ocomment 0.1.0

version = 1

# NOTE: The canonical list of the languages OComment has a built-in scanner
# NOTE: for.
# NOTE:
# NOTE: `extensions` holds every suffix that selects the language, without the
# NOTE: dot, `editor_ids` the identifiers editor clients send to the LSP, and
# NOTE: `dialects` every dialect it accepts, in the order the binary
# NOTE: lists them when it refuses one. An extension that selects a dialect
# NOTE: other than `standard` says so in `extension_dialects`; `reserved_names`
# NOTE: and `shebangs` are the whole file names and interpreter names that
# NOTE: select the language when the extension does not decide it. `notes` is a
# NOTE: remark for the human listing.
# NOTE:
# NOTE: Published pre-commit hooks pass every text file to the detector;
# NOTE: tools/check_hooks.py rejects a manifest-level extension filter.
# NOTE: docs/languages.md is generated from the whole table
# NOTE: (tools/gen_docs.py), and `ocomment languages` prints the copy the
# NOTE: binary embeds.
# NOTE: rust/ocomment/tests/spec_languages.rs checks every claim below against
# NOTE: the detector and the binary.

[[languages]]
name = "rust"
editor_ids = ["rust"]
extensions = ["rs"]
dialects = ["standard"]

[[languages]]
name = "ocaml"
editor_ids = ["ocaml"]
extensions = ["ml", "mli", "mlt"]
dialects = ["standard"]
notes = "OCaml 5.5 lexical forms"

[[languages]]
name = "c"
editor_ids = ["c", "objective-c"]
extensions = ["c", "h", "m"]
dialects = ["standard", "objective-c", "gnu-c"]
extension_dialects = { m = "objective-c" }

[[languages]]
name = "cpp"
editor_ids = ["cpp", "objective-cpp", "cuda-cpp"]
extensions = ["cc", "cpp", "cxx", "hh", "hpp", "hxx", "mm", "cu", "cuh"]
dialects = ["standard", "objective-cpp", "gnu-cpp", "cuda"]
extension_dialects = { mm = "objective-cpp", cu = "cuda", cuh = "cuda" }

[[languages]]
name = "go"
editor_ids = ["go"]
extensions = ["go"]
dialects = ["standard"]

[[languages]]
name = "java"
editor_ids = ["java"]
extensions = ["java"]
dialects = ["standard"]
notes = "Unicode escape translation"

[[languages]]
name = "javascript"
editor_ids = ["javascript", "javascriptreact"]
extensions = ["js", "mjs", "cjs", "jsx"]
dialects = ["standard", "jsx"]
extension_dialects = { jsx = "jsx" }
shebangs = ["node", "deno"]

[[languages]]
name = "typescript"
editor_ids = ["typescript", "typescriptreact"]
extensions = ["ts", "mts", "cts", "tsx"]
dialects = ["standard", "tsx"]
extension_dialects = { tsx = "tsx" }

[[languages]]
name = "python"
editor_ids = ["python"]
extensions = ["py", "pyw", "pyi"]
dialects = ["standard"]
shebangs = ["python"]

[[languages]]
name = "shell"
editor_ids = ["shellscript"]
extensions = ["sh", "bash", "zsh"]
dialects = ["standard", "posix-sh", "bash53", "zsh"]
extension_dialects = { sh = "posix-sh", bash = "bash53", zsh = "zsh" }
reserved_names = [
    "Dockerfile",
    "Containerfile",
    "Makefile",
    "GNUmakefile",
    ".profile",
    ".bashrc",
    ".zshrc",
]
shebangs = ["sh", "bash", "zsh"]

[[languages]]
name = "html"
editor_ids = ["html"]
extensions = ["html", "htm", "xhtml", "shtml"]
dialects = ["standard"]
notes = "script and style bodies are scanned as JavaScript and CSS"

[[languages]]
name = "css"
editor_ids = ["css"]
extensions = ["css", "scss", "sass"]
dialects = ["standard", "scss", "sass"]
extension_dialects = { scss = "scss", sass = "sass" }

[[languages]]
name = "jsonc"
editor_ids = ["jsonc"]
extensions = ["jsonc", "json5"]
dialects = ["standard"]
reserved_names = ["tsconfig.json", "jsconfig.json"]

[[languages]]
name = "sql"
editor_ids = ["sql"]
extensions = ["sql"]
dialects = ["standard", "postgresql", "mysql", "sqlite", "t-sql", "oracle"]

[[languages]]
name = "kotlin"
editor_ids = ["kotlin"]
extensions = ["kt", "kts"]
dialects = ["standard"]

[[languages]]
name = "toml"
editor_ids = ["toml"]
extensions = ["toml"]
dialects = ["standard"]
reserved_names = [
    "Cargo.lock",
    "Pipfile",
    "poetry.lock",
    "uv.lock",
    "pdm.lock",
]
notes = "TOML v1.0.0 lexical forms"

[[languages]]
name = "lua"
editor_ids = ["lua"]
extensions = ["lua", "rockspec"]
dialects = ["standard"]
shebangs = ["lua", "luajit"]
notes = "Lua 5.4 lexical forms"

[[languages]]
name = "yaml"
editor_ids = ["yaml"]
extensions = ["yml", "yaml"]
dialects = ["standard"]
reserved_names = [
    ".clang-format",
    ".clang-tidy",
    ".yamllint",
]
notes = "YAML 1.2.2 lexical forms"

[[languages]]
name = "php"
editor_ids = ["php"]
extensions = ["php", "phtml", "phpt"]
dialects = ["standard"]
shebangs = ["php"]
notes = "PHP 8 lexical forms; the inline HTML around the tags is opaque"

[[languages]]
name = "ruby"
editor_ids = ["ruby"]
extensions = [
    "rb",
    "rbw",
    "rake",
    "gemspec",
    "ru",
    "podspec",
    "jbuilder",
    "thor",
    "rbi",
]
dialects = ["standard"]
reserved_names = [
    "Gemfile",
    "Rakefile",
    "Guardfile",
    "Capfile",
    "Vagrantfile",
    "Brewfile",
    "Podfile",
    "Fastfile",
    "Appfile",
    "Berksfile",
    "Thorfile",
    "Dangerfile",
    ".irbrc",
    ".pryrc",
]
shebangs = ["truffleruby", "jruby", "ruby"]
notes = "Ruby 3.3 lexical forms"

[[languages]]
name = "zig"
editor_ids = ["zig"]
extensions = ["zig", "zon"]
dialects = ["standard"]
notes = "Zig 0.16 lexical forms; no block comment, so `/*` is two operators"

[[languages]]
name = "r"
editor_ids = ["r"]
extensions = ["r"]
dialects = ["standard"]
reserved_names = [".Rprofile"]
shebangs = ["rscript", "r"]
notes = "R 4.3 lexical forms; `%op%` is opaque and `r\"(...)\"` is a raw string"

[[languages]]
name = "dart"
editor_ids = ["dart"]
extensions = ["dart"]
dialects = ["standard"]
shebangs = ["dart"]
notes = "Dart 3.13 lexical forms; block comments nest and `${...}` is code"

[[languages]]
name = "swift"
editor_ids = ["swift"]
extensions = ["swift"]
dialects = ["standard"]
shebangs = ["swift"]
notes = "Swift 6 lexical forms; block comments nest and `#/.../#` is a regular expression literal"

[[languages]]
name = "csharp"
editor_ids = ["csharp"]
extensions = ["cs", "csx"]
dialects = ["standard"]
shebangs = ["dotnet-script"]
notes = "C# 13 lexical forms; a line opening with `#` is a preprocessor directive and carries at most a `//` comment"

[[languages]]
name = "scala"
editor_ids = ["scala"]
extensions = ["scala", "sc"]
dialects = ["standard"]
shebangs = ["scala-cli", "scala"]
notes = "Scala 3.8 lexical forms; block comments nest, an identifier before a quote makes the string interpolate, and an XML literal is opaque text with `{...}` code"

[[languages]]
name = "vue"
editor_ids = ["vue"]
extensions = ["vue"]
dialects = ["standard"]
notes = "Vue single-file components; the template is HTML with `{{ ... }}` code and the script and style bodies are their own languages, the `lang` attribute choosing which"

[[languages]]
name = "svelte"
editor_ids = ["svelte"]
extensions = ["svelte"]
dialects = ["standard"]
notes = "Svelte components; the template is HTML with `{ ... }` code and the script and style bodies are their own languages, the `lang` attribute choosing which"

[[languages]]
name = "markdown"
editor_ids = ["markdown"]
extensions = ["md", "markdown", "rmd"]
dialects = ["standard"]
notes = "CommonMark documents; HTML comments are comments, fenced code blocks are scanned as the language their info string names, and inline and indented code are opaque"

[[languages]]
name = "perl"
editor_ids = ["perl"]
extensions = ["pl", "pm", "t"]
dialects = ["standard"]
shebangs = ["perl"]
notes = "Perl 5.38 lexical forms; quote words and regexes hide a `#`, POD blocks are opaque, and a `/` the parse context alone settles is reported as lexically ambiguous"