berchn

Support for __attribute__ With structs

2

Currently, `__attribute__` is not parsed correctly when following or preceding a struct definition: ``` typedef struct { int x; } __attribute__((__packed__)) foo_t; translation_unit [0, 0] - [3, 0] type_definition [0,...

I'm using this string: `"#define a \n"`. The following is a list representation of the parse tree including the unnamed nodes: ```lisp (:TRANSLATION-UNIT ((0 0) (0 1)) ((:PREPROC-DEF ((0 0)...

The following is the code I'm using: ```C++ class Dummy {​​​​​​​​​​ public: int foo(); int bar(int val); }​​​​​​​​​​​; class DummyTwo; ``` ``` translation_unit [0, 0] - [7, 0] class_specifier [0,...

The following is the C++ code I am using: ```C++ int a = (x->y < u && z >= w); ``` I am expecting several binary operations for the comparison,...

"source_file" allows "BLANK" to follow "_top_level_declaration"

1

`source_file` allows `BLANK` to follow `_top_level_declaration` in `src/grammar.json`. This is resulting in an AST that doesn't appear to be valid. The following code parses without error or the addition of...

Aliasing to a Grammar Symbol without a Production Produces an AST without a Corresponding Node in node-types.json

2

I've run into this issue recently in tree-sitter-python in regards to the following commit: https://github.com/tree-sitter/tree-sitter-python/commit/3324e923485177d72c1a87d315f3b31bf2294585 Notice that `alias($.expression, $.as_pattern_target)` uses `$.as_pattern_target`, and `$.as_pattern_target` does not have a production. In node-types.json,...

Incorrect Unnamed Newline Node Range

11

Unnamed nodes frequently have incorrect ranges when newline literals are used in a grammar. This issue is related to https://github.com/tree-sitter/tree-sitter-c/issues/66 but appears to affect more than just the C parser....