assets/parser.h.erb

@@ -4,7 +4,8 @@
  * This file is generated by Propane.
  */
 
-#pragma once
+#ifndef PROPANE_PARSER_H
+#define PROPANE_PARSER_H
 
 #include <stdint.h>
 #include <stddef.h>
@@ -120,3 +121,5 @@ 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

lib/propane/generator.rb

@@ -206,19 +206,14 @@ class Propane
           index = $1.to_i
           case @language
           when "c"
-            "state_values_stack_index(statevalues, -1 - (int)n_states + #{index})->pvalue.v_#{rule.components[index - 1].ptypename}"
+            "state_value_stack_index(statevalues, -1 - (int)n_states + #{index})->pvalue.v_#{rule.components[index - 1].ptypename}"
           when "d"
             "statevalues[$-1-n_states+#{index}].pvalue.v_#{rule.components[index - 1].ptypename}"
           end
         end
       else
         code = code.gsub(/\$\$/) do |match|
-          case @language
+          "out_token_info.pvalue.v_#{pattern.ptypename}"
-          when "c"
-            "out_token_info->pvalue.v_#{pattern.ptypename}"
-          when "d"
-            "out_token_info.pvalue.v_#{pattern.ptypename}"
-          end
         end
         code = code.gsub(/\$mode\(([a-zA-Z_][a-zA-Z_0-9]*)\)/) do |match|
           mode_name = $1

spec/propane_spec.rb

@@ -29,7 +29,7 @@ describe Propane do
     end
     case options[:language]
     when "c"
-      result = system(*%w[gcc -Wall -o spec/run/testparser -Ispec -Ispec/run], *parsers, *test_files, "spec/testutils.c", "-lm")
+      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")
     end
@@ -98,25 +98,7 @@ EOF
       end
 
       it "detects a lexer error when an unknown character is seen" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-ptype int;
-token int /\\d+/ <<
-  int v = 0;
-  for (size_t i = 0u; i < match_length; i++)
-  {
-    v *= 10;
-    v += (match[i] - '0');
-  }
-  $$ = v;
->>
-Start -> int <<
-  $$ = $1;
->>
-EOF
-        when "d"
-          write_grammar <<EOF
 ptype int;
 token int /\\d+/ <<
   int v;
@@ -131,7 +113,6 @@ Start -> int <<
   $$ = $1;
 >>
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_lexer_unknown_character.#{language}", language: language)
         results = run
@@ -156,61 +137,7 @@ EOF
       end
 
       it "generates a parser that does basic math - user guide example" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <math.h>
->>
-
-ptype size_t;
-
-token plus /\\+/;
-token times /\\*/;
-token power /\\*\\*/;
-token integer /\\d+/ <<
-  size_t v = 0u;
-  for (size_t i = 0u; i < match_length; i++)
-  {
-    v *= 10;
-    v += (match[i] - '0');
-  }
-  $$ = v;
->>
-token lparen /\\(/;
-token rparen /\\)/;
-drop /\\s+/;
-
-Start -> E1 <<
-  $$ = $1;
->>
-E1 -> E2 <<
-  $$ = $1;
->>
-E1 -> E1 plus E2 <<
-  $$ = $1 + $3;
->>
-E2 -> E3 <<
-  $$ = $1;
->>
-E2 -> E2 times E3 <<
-  $$ = $1 * $3;
->>
-E3 -> E4 <<
-  $$ = $1;
->>
-E3 -> E3 power E4 <<
-  $$ = (size_t)pow($1, $3);
->>
-E4 -> integer <<
-  $$ = $1;
->>
-E4 -> lparen E1 rparen <<
-  $$ = $2;
->>
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.math;
 >>
@@ -261,7 +188,6 @@ E4 -> lparen E1 rparen <<
   $$ = $2;
 >>
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_basic_math_grammar.#{language}", language: language)
         results = run
@@ -310,22 +236,7 @@ EOF
       end
 
       it "executes user code when matching lexer token" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <stdio.h>
->>
-token abc <<
-  printf("abc!\\n");
->>
-token def;
-Start -> Abcs def;
-Abcs -> ;
-Abcs -> abc Abcs;
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.stdio;
 >>
@@ -337,7 +248,6 @@ Start -> Abcs def;
 Abcs -> ;
 Abcs -> abc Abcs;
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_user_code.#{language}", language: language)
         results = run
@@ -352,20 +262,7 @@ EOF
       end
 
       it "supports a pattern statement" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <stdio.h>
->>
-token abc;
-/def/ <<
-  printf("def!\\n");
->>
-Start -> abc;
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.stdio;
 >>
@@ -375,7 +272,6 @@ token abc;
 >>
 Start -> abc;
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_pattern.#{language}", language: language)
         results = run
@@ -390,24 +286,7 @@ EOF
       end
 
       it "supports returning tokens from pattern code blocks" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <stdio.h>
->>
-token abc;
-/def/ <<
-  printf("def!\\n");
->>
-/ghi/ <<
-  printf("ghi!\\n");
-  return $token(abc);
->>
-Start -> abc;
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.stdio;
 >>
@@ -421,7 +300,6 @@ token abc;
 >>
 Start -> abc;
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_return_token_from_pattern.#{language}", language: language)
         results = run
@@ -434,31 +312,7 @@ EOF
       end
 
       it "supports lexer modes" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <stdio.h>
->>
-token abc;
-token def;
-tokenid string;
-drop /\\s+/;
-/"/ <<
-  printf("begin string mode\\n");
-  $mode(string);
->>
-string: /[^"]+/ <<
-  printf("captured string\\n");
->>
-string: /"/ <<
-  $mode(default);
-  return $token(string);
->>
-Start -> abc string def;
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.stdio;
 >>
@@ -479,7 +333,6 @@ string: /"/ <<
 >>
 Start -> abc string def;
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_lexer_modes.#{language}", language: language)
         results = run
@@ -495,26 +348,7 @@ EOF
       end
 
       it "executes user code associated with a parser rule" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <stdio.h>
->>
-token a;
-token b;
-Start -> A B <<
-  printf("Start!\\n");
->>
-A -> a <<
-  printf("A!\\n");
->>
-B -> b <<
-  printf("B!\\n");
->>
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.stdio;
 >>
@@ -530,7 +364,6 @@ B -> b <<
   writeln("B!");
 >>
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_parser_rule_user_code.#{language}", language: language)
         results = run
@@ -544,7 +377,7 @@ EOF
 
       it "parses lists" do
         write_grammar <<EOF
-ptype #{language == "c" ? "uint32_t" : "uint"};
+ptype uint;
 token a;
 Start -> As <<
   $$ = $1;
@@ -583,23 +416,7 @@ EOF
       end
 
       it "provides matched text to user code blocks" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-<<
-#include <stdio.h>
-#include <stdlib.h>
->>
-token id /[a-zA-Z_][a-zA-Z0-9_]*/ <<
-  char * t = malloc(match_length + 1);
-  strncpy(t, (char *)match, match_length);
-  printf("Matched token is %s\\n", t);
-  free(t);
->>
-Start -> id;
-EOF
-        when "d"
-          write_grammar <<EOF
 <<
 import std.stdio;
 >>
@@ -608,7 +425,6 @@ token id /[a-zA-Z_][a-zA-Z0-9_]*/ <<
 >>
 Start -> id;
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_lexer_match_text.#{language}", language: language)
         results = run
@@ -620,19 +436,7 @@ EOF
       end
 
       it "allows storing a result value for the lexer" do
-        case language
+        write_grammar <<EOF
-        when "c"
-          write_grammar <<EOF
-ptype uint64_t;
-token word /[a-z]+/ <<
-  $$ = match_length;
->>
-Start -> word <<
-  $$ = $1;
->>
-EOF
-        when "d"
-          write_grammar <<EOF
 ptype ulong;
 token word /[a-z]+/ <<
   $$ = match.length;
@@ -641,7 +445,6 @@ Start -> word <<
   $$ = $1;
 >>
 EOF
-        end
         build_parser(language: language)
         compile("spec/test_lexer_result_value.#{language}", language: language)
         results = run

spec/test_basic_math_grammar.c

@@ -1,29 +0,0 @@
-#include "testparser.h"
-#include "testutils.h"
-#include <string.h>
-
-int main()
-{
-    char const * input = "1 + 2 * 3 + 4";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert_eq(P_SUCCESS, p_parse(&context));
-    assert_eq(11, p_result(&context));
-
-    input = "1 * 2 ** 4 * 3";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert_eq(P_SUCCESS, p_parse(&context));
-    assert_eq(48, p_result(&context));
-
-    input = "(1 + 2) * 3 + 4";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert_eq(P_SUCCESS, p_parse(&context));
-    assert_eq(13, p_result(&context));
-
-    input = "(2 * 2) ** 3 + 4 + 5";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert_eq(P_SUCCESS, p_parse(&context));
-    assert_eq(73, p_result(&context));
-
-    return 0;
-}

spec/test_error_positions.c

@@ -1,39 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "a 42";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    input = "a\n123\na  a";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_UNEXPECTED_TOKEN);
-    assert(p_position(&context).row == 2);
-    assert(p_position(&context).col == 3);
-    assert(context.token == TOKEN_a);
-
-    input = "12";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_UNEXPECTED_TOKEN);
-    assert(p_position(&context).row == 0);
-    assert(p_position(&context).col == 0);
-    assert(context.token == TOKEN_num);
-
-    input = "a 12\n\nab";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_UNEXPECTED_INPUT);
-    assert(p_position(&context).row == 2);
-    assert(p_position(&context).col == 1);
-
-    input = "a 12\n\na\n\n77\na   \xAA";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_DECODE_ERROR);
-    assert(p_position(&context).row == 5);
-    assert(p_position(&context).col == 4);
-
-    return 0;
-}

spec/test_error_positions.d

@@ -33,5 +33,6 @@ unittest
     input = "a 12\n\na\n\n77\na   \xAA";
     p_context_init(&context, input);
     assert(p_parse(&context) == P_DECODE_ERROR);
+    writeln(p_position(&context));
     assert(p_position(&context) == p_position_t(5, 4));
 }

spec/test_lexer_match_text.c

@@ -1,15 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-#include <stdio.h>
-
-int main()
-{
-    char const * input = "identifier_123";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass1\n");
-
-    return 0;
-}

spec/test_lexer_modes.c

@@ -1,20 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-#include <stdio.h>
-
-int main()
-{
-    char const * input = "abc \"a string\" def";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass1\n");
-
-    input = "abc \"abc def\" def";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass2\n");
-
-    return 0;
-}

spec/test_lexer_result_value.c

@@ -1,19 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "x";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    assert(p_result(&context) == 1u);
-
-    input = "fabulous";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    assert(p_result(&context) == 8u);
-
-    return 0;
-}

spec/test_lexer_unknown_character.c

@@ -1,18 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "x";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_UNEXPECTED_INPUT);
-
-    input = "123";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    assert(p_result(&context) == 123u);
-
-    return 0;
-}

spec/test_multiple_parsers.c

@@ -1,19 +0,0 @@
-#include "testparsermyp1.h"
-#include "testparsermyp2.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input1 = "a\n1";
-    myp1_context_t context1;
-    myp1_context_init(&context1, (uint8_t const *)input1, strlen(input1));
-    assert(myp1_parse(&context1) == MYP1_SUCCESS);
-
-    char const * input2 = "bcb";
-    myp2_context_t context2;
-    myp2_context_init(&context2, (uint8_t const *)input2, strlen(input2));
-    assert(myp2_parse(&context2) == MYP2_SUCCESS);
-
-    return 0;
-}

spec/test_parser_identical_rules_lookahead.c

@@ -1,17 +0,0 @@
-#include "testparser.h"
-#include <string.h>
-#include <assert.h>
-
-int main()
-{
-    char const * input = "aba";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    input = "abb";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    return 0;
-}

spec/test_parser_rule_from_multiple_states.c

@@ -1,24 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "a";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_UNEXPECTED_TOKEN);
-    assert(p_position(&context).row == 0);
-    assert(p_position(&context).col == 1);
-    assert(context.token == TOKEN___EOF);
-
-    input = "a b";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    input = "bb";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    return 0;
-}

spec/test_parser_rule_user_code.c

@@ -1,13 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "ab";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    return 0;
-}

spec/test_parsing_lists.c

@@ -1,24 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "a";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    assert(p_result(&context) == 1u);
-
-    input = "";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    assert(p_result(&context) == 0u);
-
-    input = "aaaaaaaaaaaaaaaa";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    assert(p_result(&context) == 16u);
-
-    return 0;
-}

spec/test_pattern.c

@@ -1,20 +0,0 @@
-#include "testparser.h"
-#include <stdio.h>
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "abcdef";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass1\n");
-
-    input = "defabcdef";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass2\n");
-
-    return 0;
-}

spec/test_return_token_from_pattern.c

@@ -1,13 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "defghidef";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-
-    return 0;
-}

spec/test_user_code.c

@@ -1,20 +0,0 @@
-#include "testparser.h"
-#include <assert.h>
-#include <stdio.h>
-#include <string.h>
-
-int main()
-{
-    char const * input = "abcdef";
-    p_context_t context;
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass1\n");
-
-    input = "abcabcdef";
-    p_context_init(&context, (uint8_t const *)input, strlen(input));
-    assert(p_parse(&context) == P_SUCCESS);
-    printf("pass2\n");
-
-    return 0;
-}

spec/testutils.c

@@ -1,12 +0,0 @@
-#include <stdio.h>
-#include <assert.h>
-#include <stdbool.h>
-
-void assert_eq_size_t_i(size_t expected, size_t actual, char const * file, size_t line)
-{
-    if (expected != actual)
-    {
-        fprintf(stderr, "%s:%lu: expected %lu, got %lu\n", file, line, expected, actual);
-        assert(false);
-    }
-}

spec/testutils.h

@@ -1,7 +0,0 @@
-#pragma once
-
-void assert_eq_size_t_i(size_t expected, size_t actual, char const * file, size_t line);
-
-#define assert_eq(expected, actual) \
-    assert_eq_size_t_i(expected, actual, __FILE__, __LINE__)
-