chore(deps): bump serde_json from 1.0.140 to 1.0.145

Bumps [serde_json](https://github.com/serde-rs/json) from 1.0.140 to 1.0.145.
- [Release notes](https://github.com/serde-rs/json/releases)
- [Commits](https://github.com/serde-rs/json/compare/v1.0.140...v1.0.145)

---
updated-dependencies:
- dependency-name: serde_json
  dependency-version: 1.0.145
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

.helix/languages.toml

@@ -1,6 +1,11 @@
 [language-server.silos]
-command = "silos"
+command = "./target/debug/silos"
+args = ["lsp"]
 
 [[language]]
 name = "go"
-language-servers = [ { name = "silos" } ]
+language-servers = [ { name = "silos" }, "gopls" ]
+
+[[language]]
+name = "rust"
+language-servers = [ ]

.vscode/settings.json

@@ -9,6 +9,7 @@
             ],
             "command": [
                 "silos"
+                "lsp"
             ]
         }
     ]

Cargo.lock

@@ -17,6 +17,20 @@ version = "2.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "512761e0bb2578dd7380c6baaa0f4ce03e84f95e960231d1dec8bf4d7d6e2627"
 
+[[package]]
+name = "ahash"
+version = "0.8.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a15f179cd60c4584b8a8c596927aadc462e27f2ca70c04e0071964a73ba7a75"
+dependencies = [
+ "cfg-if",
+ "getrandom 0.3.3",
+ "once_cell",
+ "serde",
+ "version_check",
+ "zerocopy",
+]
+
 [[package]]
 name = "aho-corasick"
 version = "1.1.3"
@@ -282,6 +296,15 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "castaway"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dec551ab6e7578819132c713a93c022a05d60159dc86e7a7050223577484c55a"
+dependencies = [
+ "rustversion",
+]
+
 [[package]]
 name = "cc"
 version = "1.2.26"
@@ -299,9 +322,9 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "clap"
-version = "4.5.41"
+version = "4.5.45"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "be92d32e80243a54711e5d7ce823c35c41c9d929dc4ab58e1276f625841aadf9"
+checksum = "1fc0e74a703892159f5ae7d3aac52c8e6c392f5ae5f359c70b5881d60aaac318"
 dependencies = [
  "clap_builder",
  "clap_derive",
@@ -309,9 +332,9 @@ dependencies = [
 
 [[package]]
 name = "clap_builder"
-version = "4.5.41"
+version = "4.5.44"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "707eab41e9622f9139419d573eca0900137718000c517d47da73045f54331c3d"
+checksum = "b3e7f4214277f3c7aa526a59dd3fbe306a370daee1f8b7b8c987069cd8e888a8"
 dependencies = [
  "anstream",
  "anstyle",
@@ -321,9 +344,9 @@ dependencies = [
 
 [[package]]
 name = "clap_derive"
-version = "4.5.41"
+version = "4.5.45"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ef4f52386a59ca4c860f7393bcf8abd8dfd91ecccc0f774635ff68e92eeef491"
+checksum = "14cb31bb0a7d536caef2639baa7fad459e15c3144efefa6dbd1c84562c4739f6"
 dependencies = [
  "heck",
  "proc-macro2",
@@ -343,6 +366,21 @@ version = "1.0.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b05b61dc5112cbb17e4b6cd61790d9845d13888356391624cbe7e41efeac1e75"
 
+[[package]]
+name = "compact_str"
+version = "0.9.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3fdb1325a1cece981e8a296ab8f0f9b63ae357bd0784a9faaf548cc7b480707a"
+dependencies = [
+ "castaway",
+ "cfg-if",
+ "itoa",
+ "rustversion",
+ "ryu",
+ "serde",
+ "static_assertions",
+]
+
 [[package]]
 name = "console"
 version = "0.15.11"
@@ -447,6 +485,15 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "dary_heap"
+version = "0.3.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "04d2cd9c18b9f454ed67da600630b021a8a80bf33f8c95896ab33aaf1c26b728"
+dependencies = [
+ "serde",
+]
+
 [[package]]
 name = "dashmap"
 version = "5.5.3"
@@ -1424,18 +1471,9 @@ checksum = "7943c866cc5cd64cbc25b2e01621d07fa8eb2a1a23160ee81ce38704e97b8ecf"
 
 [[package]]
 name = "itertools"
-version = "0.11.0"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b1c173a5686ce8bfa551b3563d0c2170bf24ca44da99c7ca4bfdab5418c3fe57"
-dependencies = [
- "either",
-]
-
-[[package]]
-name = "itertools"
-version = "0.13.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "413ee7dfc52ee1a4949ceeb7dbc8a33f2d6c088194d9f922fb8318faf1f01186"
+checksum = "2b192c782037fadd9cfa75548310488aabdbf3d2da73885b31bd0abd03351285"
 dependencies = [
  "either",
 ]
@@ -2147,12 +2185,12 @@ dependencies = [
 
 [[package]]
 name = "rayon-cond"
-version = "0.3.0"
+version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "059f538b55efd2309c9794130bc149c6a553db90e9d99c2030785c82f0bd7df9"
+checksum = "2964d0cf57a3e7a06e8183d14a8b527195c706b7983549cd5462d5aa3747438f"
 dependencies = [
  "either",
- "itertools 0.11.0",
+ "itertools",
  "rayon",
 ]
 
@@ -2411,18 +2449,28 @@ checksum = "1bc711410fbe7399f390ca1c3b60ad0f53f80e95c5eb935e52268a0e2cd49acc"
 
 [[package]]
 name = "serde"
-version = "1.0.219"
+version = "1.0.224"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5f0e2c6ed6606019b4e29e69dbaba95b11854410e5347d525002456dbbb786b6"
+checksum = "6aaeb1e94f53b16384af593c71e20b095e958dab1d26939c1b70645c5cfbcc0b"
+dependencies = [
+ "serde_core",
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_core"
+version = "1.0.224"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "32f39390fa6346e24defbcdd3d9544ba8a19985d0af74df8501fbfe9a64341ab"
 dependencies = [
  "serde_derive",
 ]
 
 [[package]]
 name = "serde_derive"
-version = "1.0.219"
+version = "1.0.224"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5b0276cf7f2c73365f7157c8123c21cd9a50fbbd844757af28ca1f5925fc2a00"
+checksum = "87ff78ab5e8561c9a675bfc1785cb07ae721f0ee53329a595cefd8c04c2ac4e0"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -2431,15 +2479,16 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.140"
+version = "1.0.145"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "20068b6e96dc6c9bd23e01df8827e6c7e1f2fddd43c21810382803c136b99373"
+checksum = "402a6f66d8c709116cf22f558eab210f5a50187f702eb4d7e5ef38d9a7f1c79c"
 dependencies = [
  "indexmap",
  "itoa",
  "memchr",
  "ryu",
  "serde",
+ "serde_core",
 ]
 
 [[package]]
@@ -2491,7 +2540,7 @@ checksum = "0fda2ff0d084019ba4d7c6f371c95d8fd75ce3524c3cb8fb653a3023f6323e64"
 
 [[package]]
 name = "silos"
-version = "5.2.2"
+version = "6.0.0"
 dependencies = [
  "anyhow",
  "candle-core",
@@ -2570,6 +2619,12 @@ version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3"
 
+[[package]]
+name = "static_assertions"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f"
+
 [[package]]
 name = "streaming-iterator"
 version = "0.1.9"
@@ -2742,23 +2797,25 @@ dependencies = [
 
 [[package]]
 name = "tokenizers"
-version = "0.21.1"
+version = "0.21.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3169b3195f925496c895caee7978a335d49218488ef22375267fba5a46a40bd7"
+checksum = "a620b996116a59e184c2fa2dfd8251ea34a36d0a514758c6f966386bd2e03476"
 dependencies = [
+ "ahash",
  "aho-corasick",
+ "compact_str",
+ "dary_heap",
  "derive_builder",
  "esaxx-rs",
- "getrandom 0.2.16",
+ "getrandom 0.3.3",
  "indicatif",
- "itertools 0.13.0",
- "lazy_static",
+ "itertools",
  "log",
  "macro_rules_attribute",
  "monostate",
  "onig",
  "paste",
- "rand 0.8.5",
+ "rand 0.9.1",
  "rayon",
  "rayon-cond",
  "regex",

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "silos"
-version = "5.2.2"
+version = "6.0.0"
 edition = "2024"
 
 [dependencies]
@@ -8,13 +8,13 @@ anyhow = "1.0.98"
 candle-core = "0.9.1"
 candle-nn = "0.9.1"
 candle-transformers = "0.9.1"
-clap = { version = "4.5.41", features = ["derive"] }
+clap = { version = "4.5.45", features = ["derive"] }
 derive_more = { version = "2.0.1", features = ["display", "error"] }
 hf-hub = "0.4.2"
 hora = "0.1.1"
 kdl = "6.3.4"
-serde_json = "1.0.140"
-tokenizers = "0.21.1"
+serde_json = "1.0.145"
+tokenizers = "0.21.4"
 tracing = "0.1.41"
 tracing-subscriber = "0.3.19"
 tree-sitter = "0.25.8"

README.md

@@ -6,9 +6,7 @@ Dumb, proomptable modular snippet search.
 
 ## Installation
 
-### Binary releases
-
-There are no binary releases yet.
+You can download a binary from releases tab or build the project from source.
 
 ### From source
 
@@ -86,7 +84,7 @@ This API parses code into an AST (Abstract Syntax Tree) via tree-sitter and can
 ``` kdl
 description "describes the mutation collection"
 mutation {
-  expression "some ((beautiful) @adjective) AST expression"
+  expression "(some ((beautiful) @adjective) AST expression) @root"
   substitute {
     literal "hello"
     capture "adjective"
@@ -95,7 +93,7 @@ mutation {
 }
 
 mutation {
-  expression "another"
+  expression "(another) @root"
   substitute {
     literal "multiple mutations work"
     literal "as long as their expression"
@@ -106,14 +104,30 @@ mutation {
 
 - `description`: A textual description of the mutation collection.
 - `mutation`:  Defines individual code changes.
-  - `expression`: Uses tree-sitter to match and capture AST nodes with `@` prefixes, The special `@root` node is reserved for the entire expression.
+  - `expression`: Uses tree-sitter to match and capture AST nodes with `@` prefixes,
+  - The special `@root` node must be specify the expression to be replaced.
   - `substitute`:  Constructs the modified code using literals and captured arguments.
 
-See the example mutation collection in `./snippets/v2/go/mutations.kdl`.
+See the example mutation collection in `./snippets/v2/go/filepath-parent.kdl`.
 
 - The API performs a single-pass substitution based on the closest matching mutation.
 - Captured groups are used within the `substitute` block and the mutated code is returned.
 
+> Every capture group must contain the largest atom to be operated on.
+For example: if you wish to operate on elements of an array, capture each identifier inside the array
+
+Correct way: Here the `array` and `identifier` only hints about where the expression `root` lies.
+
+```
+(array (identifier @root))
+```
+
+Incorrect way: Here the root expression matches the block all the array elements inside the braces, not each element.
+
+```
+(array ((identifier)*) @entire-block-capture) @root
+```
+
 **Further reading**
 
 - [tree-sitter query snytax](https://tree-sitter.github.io/tree-sitter/using-parsers/queries/1-syntax.html) to create mutation expressions.

snippets/refactor/go/base64.kdl

@@ -0,0 +1,13 @@
+description "base64 import"
+mutation {
+	expression "(import_spec_list ((import_spec)* @spec)) @root"
+	substitute {
+		literal "("
+		literal "\n"
+		capture "spec"
+		literal "\n"
+		literal #""base64""#
+		literal "\n"
+		literal ")"
+	}
+}

snippets/refactor/go/filepath-parent.kdl

@@ -4,7 +4,7 @@ mutation {
 	(call_expression
 		function: (_) @func (#eq? @func "filepath.Base")
 		arguments: (_) @args
-	)
+	) @root
 	"""
 	substitute {
 		literal "filepath.Base(filepath.Dir(filepath.Clean"

src/args.rs

@@ -1,8 +1,15 @@
-use clap::Parser;
+use clap::{Args, Parser, Subcommand};
+use std::path::PathBuf;
 
 #[derive(Parser, Debug)]
 #[command(author, version, about, long_about = None)]
-pub(crate) struct Args {
+pub(crate) struct Cli {
+    #[command(subcommand)]
+    pub command: Command,
+}
+
+#[derive(Args, Debug)]
+pub(crate) struct Lsp {
     /// Run on the Nth GPU device.
     #[arg(long)]
     pub(crate) gpu: Option<usize>,
@@ -20,7 +27,35 @@ pub(crate) struct Args {
     pub(crate) snippets: std::path::PathBuf,
 }
 
-impl Args {
+#[derive(Args, Debug)]
+pub struct DumpExpression {
+    pub path: PathBuf,
+}
+
+#[derive(Args, Debug)]
+pub struct ShowCaptures {
+    pub path: PathBuf,
+    pub expression: String,
+}
+
+#[derive(Subcommand, Debug)]
+pub enum Ast {
+    /// Dump the S expression for a given source file
+    DumpExpression(DumpExpression),
+    /// Show what parts of a source file gets captured by an S expression
+    ShowCaptures(ShowCaptures),
+}
+
+#[derive(Subcommand, Debug)]
+pub enum Command {
+    /// quick actions to dump, modify and verify abstract syntax trees
+    #[command(subcommand)]
+    Ast(Ast),
+    /// spawn a language server for use with a text editor
+    Lsp(Lsp),
+}
+
+impl Lsp {
     pub(crate) fn resolve_model_and_revision(&self) -> (String, String) {
         let default_model = "sentence-transformers/all-MiniLM-L6-v2".to_string();
         let default_revision = "refs/pr/21".to_string();

src/lsp.rs

@@ -60,12 +60,18 @@ impl LanguageServer for Backend {
     }
 
     async fn did_open(&self, params: DidOpenTextDocumentParams) {
-        self.body.lock().await.insert(params.text_document.uri, params.text_document.text);
+        self.body
+            .lock()
+            .await
+            .insert(params.text_document.uri, params.text_document.text);
     }
 
     async fn did_change(&self, params: DidChangeTextDocumentParams) {
         if let Some(body) = params.content_changes.into_iter().next() {
-            self.body.lock().await.insert(params.text_document.uri, body.text);
+            self.body
+                .lock()
+                .await
+                .insert(params.text_document.uri, body.text);
         }
     }
 

src/main.rs

@@ -3,7 +3,7 @@ use clap::Parser;
 use hora::core::{ann_index::ANNIndex, metrics::Metric::Euclidean};
 use hora::index::hnsw_idx::HNSWIndex;
 use kdl::KdlDocument;
-use state::State;
+use state::{State, dump_expression};
 use std::collections::HashMap;
 use std::sync::Arc;
 use tokio::sync::Mutex;
@@ -13,14 +13,43 @@ mod args;
 mod embed;
 mod lsp;
 mod mutation;
-mod state;
 mod sources;
+mod state;
 
 #[tokio::main]
 async fn main() -> Result<()> {
     tracing_subscriber::fmt::init();
-    let args = args::Args::parse();
+    let args = match args::Cli::parse().command {
+        args::Command::Ast(ast) => {
+            match ast {
+                args::Ast::DumpExpression(source_file) => {
+                    println!("{}", dump_expression(&source_file.path)?);
+                }
+                args::Ast::ShowCaptures(show_captures) => {
+                    let source = std::fs::read_to_string(&show_captures.path).unwrap();
+                    let source_bytes = source.as_bytes();
+                    let extension = show_captures.path.extension().unwrap().to_str().unwrap();
+                    let langfn = state::Refactor::get_lang(extension).unwrap();
+                    let mut parser = tree_sitter::Parser::new();
+                    parser.set_language(&langfn).unwrap();
+                    let tree = parser.parse(source_bytes, None).unwrap();
+                    let root_node = tree.root_node();
+                    let cooked = mutation::query(
+                        root_node,
+                        &show_captures.expression,
+                        &langfn,
+                        source_bytes,
+                    );
+                    println!("{:#?}", cooked);
+                }
+            }
+            return Ok(());
+        }
+        args::Command::Lsp(lsp) => lsp,
+    };
+
     let (model_id, revision) = args.resolve_model_and_revision();
+
     let embed = embed::Embed::new(args.gpu, &model_id, &revision)?;
     let mut dict = HashMap::default();
     let dimensions = embed.hidden_size;
@@ -64,7 +93,10 @@ async fn main() -> Result<()> {
                 .or_insert_with(|| HNSWIndex::new(dimensions, &Default::default()));
 
             current_lang_index
-                .add(&embed.embed(&mutations.description)?, mutations_collection.len())
+                .add(
+                    &embed.embed(&mutations.description)?,
+                    mutations_collection.len(),
+                )
                 .map_err(E::msg)?;
             mutations_collection.push(mutations);
         }

src/mutation.rs

@@ -72,10 +72,8 @@ pub fn from_path<P: AsRef<Path>>(path: P) -> Result<MutationCollection> {
             substitute.push(substitutor);
         }
 
-        let expression = format!("({expression}) @root");
-
         mutations.push(Mutation {
-            expression,
+            expression: expression.to_string(),
             substitute,
         })
     }
@@ -127,7 +125,7 @@ pub fn apply(
 }
 
 #[derive(Debug)]
-struct QueryCooked {
+pub struct QueryCooked {
     captures: HashMap<String, String>,
     end: usize,
     start: usize,
@@ -152,7 +150,7 @@ fn split_at_indices<'a>(c: &'a [u8], idx: &[usize]) -> SplitMap<'a> {
     SplitMap { values, indices }
 }
 
-fn query<'a>(
+pub fn query<'a>(
     node: Node<'a>,
     expr: &'a str,
     lang: &Language,
@@ -164,6 +162,7 @@ fn query<'a>(
     let mut query_matches = qc.matches(&query, node, source_bytes);
 
     let capture_names = query.capture_names();
+    // println!("names: {capture_names:#?}");
 
     let mut cooked = vec![];
 
@@ -171,19 +170,36 @@ fn query<'a>(
         let mut capture_cooked = HashMap::new();
         let mut start = 0;
         let mut end = 0;
-        for cap in matcha.captures {
-            let Some(name) = capture_names.get(cap.index as usize) else {
-                continue;
-            };
-            if *name == "root" {
-                start = cap.node.start_byte();
-                end = cap.node.end_byte();
+        if matcha.captures.is_empty() {
+            continue;
+        }
+        //     println!("match {:#?}", matcha.id());
+
+        for (ix, name) in capture_names.iter().enumerate() {
+            let nodes = matcha.nodes_for_capture_index(ix.try_into().unwrap());
+            let mut start_pos = None;
+            let mut end_pos = None;
+            //         println!("matches for {name}");
+            for node in nodes {
+                if start_pos.is_none() {
+                    start_pos.replace(node.start_byte());
+                }
+                end_pos.replace(node.end_byte());
+                //             println!("hit {node:#?}");
+            }
+            if start_pos.or(end_pos).is_none() {
                 continue;
             }
-            capture_cooked.insert(
-                name.to_string(),
-                cap.node.utf8_text(source_bytes).unwrap().to_string(),
-            );
+            if *name == "root" {
+                start = start_pos.unwrap();
+                end = end_pos.unwrap();
+            }
+            let range = start_pos.unwrap()..end_pos.unwrap();
+            //         println!("match range for {name}: {:#?}", range);
+            let text_bytes = &source_bytes[range];
+            let text = std::str::from_utf8(text_bytes).unwrap();
+            //         println!("text: {text}");
+            capture_cooked.insert(name.to_string(), text.to_string());
         }
         cooked.push(QueryCooked {
             start,

src/sources.rs

@@ -1,23 +1,32 @@
-use std::{fs, io, path::{Path, PathBuf}, collections::HashMap};
+use std::{
+    collections::HashMap,
+    fs, io,
+    path::{Path, PathBuf},
+};
 
 pub fn rule_files<P: AsRef<Path>>(path: P) -> io::Result<HashMap<String, Vec<PathBuf>>> {
     let per_language_dirs: Vec<_> = fs::read_dir(path)?
-    .filter_map(|res| res.ok())
-    .map(|direntry| direntry.path())
-    .filter(|dir| dir.is_dir()).collect();
+        .filter_map(|res| res.ok())
+        .map(|direntry| direntry.path())
+        .filter(|dir| dir.is_dir())
+        .collect();
 
     let mut basename_to_paths = HashMap::new();
 
     for language_dir in per_language_dirs {
-        let Some(dirname) = language_dir.file_stem().and_then(|v|v.to_str()).map(|v| v.to_string()) else {
+        let Some(dirname) = language_dir
+            .file_stem()
+            .and_then(|v| v.to_str())
+            .map(|v| v.to_string())
+        else {
             continue;
         };
         let rule_file_paths: Vec<_> = fs::read_dir(&language_dir)?
-        .filter_map(|res| res.ok())
-        .map(|entry| entry.path())
-        .filter(|file| file.is_file() && file.extension().is_some_and(|ext| ext == "kdl"))
-        .map(|path| path.to_path_buf())
-        .collect();
+            .filter_map(|res| res.ok())
+            .map(|entry| entry.path())
+            .filter(|file| file.is_file() && file.extension().is_some_and(|ext| ext == "kdl"))
+            .map(|path| path.to_path_buf())
+            .collect();
         basename_to_paths.insert(dirname, rule_file_paths);
     }
     Ok(basename_to_paths)

src/state.rs

@@ -4,6 +4,7 @@ use derive_more::Error;
 use hora::core::ann_index::ANNIndex;
 use hora::index::hnsw_idx::HNSWIndex;
 use std::collections::HashMap;
+use std::path::Path;
 use tree_sitter::Parser;
 
 #[derive(Debug, Display, Error)]
@@ -22,7 +23,7 @@ pub struct Refactor {
 }
 
 impl Refactor {
-    fn get_lang(s: &str) -> Result<tree_sitter::Language, Error> {
+    pub fn get_lang(s: &str) -> Result<tree_sitter::Language, Error> {
         Ok(match s {
             "go" => tree_sitter_go::LANGUAGE,
             "c" | "h" => tree_sitter_c::LANGUAGE,
@@ -81,6 +82,27 @@ impl Refactor {
         Ok(collected)
     }
 }
+
+pub fn dump_expression(path: &Path) -> Result<String, Error> {
+    let Some(lang) = path.extension() else {
+        return Err(Error::UnknownLang);
+    };
+    let lang = lang.to_str().ok_or(Error::UnknownLang)?;
+    let langfn = Refactor::get_lang(lang)?;
+    let mut parser = Parser::new();
+    parser
+        .set_language(&langfn)
+        .map_err(|_| Error::UnknownLang)?;
+
+    let source_code = std::fs::read_to_string(path).map_err(|_| Error::SnippetParsing)?;
+    let source_bytes = source_code.as_bytes();
+    let tree = parser
+        .parse(source_bytes, None)
+        .ok_or(Error::SnippetParsing)?;
+    let root_node = tree.root_node();
+    Ok(root_node.to_sexp().to_string())
+}
+
 pub struct Generate {
     pub dict: HashMap<String, HNSWIndex<f32, String>>,
 }