Commit Graph

191 Commits

Author SHA1 Message Date
gingerBill 799a56bbcb Fix column in tokenizer (due to removed line) 2021-08-13 10:50:05 +01:00
gingerBill 5fd64f48ee Minor procedure rename 2021-08-08 12:48:44 +01:00
gingerBill a3b7126875 Simplify init_tokenizer_with_data 2021-08-08 12:47:45 +01:00
gingerBill 000bda8419 Reduce superfluous error messages for return statements expecting not-1 return values 2021-08-07 16:29:00 +01:00
gingerBill 545b345eea Remove optimizations in tokenizer and default to older code (same logic) (optimize later) 2021-08-03 13:33:01 +01:00
gingerBill b0e64ca7e8 Prepare tokenizer for optimizations 2021-08-02 16:47:32 +01:00
gingerBill 7f3d4cb504 Remove the literal conversion logic to the parser from the tokenizer 2021-08-02 00:53:45 +01:00
gingerBill 97be36d18a Replace line with column_minus_one in Tokenizer; Remove dead code 2021-08-02 00:26:54 +01:00
gingerBill be76da2c90 Begin optimizing tokenizer; Replace gb_utf8_decode with utf8_decode (CC but easier to change later) 2021-08-01 23:56:17 +01:00
gingerBill b1a8357f50 Clean up a bit of the tokenizer code so that the semicolon insertion rules are in one place 2021-08-01 22:41:00 +01:00
gingerBill 700624119b Give begin_error_block its own recursive mutex 2021-07-29 12:35:11 +01:00
gingerBill a5d6fda433 Define which mutexes are blocking and recursive explicitly 2021-07-27 23:14:01 +01:00
gingerBill a98eee145d Remove try; Replace try x else y with or_else(x, y) 2021-07-05 16:23:13 +01:00
gingerBill 4b831dbddd Try try and or_else built-in procedures with operators try and try else 2021-07-04 12:37:21 +01:00
gingerBill 11c565e199 Fix semicolon insertion rule for --- 2021-06-26 22:47:12 +01:00
gingerBill 9c54ed5792 Add range-based error messages to -verbose-errors
Example:
Cannot convert '(1 + 2)' to 'untyped bool' from 'untyped integer'

	x := (1 + 2) * true;
	     ^~~~~~^
2021-05-19 14:15:57 +01:00
gingerBill 5108ebf015 Replace error calls with Token to use TokenPos 2021-05-19 13:02:44 +01:00
gingerBill 86dbcb1b20 Add -verbose-errors which shows the error in the line of code 2021-05-19 12:57:30 +01:00
gingerBill ce08e832f7 Allow ..= alongside .. as a "full range" operator; Update core:odin/parser etc 2021-05-16 12:34:35 +01:00
gingerBill 7886798156 Add space 2021-05-15 19:49:48 +01:00
gingerBill 0ad599675e Improve semicolon insertion rule for dummy tokens ++ and -- 2021-05-15 19:47:00 +01:00
gingerBill 5d03bc61b8 Tokenize ++ and -- as tokens but disallow them in the parser, and give better error messages for they are used as operators/statements 2021-05-15 19:34:46 +01:00
Gitea 011c8d5cda fix Syntax Warning 2021-04-09 13:03:01 +03:00
gingerBill 54e6c50769 Implement f16 functionality 2021-04-01 10:06:00 +01:00
gingerBill 8cc4cba06c Add support for backslash \ to consume a newline 2021-03-14 12:53:57 +00:00
gingerBill 15dbc99cb9 Minimize TokenPos size by using i32 for line/column/offset and file_id instead of String
To make `i32` safe, the parser limits the file size of odin files to a maximum of 2GiB (which will be good enough for the vast vast majority of cases)
2021-03-04 16:45:30 +00:00
gingerBill 28f279329d Remove bit_field keyword and parsing logic 2021-02-23 15:29:54 +00:00
gingerBill 01313eec7f Add flags -ignore-warnings and -warnings-as-errors 2021-02-23 14:59:28 +00:00
gingerBill a652c24ac3 Remove opaque keyboard 2021-02-23 13:02:18 +00:00
gingerBill 53d8ec4d15 Add extra error checks 2021-01-18 11:34:09 +00:00
gingerBill 98c8fde098 Remove unused tokens 2020-12-06 13:17:48 +00:00
gingerBill f0683c9102 Merge branch 'master' into parser-experiments 2020-12-06 00:49:48 +00:00
gingerBill 2a232f2397 Remove the (reserved) keyword macro 2020-12-04 16:14:11 +00:00
gingerBill b6bbe29c8f Remove const as a (reserved) keyword 2020-12-04 16:04:58 +00:00
gingerBill 54fbdabc38 Add experimental -insert-semicolon functionality to tokenizer and parser 2020-11-01 15:10:06 +00:00
gingerBill 4629754f7c Inline asm expression (-llvm-api)
See https://llvm.org/docs/LangRef.html#inline-assembler-expressions
Example:
```
x := asm(i32) -> i32 {
    "bswap $0",
    "=r,r",
}(123);
```
Allowed directives `#side_effect`, `#align_stack`, `#att`, `#intel` e.g. `asm() #side_effect #intel {...}`
2020-10-24 16:32:37 +01:00
gingerBill ca818fb857 Remove comments containing unicode characters from tokenizer 2020-07-14 23:51:46 +01:00
gingerBill 3d4a3730b0 Add notin deprecation 2020-05-27 18:47:45 +01:00
gingerBill 1f31d573e4 Fix tokenization 2020-05-27 18:43:02 +01:00
gingerBill 237962182b Fix tokenizing for %% 2020-05-27 18:41:10 +01:00
gingerBill 1a0614b0d7 Improve performance of tokenization and parsing 2020-05-27 18:23:37 +01:00
gingerBill 876820789e Add rune_is_letter_or_digit for tokenizer 2020-05-27 12:54:11 +01:00
gingerBill 4e21a4d46a Optimize rune_is_* procedures for tokenizer 2020-05-27 12:43:49 +01:00
gingerBill 6ac0fb80a6 Minor tokenizer performance improvements 2020-05-27 12:32:11 +01:00
gingerBill d6bcc25b69 Fix tokenizer for 0i #658 2020-05-24 21:43:36 +01:00
gingerBill ef539696b9 Strip carriage return \r from raw string literals 2020-05-23 11:39:29 +01:00
gingerBill 86b613fb15 Remove the need for type_of, size_of, align_of, offset_of to be keywords 2020-05-22 12:32:29 +01:00
gingerBill b13423d7f7 Global variable initialization support 2020-02-23 16:25:11 +00:00
gingerBill 7d9600b740 Ternary Expr; lbAddr extra; Phi node support 2020-02-06 23:33:41 +00:00
gingerBill 159150c6d9 Allow not_in as keyword over notin, but still allow notin to work 2020-01-16 10:00:14 +00:00