Overview
Comby provides a lightweight way of matching syntactic structures of a program’s parse tree, like expressions and function blocks. Comby is language-aware and understands basic syntax of code, strings, and comment syntax in many languages.
The bottom line: Comby lets you search and change many kinds of code structures with greater ease than what regular expressions alone allow.
The basic idea
Comby works by defining general parsers for code constructs like balanced (...),
{...}, and [...], as well as string literals and comments. It doesn’t
implement a full parser for every language, but rather a more general grammar so
that it works on a lot of languages (just like regular expressions work on the
text of every language).
Does it work on my language?
Comby implements basic parsers for the following languages:
| Bash | C/C++ | C# | Clojure | Coq | CSS | Dart | Elm | Elixir |
| Erlang | Fortran | F# | Go | Haskell | HTML | Java | JavaScript | JSX |
| JSON | Kotlin | Julia | LaTeX | Lisp | Nim | MATLAB | Move | OCaml |
| Pascal | PHP | Plain Text | Python | R | ReScript | Ruby | Rust | Scala |
| Solidity | SQL | Terraform | Swift | TSX | TypeScript | XML | Zig |
Comby also implements a generic matcher that works as a fallback parser for data formats similar to JSON, new languages, and existing ones that may not have explicit support yet (like YAML or VHDL). You can also write your own language definition if the generic matcher falls short.