belka/elna
1
0
Fork 0
Code Issues 2 Pull Requests Releases Activity

Add semicolons back

Browse Source
This commit is contained in:
Eugen Wissner 2025-03-01 14:02:01 +01:00
parent 18602d00a1
commit 87dc581679
Signed by: belka
GPG Key ID: A27FDC1E8EE902C0
2 changed files with 231 additions and 224 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
boot/parser.yy
View File

@ -126,7 +126,9 @@ along with GCC; see the file COPYING3. If not see
%type <elna::boot::variable_declaration *> formal_parameter %type <elna::boot::variable_declaration *> formal_parameter
%type <std::shared_ptr<elna::boot::top_type>> type_expression; %type <std::shared_ptr<elna::boot::top_type>> type_expression;
%type <elna::boot::traits_expression *> traits_expression; %type <elna::boot::traits_expression *> traits_expression;
%type <elna::boot::expression *> expression operand unary; %type <elna::boot::expression *> expression operand;
%type <elna::boot::unary_expression *> unary_expression;
%type <elna::boot::binary_expression *> binary_expression;
%type <std::vector<elna::boot::expression *>> expressions actual_parameter_list; %type <std::vector<elna::boot::expression *>> expressions actual_parameter_list;
%type <elna::boot::designator_expression *> designator_expression; %type <elna::boot::designator_expression *> designator_expression;
%type <elna::boot::assign_statement *> assign_statement; %type <elna::boot::assign_statement *> assign_statement;
@ -321,8 +323,11 @@ operand:
| call_expression { $$ = $1; } | call_expression { $$ = $1; }
| "(" expression ")" { $$ = $2; } | "(" expression ")" { $$ = $2; }
expression: expression:
unary { $$ = $1; } unary_expression { $$ = $1; }
| expression "*" expression | binary_expression { $$ = $1; }
| operand { $$ = $1; }
binary_expression:
expression "*" expression
{ {
$$ = new elna::boot::binary_expression(elna::boot::make_position(@2), $1, $3, $$ = new elna::boot::binary_expression(elna::boot::make_position(@2), $1, $3,
elna::boot::binary_operator::multiplication); elna::boot::binary_operator::multiplication);
@ -402,7 +407,7 @@ expression:
$$ = new elna::boot::binary_expression(elna::boot::make_position(@2), $1, $3, $$ = new elna::boot::binary_expression(elna::boot::make_position(@2), $1, $3,
elna::boot::binary_operator::shift_right); elna::boot::binary_operator::shift_right);
} }
unary: unary_expression:
"@" operand "@" operand
{ {
$$ = new elna::boot::unary_expression(elna::boot::make_position(@1), $2, $$ = new elna::boot::unary_expression(elna::boot::make_position(@1), $2,
@ -418,7 +423,6 @@ unary:
$$ = new elna::boot::unary_expression(elna::boot::make_position(@1), $2, $$ = new elna::boot::unary_expression(elna::boot::make_position(@1), $2,
elna::boot::unary_operator::minus); elna::boot::unary_operator::minus);
} }
| operand { $$ = $1; }
expressions: expressions:
expression "," expressions expression "," expressions
{ {
@ -449,11 +453,15 @@ statement:
| if_statement { $$ = $1; } | if_statement { $$ = $1; }
| call_expression { $$ = $1; } | call_expression { $$ = $1; }
statements: statements:
statement statements statement ";" statements
{ {
std::swap($$, $2); std::swap($$, $3);
$$.emplace($$.cbegin(), $1); $$.emplace($$.cbegin(), $1);
} }
| statement
{
$$.push_back($1);
}
| /* no statements */ {} | /* no statements */ {}
statement_part: statement_part:
"begin" statements { std::swap($$, $2); } "begin" statements { std::swap($$, $2); }
@ -520,15 +528,14 @@ constant_definition: identifier_definition "=" literal
$$ = new elna::boot::constant_definition(elna::boot::make_position(@1), $1.first, $1.second, $3); $$ = new elna::boot::constant_definition(elna::boot::make_position(@1), $1.first, $1.second, $3);
} }
constant_definitions: constant_definitions:
constant_definition constant_definitions constant_definition ";" constant_definitions
{ {
std::swap($$, $2); std::swap($$, $3);
$$.emplace($$.cbegin(), std::move($1)); $$.emplace($$.cbegin(), std::move($1));
} }
| constant_definition { $$.emplace_back(std::move($1)); } | /* no constant definitions */ {}
constant_part: constant_part:
/* no constant definitions */ {} {}
| "const" {}
| "const" constant_definitions { std::swap($$, $2); } | "const" constant_definitions { std::swap($$, $2); }
type_definition: identifier_definition "=" type_expression type_definition: identifier_definition "=" type_expression
{ {

424
source.elna
View File

@ -1,67 +1,67 @@
const const
SEEK_SET* = 0 SEEK_SET* = 0;
SEEK_CUR* = 1 SEEK_CUR* = 1;
SEEK_END* = 2 SEEK_END* = 2;
TOKEN_IDENTIFIER* = 1 TOKEN_IDENTIFIER* = 1;
TOKEN_IF* = 2 TOKEN_IF* = 2;
TOKEN_THEN* = 3 TOKEN_THEN* = 3;
TOKEN_ELSE* = 4 TOKEN_ELSE* = 4;
TOKEN_ELSIF* = 5 TOKEN_ELSIF* = 5;
TOKEN_WHILE* = 6 TOKEN_WHILE* = 6;
TOKEN_DO* = 7 TOKEN_DO* = 7;
TOKEN_PROC* = 8 TOKEN_PROC* = 8;
TOKEN_BEGIN* = 9 TOKEN_BEGIN* = 9;
TOKEN_END* = 10 TOKEN_END* = 10;
TOKEN_EXTERN* = 11 TOKEN_EXTERN* = 11;
TOKEN_CONST* = 12 TOKEN_CONST* = 12;
TOKEN_VAR* = 13 TOKEN_VAR* = 13;
TOKEN_ARRAY* = 14 TOKEN_ARRAY* = 14;
TOKEN_OF* = 15 TOKEN_OF* = 15;
TOKEN_TYPE* = 16 TOKEN_TYPE* = 16;
TOKEN_RECORD* = 17 TOKEN_RECORD* = 17;
TOKEN_UNION* = 18 TOKEN_UNION* = 18;
TOKEN_POINTER* = 19 TOKEN_POINTER* = 19;
TOKEN_TO* = 20 TOKEN_TO* = 20;
TOKEN_BOOLEAN* = 21 TOKEN_BOOLEAN* = 21;
TOKEN_NIL* = 22 TOKEN_NIL* = 22;
TOKEN_AND* = 23 TOKEN_AND* = 23;
TOKEN_OR* = 24 TOKEN_OR* = 24;
TOKEN_NOT* = 25 TOKEN_NOT* = 25;
TOKEN_RETURN* = 26 TOKEN_RETURN* = 26;
TOKEN_CAST* = 27 TOKEN_CAST* = 27;
TOKEN_SHIFT_LEFT* = 28 TOKEN_SHIFT_LEFT* = 28;
TOKEN_SHIFT_RIGHT* = 29 TOKEN_SHIFT_RIGHT* = 29;
TOKEN_LEFT_PAREN* = 30 TOKEN_LEFT_PAREN* = 30;
TOKEN_RIGHT_PAREN* = 31 TOKEN_RIGHT_PAREN* = 31;
TOKEN_LEFT_SQUARE* = 32 TOKEN_LEFT_SQUARE* = 32;
TOKEN_RIGHT_SQUARE* = 33 TOKEN_RIGHT_SQUARE* = 33;
TOKEN_GREATER_EQUAL* = 34 TOKEN_GREATER_EQUAL* = 34;
TOKEN_LESS_EQUAL* = 35 TOKEN_LESS_EQUAL* = 35;
TOKEN_GREATER_THAN* = 36 TOKEN_GREATER_THAN* = 36;
TOKEN_LESS_THAN* = 37 TOKEN_LESS_THAN* = 37;
TOKEN_NOT_EQUAL* = 38 TOKEN_NOT_EQUAL* = 38;
TOKEN_EQUAL* = 39 TOKEN_EQUAL* = 39;
TOKEN_SEMICOLON* = 40 TOKEN_SEMICOLON* = 40;
TOKEN_DOT* = 41 TOKEN_DOT* = 41;
TOKEN_COMMA* = 42 TOKEN_COMMA* = 42;
TOKEN_PLUS* = 43 TOKEN_PLUS* = 43;
TOKEN_MINUS* = 44 TOKEN_MINUS* = 44;
TOKEN_MULTIPLICATION* = 45 TOKEN_MULTIPLICATION* = 45;
TOKEN_DIVISION* = 46 TOKEN_DIVISION* = 46;
TOKEN_REMAINDER* = 47 TOKEN_REMAINDER* = 47;
TOKEN_ASSIGNMENT* = 48 TOKEN_ASSIGNMENT* = 48;
TOKEN_COLON* = 49 TOKEN_COLON* = 49;
TOKEN_HAT* = 50 TOKEN_HAT* = 50;
TOKEN_AT* = 51 TOKEN_AT* = 51;
TOKEN_COMMENT* = 52 TOKEN_COMMENT* = 52;
TOKEN_INTEGER* = 53 TOKEN_INTEGER* = 53;
TOKEN_WORD* = 54 TOKEN_WORD* = 54;
TOKEN_CHARACTER* = 55 TOKEN_CHARACTER* = 55;
TOKEN_STRING* = 56 TOKEN_STRING* = 56;
TOKEN_DEFER* = 57 TOKEN_DEFER* = 57;
TOKEN_EXCLAMATION* = 58 TOKEN_EXCLAMATION* = 58;
TOKEN_ARROW = 59 TOKEN_ARROW = 59;
type type
Position* = record Position* = record
@ -173,20 +173,20 @@ var
n: Word n: Word
buffer: [10]Char buffer: [10]Char
begin begin
n := 10u n := 10u;
if value = 0 then if value = 0 then
write_c('0') write_c('0')
end end;
while value <> 0 do while value <> 0 do
digit := value % 10 digit := value % 10;
value := value / 10 value := value / 10;
buffer[n] := cast(cast('0': Int) + digit: Char) buffer[n] := cast(cast('0': Int) + digit: Char);
n := n - 1u n := n - 1u
end end;
while n < 10u do while n < 10u do
n := n + 1u n := n + 1u;
write_c(buffer[n]) write_c(buffer[n])
end end
end end
@ -224,8 +224,8 @@ proc string_dup(origin: String) -> String;
var var
copy: ^Char copy: ^Char
begin begin
copy := cast(malloc(origin.length): ^Char) copy := cast(malloc(origin.length): ^Char);
strncpy(copy, origin.ptr, origin.length) strncpy(copy, origin.ptr, origin.length);
return String(copy, origin.length) return String(copy, origin.length)
end end
@ -234,9 +234,9 @@ proc string_buffer_new() -> StringBuffer;
var var
result: StringBuffer result: StringBuffer
begin begin
result.capacity := 64u result.capacity := 64u;
result.data := malloc(result.capacity) result.data := malloc(result.capacity);
result.size := 0u result.size := 0u;
return result return result
end end
@ -244,10 +244,10 @@ end
proc string_buffer_push(buffer: ^StringBuffer, char: Char); proc string_buffer_push(buffer: ^StringBuffer, char: Char);
begin begin
if buffer^.size >= buffer^.capacity then if buffer^.size >= buffer^.capacity then
buffer^.capacity := buffer^.capacity + 1024u buffer^.capacity := buffer^.capacity + 1024u;
buffer^.data := realloc(buffer^.data, buffer^.capacity) buffer^.data := realloc(buffer^.data, buffer^.capacity)
end end;
(buffer^.data + buffer^.size)^ := cast(char: Byte) (buffer^.data + buffer^.size)^ := cast(char: Byte);
buffer^.size := buffer^.size + 1u buffer^.size := buffer^.size + 1u
end end
@ -260,8 +260,8 @@ proc string_buffer_clear(buffer: ^StringBuffer) -> String;
var var
result: String result: String
begin begin
result := String(cast(buffer^.data: ^Char), buffer^.size) result := String(cast(buffer^.data: ^Char), buffer^.size);
buffer^.size := 0u buffer^.size := 0u;
return result return result
end end
@ -278,12 +278,12 @@ var
result: ^SourceFile result: ^SourceFile
file_handle: ^FILE file_handle: ^FILE
begin begin
file_handle := fopen(filename, "rb\0".ptr) file_handle := fopen(filename, "rb\0".ptr);
if file_handle <> nil then if file_handle <> nil then
result := cast(malloc(#size(SourceFile)): ^SourceFile) result := cast(malloc(#size(SourceFile)): ^SourceFile);
result^.handle := file_handle result^.handle := file_handle;
result^.size := 0u result^.size := 0u;
result^.index := 1u result^.index := 1u
end end
return result return result
@ -294,44 +294,44 @@ var
successful: Bool successful: Bool
begin begin
if escape = 'n' then if escape = 'n' then
result^ := '\n' result^ := '\n';
successful := true successful := true;
elsif escape = 'a' then elsif escape = 'a' then
result^ := '\a' result^ := '\a';
successful := true successful := true
elsif escape = 'b' then elsif escape = 'b' then
result^ := '\b' result^ := '\b';
successful := true successful := true
elsif escape = 't' then elsif escape = 't' then
result^ := '\t' result^ := '\t';
successful := true successful := true
elsif escape = 'f' then elsif escape = 'f' then
result^ := '\f' result^ := '\f';
successful := true successful := true
elsif escape = 'r' then elsif escape = 'r' then
result^ := '\r' result^ := '\r';
successful := true successful := true
elsif escape = 'v' then elsif escape = 'v' then
result^ := '\v' result^ := '\v';
successful := true successful := true
elsif escape = '\\' then elsif escape = '\\' then
result^ := '\\' result^ := '\\';
successful := true successful := true
elsif escape = '\'' then elsif escape = '\'' then
result^ := '\'' result^ := '\'';
successful := true successful := true
elsif escape = '"' then elsif escape = '"' then
result^ := '"' result^ := '"';
successful := true successful := true
elsif escape = '?' then elsif escape = '?' then
result^ := '\?' result^ := '\?';
successful := true successful := true
elsif escape = '0' then elsif escape = '0' then
result^ := '\0' result^ := '\0';
successful := true successful := true
else else
successful := false successful := false
end end;
return successful return successful
end end
@ -339,10 +339,10 @@ proc source_file_empty(source_input: ^Byte) -> Bool;
var var
source_file: ^SourceFile source_file: ^SourceFile
begin begin
source_file := cast(source_input: ^SourceFile) source_file := cast(source_input: ^SourceFile);
if source_file^.index > source_file^.size then if source_file^.index > source_file^.size then
source_file^.size := fread(cast(@source_file^.buffer: ^Byte), 1u, 1024u, source_file^.handle) source_file^.size := fread(cast(@source_file^.buffer: ^Byte), 1u, 1024u, source_file^.handle);
source_file^.index := 1u source_file^.index := 1u
end end
@ -353,7 +353,7 @@ proc source_file_head(source_input: ^Byte) -> Char;
var var
source_file: ^SourceFile source_file: ^SourceFile
begin begin
source_file := cast(source_input: ^SourceFile) source_file := cast(source_input: ^SourceFile);
return source_file^.buffer[source_file^.index] return source_file^.buffer[source_file^.index]
end end
@ -362,7 +362,7 @@ proc source_file_advance(source_input: ^Byte);
var var
source_file: ^SourceFile source_file: ^SourceFile
begin begin
source_file := cast(source_input: ^SourceFile) source_file := cast(source_input: ^SourceFile);
source_file^.index := source_file^.index + 1u source_file^.index := source_file^.index + 1u
end end
@ -377,13 +377,13 @@ end
proc source_code_advance(source_code: ^SourceCode); proc source_code_advance(source_code: ^SourceCode);
begin begin
source_code^.advance(source_code^.input) source_code^.advance(source_code^.input);
source_code^.position.column := source_code^.position.column source_code^.position.column := source_code^.position.column
end end
proc source_code_break(source_code: ^SourceCode); proc source_code_break(source_code: ^SourceCode);
begin begin
source_code^.position.line := source_code^.position.line + 1u source_code^.position.line := source_code^.position.line + 1u;
source_code^.position.column := 0u source_code^.position.column := 0u
end end
@ -396,7 +396,7 @@ begin
while not source_code_empty(source_code) and is_space(source_code_head(source_code^)) do while not source_code_empty(source_code) and is_space(source_code_head(source_code^)) do
if source_code_head(source_code^) = '\n' then if source_code_head(source_code^) = '\n' then
source_code_break(source_code) source_code_break(source_code)
end end;
source_code_advance(source_code) source_code_advance(source_code)
end end
end end
@ -410,7 +410,7 @@ var
content_length: Word content_length: Word
begin begin
while not source_code_empty(source_code) and is_ident(source_code_head(source_code^)) do while not source_code_empty(source_code) and is_ident(source_code_head(source_code^)) do
string_buffer_push(token_content, source_code_head(source_code^)) string_buffer_push(token_content, source_code_head(source_code^));
source_code_advance(source_code) source_code_advance(source_code)
end end
end end
@ -419,19 +419,19 @@ proc lex_comment(source_code: ^SourceCode, token_content: ^StringBuffer) -> Bool
var var
trailing: Word trailing: Word
begin begin
trailing := 0u trailing := 0u;
while not source_code_empty(source_code) and trailing < 2u do while not source_code_empty(source_code) and trailing < 2u do
if source_code_head(source_code^) = '*' then if source_code_head(source_code^) = '*' then
string_buffer_push(token_content, '*') string_buffer_push(token_content, '*');
trailing := 1u trailing := 1u
elsif source_code_head(source_code^) = ')' and trailing = 1u then elsif source_code_head(source_code^) = ')' and trailing = 1u then
string_buffer_pop(token_content, 1u) string_buffer_pop(token_content, 1u);
trailing := 2u trailing := 2u
else else
string_buffer_push(token_content, source_code_head(source_code^)) string_buffer_push(token_content, source_code_head(source_code^));
trailing := 0u trailing := 0u
end end;
source_code_advance(source_code) source_code_advance(source_code)
end end
@ -442,18 +442,18 @@ proc lex_character(source_code: ^SourceCode, token_content: ^Char) -> Bool;
var var
successful: Bool successful: Bool
begin begin
successful := not source_code_empty(source_code) successful := not source_code_empty(source_code);
if successful then if successful then
if source_code_head(source_code^) = '\\' then if source_code_head(source_code^) = '\\' then
source_code_advance(source_code) source_code_advance(source_code);
successful := not source_code_empty(source_code) and escape_char(source_code_head(source_code^), token_content) successful := not source_code_empty(source_code) and escape_char(source_code_head(source_code^), token_content)
else else
token_content^ := source_code_head(source_code^) token_content^ := source_code_head(source_code^);
successful := true successful := true
end end
end end;
if successful then if successful then
source_code_advance(source_code) source_code_advance(source_code)
end end
@ -467,15 +467,15 @@ var
is_valid: Bool is_valid: Bool
next_char: Char next_char: Char
begin begin
is_valid := true is_valid := true;
while is_valid and not source_code_empty(source_code) and source_code_head(source_code^) <> '"' do while is_valid and not source_code_empty(source_code) and source_code_head(source_code^) <> '"' do
is_valid := lex_character(source_code, @next_char) is_valid := lex_character(source_code, @next_char);
if is_valid then if is_valid then
string_buffer_push(token_content, next_char) string_buffer_push(token_content, next_char)
end end
end end;
if is_valid and source_code_expect(source_code, '"') then if is_valid and source_code_expect(source_code, '"') then
source_code_advance(source_code) source_code_advance(source_code)
@ -487,10 +487,10 @@ end
proc lex_number(source_code: ^SourceCode, token_content: ^Int); proc lex_number(source_code: ^SourceCode, token_content: ^Int);
begin begin
token_content^ := 0 token_content^ := 0;
while not source_code_empty(source_code) and is_digit(source_code_head(source_code^)) do while not source_code_empty(source_code) and is_digit(source_code_head(source_code^)) do
token_content^ := token_content^ * 10 + (cast(source_code_head(source_code^): Int) - cast('0': Int)) token_content^ := token_content^ * 10 + (cast(source_code_head(source_code^): Int) - cast('0': Int));
source_code_advance(source_code) source_code_advance(source_code)
end end
@ -501,9 +501,9 @@ var
current_token: ^Token current_token: ^Token
i: Word i: Word
begin begin
i := 0u i := 0u;
while i < tokens_size do while i < tokens_size do
current_token := tokens + i current_token := tokens + i;
if current_token^.kind = TOKEN_IF then if current_token^.kind = TOKEN_IF then
write_s("IF") write_s("IF")
@ -544,8 +544,8 @@ begin
elsif current_token^.kind = TOKEN_TO then elsif current_token^.kind = TOKEN_TO then
write_s("TO") write_s("TO")
elsif current_token^.kind = TOKEN_BOOLEAN then elsif current_token^.kind = TOKEN_BOOLEAN then
write_s("BOOLEAN<") write_s("BOOLEAN<");
write_b(current_token^.value.boolean_value) write_b(current_token^.value.boolean_value);
write_c('>') write_c('>')
elsif current_token^.kind = TOKEN_NIL then elsif current_token^.kind = TOKEN_NIL then
write_s("NIL") write_s("NIL")
@ -564,8 +564,8 @@ begin
elsif current_token^.kind = TOKEN_SHIFT_RIGHT then elsif current_token^.kind = TOKEN_SHIFT_RIGHT then
write_s(">>") write_s(">>")
elsif current_token^.kind = TOKEN_IDENTIFIER then elsif current_token^.kind = TOKEN_IDENTIFIER then
write_c('<') write_c('<');
write_s(current_token^.value.string) write_s(current_token^.value.string);
write_c('>') write_c('>')
elsif current_token^.kind = TOKEN_LEFT_PAREN then elsif current_token^.kind = TOKEN_LEFT_PAREN then
write_s("(") write_s("(")
@ -614,16 +614,16 @@ begin
elsif current_token^.kind = TOKEN_COMMENT then elsif current_token^.kind = TOKEN_COMMENT then
write_s("(* COMMENT *)") write_s("(* COMMENT *)")
elsif current_token^.kind = TOKEN_INTEGER then elsif current_token^.kind = TOKEN_INTEGER then
write_c('<') write_c('<');
write_i(current_token^.value.int_value) write_i(current_token^.value.int_value);
write_c('>') write_c('>')
elsif current_token^.kind = TOKEN_WORD then elsif current_token^.kind = TOKEN_WORD then
write_c('<') write_c('<');
write_i(current_token^.value.int_value) write_i(current_token^.value.int_value);
write_s("u>") write_s("u>")
elsif current_token^.kind = TOKEN_CHARACTER then elsif current_token^.kind = TOKEN_CHARACTER then
write_c('<') write_c('<');
write_i(cast(current_token^.value.char_value: Int)) write_i(cast(current_token^.value.char_value: Int));
write_s("c>") write_s("c>")
elsif current_token^.kind = TOKEN_STRING then elsif current_token^.kind = TOKEN_STRING then
write_s("\"...\"") write_s("\"...\"")
@ -634,14 +634,14 @@ begin
elsif current_token^.kind = TOKEN_ARROW then elsif current_token^.kind = TOKEN_ARROW then
write_s("->") write_s("->")
else else
write_s("UNKNOWN<") write_s("UNKNOWN<");
write_i(current_token^.kind) write_i(current_token^.kind);
write_c('>') write_c('>')
end end;
write_c(' ') write_c(' ');
i := i + 1u i := i + 1u;
end end;
write_c('\n') write_c('\n')
end end
@ -688,10 +688,10 @@ begin
elsif "to" = token_content then elsif "to" = token_content then
current_token.kind := TOKEN_TO current_token.kind := TOKEN_TO
elsif "true" = token_content then elsif "true" = token_content then
current_token.kind := TOKEN_BOOLEAN current_token.kind := TOKEN_BOOLEAN;
current_token.value.boolean_value := true current_token.value.boolean_value := true
elsif "false" = token_content then elsif "false" = token_content then
current_token.kind := TOKEN_BOOLEAN current_token.kind := TOKEN_BOOLEAN;
current_token.value.boolean_value := false current_token.value.boolean_value := false
elsif "nil" = token_content then elsif "nil" = token_content then
current_token.kind := TOKEN_NIL current_token.kind := TOKEN_NIL
@ -708,7 +708,7 @@ begin
elsif "defer" = token_content then elsif "defer" = token_content then
current_token.kind := TOKEN_DEFER current_token.kind := TOKEN_DEFER
else else
current_token.kind := TOKEN_IDENTIFIER current_token.kind := TOKEN_IDENTIFIER;
current_token.value.string := string_dup(token_content) current_token.value.string := string_dup(token_content)
end end
@ -721,39 +721,39 @@ var
first_char: Char first_char: Char
token_buffer: StringBuffer token_buffer: StringBuffer
begin begin
tokens_size^ := 0u tokens_size^ := 0u;
tokens := nil tokens := nil;
token_buffer := string_buffer_new() token_buffer := string_buffer_new();
skip_spaces(@source_code) skip_spaces(@source_code);
while not source_code_empty(@source_code) do while not source_code_empty(@source_code) do
tokens := cast(reallocarray(cast(tokens: ^Byte), tokens_size^ + 1u, #size(Token)): ^Token) tokens := cast(reallocarray(cast(tokens: ^Byte), tokens_size^ + 1u, #size(Token)): ^Token);
current_token := tokens + tokens_size^ current_token := tokens + tokens_size^;
first_char := source_code_head(source_code) first_char := source_code_head(source_code);
if is_alpha(first_char) or first_char = '_' then if is_alpha(first_char) or first_char = '_' then
lex_identifier(@source_code, @token_buffer) lex_identifier(@source_code, @token_buffer);
current_token^ := categorize_identifier(string_buffer_clear(@token_buffer)) current_token^ := categorize_identifier(string_buffer_clear(@token_buffer))
elsif is_digit(first_char) then elsif is_digit(first_char) then
lex_number(@source_code, @current_token^.value.int_value) lex_number(@source_code, @current_token^.value.int_value);
if source_code_expect(@source_code, 'u') then if source_code_expect(@source_code, 'u') then
current_token^.kind := TOKEN_WORD current_token^.kind := TOKEN_WORD;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := TOKEN_INTEGER current_token^.kind := TOKEN_INTEGER
end end
elsif first_char = '(' then elsif first_char = '(' then
source_code_advance(@source_code) source_code_advance(@source_code);
if source_code_empty(@source_code) then if source_code_empty(@source_code) then
current_token^.kind := TOKEN_LEFT_PAREN current_token^.kind := TOKEN_LEFT_PAREN
elsif source_code_head(source_code) = '*' then elsif source_code_head(source_code) = '*' then
source_code_advance(@source_code) source_code_advance(@source_code);
if lex_comment(@source_code, @token_buffer) then if lex_comment(@source_code, @token_buffer) then
current_token^.value.string := string_dup(string_buffer_clear(@token_buffer)) current_token^.value.string := string_dup(string_buffer_clear(@token_buffer));
current_token^.kind := TOKEN_COMMENT current_token^.kind := TOKEN_COMMENT
else else
current_token^.kind := 0 current_token^.kind := 0
@ -762,129 +762,129 @@ begin
current_token^.kind := TOKEN_LEFT_PAREN current_token^.kind := TOKEN_LEFT_PAREN
end end
elsif first_char = ')' then elsif first_char = ')' then
current_token^.kind := TOKEN_RIGHT_PAREN current_token^.kind := TOKEN_RIGHT_PAREN;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '\'' then elsif first_char = '\'' then
source_code_advance(@source_code) source_code_advance(@source_code);
if lex_character(@source_code, @current_token^.value.char_value) and source_code_expect(@source_code, '\'') then if lex_character(@source_code, @current_token^.value.char_value) and source_code_expect(@source_code, '\'') then
current_token^.kind := TOKEN_CHARACTER current_token^.kind := TOKEN_CHARACTER;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := 0 current_token^.kind := 0
end end
elsif first_char = '"' then elsif first_char = '"' then
source_code_advance(@source_code) source_code_advance(@source_code);
if lex_string(@source_code, @token_buffer) then if lex_string(@source_code, @token_buffer) then
current_token^.kind := TOKEN_STRING current_token^.kind := TOKEN_STRING;
current_token^.value.string := string_dup(string_buffer_clear(@token_buffer)) current_token^.value.string := string_dup(string_buffer_clear(@token_buffer))
else else
current_token^.kind := 0 current_token^.kind := 0
end end
elsif first_char = '[' then elsif first_char = '[' then
current_token^.kind := TOKEN_LEFT_SQUARE current_token^.kind := TOKEN_LEFT_SQUARE;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = ']' then elsif first_char = ']' then
current_token^.kind := TOKEN_RIGHT_SQUARE current_token^.kind := TOKEN_RIGHT_SQUARE;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '>' then elsif first_char = '>' then
source_code_advance(@source_code) source_code_advance(@source_code);
if source_code_empty(@source_code) then if source_code_empty(@source_code) then
current_token^.kind := TOKEN_GREATER_THAN current_token^.kind := TOKEN_GREATER_THAN
elsif source_code_head(source_code) = '=' then elsif source_code_head(source_code) = '=' then
current_token^.kind := TOKEN_GREATER_EQUAL current_token^.kind := TOKEN_GREATER_EQUAL;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif source_code_head(source_code) = '>' then elsif source_code_head(source_code) = '>' then
current_token^.kind := TOKEN_SHIFT_RIGHT current_token^.kind := TOKEN_SHIFT_RIGHT;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := TOKEN_GREATER_THAN current_token^.kind := TOKEN_GREATER_THAN
end end
elsif first_char = '<' then elsif first_char = '<' then
source_code_advance(@source_code) source_code_advance(@source_code);
if source_code_empty(@source_code) then if source_code_empty(@source_code) then
current_token^.kind := TOKEN_LESS_THAN current_token^.kind := TOKEN_LESS_THAN
elsif source_code_head(source_code) = '=' then elsif source_code_head(source_code) = '=' then
current_token^.kind := TOKEN_LESS_EQUAL current_token^.kind := TOKEN_LESS_EQUAL;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif source_code_head(source_code) = '<' then elsif source_code_head(source_code) = '<' then
current_token^.kind := TOKEN_SHIFT_LEFT current_token^.kind := TOKEN_SHIFT_LEFT;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif source_code_head(source_code) = '>' then elsif source_code_head(source_code) = '>' then
current_token^.kind := TOKEN_NOT_EQUAL current_token^.kind := TOKEN_NOT_EQUAL;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := TOKEN_LESS_THAN current_token^.kind := TOKEN_LESS_THAN
end end
elsif first_char = '=' then elsif first_char = '=' then
current_token^.kind := TOKEN_EQUAL current_token^.kind := TOKEN_EQUAL;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = ';' then elsif first_char = ';' then
current_token^.kind := TOKEN_SEMICOLON current_token^.kind := TOKEN_SEMICOLON;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '.' then elsif first_char = '.' then
current_token^.kind := TOKEN_DOT current_token^.kind := TOKEN_DOT;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = ',' then elsif first_char = ',' then
current_token^.kind := TOKEN_COMMA current_token^.kind := TOKEN_COMMA;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '+' then elsif first_char = '+' then
current_token^.kind := TOKEN_PLUS current_token^.kind := TOKEN_PLUS;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '-' then elsif first_char = '-' then
source_code_advance(@source_code) source_code_advance(@source_code);
if source_code_empty(@source_code) then if source_code_empty(@source_code) then
current_token^.kind := TOKEN_MINUS current_token^.kind := TOKEN_MINUS
elsif source_code_head(source_code) = '>' then elsif source_code_head(source_code) = '>' then
current_token^.kind := TOKEN_ARROW current_token^.kind := TOKEN_ARROW;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := TOKEN_MINUS current_token^.kind := TOKEN_MINUS
end end
elsif first_char = '*' then elsif first_char = '*' then
current_token^.kind := TOKEN_MULTIPLICATION current_token^.kind := TOKEN_MULTIPLICATION;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '/' then elsif first_char = '/' then
current_token^.kind := TOKEN_DIVISION current_token^.kind := TOKEN_DIVISION;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '%' then elsif first_char = '%' then
current_token^.kind := TOKEN_REMAINDER current_token^.kind := TOKEN_REMAINDER;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = ':' then elsif first_char = ':' then
source_code_advance(@source_code) source_code_advance(@source_code);
if source_code_empty(@source_code) then if source_code_empty(@source_code) then
current_token^.kind := TOKEN_COLON current_token^.kind := TOKEN_COLON
elsif source_code_head(source_code) = '=' then elsif source_code_head(source_code) = '=' then
current_token^.kind := TOKEN_ASSIGNMENT current_token^.kind := TOKEN_ASSIGNMENT;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := TOKEN_COLON current_token^.kind := TOKEN_COLON
end end
elsif first_char = '^' then elsif first_char = '^' then
current_token^.kind := TOKEN_HAT current_token^.kind := TOKEN_HAT;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '@' then elsif first_char = '@' then
current_token^.kind := TOKEN_AT current_token^.kind := TOKEN_AT;
source_code_advance(@source_code) source_code_advance(@source_code)
elsif first_char = '!' then elsif first_char = '!' then
current_token^.kind := TOKEN_EXCLAMATION current_token^.kind := TOKEN_EXCLAMATION;
source_code_advance(@source_code) source_code_advance(@source_code)
else else
current_token^.kind := 0 current_token^.kind := 0;
source_code_advance(@source_code) source_code_advance(@source_code)
end end;
if current_token^.kind <> 0 then if current_token^.kind <> 0 then
tokens_size^ := tokens_size^ + 1u tokens_size^ := tokens_size^ + 1u;
skip_spaces(@source_code) skip_spaces(@source_code)
else else
write_s("Lexical analysis error on \"") write_s("Lexical analysis error on \"");
write_c(first_char) write_c(first_char);
write_s("\".\n") write_s("\".\n")
end end
end end
@ -898,14 +898,14 @@ var
i: Int i: Int
result: ^CommandLine result: ^CommandLine
begin begin
i := 1 i := 1;
result := cast(malloc(#size(CommandLine)): ^CommandLine) result := cast(malloc(#size(CommandLine)): ^CommandLine);
result^.tokenize := false result^.tokenize := false;
result^.syntax_tree := false result^.syntax_tree := false;
result^.input := nil result^.input := nil;
while i < argc and result <> nil do while i < argc and result <> nil do
parameter := argv + i parameter := argv + i;
if strcmp(parameter^, "--tokenize\0".ptr) = 0 then if strcmp(parameter^, "--tokenize\0".ptr) = 0 then
result^.tokenize := true result^.tokenize := true
@ -914,19 +914,19 @@ begin
elsif parameter^^ <> '-' then elsif parameter^^ <> '-' then
result^.input := parameter^ result^.input := parameter^
else else
write_s("Fatal error: Unknown command line options:") write_s("Fatal error: Unknown command line options:");
write_c(' ') write_c(' ');
write_z(parameter^) write_z(parameter^);
write_s(".\n") write_s(".\n");
result := nil result := nil
end end;
i := i + 1 i := i + 1
end end;
if result <> nil and result^.input = nil then if result <> nil and result^.input = nil then
write_s("Fatal error: no input files.\n") write_s("Fatal error: no input files.\n");
result := nil result := nil
end end
@ -941,30 +941,30 @@ var
command_line: ^CommandLine command_line: ^CommandLine
return_code: Int return_code: Int
begin begin
return_code := 0 return_code := 0;
command_line := parse_command_line(argc, argv) command_line := parse_command_line(argc, argv);
if command_line = nil then if command_line = nil then
return_code := 2 return_code := 2
end end;
if return_code = 0 then if return_code = 0 then
source_code.position := make_position() source_code.position := make_position();
source_code.input := cast(read_source(command_line^.input): ^Byte) source_code.input := cast(read_source(command_line^.input): ^Byte);
source_code.empty := source_file_empty source_code.empty := source_file_empty;
source_code.head := source_file_head source_code.head := source_file_head;
source_code.advance := source_file_advance source_code.advance := source_file_advance;
if source_code.input = nil then if source_code.input = nil then
perror(command_line^.input) perror(command_line^.input);
return_code := 3 return_code := 3
end end
end end;
if return_code = 0 then if return_code = 0 then
tokens := tokenize(source_code, @tokens_size) tokens := tokenize(source_code, @tokens_size);
fclose(cast(source_code.input: ^SourceFile)^.handle) fclose(cast(source_code.input: ^SourceFile)^.handle);
if command_line^.tokenize then if command_line^.tokenize then
print_tokens(tokens, tokens_size) print_tokens(tokens, tokens_size)