wip

assets/parser.c.erb

@@ -326,8 +326,8 @@ static lexer_state_id_t check_lexer_transition(uint32_t current_state, uint32_t
 static size_t find_longest_match(<%= @grammar.prefix %>context_t * context,
     lexer_match_info_t * out_match_info, size_t * out_unexpected_input_length)
 {
-    lexer_match_info_t longest_match;
-    lexer_match_info_t attempt_match;
+    lexer_match_info_t longest_match = {0};
+    lexer_match_info_t attempt_match = {0};
     *out_match_info = longest_match;
     uint32_t current_state = lexer_mode_table[context->mode].state_table_offset;
     for (;;)
@@ -427,7 +427,7 @@ static size_t find_longest_match(<%= @grammar.prefix %>context_t * context,
  */
 static size_t attempt_lex_token(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>token_info_t * out_token_info)
 {
-    <%= @grammar.prefix %>token_info_t token_info;
+    <%= @grammar.prefix %>token_info_t token_info = {0};
     token_info.position = context->text_position;
     token_info.token = INVALID_TOKEN_ID;
     *out_token_info = token_info; // TODO: remove
@@ -437,12 +437,12 @@ static size_t attempt_lex_token(<%= @grammar.prefix %>context_t * context, <%= @
     switch (result)
     {
     case P_SUCCESS:
-        <%= @grammar.prefix %>token_t token_to_accept = match_info.accepting_state.token;
-        if (match_info.accepting_state.code_id != INVALID_USER_CODE_ID)
+        <%= @grammar.prefix %>token_t token_to_accept = match_info.accepting_state->token;
+        if (match_info.accepting_state->code_id != INVALID_USER_CODE_ID)
         {
             uint8_t const * match = &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, match_info.length, &token_info);
             /* An invalid token returned from lexer_user_code() means that the
              * user code did not explicitly return a token. So only override
              * the token to return if the user code does explicitly return a
@@ -531,7 +531,7 @@ size_t <%= @grammar.prefix %>lex(<%= @grammar.prefix %>context_t * context, <%=
  *************************************************************************/
 
 /** Reduce ID type. */
-typedef <%= get_type_for(@parser.reduce_table.size) %> = reduce_id_t;
+typedef <%= get_type_for(@parser.reduce_table.size) %> reduce_id_t;
 
 /**
  * A symbol ID can hold either a token ID or a rule set ID.
@@ -658,7 +658,7 @@ typedef struct
  * @param stack
  *   state_values stack structure.
  */
-void state_values_stack_init(state_values_stack_t * stack)
+static void state_values_stack_init(state_values_stack_t * stack)
 {
     const size_t initial_capacity = 10u;
     stack->length = 0u;
@@ -676,7 +676,7 @@ void state_values_stack_init(state_values_stack_t * stack)
  *
  * @return Pointer to the state value structure at the given index.
  */
-state_value_t * state_values_stack_index(state_values_stack_t * stack, int index)
+static state_value_t * state_values_stack_index(state_values_stack_t * stack, int index)
 {
     if (index >= 0)
     {
@@ -694,7 +694,7 @@ state_value_t * state_values_stack_index(state_values_stack_t * stack, int index
  * @param stack
  *   state_values stack structure.
  */
-void state_values_stack_push(state_values_stack_t * stack)
+static void state_values_stack_push(state_values_stack_t * stack)
 {
     size_t const current_capacity = stack->capacity;
     size_t const current_length = stack->length;
@@ -702,7 +702,7 @@ void state_values_stack_push(state_values_stack_t * stack)
     {
         size_t const new_capacity = current_capacity * 2u;
         state_value_t * new_entries = malloc(new_capacity * sizeof(state_value_t));
-        memcpy(new_entries, stack->entries, current_length * sizeof(state_value_t);
+        memcpy(new_entries, stack->entries, current_length * sizeof(state_value_t));
         free(stack->entries);
         stack->capacity = new_capacity;
         stack->entries = new_entries;
@@ -719,7 +719,7 @@ void state_values_stack_push(state_values_stack_t * stack)
  * @param n
  *   Number of states to pop.
  */
-void state_values_stack_pop(state_values_stack_t * stack, size_t n)
+static void state_values_stack_pop(state_values_stack_t * stack, size_t n)
 {
     stack->length -= n;
 }
@@ -730,7 +730,7 @@ void state_values_stack_pop(state_values_stack_t * stack, size_t n)
  * @param stack
  *   state_values stack structure.
  */
-void state_values_stack_free(state_values_stack_t * stack)
+static void state_values_stack_free(state_values_stack_t * stack)
 {
      free(stack->entries);
 }
@@ -744,7 +744,7 @@ void state_values_stack_free(state_values_stack_t * stack)
  */
 static <%= @grammar.prefix %>value_t parser_user_code(uint32_t rule, state_values_stack_t * statevalues, uint32_t n_states)
 {
-    <%= @grammar.prefix %>value_t _pvalue;
+    <%= @grammar.prefix %>value_t _pvalue = {0};
 
     switch (rule)
     {
@@ -821,7 +821,7 @@ static size_t check_reduce(size_t state_id, <%= @grammar.prefix %>token_t token)
  *   can be accessed with <%= @grammar.prefix %>result().
  * @retval P_UNEXPECTED_TOKEN
  *   An unexpected token was encountered that does not match any grammar rule.
- *   The value context.token holds the unexpected token.
+ *   The value context->token holds the unexpected token.
  * @reval P_DECODE_ERROR
  *   The decoder encountered invalid text encoding.
  * @reval P_UNEXPECTED_INPUT
@@ -880,7 +880,7 @@ size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context)
             {
                 /* We shifted a RuleSet. */
                 state_values_stack_index(&statevalues, -1)->pvalue = reduced_parser_value;
-                <%= @grammar.prefix %>value_t new_parse_result;
+                <%= @grammar.prefix %>value_t new_parse_result = {0};
                 reduced_parser_value = new_parse_result;
                 reduced_rule_set = INVALID_ID;
             }
@@ -908,6 +908,7 @@ size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context)
         break;
     }
     state_values_stack_free(&statevalues);
+    return result;
 }
 
 /**
@@ -920,7 +921,7 @@ size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context)
  */
 <%= start_rule_type[1] %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context)
 {
-    return context.parse_result.v_<%= start_rule_type[0] %>;
+    return context->parse_result.v_<%= start_rule_type[0] %>;
 }
 
 /**
@@ -933,5 +934,5 @@ size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context)
  */
 <%= @grammar.prefix %>position_t <%= @grammar.prefix %>position(<%= @grammar.prefix %>context_t * context)
 {
-    return context.text_position;
+    return context->text_position;
 }

assets/parser.h.erb

@@ -8,6 +8,7 @@
 #define PROPANE_PARSER_H
 
 #include <stdint.h>
+#include <stddef.h>
 
 /**************************************************************************
  * Public types
@@ -87,6 +88,9 @@ typedef struct
     /** Input text. */
     uint8_t const * input;
 
+    /** Input text length. */
+    size_t input_length;
+
     /** Input text index (byte offset). */
     size_t input_index;
 
@@ -105,4 +109,17 @@ typedef struct
     <%= @grammar.prefix %>token_t token;
 } <%= @grammar.prefix %>context_t;
 
+void <%= @grammar.prefix %>context_init(<%= @grammar.prefix %>context_t * context, uint8_t const * input, size_t input_length);
+
+size_t <%= @grammar.prefix %>decode_code_point(uint8_t const * input, size_t input_length,
+        <%= @grammar.prefix %>code_point_t * out_code_point, uint8_t * out_code_point_length);
+
+size_t <%= @grammar.prefix %>lex(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>token_info_t * out_token_info);
+
+size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context);
+
+<%= start_rule_type[1] %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context);
+
+<%= @grammar.prefix %>position_t <%= @grammar.prefix %>position(<%= @grammar.prefix %>context_t * context);
+
 #endif

spec/propane_spec.rb

@@ -28,10 +28,10 @@ describe Propane do
       "spec/run/testparser#{name}.#{options[:language]}"
     end
     case options[:language]
+    when "c"
+      result = system(*%w[gcc -Wall -o spec/run/testparser -Ispec -Ispec/run], *parsers, *test_files)
     when "d"
       result = system(*%w[ldc2 --unittest -of spec/run/testparser -Ispec], *parsers, *test_files, "spec/testutils.d")
-    when "c"
-      result = system(*%w[gcc -o spec/run/testparser -Ispec], *parsers, *test_files)
     end
     expect(result).to be_truthy
   end

spec/test_lexer.c

@@ -0,0 +1,92 @@
+#include "testparser.h"
+#include <assert.h>
+#include <string.h>
+
+int main()
+{
+    size_t result;
+    p_code_point_t code_point;
+    uint8_t code_point_length;
+
+    result = p_decode_code_point((uint8_t const *)"5", 1u, &code_point, &code_point_length);
+    assert(result == P_SUCCESS);
+    assert(code_point == '5');
+    assert(code_point_length == 1u);
+
+    result = p_decode_code_point((uint8_t const *)"", 0u, &code_point, &code_point_length);
+    assert(result == P_EOF);
+
+    result = p_decode_code_point((uint8_t const *)"\xC2\xA9", 2u, &code_point, &code_point_length);
+    assert(result == P_SUCCESS);
+    assert(code_point == 0xA9u);
+    assert(code_point_length == 2u);
+
+    result = p_decode_code_point((uint8_t const *)"\xf0\x9f\xa7\xa1", 4u, &code_point, &code_point_length);
+    assert(result == P_SUCCESS);
+    assert(code_point == 0x1F9E1u);
+    assert(code_point_length == 4u);
+
+    result = p_decode_code_point((uint8_t const *)"\xf0\x9f\x27", 3u, &code_point, &code_point_length);
+    assert(result == P_DECODE_ERROR);
+
+    result = p_decode_code_point((uint8_t const *)"\xf0\x9f\xa7\xFF", 4u, &code_point, &code_point_length);
+    assert(result == P_DECODE_ERROR);
+
+    result = p_decode_code_point((uint8_t const *)"\xfe", 1u, &code_point, &code_point_length);
+    assert(result == P_DECODE_ERROR);
+
+
+    p_token_info_t token_info;
+    char const * input = "5 + 4 * \n677 + 567";
+    p_context_t context;
+    p_context_init(&context, (uint8_t const *)input, strlen(input));
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 0u);
+    assert(token_info.position.col == 0u);
+    assert(token_info.length == 1u);
+    assert(token_info.token == TOKEN_int);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 0u);
+    assert(token_info.position.col == 2u);
+    assert(token_info.length == 1u);
+    assert(token_info.token == TOKEN_plus);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 0u);
+    assert(token_info.position.col == 4u);
+    assert(token_info.length == 1u);
+    assert(token_info.token == TOKEN_int);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 0u);
+    assert(token_info.position.col == 6u);
+    assert(token_info.length == 1u);
+    assert(token_info.token == TOKEN_times);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 1u);
+    assert(token_info.position.col == 0u);
+    assert(token_info.length == 3u);
+    assert(token_info.token == TOKEN_int);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 1u);
+    assert(token_info.position.col == 4u);
+    assert(token_info.length == 1u);
+    assert(token_info.token == TOKEN_plus);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 1u);
+    assert(token_info.position.col == 6u);
+    assert(token_info.length == 3u);
+    assert(token_info.token == TOKEN_int);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 1u);
+    assert(token_info.position.col == 9u);
+    assert(token_info.length == 0u);
+    assert(token_info.token == TOKEN___EOF);
+
+    p_context_init(&context, (uint8_t const *)"", 0u);
+    assert(p_lex(&context, &token_info) == P_SUCCESS);
+    assert(token_info.position.row == 0u);
+    assert(token_info.position.col == 0u);
+    assert(token_info.length == 0u);
+    assert(token_info.token == TOKEN___EOF);
+
+    return 0;
+}