Open Source Haskell Source Code Analysis Tools

ShellCheck is a GPLv3 tool that provides warnings and possible suggestions for bash/sh shell scripts. ShellCheck finds bugs in your shell scripts. You can cabal, apt, dnf, pkg or brew install it locally right now. ShellCheck highlights and clarifies typical beginner's syntax mistakes and issues that cause a shell to give a cryptic error message. It shows typical intermediate level semantic problems that cause a shell to behave in a abnormally and counter-intuitively. It can also discover ssubtle caveats, corner cases and pitfalls that may cause an user's working script to fail under probable future circumstances. ShellCheck.net is always synchronized to the latest git version, and is the simplest way to give ShellCheck a go.

Unused identifies unused code in Rails, Phoenix, and other types of applications, improving developer productivity. By default, unused leverages a different memory allocator called mimalloc. For my local benchmarks, it speeds up execution by a significant amount (which is documented in the commit introducing mimalloc), but currently runs into sporadic issues on Apple M1 devices. If you run into issues with segmentation faults, consider reinstalling unused with the stock Rust allocator. It is strongly recommended you install Universal Ctags to generate tags files. Universal Ctags supports more languages and has native parsers for a good number of them, resulting in faster tags generation time.

semantic is a Haskell library and command line tool for parsing, analyzing, and comparing source code. Run semantic --help for complete list of up-to-date options. Semantic uses tree-sitter to generate parse trees, but layers in a more generalized notion of syntax terms across all supported programming languages. We'll see why this is important when we get to diffs and program analysis, but for now let's just inspect some output. It helps to have a simple program to parse. Symbols are named identifiers driven by the ASTs. This is the format that github.com uses to generate code navigation information allowing c-tags style lookup of symbolic names for fast, incremental navigation in all the supported languages. The incremental part is important because files change often so we want to be able to parse just what's changed and not have to analyze the entire project again.