[grammar test] before_comment.

tests/grammar.c

@@ -144,9 +144,10 @@ static void PrintRule(MD_Node *rule)
 typedef enum OperationFlags OperationFlags;
 enum OperationFlags
 {
-    OperationFlag_Fill   = 1<<0,
+    OperationFlag_Fill          = 1<<0,
-    OperationFlag_Markup = 1<<1,
+    OperationFlag_Markup        = 1<<1,
-    OperationFlag_Tag    = 1<<2,
+    OperationFlag_Tag           = 1<<2,
+    OperationFlag_PreComment    = 1<<3,
 };
 
 static void Extend(MD_String8 *s, char c)
@@ -160,6 +161,7 @@ static void ExpandProduction(MD_Node *production, MD_String8List *out, MD_Node *
 static void ExpandRule(MD_Node *rule, MD_String8List *out_strings, MD_Node *cur_node, OperationFlags op_flags,
                        MD_u32 max_depth, MD_u32 depth)
 {
+    MD_String8 pre_comment = {0};
     for(MD_EachNode(rule_element, rule->first_child))
     {
         MD_b32 expand = 1;
@@ -179,6 +181,8 @@ static void ExpandRule(MD_Node *rule, MD_String8List *out_strings, MD_Node *cur_
         if(expand)
         {
             MD_Node *node_to_tag = 0;
+            MD_Node *node_to_precomment = 0;
+            MD_String8 precomment = {0};
             OperationFlags old_op_flags = op_flags;
             for(MD_EachNode(tag_node, rule_element->first_tag)){
                 if(MD_StringMatch(tag_node->string, MD_S8Lit("child"), 0))
@@ -205,6 +209,12 @@ static void ExpandRule(MD_Node *rule, MD_String8List *out_strings, MD_Node *cur_
                 {
                     op_flags |= OperationFlag_Markup;
                 }
+                else if(MD_StringMatch(tag_node->string, MD_S8Lit("pre_comment"), 0))
+                {
+                    op_flags |= OperationFlag_PreComment;
+                    node_to_precomment = cur_node;
+                    cur_node = NewChild(0); // NOTE(mal): Only used to store the comment, won't be linked as a node
+                }
                 else if(MD_StringMatch(tag_node->string, MD_S8Lit(OPTIONAL_TAG), 0))
                 {
                 }
@@ -264,6 +274,12 @@ static void ExpandRule(MD_Node *rule, MD_String8List *out_strings, MD_Node *cur_
                 cur_node = node_to_tag;
             }
 
+            if(node_to_precomment)
+            {
+                node_to_precomment->comment_before = cur_node->string;
+                cur_node = node_to_precomment;
+            }
+
             op_flags = old_op_flags;
         }
     }
@@ -363,6 +379,7 @@ static MD_b32 EqualTrees(MD_Node *a, MD_Node *b)
                      MD_StringMatch(a->whole_string, b->whole_string, 0));
     result &= EqualList(a->first_tag, b->first_tag);
     result &= EqualList(a->first_child, b->first_child);
+    result &= MD_StringMatch(a->comment_before, b->comment_before, 0);
     return result;
 }
 

tests/grammar.md

@@ -88,13 +88,53 @@
 // digit           : '0'|'1'|'2'|'3'|'4'|'5'|'6'|'7'|'8'|'9'
 
 //// General labels
+//file            : general_set
+//general_set     : { [@child set] | [@child uns_set] }
+//set             : {[tag_list] untagged_set}
+//uns_set         : {[tag_list] untagged_uns_set}
+//tag_list        : '@' @tag tag ' ' [tag_list]
+//tag             : @fill id['(' general_set ')']
+//untagged_set    : @fill label | untagged_set [sep1 @sibling set] | scoped_set | untagged_uns_set sep2 @sibling set
+//untagged_uns_set: @fill label [uns_set_tail]
+//uns_set_tail    : ':' @child uns_set | ' ' @sibling uns_set | ':'  scoped_set | ' ' @sibling scoped_set
+//scoped_set      : '{' general_set '}' | alt_scope_beg general_set alt_scope_end
+//alt_scope_beg   : '(' | '['
+//alt_scope_end   : ')' | ']'
+//sep1            : ' ' | '\n' | ',' | ';'
+//sep2            : '\n'| ',' | ';'
+//id              : alpha [alphanumeric] | '_' [alphanumeric]
+//alphanumeric    : alpha [alphanumeric] | digit [alphanumeric] | '_' [alphanumeric]
+//alpha           : lowercase | uppercase
+//lowercase       : 'a'|'b'|'c'|'d'|'e'|'f'|'g'|'h'|'i'|'j'|'k'|'l'|'m'|'n'|'o'|'p'|'q'|'r'|'s'|'t'|'u'|'v'|'w'|'z'|'y'|'z'
+//uppercase       : 'A'|'B'|'C'|'D'|'E'|'F'|'G'|'H'|'I'|'J'|'K'|'L'|'M'|'N'|'O'|'P'|'Q'|'R'|'S'|'T'|'U'|'V'|'W'|'Z'|'Y'|'Z'
+//digit           : '0'|'1'|'2'|'3'|'4'|'5'|'6'|'7'|'8'|'9'
+//label           : id | integer_literal | char_literal | string_literal | symbol_label
+//integer_literal : { ['-'] natural_literal }
+//natural_literal : digit [natural_literal]
+//char_literal                    : @markup '\'' [char_literal_items] @markup '\''
+//char_literal_items              : char_literal_item [char_literal_items]
+//char_literal_item               : ascii_no_backslash_no_quotes | '"' | '\\' ascii
+//ascii                           : ascii_no_backslash_no_quotes | '\'' | '"' | '`' | '\\'
+//ascii_no_backslash_no_quotes    : digit | alpha | symbol_no_backslash_no_quotes | ' '
+//symbol_no_backslash_no_quotes   : symbol_no_backslash_no_quotes_1 | symbol_no_backslash_no_quotes_2
+//symbol_no_backslash_no_quotes_1 : '!'|'#'|'$'|'%'|'&'|'('|')'|'*'|'+'|','|'-'|'.'|'/'|':'
+//symbol_no_backslash_no_quotes_2 : ';'|'<'|'='|'>'|'?'|'@'|'['|']'|'^'|'_'|'{'|'|'|'}'|'~'
+//string_literal                  : @markup '"' [string_literal_items] @markup '"' | @markup '`' [string_literal_items] @markup '`'
+//string_literal_items            : string_literal_item [string_literal_items]
+//string_literal_item             : ascii_no_backslash_no_quotes | '\'' | '\\' ascii
+//symbol_label                    : '~'|'!'|'%'|'^'|'&'|'*'|'+'|'-'|'/'|'|'|'<'|'>'|'$'|'='|'.'|'?'|'$'
+
+//// Comments before
 file            : general_set
 general_set     : { [@child set] | [@child uns_set] }
-set             : {[tag_list] untagged_set}
+set             : {[tag_list] ['\n' @pre_comment pre_comment] untagged_set}
+pre_comment     : '/' '/' [' '] [@fill c_code_content] '\n' | '/' '*' [@fill c_code_content] '*' '/'
+c_code_content  : 'c''o''m''m''e''n''t' // TODO: Arbitrary strings, including C-style comments, as long as /* */ pairs are balanced
+cpp_code_content: 'c''o''m''m''e''n''t' // TODO: Arbitrary strings that don't start with space
 uns_set         : {[tag_list] untagged_uns_set}
 tag_list        : '@' @tag tag ' ' [tag_list]
 tag             : @fill id['(' general_set ')']
-untagged_set    : @fill label | set [sep1 @sibling set] | scoped_set | uns_set sep2 @sibling set
+untagged_set    : @fill label | untagged_set [sep1 @sibling set] | scoped_set | untagged_uns_set sep2 @sibling set
 untagged_uns_set: @fill label [uns_set_tail]
 uns_set_tail    : ':' @child uns_set | ' ' @sibling uns_set | ':'  scoped_set | ' ' @sibling scoped_set
 scoped_set      : '{' general_set '}' | alt_scope_beg general_set alt_scope_end