if let JSONValue::Object(kve) = $3 {
+      if let JSONValue::Object(entries) = &mut obj {
+          entries.extend(kve);
+      }
+  }
+  $$ = obj;
+>>
+KeyValue -> string colon Value <<
+  let name = if let JSONValue::StringVal(s) = $1 { s } else { String::new() };
+  $$ = JSONValue::Object(vec![(name, $3)]);
+>>
+Array -> lbracket rbracket <<
+  $$ = JSONValue::Array(Vec::new());
+>>
+Array -> lbracket Values rbracket <<
+  $$ = $2;
+>>
+Values -> Value <<
+  $$ = $1;
+>>
+Values -> Values comma Value <<
+  let mut arr = $1;
+  if let JSONValue::Array(elems) = &mut arr {
+      elems.push($3);
+  }
+  $$ = arr;
+>>

spec/macros.c.propane

@@ -0,0 +1,25 @@
+<<
+#include <stdlib.h>
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info);
+void record(int v);
+>>
+
+ptype int;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lbrace /\{/;
+token rbrace /\}/;
+token plus /\+/;
+token macro;
+token macroname /@[a-zA-Z_]\w*/;
+token num /\d+/ << char b[100]; memcpy(b, match_text, match_length); b[match_length] = '\0'; $$ = atoi(b); >>
+
+Start -> Statements;
+Statements -> ;
+Statements -> Statement Statements;
+Statement -> Add;
+Statement -> MacroStart;
+Add -> num plus num << $$ = $1 + $3; record($$); >>
+MacroStart -> macro macroname lbrace;

spec/macros.d.propane

@@ -0,0 +1,31 @@
+<<
+import test_macros;
+>>
+
+ptype int;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lbrace /\{/;
+token rbrace /\}/;
+token plus /\+/;
+token macro;
+token macroname /@[a-zA-Z_]\w*/;
+token num /\d+/ <<
+    int n = 0;
+    foreach (c; match_text)
+    {
+        n *= 10;
+        n += (c - '0');
+    }
+    $$ = n;
+>>
+
+Start -> Statements;
+Statements -> ;
+Statements -> Statement Statements;
+Statement -> Add;
+Statement -> MacroStart;
+Add -> num plus num << $$ = $1 + $3; record($$); >>
+MacroStart -> macro macroname lbrace;

spec/macros.rust.propane

@@ -0,0 +1,80 @@
+<<
+fn mylexfn(context: &mut p_context_t, out_token_info: &mut p_token_info_t) -> usize {
+    loop {
+        if context.expanding {
+            let ei = context.expand_i;
+            context.expand_i += 1;
+            if context.expand_i >= context.token_infos.len() {
+                context.expanding = false;
+            }
+            *out_token_info = context.token_infos[ei].clone();
+            return P_SUCCESS;
+        }
+
+        let lex_result = p_lex(context, out_token_info);
+        if lex_result != P_SUCCESS {
+            return lex_result;
+        }
+
+        if out_token_info.token == TOKEN_macro {
+            context.defining = true;
+        } else if out_token_info.token == TOKEN_macroname {
+            if !context.defining {
+                context.expanding = true;
+                context.expand_i = 0;
+                continue;
+            }
+        } else if out_token_info.token == TOKEN_lbrace {
+            if context.defining {
+                /* Capture the macro body tokens (up to the closing '}'). */
+                let mut infos: Vec<p_token_info_t> = Vec::new();
+                loop {
+                    let mut ti = p_token_info_t::default();
+                    assert_eq!(P_SUCCESS, p_lex(context, &mut ti));
+                    if ti.token == TOKEN_rbrace {
+                        break;
+                    }
+                    infos.push(ti);
+                }
+                context.token_infos = infos;
+                context.defining = false;
+            }
+        } else {
+            context.defining = false;
+        }
+        return lex_result;
+    }
+}
+>>
+
+context_user_fields <<
+    pub defining: bool,
+    pub expanding: bool,
+    pub expand_i: usize,
+    pub token_infos: Vec<p_token_info_t>,
+    pub nums: Vec<i64>,
+>>
+
+ptype i64;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lbrace /\{/;
+token rbrace /\}/;
+token plus /\+/;
+token macro;
+token macroname /@[a-zA-Z_]\w*/;
+token num /\d+/ <<
+  let mut v: i64 = 0;
+  for c in match_text { v = v * 10 + (*c - b'0') as i64; }
+  $$ = v;
+>>
+
+Start -> Statements;
+Statements -> ;
+Statements -> Statement Statements;
+Statement -> Add;
+Statement -> MacroStart;
+Add -> num plus num << $$ = $1 + $3; ${context.nums}.push($$); >>
+MacroStart -> macro macroname lbrace;

spec/parse_inner_nested.c.propane

@@ -0,0 +1,19 @@
+<<
+#include <stdlib.h>
+#include <string.h>
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info);
+>>
+
+ptype int;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lparen /\(/;
+token rparen /\)/;
+token plus /\+/;
+token num /\d+/ << char b[32]; memcpy(b, match_text, match_length); b[match_length] = '\0'; $$ = atoi(b); >>
+
+Start -> Expr << $$ = $1; >>
+Expr -> num << $$ = $1; >>
+Expr -> Expr plus num << $$ = $1 + $3; >>

spec/parse_inner_nested.d.propane

@@ -0,0 +1,17 @@
+<<
+import test_parse_inner_nested;
+>>
+
+ptype int;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lparen /\(/;
+token rparen /\)/;
+token plus /\+/;
+token num /\d+/ << int n = 0; foreach (ch; match_text) { n *= 10; n += (ch - '0'); } $$ = n; >>
+
+Start -> Expr << $$ = $1; >>
+Expr -> num << $$ = $1; >>
+Expr -> Expr plus num << $$ = $1 + $3; >>

spec/parse_inner_nested.rust.propane

@@ -0,0 +1,41 @@
+<<
+fn mylexfn(context: &mut p_context_t, out_token_info: &mut p_token_info_t) -> usize {
+    let result = p_lex(context, out_token_info);
+    if result != P_SUCCESS {
+        return result;
+    }
+    if out_token_info.token == TOKEN_lparen {
+        /* Reentrant nested parse of the parenthesized sub-expression. */
+        let inner_result = p_parse_inner_Start(context, &[TOKEN_rparen]);
+        if inner_result != P_SUCCESS {
+            return inner_result;
+        }
+        let value = p_result_Start(context);
+        /* p_parse_inner rewound the input so ')' was not consumed; consume it. */
+        let mut rparen_info = p_token_info_t::default();
+        assert_eq!(P_SUCCESS, p_lex(context, &mut rparen_info));
+        assert_eq!(TOKEN_rparen, rparen_info.token);
+        out_token_info.token = TOKEN_num;
+        out_token_info.pvalue = p_value(value);
+    }
+    P_SUCCESS
+}
+>>
+
+ptype i64;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lparen /\(/;
+token rparen /\)/;
+token plus /\+/;
+token num /\d+/ <<
+  let mut v: i64 = 0;
+  for c in match_text { v = v * 10 + (*c - b'0') as i64; }
+  $$ = v;
+>>
+
+Start -> Expr << $$ = $1; >>
+Expr -> num << $$ = $1; >>
+Expr -> Expr plus num << $$ = $1 + $3; >>

spec/parse_inner_nested_tree.c.propane

@@ -0,0 +1,17 @@
+<<
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info);
+>>
+
+tree;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lparen /\(/;
+token rparen /\)/;
+token plus /\+/;
+token num /\d+/;
+
+Start -> Expr;
+Expr -> num;
+Expr -> Expr plus num;

spec/parse_inner_nested_tree.d.propane

@@ -0,0 +1,17 @@
+<<
+import test_parse_inner_nested_tree;
+>>
+
+tree;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lparen /\(/;
+token rparen /\)/;
+token plus /\+/;
+token num /\d+/;
+
+Start -> Expr;
+Expr -> num;
+Expr -> Expr plus num;

spec/parse_inner_nested_tree.rust.propane

@@ -0,0 +1,44 @@
+<<
+fn mylexfn(context: &mut p_context_t, out_token_info: &mut p_token_info_t) -> usize {
+    let result = p_lex(context, out_token_info);
+    if result != P_SUCCESS {
+        return result;
+    }
+    if out_token_info.token == TOKEN_lparen {
+        let start_position = out_token_info.position;
+        let inner_result = p_parse_inner_Start(context, &[TOKEN_rparen]);
+        if inner_result != P_SUCCESS {
+            return inner_result;
+        }
+        /* Read the inner subtree's span before re-borrowing context to lex. */
+        let inner = p_result_Start(context);
+        assert!(inner.valid());
+        let inner_start_col = inner.position().col;
+        let inner_end_col = inner.end_position().col;
+        let mut rparen_info = p_token_info_t::default();
+        assert_eq!(P_SUCCESS, p_lex(context, &mut rparen_info));
+        assert_eq!(TOKEN_rparen, rparen_info.token);
+        assert_eq!(start_position.col + 1, inner_start_col);
+        assert_eq!(rparen_info.position.col - 1, inner_end_col);
+        /* Synthesize a num token spanning the whole "( ... )" group. */
+        out_token_info.token = TOKEN_num;
+        out_token_info.position = start_position;
+        out_token_info.end_position = rparen_info.end_position;
+    }
+    P_SUCCESS
+}
+>>
+
+tree;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token lparen /\(/;
+token rparen /\)/;
+token plus /\+/;
+token num /\d+/;
+
+Start -> Expr;
+Expr -> num;
+Expr -> Expr plus num;

spec/propane/grammar_spec.rb

@@ -28,7 +28,7 @@ B -> <<
   b = 0;
 >>
 EOF
-      grammar = Grammar.new(input)
+      grammar = Grammar.new(input, "test.propane")
       expect(grammar.modulename).to eq "a.b"
       expect(grammar.ptype).to eq "XYZ  *"
       expect(grammar.ptypes).to eq("default" => "XYZ  *")
@@ -62,7 +62,7 @@ EOF
       expect(o).to_not be_nil
       expect(o.pattern).to eq "token_with_code"
       expect(o.line_number).to eq 11
-      expect(o.code).to eq "Code for the token\n"
+      expect(o.code).to eq %[#line 12 "test.propane"\nCode for the token\n#linereset\n]
 
       o = grammar.tokens.find {|token| token.name == "token_with_no_pattern"}
       expect(o).to_not be_nil
@@ -83,7 +83,7 @@ EOF
       expect(o.name).to eq "A"
       expect(o.components).to eq %w[B]
       expect(o.line_number).to eq 19
-      expect(o.code).to eq "  a = 42;\n"
+      expect(o.code).to eq %[#line 20 "test.propane"\n  a = 42;\n#linereset\n]
 
       o = grammar.rules[1]
       expect(o.name).to eq "B"
@@ -95,7 +95,7 @@ EOF
       expect(o.name).to eq "B"
       expect(o.components).to eq []
       expect(o.line_number).to eq 23
-      expect(o.code).to eq "  b = 0;\n"
+      expect(o.code).to eq %[#line 24 "test.propane"\n  b = 0;\n#linereset\n]
     end
 
     it "parses code segments with semicolons" do
@@ -113,7 +113,7 @@ tokenid token_with_no_pattern;
 
 prefix myparser_;
 EOF
-      grammar = Grammar.new(input)
+      grammar = Grammar.new(input, "test.propane")
       expect(grammar.prefix).to eq "myparser_"
 
       o = grammar.tokens.find {|token| token.name == "code1"}
@@ -122,7 +122,7 @@ EOF
 
       o = grammar.patterns.find {|pattern| pattern.token == o}
       expect(o).to_not be_nil
-      expect(o.code).to eq "  a = b;\n  return c;\n"
+      expect(o.code).to eq %[#line 2 "test.propane"\n  a = b;\n  return c;\n#linereset\n]
 
       o = grammar.tokens.find {|token| token.name == "code2"}
       expect(o).to_not be_nil
@@ -130,7 +130,42 @@ EOF
 
       o = grammar.patterns.find {|pattern| pattern.token == o}
       expect(o).to_not be_nil
-      expect(o.code).to eq %[  writeln("Hello there");\n]
+      expect(o.code).to eq %[#line 7 "test.propane"\n  writeln("Hello there");\n#linereset\n]
+    end
+
+    it "does not emit #line directives with noline statement" do
+      input = <<EOF
+noline;
+
+token code1 <<
+  a = b;
+  return c;
+>>
+
+token code2 <<
+  writeln("Hello there");
+>>
+
+tokenid token_with_no_pattern;
+
+prefix myparser_;
+EOF
+      grammar = Grammar.new(input, "test.propane")
+      expect(grammar.prefix).to eq "myparser_"
+
+      o = grammar.tokens.find {|token| token.name == "code1"}
+      expect(o).to_not be_nil
+
+      o = grammar.patterns.find {|pattern| pattern.token == o}
+      expect(o).to_not be_nil
+      expect(o.code).to eq %[\n  a = b;\n  return c;]
+
+      o = grammar.tokens.find {|token| token.name == "code2"}
+      expect(o).to_not be_nil
+
+      o = grammar.patterns.find {|pattern| pattern.token == o}
+      expect(o).to_not be_nil
+      expect(o.code).to eq %[\n  writeln("Hello there");]
     end
 
     it "supports mode labels" do
@@ -144,7 +179,7 @@ m2: /bar/ <<
 drop /q/;
 m3: drop /r/;
 EOF
-      grammar = Grammar.new(input)
+      grammar = Grammar.new(input, "test.propane")
 
       o = grammar.tokens.find {|token| token.name == "a"}
       expect(o).to_not be_nil
@@ -197,7 +232,7 @@ tokenid int(integer);
 Start (node) -> R;
 R -> abc int;
 EOF
-      grammar = Grammar.new(input)
+      grammar = Grammar.new(input, "test.propane")
 
       o = grammar.tokens.find {|token| token.name == "abc"}
       expect(o).to_not be_nil

spec/propane/lexer/dfa_spec.rb

@@ -51,7 +51,7 @@ class TestLexer
 end
 
 def run(grammar, input)
-  grammar = Propane::Grammar.new(grammar)
+  grammar = Propane::Grammar.new(grammar, "test.propane")
   token_dfa = Propane::Lexer::DFA.new(grammar.patterns)
   test_lexer = TestLexer.new(token_dfa)
   test_lexer.lex(input)

spec/rewind.c.propane

@@ -0,0 +1,23 @@
+<<
+#include <stdlib.h>
+#include <string.h>
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info);
+void record(int value);
+>>
+
+ptype int;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token repeat /repeat/;
+token lbrace /\{/;
+token rbrace /\}/;
+token plus /\+/;
+token num /\d+/ << char b[32]; memcpy(b, match_text, match_length); b[match_length] = '\0'; $$ = atoi(b); >>
+
+Start -> Statements;
+Statements -> ;
+Statements -> Statement Statements;
+Statement -> Add;
+Add -> num plus num << record($1 + $3); >>

spec/rewind.d.propane

@@ -0,0 +1,20 @@
+<<
+import test_rewind;
+>>
+
+ptype int;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token repeat /repeat/;
+token lbrace /\{/;
+token rbrace /\}/;
+token plus /\+/;
+token num /\d+/ << int n = 0; foreach (ch; match_text) { n *= 10; n += (ch - '0'); } $$ = n; >>
+
+Start -> Statements;
+Statements -> ;
+Statements -> Statement Statements;
+Statement -> Add;
+Add -> num plus num << record($1 + $3); >>

spec/rewind.rust.propane

@@ -0,0 +1,67 @@
+<<
+fn mylexfn(context: &mut p_context_t, out_token_info: &mut p_token_info_t) -> usize {
+    loop {
+        let result = p_lex(context, out_token_info);
+        if result != P_SUCCESS {
+            return result;
+        }
+        if out_token_info.token == TOKEN_repeat {
+            let mut count_info = p_token_info_t::default();
+            assert_eq!(P_SUCCESS, p_lex(context, &mut count_info));
+            assert_eq!(TOKEN_num, count_info.token);
+            let mut brace_info = p_token_info_t::default();
+            assert_eq!(P_SUCCESS, p_lex(context, &mut brace_info));
+            assert_eq!(TOKEN_lbrace, brace_info.token);
+            context.remaining = p_value_get(&count_info.pvalue);
+            context.body_index = p_input_index(context);
+            context.body_position = p_position(context);
+            continue;
+        }
+        if out_token_info.token == TOKEN_rbrace {
+            if context.remaining > 1 {
+                context.remaining -= 1;
+                let bi = context.body_index;
+                let bp = context.body_position;
+                p_set_input_index(context, bi);
+                p_set_position(context, bp);
+                continue;
+            }
+            context.remaining = 0;
+            continue;
+        }
+        if out_token_info.token == TOKEN_num {
+            context.num_cols.push(out_token_info.position.col);
+        }
+        return result;
+    }
+}
+>>
+
+context_user_fields <<
+    pub nums: Vec<i64>,
+    pub num_cols: Vec<u32>,
+    pub remaining: i64,
+    pub body_index: usize,
+    pub body_position: p_position_t,
+>>
+
+ptype i64;
+
+lex_fn mylexfn;
+
+drop /\s+/;
+token repeat /repeat/;
+token lbrace /\{/;
+token rbrace /\}/;
+token plus /\+/;
+token num /\d+/ <<
+  let mut v: i64 = 0;
+  for c in match_text { v = v * 10 + (*c - b'0') as i64; }
+  $$ = v;
+>>
+
+Start -> Statements;
+Statements -> ;
+Statements -> Statement Statements;
+Statement -> Add;
+Add -> num plus num << ${context.nums}.push($1 + $3); >>

spec/spec_helper.rb

@@ -15,6 +15,10 @@ unless ENV["dist_specs"]
       command_name "RSpec"
     end
     project_name "Propane"
+    # Keep this process's results separate from the propane subprocess results
+    # so that nothing has to merge on the fly; the spec Rake task collates all
+    # of the parts once the suite is done.
+    coverage_dir "coverage/parts/rspec"
     merge_timeout 3600
     formatter(MyFormatter)
   end

spec/test_basic_math_grammar.rs

@@ -0,0 +1,16 @@
+use testparser::*;
+
+fn main() {
+    let cases: [(&[u8], u64); 4] = [
+        (b"1 + 2 * 3 + 4", 11),
+        (b"1 * 2 ** 4 * 3", 48),
+        (b"(1 + 2) * 3 + 4", 13),
+        (b"(2 * 2) ** 3 + 4 + 5", 73),
+    ];
+    for (input, expected) in cases {
+        let mut context = p_context_new(input);
+        assert_eq!(P_SUCCESS, p_parse(&mut context));
+        assert_eq!(expected, p_result(&context));
+        p_context_delete(context);
+    }
+}

spec/test_context_field_handle.c

@@ -0,0 +1,39 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    char input[128];
+    size_t i;
+    p_context_t * context;
+    Token token;
+
+    /* Enough tokens that the tree node arena is reallocated during the parse. */
+    memset(input, 0, sizeof(input));
+    for (i = 0u; i < 40u; i++)
+    {
+        input[i] = 'a';
+    }
+
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert_eq(P_SUCCESS, p_parse(context));
+
+    /* The handle was stored in a context user field during the parse, before
+     * the remaining nodes were created. It still refers to the same node. */
+    assert_eq(1u, context->have_first);
+    assert(p_node_valid(context->first_item));
+    token = p_Item_pToken1(context->first_item);
+    assert(p_node_valid(token));
+    assert_eq(TOKEN_a, p_Token_token(token));
+    assert_eq(7u, p_Token_pvalue(token));
+
+    /* The stored handle refers to the first Item, which starts at column 1. */
+    assert_eq(1u, p_node_position(context->first_item).row);
+    assert_eq(1u, p_node_position(context->first_item).col);
+
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_context_field_handle.cpp

@@ -0,0 +1,36 @@
+#include "testparser.h"
+#include <cassert>
+#include <cstring>
+#include "testutils.h"
+
+int main()
+{
+    char input[128];
+
+    /* Enough tokens that the tree node arena is reallocated during the parse. */
+    memset(input, 0, sizeof(input));
+    for (size_t i = 0u; i < 40u; i++)
+    {
+        input[i] = 'a';
+    }
+
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert_eq(P_SUCCESS, p_parse(context));
+
+    /* The handle was stored in a context user field during the parse, before
+     * the remaining nodes were created. It still refers to the same node. */
+    assert_eq(1u, context->have_first);
+    assert(context->first_item.valid());
+    Token token = context->first_item.pToken1();
+    assert(token.valid());
+    assert_eq(TOKEN_a, token.token());
+    assert_eq(7u, token.pvalue());
+
+    /* The stored handle refers to the first Item, which starts at column 1. */
+    assert_eq(1u, context->first_item.position().row);
+    assert_eq(1u, context->first_item.position().col);
+
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_context_field_handle.d

@@ -0,0 +1,35 @@
+import testparser;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* Enough tokens that the tree node array is reallocated during the parse. */
+    string input;
+    foreach (i; 0 .. 40)
+    {
+        input ~= "a";
+    }
+
+    p_context_t * context = p_context_new(input);
+    assert_eq(P_SUCCESS, p_parse(context));
+
+    /* The handle was stored in a context user field during the parse, before
+     * the remaining nodes were created. It still refers to the same node. */
+    assert_eq(1, context.have_first);
+    assert(context.first_item.valid);
+    Token token = context.first_item.pToken1;
+    assert(token.valid);
+    assert_eq(TOKEN_a, token.token);
+    assert_eq(7, token.pvalue);
+
+    /* The stored handle refers to the first Item, which starts at column 1. */
+    assert_eq(1u, context.first_item.position.row);
+    assert_eq(1u, context.first_item.position.col);
+
+    p_context_delete(context);
+}

spec/test_custom_lex_fn.c

@@ -0,0 +1,15 @@
+#include "testparser.h"
+#include "testutils.h"
+#include <string.h>
+
+int main()
+{
+    char const * input = "cbacba";
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert_eq(P_SUCCESS, p_parse(context));
+    size_t result = p_result(context);
+    assert_eq(0x932187932187, result);
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_custom_lex_fn.d

@@ -0,0 +1,18 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    string input = "cbacba";
+    p_context_t * context = p_context_new(input);
+    assert_eq(P_SUCCESS, p_parse(context));
+    size_t result = p_result(context);
+    assert_eq(0x932187932187, result);
+    p_context_delete(context);
+}

spec/test_custom_lex_fn.rs

@@ -0,0 +1,8 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"cbacba");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    assert_eq!(0x932187932187, p_result(&c));
+    p_context_delete(c);
+}

spec/test_drop_code_block.rs

@@ -0,0 +1,7 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"    # comment 1\n#    comment 2\na\n");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+}

spec/test_error_positions.rs

@@ -0,0 +1,35 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"a 42");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"a\n123\na  a");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse(&mut c));
+    assert_eq!(3, p_position(&c).row);
+    assert_eq!(4, p_position(&c).col);
+    assert_eq!(TOKEN_a, p_token(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"12");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse(&mut c));
+    assert_eq!(1, p_position(&c).row);
+    assert_eq!(1, p_position(&c).col);
+    assert_eq!(TOKEN_num, p_token(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"a 12\n\nab");
+    assert_eq!(P_UNEXPECTED_INPUT, p_parse(&mut c));
+    assert_eq!(3, p_position(&c).row);
+    assert_eq!(2, p_position(&c).col);
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"a 12\n\na\n\n77\na   \xAA");
+    assert_eq!(P_DECODE_ERROR, p_parse(&mut c));
+    assert_eq!(6, p_position(&c).row);
+    assert_eq!(5, p_position(&c).col);
+    assert_eq!("a", p_token_names[TOKEN_a as usize]);
+    assert_eq!("num", p_token_names[TOKEN_num as usize]);
+    p_context_delete(c);
+}

spec/test_field_aliases.rs

@@ -0,0 +1,7 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"foo1\nbar2");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+}

spec/test_input_index.c

@@ -0,0 +1,60 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    /* Grammar (simple):
+     *   drop /\\s+/;
+     *   token a; token b;
+     *   Start -> a b;
+     *
+     * Verifies that p_input_index() reports the parser/lexer's current byte
+     * offset into the input text. */
+
+    /* Fresh context: input_index starts at 0. */
+    {
+        char const * input = "ab";
+        p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+        assert_eq(0u, p_input_index(context));
+        p_context_delete(context);
+    }
+
+    /* After each successful lex the byte offset advances past the token. */
+    {
+        char const * input = "a b";
+        p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_info_t token_info;
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert_eq((size_t)TOKEN_a, (size_t)token_info.token);
+        assert_eq(1u, p_input_index(context));
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert_eq((size_t)TOKEN_b, (size_t)token_info.token);
+        /* The dropped space between `a` and `b` advances input_index too. */
+        assert_eq(3u, p_input_index(context));
+        p_context_delete(context);
+    }
+
+    /* After a full successful parse, input_index has reached the end. */
+    {
+        char const * input = "ab";
+        p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+        assert(p_parse_Start(context) == P_SUCCESS);
+        assert_eq(2u, p_input_index(context));
+        p_context_delete(context);
+    }
+
+    /* When parse_inner completes via a follow token, the follow token is not
+     * consumed, so input_index points at the start of the follow token. */
+    {
+        char const * input = "abb";
+        p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_Start(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(2u, p_input_index(context));
+        p_context_delete(context);
+    }
+
+    return 0;
+}

spec/test_input_index.d

@@ -0,0 +1,51 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* See test_input_index.c for details on the grammar and cases. */
+
+    /* Fresh context: input_index starts at 0. */
+    {
+        string input = "ab";
+        p_context_t * context = p_context_new(input);
+        assert(p_input_index(context) == 0);
+    }
+
+    /* After each successful lex the byte offset advances past the token. */
+    {
+        string input = "a b";
+        p_context_t * context = p_context_new(input);
+        p_token_info_t token_info;
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert(token_info.token == TOKEN_a);
+        assert(p_input_index(context) == 1);
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert(token_info.token == TOKEN_b);
+        assert(p_input_index(context) == 3);
+    }
+
+    /* After a full successful parse, input_index has reached the end. */
+    {
+        string input = "ab";
+        p_context_t * context = p_context_new(input);
+        assert(p_parse_Start(context) == P_SUCCESS);
+        assert(p_input_index(context) == 2);
+    }
+
+    /* When parse_inner completes via a follow token, the follow token is not
+     * consumed, so input_index points at the start of the follow token. */
+    {
+        string input = "abb";
+        p_context_t * context = p_context_new(input);
+        p_token_t[] follow_tokens = [TOKEN_b];
+        assert(p_parse_inner_Start(context, follow_tokens) == P_SUCCESS);
+        assert(p_input_index(context) == 2);
+    }
+}

spec/test_input_index.rs

@@ -0,0 +1,28 @@
+use testparser::*;
+
+fn main() {
+    let c = p_context_new(b"ab");
+    assert_eq!(0, p_input_index(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"a b");
+    let mut ti = p_token_info_t::default();
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_a, ti.token);
+    assert_eq!(1, p_input_index(&c));
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_b, ti.token);
+    assert_eq!(3, p_input_index(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse_Start(&mut c));
+    assert_eq!(2, p_input_index(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"abb");
+    let follow = [TOKEN_b];
+    assert_eq!(P_SUCCESS, p_parse_inner_Start(&mut c, &follow));
+    assert_eq!(2, p_input_index(&c));
+    p_context_delete(c);
+}

spec/test_lexer.rs

@@ -0,0 +1,49 @@
+use testparser::*;
+
+fn chk(ti: &p_token_info_t, row: u32, col: u32, erow: u32, ecol: u32, len: usize, token: p_token_t) {
+    assert_eq!(row, ti.position.row);
+    assert_eq!(col, ti.position.col);
+    assert_eq!(erow, ti.end_position.row);
+    assert_eq!(ecol, ti.end_position.col);
+    assert_eq!(len, ti.length);
+    assert_eq!(token, ti.token);
+}
+
+fn main() {
+    let mut cp: p_code_point_t = 0;
+    let mut cpl: u8 = 0;
+
+    assert_eq!(P_SUCCESS, p_decode_code_point(b"5", &mut cp, &mut cpl));
+    assert_eq!('5' as u32, cp);
+    assert_eq!(1, cpl);
+
+    assert_eq!(P_EOF, p_decode_code_point(b"", &mut cp, &mut cpl));
+
+    assert_eq!(P_SUCCESS, p_decode_code_point(b"\xC2\xA9", &mut cp, &mut cpl));
+    assert_eq!(0xA9, cp);
+    assert_eq!(2, cpl);
+
+    assert_eq!(P_SUCCESS, p_decode_code_point(b"\xf0\x9f\xa7\xa1", &mut cp, &mut cpl));
+    assert_eq!(0x1F9E1, cp);
+    assert_eq!(4, cpl);
+
+    assert_eq!(P_DECODE_ERROR, p_decode_code_point(b"\xf0\x9f\x27", &mut cp, &mut cpl));
+    assert_eq!(P_DECODE_ERROR, p_decode_code_point(b"\xf0\x9f\xa7\xFF", &mut cp, &mut cpl));
+    assert_eq!(P_DECODE_ERROR, p_decode_code_point(b"\xfe", &mut cp, &mut cpl));
+
+    let mut context = p_context_new(b"5 + 4 * \n677 + 567");
+    let mut ti = p_token_info_t::default();
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 1, 1, 1, 1, 1, TOKEN_int);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 1, 3, 1, 3, 1, TOKEN_plus);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 1, 5, 1, 5, 1, TOKEN_int);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 1, 7, 1, 7, 1, TOKEN_times);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 2, 1, 2, 3, 3, TOKEN_int);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 2, 5, 2, 5, 1, TOKEN_plus);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 2, 7, 2, 9, 3, TOKEN_int);
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 2, 10, 2, 10, 0, TOKEN___EOF);
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"");
+    assert_eq!(P_SUCCESS, p_lex(&mut context, &mut ti)); chk(&ti, 1, 1, 1, 1, 0, TOKEN___EOF);
+    p_context_delete(context);
+}

spec/test_lexer_match_text.rs

@@ -0,0 +1,8 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"identifier_123");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass1");
+    p_context_delete(context);
+}

spec/test_lexer_modes.rs

@@ -0,0 +1,13 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"abc \"a string\" def");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass1");
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"abc \"abc def\" def");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass2");
+    p_context_delete(context);
+}

spec/test_lexer_multiple_modes.rs

@@ -0,0 +1,13 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"abc.def");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass1");
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"abc .  abc");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass2");
+    p_context_delete(context);
+}

spec/test_lexer_positions.c

@@ -0,0 +1,50 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+
+int main()
+{
+    char const * input = "abc\n  defg hi\n!";
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    p_token_info_t token_info;
+
+    /* First token "abc" on row 1, cols 1-3. */
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_word);
+    assert(context->last_start.row == 1u);
+    assert(context->last_start.col == 1u);
+    assert(context->last_end.row == 1u);
+    assert(context->last_end.col == 3u);
+    /* The lexer code block observed the same positions reported to the caller. */
+    assert(context->last_start.row == token_info.position.row);
+    assert(context->last_start.col == token_info.position.col);
+    assert(context->last_end.row == token_info.end_position.row);
+    assert(context->last_end.col == token_info.end_position.col);
+
+    /* Second token "defg" on row 2, cols 3-6. */
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_word);
+    assert(context->last_start.row == 2u);
+    assert(context->last_start.col == 3u);
+    assert(context->last_end.row == 2u);
+    assert(context->last_end.col == 6u);
+
+    /* Third token "hi" on row 2, cols 8-9. */
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_word);
+    assert(context->last_start.row == 2u);
+    assert(context->last_start.col == 8u);
+    assert(context->last_end.row == 2u);
+    assert(context->last_end.col == 9u);
+
+    /* The "!" stop token terminates the lexer. The context input text position
+     * must not be updated when the lexer user code requests termination, so it
+     * still points at the "!" token on row 3, col 1. */
+    assert(p_lex(context, &token_info) == P_USER_TERMINATED);
+    assert(p_user_terminate_code(context) == 42u);
+    assert(context->text_position.row == 3u);
+    assert(context->text_position.col == 1u);
+
+    p_context_delete(context);
+    return 0;
+}

spec/test_lexer_positions.d

@@ -0,0 +1,42 @@
+import testparser;
+import std.stdio;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    string input = "abc\n  defg hi\n!";
+    p_context_t * context = p_context_new(input);
+    p_token_info_t token_info;
+
+    /* First token "abc" on row 1, cols 1-3. */
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_word);
+    assert(context.last_start == p_position_t(1, 1));
+    assert(context.last_end == p_position_t(1, 3));
+    /* The lexer code block observed the same positions reported to the caller. */
+    assert(context.last_start == token_info.position);
+    assert(context.last_end == token_info.end_position);
+
+    /* Second token "defg" on row 2, cols 3-6. */
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_word);
+    assert(context.last_start == p_position_t(2, 3));
+    assert(context.last_end == p_position_t(2, 6));
+
+    /* Third token "hi" on row 2, cols 8-9. */
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_word);
+    assert(context.last_start == p_position_t(2, 8));
+    assert(context.last_end == p_position_t(2, 9));
+
+    /* The "!" stop token terminates the lexer. The context input text position
+     * must not be updated when the lexer user code requests termination, so it
+     * still points at the "!" token on row 3, col 1. */
+    assert(p_lex(context, &token_info) == P_USER_TERMINATED);
+    assert(p_user_terminate_code(context) == 42u);
+    assert(context.text_position == p_position_t(3, 1));
+}

spec/test_lexer_positions.rs

@@ -0,0 +1,38 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"abc\n  defg hi\n!");
+    let mut ti = p_token_info_t::default();
+
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_word, ti.token);
+    assert_eq!(1, c.last_start.row);
+    assert_eq!(1, c.last_start.col);
+    assert_eq!(1, c.last_end.row);
+    assert_eq!(3, c.last_end.col);
+    assert_eq!(c.last_start.row, ti.position.row);
+    assert_eq!(c.last_start.col, ti.position.col);
+    assert_eq!(c.last_end.row, ti.end_position.row);
+    assert_eq!(c.last_end.col, ti.end_position.col);
+
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_word, ti.token);
+    assert_eq!(2, c.last_start.row);
+    assert_eq!(3, c.last_start.col);
+    assert_eq!(2, c.last_end.row);
+    assert_eq!(6, c.last_end.col);
+
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_word, ti.token);
+    assert_eq!(2, c.last_start.row);
+    assert_eq!(8, c.last_start.col);
+    assert_eq!(2, c.last_end.row);
+    assert_eq!(9, c.last_end.col);
+
+    assert_eq!(P_USER_TERMINATED, p_lex(&mut c, &mut ti));
+    assert_eq!(42, p_user_terminate_code(&c));
+    assert_eq!(3, p_position(&c).row);
+    assert_eq!(1, p_position(&c).col);
+
+    p_context_delete(c);
+}

spec/test_lexer_result_value.rs

@@ -0,0 +1,13 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"x");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    assert_eq!(1, p_result(&context));
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"fabulous");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    assert_eq!(8, p_result(&context));
+    p_context_delete(context);
+}

spec/test_lexer_unknown_character.rs

@@ -0,0 +1,12 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"x");
+    assert_eq!(P_UNEXPECTED_INPUT, p_parse(&mut context));
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"123");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    assert_eq!(123, p_result(&context));
+    p_context_delete(context);
+}

spec/test_macros.c

@@ -0,0 +1,118 @@
+#include "testparser.h"
+#include "testutils.h"
+#include <string.h>
+#include <assert.h>
+#include <stddef.h>
+#include <stdbool.h>
+
+static p_context_t * context;
+size_t n_tokens;
+p_token_info_t token_infos[10];
+
+/* Capture the macro body tokens (everything up to the closing '}') into
+ * token_infos[]. Called from mylexfn() right after the definition's '{' has
+ * been lexed, so the input cursor is positioned at the first body token. */
+static void capture_macro_body(void)
+{
+    n_tokens = 0u;
+    for (;;)
+    {
+        size_t result = p_lex(context, &token_infos[n_tokens]);
+        assert_eq(result, P_SUCCESS);
+        if (token_infos[n_tokens].token == TOKEN_rbrace)
+        {
+            break;
+        }
+        n_tokens++;
+        assert(n_tokens < sizeof(token_infos) / sizeof(token_infos[0]));
+    }
+}
+
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info)
+{
+    static bool defining;
+    static bool expanding;
+    static size_t expand_i;
+
+    for (;;)
+    {
+        if (expanding)
+        {
+            size_t ei = expand_i++;
+            if (expand_i >= n_tokens)
+            {
+                expanding = false;
+            }
+            *out_token_info = token_infos[ei];
+            return P_SUCCESS;
+        }
+
+        size_t lex_result = p_lex(context, out_token_info);
+        if (lex_result != P_SUCCESS)
+        {
+            return lex_result;
+        }
+
+        switch (out_token_info->token)
+        {
+        case TOKEN_macro:
+            /* Start of a macro definition: "macro macroname { ... }". */
+            defining = true;
+            break;
+        case TOKEN_macroname:
+            if (!defining)
+            {
+                /* Use of a macro: replay its captured body tokens instead of
+                 * returning the macroname to the parser. */
+                expanding = true;
+                expand_i = 0u;
+                continue;
+            }
+            /* Definition name: pass through and keep waiting for '{'. */
+            break;
+        case TOKEN_lbrace:
+            if (defining)
+            {
+                /* Consume and store the macro body now, before the parser gets
+                 * a chance to read its lookahead token (which would otherwise
+                 * swallow the first body token). */
+                capture_macro_body();
+                defining = false;
+            }
+            break;
+        default:
+            defining = false;
+            break;
+        }
+        return lex_result;
+    }
+}
+
+size_t n_nums;
+int nums[10];
+
+void record(int v)
+{
+    nums[n_nums++] = v;
+}
+
+int main()
+{
+    char const * input =
+        "macro @m { 23 + 200 }\n"
+        "66 + 100\n"
+        "@m\n"
+        "33 + 55\n"
+        "@m\n";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    assert_eq(n_nums, 4);
+    assert_eq(nums[0], 166);
+    assert_eq(nums[1], 223);
+    assert_eq(nums[2], 88);
+    assert_eq(nums[3], 223);
+
+    return 0;
+}

spec/test_macros.d

@@ -0,0 +1,116 @@
+import testparser;
+import testutils;
+
+size_t n_tokens;
+p_token_info_t[10] token_infos;
+
+// Capture the macro body tokens (everything up to the closing '}') into
+// token_infos[]. Called from mylexfn() right after the definition's '{' has
+// been lexed, so the input cursor is positioned at the first body token.
+void capture_macro_body(p_context_t * context)
+{
+    n_tokens = 0u;
+    for (;;)
+    {
+        size_t result = p_lex(context, &token_infos[n_tokens]);
+        assert(result == P_SUCCESS);
+        if (token_infos[n_tokens].token == TOKEN_rbrace)
+        {
+            break;
+        }
+        n_tokens++;
+        assert(n_tokens < token_infos.length);
+    }
+}
+
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info)
+{
+    static bool defining;
+    static bool expanding;
+    static size_t expand_i;
+
+    for (;;)
+    {
+        if (expanding)
+        {
+            size_t ei = expand_i++;
+            if (expand_i >= n_tokens)
+            {
+                expanding = false;
+            }
+            *out_token_info = token_infos[ei];
+            return P_SUCCESS;
+        }
+
+        size_t lex_result = p_lex(context, out_token_info);
+        if (lex_result != P_SUCCESS)
+        {
+            return lex_result;
+        }
+
+        switch (out_token_info.token)
+        {
+        case TOKEN_macro:
+            // Start of a macro definition: "macro macroname { ... }".
+            defining = true;
+            break;
+        case TOKEN_macroname:
+            if (!defining)
+            {
+                // Use of a macro: replay its captured body tokens instead of
+                // returning the macroname to the parser.
+                expanding = true;
+                expand_i = 0u;
+                continue;
+            }
+            // Definition name: pass through and keep waiting for '{'.
+            break;
+        case TOKEN_lbrace:
+            if (defining)
+            {
+                // Consume and store the macro body now, before the parser gets
+                // a chance to read its lookahead token (which would otherwise
+                // swallow the first body token).
+                capture_macro_body(context);
+                defining = false;
+            }
+            break;
+        default:
+            defining = false;
+            break;
+        }
+        return lex_result;
+    }
+}
+
+size_t n_nums;
+int[10] nums;
+
+void record(int v)
+{
+    nums[n_nums++] = v;
+}
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    string input =
+        "macro @m { 23 + 200 }\n" ~
+        "66 + 100\n" ~
+        "@m\n" ~
+        "33 + 55\n" ~
+        "@m\n";
+    p_context_t * context = p_context_new(input);
+    assert(p_parse(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    assert(n_nums == 4);
+    assert(nums[0] == 166);
+    assert(nums[1] == 223);
+    assert(nums[2] == 88);
+    assert(nums[3] == 223);
+}

spec/test_macros.rs

@@ -0,0 +1,9 @@
+use testparser::*;
+
+fn main() {
+    let input = b"macro @m { 23 + 200 }\n66 + 100\n@m\n33 + 55\n@m\n";
+    let mut c = p_context_new(input);
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    assert_eq!(vec![166, 223, 88, 223], c.nums);
+    p_context_delete(c);
+}

spec/test_match_backslashes.rs

@@ -0,0 +1,7 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"\x07\x08\t\n\x0b\x0c\rt");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+}

spec/test_multiple_parsers.rs

@@ -0,0 +1,12 @@
+use testparsermyp1 as m1;
+use testparsermyp2 as m2;
+
+fn main() {
+    let mut context1 = m1::myp1_context_new(b"a\n1");
+    assert_eq!(m1::MYP1_SUCCESS, m1::myp1_parse(&mut context1));
+    m1::myp1_context_delete(context1);
+
+    let mut context2 = m2::myp2_context_new(b"bcb");
+    assert_eq!(m2::MYP2_SUCCESS, m2::myp2_parse(&mut context2));
+    m2::myp2_context_delete(context2);
+}

spec/test_named_optional_rule_component_tree.c

@@ -9,46 +9,42 @@ int main()
     p_context_t * context;
     context = p_context_new((uint8_t const *)input, strlen(input));
     assert(p_parse(context) == P_SUCCESS);
-    Start * start = p_result(context);
-    assert(start->a == NULL);
-    assert(start->pToken2 != NULL);
-    assert_eq(TOKEN_b, start->pToken2->token);
-    assert(start->pR3 == NULL);
-    assert(start->pR == NULL);
-    assert(start->r == NULL);
+    Start start = p_result(context);
+    assert(!p_node_valid(p_Start_a(start)));
+    assert(p_node_valid(p_Start_pToken2(start)));
+    assert_eq(TOKEN_b, p_tree_walk_Start(start, pToken2, token));
+    assert(!p_node_valid(p_Start_pR3(start)));
+    assert(!p_node_valid(p_Start_pR(start)));
+    assert(!p_node_valid(p_Start_r(start)));
 
-    p_tree_delete(start);
     p_context_delete(context);
 
     input = "abcd";
     context = p_context_new((uint8_t const *)input, strlen(input));
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start->a != NULL);
-    assert_eq(TOKEN_a, start->pToken1->token);
-    assert(start->pToken2 != NULL);
-    assert(start->pR3 != NULL);
-    assert(start->pR != NULL);
-    assert(start->r != NULL);
-    assert(start->pR == start->pR3);
-    assert(start->pR == start->r);
-    assert_eq(TOKEN_c, start->pR->pToken1->token);
+    assert(p_node_valid(p_Start_a(start)));
+    assert_eq(TOKEN_a, p_tree_walk_Start(start, pToken1, token));
+    assert(p_node_valid(p_Start_pToken2(start)));
+    assert(p_node_valid(p_Start_pR3(start)));
+    assert(p_node_valid(p_Start_pR(start)));
+    assert(p_node_valid(p_Start_r(start)));
+    assert(p_node_id(p_Start_pR(start)) == p_node_id(p_Start_pR3(start)));
+    assert(p_node_id(p_Start_pR(start)) == p_node_id(p_Start_r(start)));
+    assert_eq(TOKEN_c, p_tree_walk_Start(start, pR, pToken1, token));
 
-    p_tree_delete(start);
     p_context_delete(context);
 
     input = "bdc";
     context = p_context_new((uint8_t const *)input, strlen(input));
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start->a == NULL);
-    assert(start->pToken2 != NULL);
-    assert(start->r != NULL);
-    assert_eq(TOKEN_d, start->pR->pToken1->token);
+    assert(!p_node_valid(p_Start_a(start)));
+    assert(p_node_valid(p_Start_pToken2(start)));
+    assert(p_node_valid(p_Start_r(start)));
+    assert_eq(TOKEN_d, p_tree_walk_Start(start, pR, pToken1, token));
 
-    p_tree_delete(start);
     p_context_delete(context);
 
     return 0;
 }
-

spec/test_named_optional_rule_component_tree.d

@@ -12,40 +12,40 @@ unittest
     string input = "b";
     p_context_t * context = p_context_new(input);
     assert(p_parse(context) == P_SUCCESS);
-    Start * start = p_result(context);
-    assert(start.pToken1 is null);
-    assert(start.pToken2 !is null);
+    Start start = p_result(context);
+    assert(!start.pToken1.valid);
+    assert(start.pToken2.valid);
     assert_eq(TOKEN_b, start.pToken2.token);
-    assert(start.pR3 is null);
-    assert(start.pR is null);
-    assert(start.r is null);
+    assert(!start.pR3.valid);
+    assert(!start.pR.valid);
+    assert(!start.r.valid);
 
-    p_tree_delete(start);
+    p_context_delete(context);
 
     input = "abcd";
     context = p_context_new(input);
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start.pToken1 != null);
+    assert(start.pToken1.valid);
     assert_eq(TOKEN_a, start.pToken1.token);
-    assert(start.pToken2 != null);
-    assert(start.pR3 != null);
-    assert(start.pR != null);
-    assert(start.r != null);
+    assert(start.pToken2.valid);
+    assert(start.pR3.valid);
+    assert(start.pR.valid);
+    assert(start.r.valid);
     assert(start.pR == start.pR3);
     assert(start.pR == start.r);
     assert_eq(TOKEN_c, start.pR.pToken1.token);
 
-    p_tree_delete(start);
+    p_context_delete(context);
 
     input = "bdc";
     context = p_context_new(input);
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start.pToken1 is null);
-    assert(start.pToken2 !is null);
-    assert(start.pR !is null);
+    assert(!start.pToken1.valid);
+    assert(start.pToken2.valid);
+    assert(start.pR.valid);
     assert_eq(TOKEN_d, start.pR.pToken1.token);
 
-    p_tree_delete(start);
+    p_context_delete(context);
 }

spec/test_named_optional_rule_component_tree.rs

@@ -0,0 +1,43 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"b");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    {
+        let start = p_result(&context);
+        assert!(!start.a().valid());
+        assert!(start.pToken2().valid());
+        assert_eq!(TOKEN_b, start.pToken2().token());
+        assert!(!start.pR3().valid());
+        assert!(!start.pR().valid());
+        assert!(!start.r().valid());
+    }
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"abcd");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    {
+        let start = p_result(&context);
+        assert!(start.a().valid());
+        assert_eq!(TOKEN_a, start.pToken1().token());
+        assert!(start.pToken2().valid());
+        assert!(start.pR3().valid());
+        assert!(start.pR().valid());
+        assert!(start.r().valid());
+        assert_eq!(start.pR().node_id(), start.pR3().node_id());
+        assert_eq!(start.pR().node_id(), start.r().node_id());
+        assert_eq!(TOKEN_c, start.pR().pToken1().token());
+    }
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"bdc");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    {
+        let start = p_result(&context);
+        assert!(!start.a().valid());
+        assert!(start.pToken2().valid());
+        assert!(start.r().valid());
+        assert_eq!(TOKEN_d, start.pR().pToken1().token());
+    }
+    p_context_delete(context);
+}

spec/test_optional_rule_component.rs

@@ -0,0 +1,9 @@
+use testparser::*;
+
+fn main() {
+    for input in [&b"b"[..], &b"abcd"[..], &b"abdc"[..]] {
+        let mut context = p_context_new(input);
+        assert_eq!(P_SUCCESS, p_parse(&mut context));
+        p_context_delete(context);
+    }
+}

spec/test_optional_rule_component_tree.c

@@ -9,43 +9,39 @@ int main()
     p_context_t * context;
     context = p_context_new((uint8_t const *)input, strlen(input));
     assert(p_parse(context) == P_SUCCESS);
-    Start * start = p_result(context);
-    assert(start->pToken1 == NULL);
-    assert(start->pToken2 != NULL);
-    assert_eq(TOKEN_b, start->pToken2->token);
-    assert(start->pR3 == NULL);
-    assert(start->pR == NULL);
+    Start start = p_result(context);
+    assert(!p_node_valid(p_Start_pToken1(start)));
+    assert(p_node_valid(p_Start_pToken2(start)));
+    assert_eq(TOKEN_b, p_tree_walk_Start(start, pToken2, token));
+    assert(!p_node_valid(p_Start_pR3(start)));
+    assert(!p_node_valid(p_Start_pR(start)));
 
-    p_tree_delete(start);
     p_context_delete(context);
 
     input = "abcd";
     context = p_context_new((uint8_t const *)input, strlen(input));
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start->pToken1 != NULL);
-    assert_eq(TOKEN_a, start->pToken1->token);
-    assert(start->pToken2 != NULL);
-    assert(start->pR3 != NULL);
-    assert(start->pR != NULL);
-    assert(start->pR == start->pR3);
-    assert_eq(TOKEN_c, start->pR->pToken1->token);
+    assert(p_node_valid(p_Start_pToken1(start)));
+    assert_eq(TOKEN_a, p_tree_walk_Start(start, pToken1, token));
+    assert(p_node_valid(p_Start_pToken2(start)));
+    assert(p_node_valid(p_Start_pR3(start)));
+    assert(p_node_valid(p_Start_pR(start)));
+    assert(p_node_id(p_Start_pR(start)) == p_node_id(p_Start_pR3(start)));
+    assert_eq(TOKEN_c, p_tree_walk_Start(start, pR, pToken1, token));
 
-    p_tree_delete(start);
     p_context_delete(context);
 
     input = "bdc";
     context = p_context_new((uint8_t const *)input, strlen(input));
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start->pToken1 == NULL);
-    assert(start->pToken2 != NULL);
-    assert(start->pR != NULL);
-    assert_eq(TOKEN_d, start->pR->pToken1->token);
+    assert(!p_node_valid(p_Start_pToken1(start)));
+    assert(p_node_valid(p_Start_pToken2(start)));
+    assert(p_node_valid(p_Start_pR(start)));
+    assert_eq(TOKEN_d, p_tree_walk_Start(start, pR, pToken1, token));
 
-    p_tree_delete(start);
     p_context_delete(context);
 
     return 0;
 }
-

spec/test_optional_rule_component_tree.d

@@ -12,37 +12,38 @@ unittest
     string input = "b";
     p_context_t * context = p_context_new(input);
     assert(p_parse(context) == P_SUCCESS);
-    Start * start = p_result(context);
-    assert(start.pToken1 is null);
-    assert(start.pToken2 !is null);
+    Start start = p_result(context);
+    assert(!start.pToken1.valid);
+    assert(start.pToken2.valid);
     assert_eq(TOKEN_b, start.pToken2.token);
-    assert(start.pR3 is null);
-    assert(start.pR is null);
+    assert(!start.pR3.valid);
+    assert(!start.pR.valid);
 
-    p_tree_delete(start);
+    p_context_delete(context);
 
     input = "abcd";
     context = p_context_new(input);
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start.pToken1 != null);
+    assert(start.pToken1.valid);
     assert_eq(TOKEN_a, start.pToken1.token);
-    assert(start.pToken2 != null);
-    assert(start.pR3 != null);
-    assert(start.pR != null);
+    assert(start.pToken2.valid);
+    assert(start.pR3.valid);
+    assert(start.pR.valid);
     assert(start.pR == start.pR3);
+    assert_eq(start.pR.node_id, start.pR3.node_id);
     assert_eq(TOKEN_c, start.pR.pToken1.token);
 
-    p_tree_delete(start);
+    p_context_delete(context);
 
     input = "bdc";
     context = p_context_new(input);
     assert(p_parse(context) == P_SUCCESS);
     start = p_result(context);
-    assert(start.pToken1 is null);
-    assert(start.pToken2 !is null);
-    assert(start.pR !is null);
+    assert(!start.pToken1.valid);
+    assert(start.pToken2.valid);
+    assert(start.pR.valid);
     assert_eq(TOKEN_d, start.pR.pToken1.token);
 
-    p_tree_delete(start);
+    p_context_delete(context);
 }

spec/test_optional_rule_component_tree.rs

@@ -0,0 +1,40 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"b");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    {
+        let start = p_result(&context);
+        assert!(!start.pToken1().valid());
+        assert!(start.pToken2().valid());
+        assert_eq!(TOKEN_b, start.pToken2().token());
+        assert!(!start.pR3().valid());
+        assert!(!start.pR().valid());
+    }
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"abcd");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    {
+        let start = p_result(&context);
+        assert!(start.pToken1().valid());
+        assert_eq!(TOKEN_a, start.pToken1().token());
+        assert!(start.pToken2().valid());
+        assert!(start.pR3().valid());
+        assert!(start.pR().valid());
+        assert_eq!(start.pR().node_id(), start.pR3().node_id());
+        assert_eq!(TOKEN_c, start.pR().pToken1().token());
+    }
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"bdc");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    {
+        let start = p_result(&context);
+        assert!(!start.pToken1().valid());
+        assert!(start.pToken2().valid());
+        assert!(start.pR().valid());
+        assert_eq!(TOKEN_d, start.pR().pToken1().token());
+    }
+    p_context_delete(context);
+}

spec/test_parse_inner.c

@@ -0,0 +1,73 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    /* Grammar (chain reduce):
+     *   Start -> Y      << $$ = $1; >>
+     *   Y     -> a      << $$ = $1; >>
+     *   token a         << $$ = 1; >>
+     *
+     * The reduce lookahead for both `Y -> a` and `Start -> Y` is only $EOF,
+     * so `p_parse_Start("ab")` fails at token `b`. p_parse_inner_Start with
+     * `b` as a follow token should succeed via the reduce-side retry chain
+     * (Y then Start) followed by the shift-side retry hitting $EOF at the
+     * final state. */
+
+    /* Standard parse succeeds on complete input. */
+    char const * input = "a";
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_SUCCESS);
+    assert_eq(1u, (size_t)p_result_Start(context));
+    p_context_delete(context);
+
+    /* Standard parse fails when there's an unexpected trailing token. */
+    input = "ab";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_UNEXPECTED_TOKEN);
+    p_context_delete(context);
+
+    /* parse_inner succeeds via a chain of reduce retries (Y, then Start),
+     * followed by the shift-side retry hitting $EOF at the final state. */
+    {
+        input = "ab";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_Start(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(1u, (size_t)p_result_Start(context));
+        p_context_delete(context);
+    }
+
+    /* parse_inner with an empty (NULL) follow-token vector behaves like a
+     * standard parse. */
+    input = "ab";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_inner_Start(context, NULL, 0u) == P_UNEXPECTED_TOKEN);
+    p_context_delete(context);
+
+    /* parse_inner behaves like a standard parse when the input matches the
+     * grammar fully. */
+    input = "a";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    {
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_Start(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(1u, (size_t)p_result_Start(context));
+    }
+    p_context_delete(context);
+
+    /* parse_inner with a non-matching follow token still fails. The grammar
+     * can't consume `b` and it isn't listed as a follow token, so the retries
+     * do not fire. */
+    {
+        input = "ab";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN___EOF };
+        assert(p_parse_inner_Start(context, follow_tokens, 1u) == P_UNEXPECTED_TOKEN);
+        p_context_delete(context);
+    }
+
+    return 0;
+}

spec/test_parse_inner.d

@@ -0,0 +1,51 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* See test_parse_inner.c for details on the grammar and cases. */
+
+    /* Standard parse succeeds on complete input. */
+    string input = "a";
+    p_context_t * context = p_context_new(input);
+    assert(p_parse_Start(context) == P_SUCCESS);
+    assert(p_result_Start(context) == 1);
+
+    /* Standard parse fails when there's an unexpected trailing token. */
+    input = "ab";
+    context = p_context_new(input);
+    assert(p_parse_Start(context) == P_UNEXPECTED_TOKEN);
+
+    /* parse_inner succeeds via a chain of reduce retries (Y, then Start),
+     * followed by the shift-side retry hitting $EOF at the final state. */
+    input = "ab";
+    context = p_context_new(input);
+    p_token_t[] follow_tokens_b = [TOKEN_b];
+    assert(p_parse_inner_Start(context, follow_tokens_b) == P_SUCCESS);
+    assert(p_result_Start(context) == 1);
+
+    /* parse_inner with a null follow-token slice behaves like a standard
+     * parse. */
+    input = "ab";
+    context = p_context_new(input);
+    assert(p_parse_inner_Start(context, null) == P_UNEXPECTED_TOKEN);
+
+    /* parse_inner behaves like a standard parse when the input matches the
+     * grammar fully. */
+    input = "a";
+    context = p_context_new(input);
+    assert(p_parse_inner_Start(context, follow_tokens_b) == P_SUCCESS);
+    assert(p_result_Start(context) == 1);
+
+    /* parse_inner with a non-matching follow token still fails. */
+    input = "ab";
+    context = p_context_new(input);
+    p_token_t[] follow_tokens_eof = [TOKEN___EOF];
+    assert(p_parse_inner_Start(context, follow_tokens_eof) == P_UNEXPECTED_TOKEN);
+}

spec/test_parse_inner.rs

@@ -0,0 +1,30 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"a");
+    assert_eq!(P_SUCCESS, p_parse_Start(&mut c));
+    assert_eq!(1, p_result_Start(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse_Start(&mut c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse_inner_Start(&mut c, &[TOKEN_b]));
+    assert_eq!(1, p_result_Start(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse_inner_Start(&mut c, &[]));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"a");
+    assert_eq!(P_SUCCESS, p_parse_inner_Start(&mut c, &[TOKEN_b]));
+    assert_eq!(1, p_result_Start(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse_inner_Start(&mut c, &[TOKEN___EOF]));
+    p_context_delete(c);
+}

spec/test_parse_inner_nested.c

@@ -0,0 +1,78 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+/* Grammar (integer evaluator; parentheses handled by the lex function):
+ *   ptype int;
+ *   lex_fn mylexfn;
+ *   token lparen /\(/; token rparen /\)/; token plus /\+/;
+ *   token num /\d+/          << ... atoi ... >>
+ *   Start -> Expr            << $$ = $1; >>
+ *   Expr  -> num             << $$ = $1; >>
+ *   Expr  -> Expr plus num   << $$ = $1 + $3; >>
+ *
+ * The tokens lparen and rparen appear in no grammar rule. Instead, when the
+ * lex function lexes a '(', it performs a nested parse (p_parse_inner_Start)
+ * of the parenthesized sub-expression -- reentrantly, while the outer parse is
+ * still suspended in this callback -- reads the computed value with
+ * p_result_Start, consumes the ')' that p_parse_inner deliberately left in the
+ * input, and hands a single synthesized num token carrying that value back to
+ * the outer parse. Nested groups recurse this process to arbitrary depth. */
+
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info)
+{
+    size_t result = p_lex(context, out_token_info);
+    if (result != P_SUCCESS)
+    {
+        return result;
+    }
+    if (out_token_info->token == TOKEN_lparen)
+    {
+        /* Nested parse of the parenthesized sub-expression, stopping at the
+         * closing ')' follow token. This re-enters the parser while the outer
+         * parse is suspended in this lex callback. */
+        p_token_t follow_tokens[] = { TOKEN_rparen };
+        size_t inner_result = p_parse_inner_Start(context, follow_tokens, 1u);
+        if (inner_result != P_SUCCESS)
+        {
+            return inner_result;
+        }
+        int value = p_result_Start(context);
+        /* p_parse_inner rewound the input so that ')' was not consumed; consume
+         * it now. */
+        p_token_info_t rparen_info;
+        size_t rparen_result = p_lex(context, &rparen_info);
+        assert(rparen_result == P_SUCCESS);
+        assert(rparen_info.token == TOKEN_rparen);
+        /* Replace the '(' token with a synthesized num carrying the nested
+         * parse result. */
+        out_token_info->token = TOKEN_num;
+        out_token_info->pvalue = p_value(value);
+    }
+    return P_SUCCESS;
+}
+
+static int eval(char const * input)
+{
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse(context) == P_SUCCESS);
+    int value = p_result(context);
+    p_context_delete(context);
+    return value;
+}
+
+int main()
+{
+    /* No parentheses: plain outer parse. */
+    assert_eq(5u, (size_t)eval("2 + 3"));
+    /* A single group evaluated by the nested parse. */
+    assert_eq(3u, (size_t)eval("(1 + 2)"));
+    /* A group in the middle of an outer expression. */
+    assert_eq(14u, (size_t)eval("2 + (3 + 4) + 5"));
+    /* Nested groups: the nested parse re-enters itself. */
+    assert_eq(37u, (size_t)eval("2 + (10 + (20 + 5))"));
+    assert_eq(15u, (size_t)eval("(1 + 2) + (3 + (4 + 5))"));
+
+    return 0;
+}

spec/test_parse_inner_nested.d

@@ -0,0 +1,64 @@
+import testparser;
+import testutils;
+
+/* Grammar: see test_parse_inner_nested.c. */
+
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info)
+{
+    size_t result = p_lex(context, out_token_info);
+    if (result != P_SUCCESS)
+    {
+        return result;
+    }
+    if (out_token_info.token == TOKEN_lparen)
+    {
+        /* Nested parse of the parenthesized sub-expression, stopping at the
+         * closing ')' follow token. This re-enters the parser while the outer
+         * parse is suspended in this lex callback. */
+        p_token_t[] follow_tokens = [TOKEN_rparen];
+        size_t inner_result = p_parse_inner_Start(context, follow_tokens);
+        if (inner_result != P_SUCCESS)
+        {
+            return inner_result;
+        }
+        int value = p_result_Start(context);
+        /* p_parse_inner rewound the input so that ')' was not consumed; consume
+         * it now. */
+        p_token_info_t rparen_info;
+        size_t rparen_result = p_lex(context, &rparen_info);
+        assert(rparen_result == P_SUCCESS);
+        assert(rparen_info.token == TOKEN_rparen);
+        /* Replace the '(' token with a synthesized num carrying the nested
+         * parse result. */
+        out_token_info.token = TOKEN_num;
+        out_token_info.pvalue = p_value(value);
+    }
+    return P_SUCCESS;
+}
+
+int eval(string input)
+{
+    p_context_t * context = p_context_new(input);
+    assert(p_parse(context) == P_SUCCESS);
+    int value = p_result(context);
+    p_context_delete(context);
+    return value;
+}
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* No parentheses: plain outer parse. */
+    assert_eq(5, eval("2 + 3"));
+    /* A single group evaluated by the nested parse. */
+    assert_eq(3, eval("(1 + 2)"));
+    /* A group in the middle of an outer expression. */
+    assert_eq(14, eval("2 + (3 + 4) + 5"));
+    /* Nested groups: the nested parse re-enters itself. */
+    assert_eq(37, eval("2 + (10 + (20 + 5))"));
+    assert_eq(15, eval("(1 + 2) + (3 + (4 + 5))"));
+}

spec/test_parse_inner_nested.rs

@@ -0,0 +1,17 @@
+use testparser::*;
+
+fn eval(input: &[u8]) -> i64 {
+    let mut c = p_context_new(input);
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    let v = p_result(&c);
+    p_context_delete(c);
+    v
+}
+
+fn main() {
+    assert_eq!(5, eval(b"2 + 3"));
+    assert_eq!(3, eval(b"(1 + 2)"));
+    assert_eq!(14, eval(b"2 + (3 + 4) + 5"));
+    assert_eq!(37, eval(b"2 + (10 + (20 + 5))"));
+    assert_eq!(15, eval(b"(1 + 2) + (3 + (4 + 5))"));
+}

spec/test_parse_inner_nested_tree.c

@@ -0,0 +1,91 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+/* Grammar: see the D variant / spec. Tree generation mode; parentheses handled
+ * by the lex function. Tree nodes live in the context arena. */
+
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info)
+{
+    size_t result = p_lex(context, out_token_info);
+    if (result != P_SUCCESS)
+    {
+        return result;
+    }
+    if (out_token_info->token == TOKEN_lparen)
+    {
+        p_position_t start_position = out_token_info->position;
+        /* Reentrant nested parse of the parenthesized sub-expression. */
+        p_token_t follow_tokens[] = { TOKEN_rparen };
+        size_t inner_result = p_parse_inner_Start(context, follow_tokens, 1u);
+        if (inner_result != P_SUCCESS)
+        {
+            return inner_result;
+        }
+        Start inner = p_result_Start(context);
+        assert(p_node_valid(inner));
+        /* p_parse_inner rewound the input so that ')' was not consumed; consume
+         * it now. */
+        p_token_info_t rparen_info;
+        size_t rparen_result = p_lex(context, &rparen_info);
+        assert(rparen_result == P_SUCCESS);
+        assert(rparen_info.token == TOKEN_rparen);
+        /* The subtree covers the region strictly between the parentheses. */
+        assert_eq((size_t)(start_position.col + 1u), (size_t)p_node_position(inner).col);
+        assert_eq((size_t)(rparen_info.position.col - 1u), (size_t)p_node_end_position(inner).col);
+        /* The inner subtree is discarded (the lexer synthesizes a num token in
+         * its place), but its nodes remain in the shared context arena and are
+         * freed with the context. */
+        out_token_info->token = TOKEN_num;
+        out_token_info->position = start_position;
+        out_token_info->end_position = rparen_info.end_position;
+    }
+    return P_SUCCESS;
+}
+
+int main()
+{
+    /* "(3 + 4) + (5 + 6)": two parenthesized groups, each collapsed by the
+     * lexer into a single num token spanning its group. */
+    char const * input = "(3 + 4) + (5 + 6)";
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse(context) == P_SUCCESS);
+
+    Start tree = p_result(context);
+    assert(p_node_valid(tree));
+
+    /* Start -> Expr, where the top Expr is "Expr plus num". */
+    Expr top = p_Start_pExpr(tree);
+    assert(p_node_valid(top));
+    assert(p_node_valid(p_Expr_pExpr(top)));
+    assert(p_node_valid(p_Expr_pToken2(top)));
+    assert(p_node_valid(p_Expr_pToken3(top)));
+
+    /* The '+' joining the two groups is at column 9. */
+    assert_eq(1u, (size_t)p_node_position(p_Expr_pToken2(top)).row);
+    assert_eq(9u, (size_t)p_node_position(p_Expr_pToken2(top)).col);
+
+    /* Right operand: synthesized num for "(5 + 6)", spanning columns 11..17. */
+    assert_eq(1u, (size_t)p_node_position(p_Expr_pToken3(top)).row);
+    assert_eq(11u, (size_t)p_node_position(p_Expr_pToken3(top)).col);
+    assert_eq(1u, (size_t)p_node_end_position(p_Expr_pToken3(top)).row);
+    assert_eq(17u, (size_t)p_node_end_position(p_Expr_pToken3(top)).col);
+
+    /* Left operand: Expr -> num, the synthesized num for "(3 + 4)", spanning
+     * columns 1..7. */
+    Expr left = p_Expr_pExpr(top);
+    assert(p_node_valid(p_Expr_pToken1(left)));
+    assert_eq(1u, (size_t)p_node_position(p_Expr_pToken1(left)).row);
+    assert_eq(1u, (size_t)p_node_position(p_Expr_pToken1(left)).col);
+    assert_eq(1u, (size_t)p_node_end_position(p_Expr_pToken1(left)).row);
+    assert_eq(7u, (size_t)p_node_end_position(p_Expr_pToken1(left)).col);
+
+    /* The whole tree spans columns 1..17. */
+    assert_eq(1u, (size_t)p_node_position(tree).col);
+    assert_eq(17u, (size_t)p_node_end_position(tree).col);
+
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_parse_inner_nested_tree.d

@@ -0,0 +1,92 @@
+import testparser;
+import testutils;
+
+/* Grammar: see test_parse_inner_nested_tree.c. */
+
+size_t mylexfn(p_context_t * context, p_token_info_t * out_token_info)
+{
+    size_t result = p_lex(context, out_token_info);
+    if (result != P_SUCCESS)
+    {
+        return result;
+    }
+    if (out_token_info.token == TOKEN_lparen)
+    {
+        p_position_t start_position = out_token_info.position;
+        /* Reentrant nested parse of the parenthesized sub-expression. */
+        p_token_t[] follow_tokens = [TOKEN_rparen];
+        size_t inner_result = p_parse_inner_Start(context, follow_tokens);
+        if (inner_result != P_SUCCESS)
+        {
+            return inner_result;
+        }
+        Start inner = p_result_Start(context);
+        assert(inner.valid);
+        /* p_parse_inner rewound the input so that ')' was not consumed; consume
+         * it now. */
+        p_token_info_t rparen_info;
+        size_t rparen_result = p_lex(context, &rparen_info);
+        assert(rparen_result == P_SUCCESS);
+        assert(rparen_info.token == TOKEN_rparen);
+        /* The subtree covers the region strictly between the parentheses. */
+        assert_eq(start_position.col + 1u, inner.position.col);
+        assert_eq(rparen_info.position.col - 1u, inner.end_position.col);
+        /* The inner subtree is discarded (the lexer synthesizes a num token in
+         * its place), but its nodes remain in the shared context arena and are
+         * freed with the context. */
+        /* Synthesize a num token spanning the entire "( ... )" group. */
+        out_token_info.token = TOKEN_num;
+        out_token_info.position = start_position;
+        out_token_info.end_position = rparen_info.end_position;
+    }
+    return P_SUCCESS;
+}
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* "(3 + 4) + (5 + 6)": two parenthesized groups, each collapsed by the
+     * lexer into a single num token spanning its group. */
+    string input = "(3 + 4) + (5 + 6)";
+    p_context_t * context = p_context_new(input);
+    assert(p_parse(context) == P_SUCCESS);
+
+    Start tree = p_result(context);
+    assert(tree.valid);
+
+    /* Start -> Expr, where the top Expr is "Expr plus num". */
+    Expr top = tree.pExpr;
+    assert(top.valid);
+    assert(top.pExpr.valid);
+    assert(top.pToken2.valid);
+    assert(top.pToken3.valid);
+
+    /* The '+' joining the two groups is at column 9. */
+    assert_eq(1u, top.pToken2.position.row);
+    assert_eq(9u, top.pToken2.position.col);
+
+    /* Right operand: synthesized num for "(5 + 6)", spanning columns 11..17. */
+    assert_eq(1u, top.pToken3.position.row);
+    assert_eq(11u, top.pToken3.position.col);
+    assert_eq(1u, top.pToken3.end_position.row);
+    assert_eq(17u, top.pToken3.end_position.col);
+
+    /* Left operand: Expr -> num, the synthesized num for "(3 + 4)", spanning
+     * columns 1..7. */
+    Expr left = top.pExpr;
+    assert(left.pToken1.valid);
+    assert_eq(1u, left.pToken1.position.row);
+    assert_eq(1u, left.pToken1.position.col);
+    assert_eq(1u, left.pToken1.end_position.row);
+    assert_eq(7u, left.pToken1.end_position.col);
+
+    /* The whole tree spans columns 1..17. */
+    assert_eq(1u, tree.position.col);
+    assert_eq(17u, tree.end_position.col);
+
+    p_context_delete(context);
+}

spec/test_parse_inner_nested_tree.rs

@@ -0,0 +1,36 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"(3 + 4) + (5 + 6)");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    {
+        let tree = p_result(&c);
+        assert!(tree.valid());
+
+        /* Start -> Expr, where the top Expr is "Expr plus num". */
+        let top = tree.pExpr();
+        assert!(top.valid());
+        assert!(top.pExpr().valid());
+        assert!(top.pToken2().valid());
+        assert!(top.pToken3().valid());
+
+        /* The '+' joining the two groups is at column 9. */
+        assert_eq!(1, top.pToken2().position().row);
+        assert_eq!(9, top.pToken2().position().col);
+
+        /* Right operand: synthesized num for "(5 + 6)", columns 11..17. */
+        assert_eq!(11, top.pToken3().position().col);
+        assert_eq!(17, top.pToken3().end_position().col);
+
+        /* Left operand: synthesized num for "(3 + 4)", columns 1..7. */
+        let left = top.pExpr();
+        assert!(left.pToken1().valid());
+        assert_eq!(1, left.pToken1().position().col);
+        assert_eq!(7, left.pToken1().end_position().col);
+
+        /* The whole tree spans columns 1..17. */
+        assert_eq!(1, tree.position().col);
+        assert_eq!(17, tree.end_position().col);
+    }
+    p_context_delete(c);
+}

spec/test_parse_inner_recursive.c

@@ -0,0 +1,77 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    /* Grammar (recursive):
+     *   Start -> a Start b   << $$ = $2; >>
+     *   Start -> c           << $$ = $1; >>
+     *   token a << $$ = 1; >>
+     *   token b << $$ = 2; >>
+     *   token c << $$ = 3; >>
+     *
+     * Here `Start` can appear in the middle of another `Start` rule, so the
+     * inner-parse follow-token success must be blocked whenever an unfinished
+     * outer `Start -> a Start b` remains on the parse stack (i.e. the parse
+     * stack contains more than just the initial state and the reduced start
+     * rule set). */
+
+    /* Standard parse of `c` succeeds. */
+    char const * input = "c";
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_SUCCESS);
+    assert_eq(3u, (size_t)p_result_Start(context));
+    p_context_delete(context);
+
+    /* Standard parse of `acb` succeeds (full outer rule). */
+    input = "acb";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_SUCCESS);
+    assert_eq(3u, (size_t)p_result_Start(context));
+    p_context_delete(context);
+
+    /* Standard parse of `ac` fails (`b` missing). */
+    input = "ac";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_UNEXPECTED_TOKEN);
+    p_context_delete(context);
+
+    /* parse_inner with `ac` and follow token `b` also fails: even though the
+     * inner `Start -> c` reduces and `Start` is shifted, the outer
+     * `Start -> a Start . b` is still on the stack (stack length > 2), so the
+     * "reduced start rule is the only thing on the parse stack" invariant
+     * blocks the shift-side follow-token success. */
+    {
+        input = "ac";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b, TOKEN___EOF };
+        assert(p_parse_inner_Start(context, follow_tokens, 2u) == P_UNEXPECTED_TOKEN);
+        p_context_delete(context);
+    }
+
+    /* parse_inner with `acb` (complete outer rule) succeeds via the standard
+     * path. */
+    {
+        input = "acb";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_Start(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(3u, (size_t)p_result_Start(context));
+        p_context_delete(context);
+    }
+
+    /* parse_inner with just `c` succeeds via the standard path even when a
+     * follow-token vector is supplied. */
+    {
+        input = "c";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_Start(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(3u, (size_t)p_result_Start(context));
+        p_context_delete(context);
+    }
+
+    return 0;
+}

spec/test_parse_inner_recursive.d

@@ -0,0 +1,49 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* See test_parse_inner_recursive.c for details on the grammar. */
+
+    /* Standard parse of `c` succeeds. */
+    string input = "c";
+    p_context_t * context = p_context_new(input);
+    assert(p_parse_Start(context) == P_SUCCESS);
+    assert(p_result_Start(context) == 3);
+
+    /* Standard parse of `acb` succeeds. */
+    input = "acb";
+    context = p_context_new(input);
+    assert(p_parse_Start(context) == P_SUCCESS);
+    assert(p_result_Start(context) == 3);
+
+    /* Standard parse of `ac` fails. */
+    input = "ac";
+    context = p_context_new(input);
+    assert(p_parse_Start(context) == P_UNEXPECTED_TOKEN);
+
+    /* parse_inner with `ac` fails: outer rule still on the stack. */
+    input = "ac";
+    context = p_context_new(input);
+    p_token_t[] follow_tokens_bothway = [TOKEN_b, TOKEN___EOF];
+    assert(p_parse_inner_Start(context, follow_tokens_bothway) == P_UNEXPECTED_TOKEN);
+
+    /* parse_inner with `acb` succeeds via the standard path. */
+    input = "acb";
+    context = p_context_new(input);
+    p_token_t[] follow_tokens_b = [TOKEN_b];
+    assert(p_parse_inner_Start(context, follow_tokens_b) == P_SUCCESS);
+    assert(p_result_Start(context) == 3);
+
+    /* parse_inner with just `c` succeeds via the standard path. */
+    input = "c";
+    context = p_context_new(input);
+    assert(p_parse_inner_Start(context, follow_tokens_b) == P_SUCCESS);
+    assert(p_result_Start(context) == 3);
+}

spec/test_parse_inner_recursive.rs

@@ -0,0 +1,31 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"c");
+    assert_eq!(P_SUCCESS, p_parse_Start(&mut c));
+    assert_eq!(3, p_result_Start(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"acb");
+    assert_eq!(P_SUCCESS, p_parse_Start(&mut c));
+    assert_eq!(3, p_result_Start(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ac");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse_Start(&mut c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ac");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse_inner_Start(&mut c, &[TOKEN_b, TOKEN___EOF]));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"acb");
+    assert_eq!(P_SUCCESS, p_parse_inner_Start(&mut c, &[TOKEN_b]));
+    assert_eq!(3, p_result_Start(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"c");
+    assert_eq!(P_SUCCESS, p_parse_inner_Start(&mut c, &[TOKEN_b]));
+    assert_eq!(3, p_result_Start(&c));
+    p_context_delete(c);
+}

spec/test_parse_inner_shared.c

@@ -0,0 +1,104 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    /* Grammar:
+     *   start Start;
+     *   start R1;
+     *   Start -> R1 a;
+     *   Start -> R2 b;
+     *   R1 -> a b   << $$ = 11; >>
+     *   R2 -> a b   << $$ = 22; >>
+     *   token a; token b;
+     *
+     * The rules `R1 -> a b` and `R2 -> a b` produce identical input. Within
+     * parse_Start, the generated parser differentiates the reduce by
+     * lookahead: `a` selects R1 (because `Start -> R1 a`) and `b` selects R2
+     * (because `Start -> R2 b`). Within parse_R1, the reduce is unconditional
+     * on any lookahead. This test exercises p_parse_inner_R1() to confirm
+     * that reductions to R1 succeed even when the incoming follow token is
+     * not the natural lookahead used by parse_Start's disambiguation. */
+
+    /* Sanity-check that parse_Start resolves R1 vs R2 via lookahead in the
+     * shared "a b" state. */
+    char const * input = "aba";
+    p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    input = "abb";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_Start(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    /* Standard parse of R1 succeeds on "ab". */
+    input = "ab";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_R1(context) == P_SUCCESS);
+    assert_eq(11u, (size_t)p_result_R1(context));
+    p_context_delete(context);
+
+    /* Standard parse of R1 fails on "abb" (unexpected trailing token). */
+    input = "abb";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_R1(context) == P_UNEXPECTED_TOKEN);
+    p_context_delete(context);
+
+    /* parse_inner_R1("abb", [b]) succeeds: even though `b` is the lookahead
+     * that parse_Start uses to select R2 over R1 in the ambiguous state, from
+     * R1's start state the reduce to R1 is unconditional, and the follow-
+     * token shift retry at the R1-accepting state completes the parse.
+     *
+     * The follow token that completed the parse must not be consumed from
+     * the input: p_position() should point to the follow token, and a
+     * subsequent p_lex() should return it. */
+    {
+        input = "abb";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_R1(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(11u, (size_t)p_result_R1(context));
+        /* Follow token `b` is at column 3 (1-based). */
+        p_position_t pos = p_position(context);
+        assert_eq(1u, (size_t)pos.row);
+        assert_eq(3u, (size_t)pos.col);
+        p_token_info_t token_info;
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert_eq((size_t)TOKEN_b, (size_t)token_info.token);
+        assert_eq(1u, (size_t)token_info.position.row);
+        assert_eq(3u, (size_t)token_info.position.col);
+        p_context_delete(context);
+    }
+
+    /* parse_inner_R1("aba", [a]) also succeeds: `a` is the follow token
+     * parse_Start uses to select R1, and it works here as a follow token
+     * too. */
+    {
+        input = "aba";
+        context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_a };
+        assert(p_parse_inner_R1(context, follow_tokens, 1u) == P_SUCCESS);
+        assert_eq(11u, (size_t)p_result_R1(context));
+        /* Follow token `a` is at column 3 (1-based) and remains in the
+         * input. */
+        p_position_t pos = p_position(context);
+        assert_eq(1u, (size_t)pos.row);
+        assert_eq(3u, (size_t)pos.col);
+        p_token_info_t token_info;
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert_eq((size_t)TOKEN_a, (size_t)token_info.token);
+        p_context_delete(context);
+    }
+
+    /* parse_inner_R1("ab", NULL) behaves like p_parse_R1("ab"). */
+    input = "ab";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse_inner_R1(context, NULL, 0u) == P_SUCCESS);
+    assert_eq(11u, (size_t)p_result_R1(context));
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_parse_inner_shared.d

@@ -0,0 +1,72 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* See test_parse_inner_shared.c for details on the grammar. */
+
+    /* Sanity-check that parse_Start resolves R1 vs R2 via lookahead. */
+    string input = "aba";
+    p_context_t * context = p_context_new(input);
+    assert(p_parse_Start(context) == P_SUCCESS);
+
+    input = "abb";
+    context = p_context_new(input);
+    assert(p_parse_Start(context) == P_SUCCESS);
+
+    /* Standard parse of R1 succeeds on "ab". */
+    input = "ab";
+    context = p_context_new(input);
+    assert(p_parse_R1(context) == P_SUCCESS);
+    assert(p_result_R1(context) == 11);
+
+    /* Standard parse of R1 fails on "abb". */
+    input = "abb";
+    context = p_context_new(input);
+    assert(p_parse_R1(context) == P_UNEXPECTED_TOKEN);
+
+    /* parse_inner_R1("abb", [b]) succeeds: `b` is the lookahead that
+     * parse_Start would use to select R2 over R1, but from R1's own start
+     * state R1 reduces unconditionally, and the follow-token shift retry at
+     * the R1-accepting state completes the parse.
+     *
+     * The follow token that completed the parse must not be consumed: a
+     * subsequent p_lex() should return it. */
+    input = "abb";
+    context = p_context_new(input);
+    p_token_t[] follow_tokens_b = [TOKEN_b];
+    assert(p_parse_inner_R1(context, follow_tokens_b) == P_SUCCESS);
+    assert(p_result_R1(context) == 11);
+    p_position_t pos = p_position(context);
+    assert(pos.row == 1);
+    assert(pos.col == 3);
+    p_token_info_t token_info;
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_b);
+    assert(token_info.position.row == 1);
+    assert(token_info.position.col == 3);
+
+    /* parse_inner_R1("aba", [a]) also succeeds. */
+    input = "aba";
+    context = p_context_new(input);
+    p_token_t[] follow_tokens_a = [TOKEN_a];
+    assert(p_parse_inner_R1(context, follow_tokens_a) == P_SUCCESS);
+    assert(p_result_R1(context) == 11);
+    pos = p_position(context);
+    assert(pos.row == 1);
+    assert(pos.col == 3);
+    assert(p_lex(context, &token_info) == P_SUCCESS);
+    assert(token_info.token == TOKEN_a);
+
+    /* parse_inner_R1("ab", null) behaves like p_parse_R1("ab"). */
+    input = "ab";
+    context = p_context_new(input);
+    assert(p_parse_inner_R1(context, null) == P_SUCCESS);
+    assert(p_result_R1(context) == 11);
+}

spec/test_parse_inner_shared.rs

@@ -0,0 +1,47 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"aba");
+    assert_eq!(P_SUCCESS, p_parse_Start(&mut c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"abb");
+    assert_eq!(P_SUCCESS, p_parse_Start(&mut c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse_R1(&mut c));
+    assert_eq!(11, p_result_R1(&c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"abb");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse_R1(&mut c));
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"abb");
+    assert_eq!(P_SUCCESS, p_parse_inner_R1(&mut c, &[TOKEN_b]));
+    assert_eq!(11, p_result_R1(&c));
+    let pos = p_position(&c);
+    assert_eq!(1, pos.row);
+    assert_eq!(3, pos.col);
+    let mut ti = p_token_info_t::default();
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_b, ti.token);
+    assert_eq!(3, ti.position.col);
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"aba");
+    assert_eq!(P_SUCCESS, p_parse_inner_R1(&mut c, &[TOKEN_a]));
+    assert_eq!(11, p_result_R1(&c));
+    let pos = p_position(&c);
+    assert_eq!(3, pos.col);
+    let mut ti = p_token_info_t::default();
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_a, ti.token);
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse_inner_R1(&mut c, &[]));
+    assert_eq!(11, p_result_R1(&c));
+    p_context_delete(c);
+}

spec/test_parse_inner_tree.c

@@ -0,0 +1,68 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    /* See the D variant / grammar comments for details. In tree mode the tree
+     * nodes live in the context arena and are freed with p_context_delete(). */
+
+    /* Baseline: p_parse_R1 works on "ab" and the returned tree is
+     * well-formed. */
+    {
+        char const * input = "ab";
+        p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+        assert(p_parse_R1(context) == P_SUCCESS);
+        R1 tree = p_result_R1(context);
+        assert(p_node_valid(tree));
+        assert(p_node_valid(p_R1_pToken1(tree)));
+        assert_eq((size_t)TOKEN_a, (size_t)p_tree_walk_R1(tree, pToken1, token));
+        assert(p_node_valid(p_R1_pToken2(tree)));
+        assert_eq((size_t)TOKEN_b, (size_t)p_tree_walk_R1(tree, pToken2, token));
+        p_context_delete(context);
+    }
+
+    /* Primary case: p_parse_inner_R1 with a non-EOF follow token completes
+     * the parse, returns a well-formed tree, and leaves the follow token
+     * unconsumed. */
+    {
+        char const * input = "abb";
+        p_context_t * context = p_context_new((uint8_t const *)input, strlen(input));
+        p_token_t follow_tokens[] = { TOKEN_b };
+        assert(p_parse_inner_R1(context, follow_tokens, 1u) == P_SUCCESS);
+
+        /* Tree is well-formed. */
+        R1 tree = p_result_R1(context);
+        assert(p_node_valid(tree));
+        assert(p_node_valid(p_R1_pToken1(tree)));
+        assert_eq((size_t)TOKEN_a, (size_t)p_tree_walk_R1(tree, pToken1, token));
+        assert_eq(1u, (size_t)p_node_position(p_R1_pToken1(tree)).row);
+        assert_eq(1u, (size_t)p_node_position(p_R1_pToken1(tree)).col);
+        assert(p_node_valid(p_R1_pToken2(tree)));
+        assert_eq((size_t)TOKEN_b, (size_t)p_tree_walk_R1(tree, pToken2, token));
+        assert_eq(1u, (size_t)p_node_position(p_R1_pToken2(tree)).row);
+        assert_eq(2u, (size_t)p_node_position(p_R1_pToken2(tree)).col);
+
+        /* The R1 tree covers positions 1..2 - the third `b` at column 3 is
+         * the follow token and is not part of the tree. */
+        assert_eq(1u, (size_t)p_node_position(tree).row);
+        assert_eq(1u, (size_t)p_node_position(tree).col);
+        assert_eq(1u, (size_t)p_node_end_position(tree).row);
+        assert_eq(2u, (size_t)p_node_end_position(tree).col);
+
+        /* Follow token remains in the input. */
+        p_position_t pos = p_position(context);
+        assert_eq(1u, (size_t)pos.row);
+        assert_eq(3u, (size_t)pos.col);
+        p_token_info_t token_info;
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert_eq((size_t)TOKEN_b, (size_t)token_info.token);
+        assert_eq(1u, (size_t)token_info.position.row);
+        assert_eq(3u, (size_t)token_info.position.col);
+
+        p_context_delete(context);
+    }
+
+    return 0;
+}

spec/test_parse_inner_tree.d

@@ -0,0 +1,68 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    /* See test_parse_inner_tree.c for details on the grammar and cases. */
+
+    /* Baseline: p_parse_R1 works on "ab". */
+    {
+        string input = "ab";
+        p_context_t * context = p_context_new(input);
+        assert(p_parse_R1(context) == P_SUCCESS);
+        R1 tree = p_result_R1(context);
+        assert(tree.valid);
+        assert(tree.pToken1.valid);
+        assert(tree.pToken1.token == TOKEN_a);
+        assert(tree.pToken2.valid);
+        assert(tree.pToken2.token == TOKEN_b);
+        p_context_delete(context);
+    }
+
+    /* Primary case: p_parse_inner_R1 with a non-EOF follow token completes
+     * the parse, returns a well-formed tree, and leaves the follow token
+     * unconsumed. */
+    {
+        string input = "abb";
+        p_context_t * context = p_context_new(input);
+        p_token_t[] follow_tokens = [TOKEN_b];
+        assert(p_parse_inner_R1(context, follow_tokens) == P_SUCCESS);
+
+        /* Tree is well-formed. */
+        R1 tree = p_result_R1(context);
+        assert(tree.valid);
+        assert(tree.pToken1.valid);
+        assert(tree.pToken1.token == TOKEN_a);
+        assert(tree.pToken1.position.row == 1);
+        assert(tree.pToken1.position.col == 1);
+        assert(tree.pToken2.valid);
+        assert(tree.pToken2.token == TOKEN_b);
+        assert(tree.pToken2.position.row == 1);
+        assert(tree.pToken2.position.col == 2);
+
+        /* The R1 tree covers positions 1..2. The third `b` at column 3 is
+         * the follow token and is not part of the tree. */
+        assert(tree.position.row == 1);
+        assert(tree.position.col == 1);
+        assert(tree.end_position.row == 1);
+        assert(tree.end_position.col == 2);
+
+        /* Follow token remains in the input. */
+        p_position_t pos = p_position(context);
+        assert(pos.row == 1);
+        assert(pos.col == 3);
+        p_token_info_t token_info;
+        assert(p_lex(context, &token_info) == P_SUCCESS);
+        assert(token_info.token == TOKEN_b);
+        assert(token_info.position.row == 1);
+        assert(token_info.position.col == 3);
+
+        p_context_delete(context);
+    }
+}

spec/test_parse_inner_tree.rs

@@ -0,0 +1,36 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse_R1(&mut c));
+    {
+        let tree = p_result_R1(&c);
+        assert!(tree.valid());
+        assert!(tree.pToken1().valid());
+        assert_eq!(TOKEN_a, tree.pToken1().token());
+        assert!(tree.pToken2().valid());
+        assert_eq!(TOKEN_b, tree.pToken2().token());
+    }
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"abb");
+    assert_eq!(P_SUCCESS, p_parse_inner_R1(&mut c, &[TOKEN_b]));
+    {
+        let tree = p_result_R1(&c);
+        assert!(tree.valid());
+        assert_eq!(TOKEN_a, tree.pToken1().token());
+        assert_eq!(1, tree.pToken1().position().row);
+        assert_eq!(1, tree.pToken1().position().col);
+        assert_eq!(TOKEN_b, tree.pToken2().token());
+        assert_eq!(2, tree.pToken2().position().col);
+        assert_eq!(1, tree.position().col);
+        assert_eq!(2, tree.end_position().col);
+    }
+    let pos = p_position(&c);
+    assert_eq!(3, pos.col);
+    let mut ti = p_token_info_t::default();
+    assert_eq!(P_SUCCESS, p_lex(&mut c, &mut ti));
+    assert_eq!(TOKEN_b, ti.token);
+    assert_eq!(3, ti.position.col);
+    p_context_delete(c);
+}

spec/test_parser_identical_rules_lookahead.rs

@@ -0,0 +1,9 @@
+use testparser::*;
+
+fn main() {
+    for input in [&b"aba"[..], &b"abb"[..]] {
+        let mut context = p_context_new(input);
+        assert_eq!(P_SUCCESS, p_parse(&mut context));
+        p_context_delete(context);
+    }
+}

spec/test_parser_rule_from_multiple_states.rs

@@ -0,0 +1,18 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"a");
+    assert_eq!(P_UNEXPECTED_TOKEN, p_parse(&mut context));
+    assert_eq!(1, p_position(&context).row);
+    assert_eq!(2, p_position(&context).col);
+    assert_eq!(TOKEN___EOF, p_token(&context));
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"a b");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"bb");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    p_context_delete(context);
+}

spec/test_parser_rule_user_code.rs

@@ -0,0 +1,7 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    p_context_delete(context);
+}

spec/test_parser_user_code_tree.c

@@ -0,0 +1,38 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include "testutils.h"
+
+int main()
+{
+    char const * input = "ab";
+    p_context_t * context;
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert_eq(P_SUCCESS, p_parse(context));
+
+    /* The parser user code recorded values accessed via $$, $1, and $2 while
+     * the tree node for the Start rule was being formed. */
+    assert_eq(3, context->start_n_fields);
+    assert_eq(11, context->start_a_value);
+    assert_eq(11, context->a_value);
+    assert_eq(22, context->b_value);
+    assert_eq(TOKEN_b, context->b_token);
+
+    /* The empty-matched rule C has a null $$ tree node, and its field in the
+     * Start node is null as well. */
+    assert_eq(1, context->c_is_null);
+    assert_eq(1, context->c_field_is_null);
+
+    /* Field aliases reference the same component tree nodes as the positional
+     * references. */
+    assert_eq(11, context->alias_a_value);
+    assert_eq(22, context->alias_b_value);
+
+    Start start = p_result(context);
+    assert(p_node_valid(p_Start_pA(start)));
+    assert(p_node_valid(p_Start_pB(start)));
+    assert(!p_node_valid(p_Start_pC(start)));
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_parser_user_code_tree.d

@@ -0,0 +1,39 @@
+import testparser;
+import std.stdio;
+import testutils;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    string input = "ab";
+    p_context_t * context = p_context_new(input);
+    assert_eq(P_SUCCESS, p_parse(context));
+
+    /* The parser user code recorded values accessed via $$, $1, and $2 while
+     * the tree node for the Start rule was being formed. */
+    assert_eq(3, context.start_n_fields);
+    assert_eq(11, context.start_a_value);
+    assert_eq(11, context.a_value);
+    assert_eq(22, context.b_value);
+    assert_eq(TOKEN_b, context.b_token);
+
+    /* The empty-matched rule C has a null $$ tree node, and its field in the
+     * Start node is null as well. */
+    assert_eq(1, context.c_is_null);
+    assert_eq(1, context.c_field_is_null);
+
+    /* Field aliases reference the same component tree nodes as the positional
+     * references. */
+    assert_eq(11, context.alias_a_value);
+    assert_eq(22, context.alias_b_value);
+
+    Start start = p_result(context);
+    assert(start.pA.valid);
+    assert(start.pB.valid);
+    assert(!start.pC.valid);
+    p_context_delete(context);
+}

spec/test_parser_user_code_tree.rs

@@ -0,0 +1,24 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"ab");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+
+    assert_eq!(3, context.start_n_fields);
+    assert_eq!(11, context.start_a_value);
+    assert_eq!(11, context.a_value);
+    assert_eq!(22, context.b_value);
+    assert_eq!(TOKEN_b, context.b_token);
+    assert_eq!(1, context.c_is_null);
+    assert_eq!(1, context.c_field_is_null);
+    assert_eq!(11, context.alias_a_value);
+    assert_eq!(22, context.alias_b_value);
+
+    {
+        let start = p_result(&context);
+        assert!(start.pA().valid());
+        assert!(start.pB().valid());
+        assert!(!start.pC().valid());
+    }
+    p_context_delete(context);
+}

spec/test_parsing_json.rs

@@ -0,0 +1,31 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"{}");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    assert_eq!(JSON_OBJECT, p_result(&c).id());
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"[]");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    assert_eq!(JSON_ARRAY, p_result(&c).id());
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"-45.6");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    assert_eq!(JSON_NUMBER, p_result(&c).id());
+    assert_eq!(-45.6, p_result(&c).number());
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"{\"hi\":true}");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    assert_eq!(JSON_OBJECT, p_result(&c).id());
+    p_context_delete(c);
+
+    let mut c = p_context_new(b"[1, 2, \"three\", [4, 5], {\"six\": 6}]");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    let v = p_result(&c);
+    assert_eq!(JSON_ARRAY, v.id());
+    assert_eq!(5, v.array_len());
+    p_context_delete(c);
+}

spec/test_parsing_lists.rs

@@ -0,0 +1,11 @@
+use testparser::*;
+
+fn main() {
+    let cases: [(&[u8], u32); 3] = [(b"a", 1), (b"", 0), (b"aaaaaaaaaaaaaaaa", 16)];
+    for (input, expected) in cases {
+        let mut context = p_context_new(input);
+        assert_eq!(P_SUCCESS, p_parse(&mut context));
+        assert_eq!(expected, p_result(&context));
+        p_context_delete(context);
+    }
+}

spec/test_pattern.rs

@@ -0,0 +1,13 @@
+use testparser::*;
+
+fn main() {
+    let mut context = p_context_new(b"abcdef");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass1");
+    p_context_delete(context);
+
+    let mut context = p_context_new(b"defabcdef");
+    assert_eq!(P_SUCCESS, p_parse(&mut context));
+    println!("pass2");
+    p_context_delete(context);
+}

spec/test_positions.c

@@ -0,0 +1,29 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+#include <stdio.h>
+
+int main()
+{
+    char const * input = "    Hello\n\n        4200\n";
+    p_context_t * context;
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    printf("\n");
+
+    input = "\n tok2";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    printf("\n");
+
+    input = "  tok1";
+    context = p_context_new((uint8_t const *)input, strlen(input));
+    assert(p_parse(context) == P_SUCCESS);
+    p_context_delete(context);
+
+    return 0;
+}

spec/test_positions.d

@@ -0,0 +1,27 @@
+import testparser;
+import std.stdio;
+
+int main()
+{
+    return 0;
+}
+
+unittest
+{
+    string input = "    Hello\n\n        4200\n";
+    p_context_t * context;
+    context = p_context_new(input);
+    assert(p_parse(context) == P_SUCCESS);
+
+    writeln();
+
+    input = "\n tok2";
+    context = p_context_new(input);
+    assert(p_parse(context) == P_SUCCESS);
+
+    writeln();
+
+    input = "  tok1";
+    context = p_context_new(input);
+    assert(p_parse(context) == P_SUCCESS);
+}

spec/test_positions.rs

@@ -0,0 +1,19 @@
+use testparser::*;
+
+fn main() {
+    let mut c = p_context_new(b"    Hello\n\n        4200\n");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+
+    println!();
+
+    let mut c = p_context_new(b"\n tok2");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+
+    println!();
+
+    let mut c = p_context_new(b"  tok1");
+    assert_eq!(P_SUCCESS, p_parse(&mut c));
+    p_context_delete(c);
+}