Auto generate docs for language support

Cargo.lock

@@ -1264,5 +1264,7 @@ checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
 name = "xtask"
 version = "0.5.0"
 dependencies = [
+ "helix-core",
  "helix-term",
+ "toml",
 ]

book/src/SUMMARY.md

@@ -4,6 +4,7 @@
 - [Usage](./usage.md)
   - [Keymap](./keymap.md)
   - [Commands](./commands.md)
+  - [Language Support](./lang-support.md)
 - [Migrating from Vim](./from-vim.md)
 - [Configuration](./configuration.md)
   - [Themes](./themes.md)

book/src/generated/lang-support.md

@@ -0,0 +1,41 @@
+| Language | Syntax Highlighting | Treesitter Textobjects | Auto Indent | Default LSP |
+| --- | --- | --- | --- | --- |
+| Bash | ✓ |  |  | `bash-language-server` |
+| C | ✓ |  |  | `clangd` |
+| C# | ✓ |  |  |  |
+| CMake | ✓ |  |  | `cmake-language-server` |
+| C++ | ✓ |  |  | `clangd` |
+| CSS | ✓ |  |  |  |
+| Elixir | ✓ |  |  | `elixir-ls` |
+| GLSL | ✓ |  | ✓ |  |
+| Go | ✓ | ✓ | ✓ | `gopls` |
+| HTML | ✓ |  |  |  |
+| Java | ✓ |  |  |  |
+| JavaScript | ✓ |  | ✓ |  |
+| JSON | ✓ |  | ✓ |  |
+| Julia | ✓ |  |  | `julia` |
+| LaTeX | ✓ |  |  |  |
+| Ledger | ✓ |  |  |  |
+| LLVM | ✓ |  |  |  |
+| Lua | ✓ |  | ✓ |  |
+| Mint |  |  |  | `mint` |
+| Nix | ✓ |  | ✓ | `rnix-lsp` |
+| OCaml | ✓ |  | ✓ |  |
+| OCaml-Interface | ✓ |  |  |  |
+| Perl | ✓ | ✓ |  |  |
+| PHP | ✓ |  | ✓ |  |
+| Prolog |  |  |  | `swipl` |
+| Protobuf | ✓ |  | ✓ |  |
+| Python | ✓ | ✓ | ✓ | `pylsp` |
+| Racket |  |  |  | `racket` |
+| Ruby | ✓ |  |  | `solargraph` |
+| Rust | ✓ | ✓ | ✓ | `rust-analyzer` |
+| Svelte | ✓ |  | ✓ | `svelteserver` |
+| TOML | ✓ |  |  |  |
+| TSQ | ✓ |  |  |  |
+| TSX | ✓ |  |  | `typescript-language-server` |
+| TypeScript | ✓ |  | ✓ | `typescript-language-server` |
+| Vue | ✓ |  |  |  |
+| WGSL | ✓ |  |  |  |
+| YAML | ✓ |  | ✓ |  |
+| Zig | ✓ |  | ✓ | `zls` |

book/src/lang-support.md

@@ -0,0 +1,10 @@
+# Language Support
+
+For more information like arguments passed to default LSP server,
+extensions assosciated with a filetype, custom LSP settings, filetype
+specific indent settings, etc see the default
+[`languages.toml`][languages.toml] file.
+
+{{#include ./generated/lang-support.md}}
+
+[languages.toml]: https://github.com/helix-editor/helix/blob/master/languages.toml

docs/CONTRIBUTING.md

@@ -23,7 +23,7 @@ like the list of `:commands` and supported languages. To generate these
 files, run
 
 ```shell
-cargo xtask bookgen
+cargo xtask docgen
 ```
 
 inside the project. We use [xtask][xtask] as an ad-hoc task runner and

helix-core/src/indent.rs

@@ -452,6 +452,7 @@ where
                 file_types: vec!["rs".to_string()],
                 shebangs: vec![],
                 language_id: "Rust".to_string(),
+                display_name: "Rust".to_string(),
                 highlight_config: OnceCell::new(),
                 config: None,
                 //

helix-core/src/syntax.rs

@@ -50,7 +50,8 @@ pub struct Configuration {
 #[serde(rename_all = "kebab-case", deny_unknown_fields)]
 pub struct LanguageConfiguration {
     #[serde(rename = "name")]
-    pub language_id: String,
+    pub language_id: String, // c-sharp, rust
+    pub display_name: String,    // C#, Rust
     pub scope: String,           // source.rust
     pub file_types: Vec<String>, // filename ends_with? <Gemfile, rb, etc>
     #[serde(default)]

languages.toml

@@ -1,5 +1,6 @@
 [[language]]
 name = "rust"
+display-name = "Rust"
 scope = "source.rust"
 injection-regex = "rust"
 file-types = ["rs"]
@@ -14,6 +15,7 @@ procMacro = { enable = false }
 
 [[language]]
 name = "toml"
+display-name = "TOML"
 scope = "source.toml"
 injection-regex = "toml"
 file-types = ["toml"]
@@ -24,6 +26,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "protobuf"
+display-name = "Protobuf"
 scope = "source.proto"
 injection-regex = "protobuf"
 file-types = ["proto"]
@@ -34,6 +37,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "elixir"
+display-name = "Elixir"
 scope = "source.elixir"
 injection-regex = "elixir"
 file-types = ["ex", "exs"]
@@ -46,6 +50,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "mint"
+display-name = "Mint"
 scope = "source.mint"
 injection-regex = "mint"
 file-types = ["mint"]
@@ -58,6 +63,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "json"
+display-name = "JSON"
 scope = "source.json"
 injection-regex = "json"
 file-types = ["json"]
@@ -67,6 +73,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "c"
+display-name = "C"
 scope = "source.c"
 injection-regex = "c"
 file-types = ["c"] # TODO: ["h"]
@@ -78,6 +85,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "cpp"
+display-name = "C++"
 scope = "source.cpp"
 injection-regex = "cpp"
 file-types = ["cc", "hh", "cpp", "hpp", "h", "ipp", "tpp", "cxx", "hxx", "ixx", "txx", "ino"]
@@ -89,6 +97,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "c-sharp"
+display-name = "C#"
 scope = "source.csharp"
 injection-regex = "c-?sharp"
 file-types = ["cs"]
@@ -99,6 +108,7 @@ indent = { tab-width = 4, unit = "\t" }
 
 [[language]]
 name = "go"
+display-name = "Go"
 scope = "source.go"
 injection-regex = "go"
 file-types = ["go"]
@@ -112,6 +122,7 @@ indent = { tab-width = 4, unit = "\t" }
 
 [[language]]
 name = "javascript"
+display-name = "JavaScript"
 scope = "source.js"
 injection-regex = "^(js|javascript)$"
 file-types = ["js", "mjs"]
@@ -124,6 +135,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "typescript"
+display-name = "TypeScript"
 scope = "source.ts"
 injection-regex = "^(ts|typescript)$"
 file-types = ["ts"]
@@ -136,6 +148,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "tsx"
+display-name = "TSX"
 scope = "source.tsx"
 injection-regex = "^(tsx)$" # |typescript
 file-types = ["tsx"]
@@ -147,6 +160,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "css"
+display-name = "CSS"
 scope = "source.css"
 injection-regex = "css"
 file-types = ["css"]
@@ -156,6 +170,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "html"
+display-name = "HTML"
 scope = "text.html.basic"
 injection-regex = "html"
 file-types = ["html"]
@@ -165,6 +180,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "python"
+display-name = "Python"
 scope = "source.python"
 injection-regex = "python"
 file-types = ["py"]
@@ -178,6 +194,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "nix"
+display-name = "Nix"
 scope = "source.nix"
 injection-regex = "nix"
 file-types = ["nix"]
@@ -190,6 +207,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "ruby"
+display-name = "Ruby"
 scope = "source.ruby"
 injection-regex = "ruby"
 file-types = ["rb"]
@@ -202,6 +220,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "bash"
+display-name = "Bash"
 scope = "source.bash"
 injection-regex = "bash"
 file-types = ["sh", "bash"]
@@ -214,6 +233,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "php"
+display-name = "PHP"
 scope = "source.php"
 injection-regex = "php"
 file-types = ["php"]
@@ -224,6 +244,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "latex"
+display-name = "LaTeX"
 scope = "source.tex"
 injection-regex = "tex"
 file-types = ["tex"]
@@ -234,6 +255,7 @@ indent = { tab-width = 4, unit = "\t" }
 
 [[language]]
 name = "julia"
+display-name = "Julia"
 scope = "source.julia"
 injection-regex = "julia"
 file-types = ["jl"]
@@ -259,6 +281,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "java"
+display-name = "Java"
 scope = "source.java"
 injection-regex = "java"
 file-types = ["java"]
@@ -267,6 +290,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "ledger"
+display-name = "Ledger"
 scope = "source.ledger"
 injection-regex = "ledger"
 file-types = ["ldg", "ledger", "journal"]
@@ -276,6 +300,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "ocaml"
+display-name = "OCaml"
 scope = "source.ocaml"
 injection-regex = "ocaml"
 file-types = ["ml"]
@@ -286,6 +311,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "ocaml-interface"
+display-name = "OCaml-Interface"
 scope = "source.ocaml.interface"
 file-types = ["mli"]
 shebangs = []
@@ -295,6 +321,7 @@ indent = { tab-width = 2, unit = "  "}
 
 [[language]]
 name = "lua"
+display-name = "Lua"
 scope = "source.lua"
 file-types = ["lua"]
 shebangs = ["lua"]
@@ -304,6 +331,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "svelte"
+display-name = "Svelte"
 scope = "source.svelte"
 injection-regex = "svelte"
 file-types = ["svelte"]
@@ -314,6 +342,7 @@ language-server = { command = "svelteserver", args = ["--stdio"] }
 
 [[language]]
 name = "vue"
+display-name = "Vue"
 scope = "source.vue"
 injection-regex = "vue"
 file-types = ["vue"]
@@ -322,6 +351,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "yaml"
+display-name = "YAML"
 scope = "source.yaml"
 file-types = ["yml", "yaml"]
 roots = []
@@ -330,6 +360,7 @@ indent = { tab-width = 2, unit = "  " }
 
 # [[language]]
 # name = "haskell"
+# display-name = "Haskell"
 # scope = "source.haskell"
 # injection-regex = "haskell"
 # file-types = ["hs"]
@@ -340,6 +371,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "zig"
+display-name = "Zig"
 scope = "source.zig"
 injection-regex = "zig"
 file-types = ["zig"]
@@ -352,6 +384,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "prolog"
+display-name = "Prolog"
 scope = "source.prolog"
 roots = []
 file-types = ["pl", "prolog"]
@@ -365,6 +398,7 @@ language-server = { command = "swipl", args = [
 
 [[language]]
 name = "tsq"
+display-name = "TSQ"
 scope = "source.tsq"
 file-types = ["scm"]
 roots = []
@@ -373,6 +407,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "cmake"
+display-name = "CMake"
 scope = "source.cmake"
 file-types = ["cmake", "CMakeLists.txt"]
 roots = []
@@ -382,6 +417,7 @@ language-server = { command = "cmake-language-server" }
 
 [[language]]
 name = "glsl"
+display-name = "GLSL"
 scope = "source.glsl"
 file-types = ["glsl", "vert", "tesc", "tese", "geom", "frag", "comp" ]
 roots = []
@@ -390,6 +426,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "perl"
+display-name = "Perl"
 scope = "source.perl"
 file-types = ["pl", "pm"]
 shebangs = ["perl"]
@@ -399,6 +436,7 @@ indent = { tab-width = 2, unit = "  " }
 
 [[language]]
 name = "racket"
+display-name = "Racket"
 scope = "source.rkt"
 roots = []
 file-types = ["rkt"]
@@ -408,6 +446,7 @@ language-server = { command = "racket", args = ["-l", "racket-langserver"] }
 
 [[language]]
 name = "wgsl"
+display-name = "WGSL"
 scope = "source.wgsl"
 file-types = ["wgsl"]
 roots = []
@@ -416,6 +455,7 @@ indent = { tab-width = 4, unit = "    " }
 
 [[language]]
 name = "llvm"
+display-name = "LLVM"
 scope = "source.llvm"
 roots = []
 file-types = ["ll"]

xtask/Cargo.toml

@@ -7,3 +7,5 @@ edition = "2021"
 
 [dependencies]
 helix-term = { version = "0.5", path = "../helix-term" }
+helix-core = { version = "0.5", path = "../helix-core" }
+toml = "0.5"

xtask/src/main.rs

@@ -1,17 +1,138 @@
-use std::env;
+use std::{env, error::Error};
+
+type DynError = Box<dyn Error>;
+
+pub mod helpers {
+    use std::{
+        fmt::Display,
+        path::{Path, PathBuf},
+    };
+
+    use crate::path;
+    use helix_core::syntax::Configuration as LangConfig;
+
+    #[derive(Copy, Clone)]
+    pub enum TsFeature {
+        Highlight,
+        TextObjects,
+        AutoIndent,
+    }
+
+    impl TsFeature {
+        pub fn all() -> &'static [Self] {
+            &[Self::Highlight, Self::TextObjects, Self::AutoIndent]
+        }
+
+        pub fn runtime_filename(&self) -> &'static str {
+            match *self {
+                Self::Highlight => "highlights.scm",
+                Self::TextObjects => "textobjects.scm",
+                Self::AutoIndent => "indents.toml",
+            }
+        }
+    }
+
+    impl Display for TsFeature {
+        fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+            write!(
+                f,
+                "{}",
+                match *self {
+                    Self::Highlight => "Syntax Highlighting",
+                    Self::TextObjects => "Treesitter Textobjects",
+                    Self::AutoIndent => "Auto Indent",
+                }
+            )
+        }
+    }
+
+    /// Get the list of languages that support a particular tree-sitter
+    /// based feature.
+    pub fn ts_lang_support(feat: TsFeature) -> Vec<String> {
+        let queries_dir = path::ts_queries();
+
+        find_files(&queries_dir, feat.runtime_filename())
+            .iter()
+            .map(|f| {
+                // .../helix/runtime/queries/python/highlights.scm
+                let tail = f.strip_prefix(&queries_dir).unwrap(); // python/highlights.scm
+                let lang = tail.components().next().unwrap(); // python
+                lang.as_os_str().to_string_lossy().to_string()
+            })
+            .collect()
+    }
+
+    /// Get the list of languages that have any form of tree-sitter
+    /// queries defined in the runtime directory.
+    pub fn langs_with_ts_queries() -> Vec<String> {
+        std::fs::read_dir(path::ts_queries())
+            .unwrap()
+            .filter_map(|entry| {
+                let entry = entry.ok()?;
+                entry
+                    .file_type()
+                    .ok()?
+                    .is_dir()
+                    .then(|| entry.file_name().to_string_lossy().to_string())
+            })
+            .collect()
+    }
+
+    // naive implementation, but suffices for our needs
+    pub fn find_files(dir: &Path, filename: &str) -> Vec<PathBuf> {
+        std::fs::read_dir(dir)
+            .unwrap()
+            .filter_map(|entry| {
+                let path = entry.ok()?.path();
+                if path.is_dir() {
+                    Some(find_files(&path, filename))
+                } else {
+                    (path.file_name()?.to_string_lossy() == filename).then(|| vec![path])
+                }
+            })
+            .flatten()
+            .collect()
+    }
+
+    pub fn lang_config() -> LangConfig {
+        let bytes = std::fs::read(path::lang_config()).unwrap();
+        toml::from_slice(&bytes).unwrap()
+    }
+}
 
 pub mod md_gen {
-    use super::path;
+    use crate::DynError;
+
+    use crate::helpers;
+    use crate::path;
     use std::fs;
 
     use helix_term::commands::cmd::TYPABLE_COMMAND_LIST;
 
     pub const TYPABLE_COMMANDS_MD_OUTPUT: &str = "typable-cmd.md";
+    pub const LANG_SUPPORT_MD_OUTPUT: &str = "lang-support.md";
 
-    pub fn typable_commands() -> String {
+    fn md_table_heading(cols: &[String]) -> String {
+        let mut header = String::new();
+        header += &md_table_row(cols);
+        header += &md_table_row(&vec!["---".to_string(); cols.len()]);
+        header
+    }
+
+    fn md_table_row(cols: &[String]) -> String {
+        "| ".to_owned() + &cols.join(" | ") + " |\n"
+    }
+
+    fn md_mono(s: &str) -> String {
+        format!("`{}`", s)
+    }
+
+    pub fn typable_commands() -> Result<String, DynError> {
         let mut md = String::new();
-        md.push_str("| Name | Description |\n");
+        md.push_str(&md_table_heading(&[
-        md.push_str("| ---  | ---         |\n");
+            "Name".to_owned(),
+            "Description".to_owned(),
+        ]));
 
         let cmdify = |s: &str| format!("`:{}`", s);
 
@@ -22,11 +143,72 @@ pub mod md_gen {
                 .collect::<Vec<_>>()
                 .join(", ");
 
-            let entry = format!("| {} | {} |\n", names, cmd.doc);
+            md.push_str(&md_table_row(&[names.to_owned(), cmd.doc.to_owned()]));
-            md.push_str(&entry);
+        }
+
+        Ok(md)
+    }
+
+    pub fn lang_features() -> Result<String, DynError> {
+        let mut md = String::new();
+        let ts_features = helpers::TsFeature::all();
+
+        let mut cols = vec!["Language".to_owned()];
+        cols.append(
+            &mut ts_features
+                .iter()
+                .map(|t| t.to_string())
+                .collect::<Vec<_>>(),
+        );
+        cols.push("Default LSP".to_owned());
+
+        md.push_str(&md_table_heading(&cols));
+        let config = helpers::lang_config();
+
+        let mut langs = config
+            .language
+            .iter()
+            .map(|l| l.language_id.clone())
+            .collect::<Vec<_>>();
+        langs.sort_unstable();
+
+        let mut ts_features_to_langs = Vec::new();
+        for &feat in ts_features {
+            ts_features_to_langs.push((feat, helpers::ts_lang_support(feat)));
+        }
+
+        let mut row = Vec::new();
+        for lang in langs {
+            let lc = config
+                .language
+                .iter()
+                .find(|l| l.language_id == lang)
+                .unwrap(); // lang comes from config
+            row.push(lc.display_name.clone());
+
+            for (_feat, support_list) in &ts_features_to_langs {
+                row.push(
+                    if support_list.contains(&lang) {
+                        "✓"
+                    } else {
+                        ""
                     }
+                    .to_owned(),
+                );
+            }
+            row.push(
+                lc.language_server
+                    .as_ref()
+                    .map(|s| s.command.clone())
+                    .map(|c| md_mono(&c))
+                    .unwrap_or_default(),
+            );
 
-        md
+            md.push_str(&md_table_row(&row));
+            row.clear();
+        }
+
+        Ok(md)
     }
 
     pub fn write(filename: &str, data: &str) {
@@ -49,37 +231,46 @@ pub mod path {
     pub fn book_gen() -> PathBuf {
         project_root().join("book/src/generated/")
     }
+
+    pub fn ts_queries() -> PathBuf {
+        project_root().join("runtime/queries")
+    }
+
+    pub fn lang_config() -> PathBuf {
+        project_root().join("languages.toml")
+    }
 }
 
 pub mod tasks {
-    use super::md_gen;
+    use crate::md_gen;
+    use crate::DynError;
 
-    pub fn bookgen() {
+    pub fn docgen() -> Result<(), DynError> {
-        md_gen::write(
+        use md_gen::*;
-            md_gen::TYPABLE_COMMANDS_MD_OUTPUT,
+        write(TYPABLE_COMMANDS_MD_OUTPUT, &typable_commands()?);
-            &md_gen::typable_commands(),
+        write(LANG_SUPPORT_MD_OUTPUT, &lang_features()?);
-        );
+        Ok(())
     }
 
     pub fn print_help() {
         println!(
             "
-Usage: Run with `cargo xtask <task>`, eg. `cargo xtask bookgen`.
+Usage: Run with `cargo xtask <task>`, eg. `cargo xtask docgen`.
 
     Tasks:
-        bookgen: Generate files to be included in the mdbook output.
+        docgen: Generate files to be included in the mdbook output.
 "
         );
     }
 }
 
-fn main() -> Result<(), String> {
+fn main() -> Result<(), DynError> {
     let task = env::args().nth(1);
     match task {
         None => tasks::print_help(),
         Some(t) => match t.as_str() {
-            "bookgen" => tasks::bookgen(),
+            "docgen" => tasks::docgen()?,
-            invalid => return Err(format!("Invalid task name: {}", invalid)),
+            invalid => return Err(format!("Invalid task name: {}", invalid).into()),
         },
     };
     Ok(())