feat(tree-sitter): add tree-sitter devcontainer feature

src/tree-sitter/devcontainer-feature.json

@@ -0,0 +1,57 @@
+{
+  "id": "tree-sitter",
+  "version": "1.0.0",
+  "name": "Tree-sitter",
+  "description": "Installs the Tree-sitter CLI for incremental parsing. Optionally compiles grammars (PHP, TypeScript, JavaScript, Go, Dart, Python, YAML) from official repos.",
+  "documentationURL": "https://tree-sitter.github.io/tree-sitter/",
+  "licenseURL": "https://github.com/tree-sitter/tree-sitter/blob/master/LICENSE",
+  "options": {
+    "version": {
+      "type": "string",
+      "default": "latest",
+      "description": "Tree-sitter CLI version: 'latest' or a specific version (e.g. '0.25.4')"
+    },
+    "grammarPhp": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the PHP grammar (tree-sitter/tree-sitter-php)"
+    },
+    "grammarTypescript": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the TypeScript grammar (tree-sitter/tree-sitter-typescript)"
+    },
+    "grammarJavascript": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the JavaScript grammar (tree-sitter/tree-sitter-javascript)"
+    },
+    "grammarGo": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the Go grammar (tree-sitter/tree-sitter-go)"
+    },
+    "grammarDart": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the Dart grammar (UserNobody14/tree-sitter-dart)"
+    },
+    "grammarPython": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the Python grammar (tree-sitter/tree-sitter-python)"
+    },
+    "grammarYaml": {
+      "type": "boolean",
+      "default": false,
+      "description": "Compile and install the YAML grammar (tree-sitter-grammars/tree-sitter-yaml)"
+    }
+  },
+  "containerEnv": {
+    "TREE_SITTER_DIR": "/usr/local/lib/tree-sitter"
+  },
+  "installsAfter": [
+    "ghcr.io/devcontainers/features/common-utils",
+    "ghcr.io/devcontainers/features/git"
+  ]
+}

src/tree-sitter/install.sh

@@ -0,0 +1,128 @@
+#!/bin/bash
+set -e
+
+# --- Options injected by devcontainer feature engine ---
+VERSION="${VERSION:-latest}"
+GRAMMAR_PHP="${GRAMMARPHP:-false}"
+GRAMMAR_TYPESCRIPT="${GRAMMARTYPESCRIPT:-false}"
+GRAMMAR_JAVASCRIPT="${GRAMMARJAVASCRIPT:-false}"
+GRAMMAR_GO="${GRAMMARGO:-false}"
+GRAMMAR_DART="${GRAMMARDART:-false}"
+GRAMMAR_PYTHON="${GRAMMARPYTHON:-false}"
+GRAMMAR_YAML="${GRAMMARYAML:-false}"
+
+TREE_SITTER_DIR="/usr/local/lib/tree-sitter"
+
+# --- Helpers ---
+check_packages() {
+    if ! dpkg -s "$@" >/dev/null 2>&1; then
+        if [ "$(find /var/lib/apt/lists/* 2>/dev/null | head -1)" = "" ]; then
+            echo "==> Running apt-get update..."
+            apt-get update -y
+        fi
+        apt-get install -y --no-install-recommends "$@"
+    fi
+}
+
+any_grammar_enabled() {
+    [ "$GRAMMAR_PHP" = "true" ] || [ "$GRAMMAR_TYPESCRIPT" = "true" ] || \
+    [ "$GRAMMAR_JAVASCRIPT" = "true" ] || [ "$GRAMMAR_GO" = "true" ] || \
+    [ "$GRAMMAR_DART" = "true" ] || [ "$GRAMMAR_PYTHON" = "true" ] || \
+    [ "$GRAMMAR_YAML" = "true" ]
+}
+
+# Clone a grammar repo, build the .so, clean up.
+# Usage: build_grammar <github-org/repo> <output-name> [subdir]
+build_grammar() {
+    local repo="$1"
+    local name="$2"
+    local subdir="${3:-}"
+    local clone_dir="/tmp/ts-${name}"
+
+    echo "==> Building grammar: ${name} (${repo})..."
+    git clone --depth 1 "https://github.com/${repo}.git" "$clone_dir"
+
+    local build_dir="$clone_dir"
+    if [ -n "$subdir" ]; then
+        build_dir="${clone_dir}/${subdir}"
+    fi
+
+    tree-sitter build --output "${TREE_SITTER_DIR}/${name}.so" "$build_dir"
+    rm -rf "$clone_dir"
+    echo "==> Grammar ${name} installed: ${TREE_SITTER_DIR}/${name}.so"
+}
+
+# --- Main ---
+echo "==> Tree-sitter feature: version=${VERSION}"
+
+# 1. Base dependencies
+check_packages curl ca-certificates
+
+# 2. Resolve version
+if [ "$VERSION" = "latest" ]; then
+    # NOTE: Anonymous GitHub API rate limit is 60 req/hour.
+    # If builds fail here, pin a specific version instead.
+    VERSION=$(curl -fsSL "https://api.github.com/repos/tree-sitter/tree-sitter/releases/latest" \
+        | sed -n 's/.*"tag_name": *"v\([^"]*\)".*/\1/p')
+    if [ -z "$VERSION" ]; then
+        echo "ERROR: Failed to resolve latest tree-sitter version from GitHub API"
+        exit 1
+    fi
+    echo "==> Resolved latest version: ${VERSION}"
+fi
+
+# 3. Download pre-built CLI binary
+ARCH=$(dpkg --print-architecture)
+case "$ARCH" in
+    amd64) TS_ARCH="x64" ;;
+    arm64) TS_ARCH="arm64" ;;
+    *)
+        echo "ERROR: Unsupported architecture: $ARCH"
+        exit 1
+        ;;
+esac
+
+DOWNLOAD_URL="https://github.com/tree-sitter/tree-sitter/releases/download/v${VERSION}/tree-sitter-linux-${TS_ARCH}.gz"
+echo "==> Downloading tree-sitter v${VERSION} for ${TS_ARCH}..."
+curl -fsSL "$DOWNLOAD_URL" | gunzip > /usr/local/bin/tree-sitter
+chmod +x /usr/local/bin/tree-sitter
+
+echo "==> Tree-sitter $(tree-sitter --version) installed at $(command -v tree-sitter)"
+
+# 4. Grammar compilation (only if at least one grammar is enabled)
+if any_grammar_enabled; then
+    echo "==> Installing build dependencies for grammars..."
+    check_packages gcc libc6-dev git
+
+    mkdir -p "$TREE_SITTER_DIR"
+
+    if [ "$GRAMMAR_PHP" = "true" ]; then
+        build_grammar "tree-sitter/tree-sitter-php" "php" "php"
+    fi
+    if [ "$GRAMMAR_TYPESCRIPT" = "true" ]; then
+        build_grammar "tree-sitter/tree-sitter-typescript" "typescript" "typescript"
+    fi
+    if [ "$GRAMMAR_JAVASCRIPT" = "true" ]; then
+        build_grammar "tree-sitter/tree-sitter-javascript" "javascript"
+    fi
+    if [ "$GRAMMAR_GO" = "true" ]; then
+        build_grammar "tree-sitter/tree-sitter-go" "go"
+    fi
+    if [ "$GRAMMAR_DART" = "true" ]; then
+        build_grammar "UserNobody14/tree-sitter-dart" "dart"
+    fi
+    if [ "$GRAMMAR_PYTHON" = "true" ]; then
+        build_grammar "tree-sitter/tree-sitter-python" "python"
+    fi
+    if [ "$GRAMMAR_YAML" = "true" ]; then
+        build_grammar "tree-sitter-grammars/tree-sitter-yaml" "yaml"
+    fi
+
+    # Give remote user ownership so tree-sitter can write at runtime
+    REMOTE_USER="${_REMOTE_USER:-vscode}"
+    if id "$REMOTE_USER" &>/dev/null; then
+        chown -R "${REMOTE_USER}:${REMOTE_USER}" "$TREE_SITTER_DIR"
+    fi
+fi
+
+echo "==> Tree-sitter feature install complete"

test-local.sh

@@ -8,9 +8,11 @@ SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
 NO_CACHE=false
 
 # Parse flags
+BASE_IMAGE="mcr.microsoft.com/devcontainers/base:ubuntu"
 while [[ "$1" == -* ]]; do
     case "$1" in
         --no-cache) NO_CACHE=true; shift ;;
+        --base-image) BASE_IMAGE="$2"; shift 2 ;;
         *) echo -e "${RED}Unknown option: $1${NC}"; exit 1 ;;
     esac
 done
@@ -42,6 +44,10 @@ run_tests() {
         --project-folder "$SCRIPT_DIR"
     )
 
+    if [ -n "$BASE_IMAGE" ]; then
+        args+=(--base-image "$BASE_IMAGE")
+    fi
+
     if [ -n "$SCENARIO" ]; then
         args+=(--filter "$SCENARIO")
     fi

test/tree-sitter/base-image.txt

@@ -0,0 +1 @@
+mcr.microsoft.com/devcontainers/base:ubuntu

test/tree-sitter/install_tree_sitter_latest.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+set -e
+
+# Scenario: install latest tree-sitter CLI only (no grammars)
+# No additional scenario-specific checks beyond test.sh

test/tree-sitter/install_tree_sitter_with_all_grammars.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+set -e
+
+# Scenario: install tree-sitter with all grammars
+source "$(dirname "$0")/test.sh"
+
+GRAMMARS="php typescript javascript go dart python yaml"
+for grammar in $GRAMMARS; do
+    if [ ! -f "${TREE_SITTER_DIR}/${grammar}.so" ]; then
+        echo "FAIL: ${grammar} grammar not found at ${TREE_SITTER_DIR}/${grammar}.so"
+        exit 1
+    fi
+    echo "PASS: ${grammar} grammar installed"
+done
+
+echo "==> All grammar tests passed"

test/tree-sitter/install_tree_sitter_with_php.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+set -e
+
+# Scenario: install tree-sitter with PHP grammar
+source "$(dirname "$0")/test.sh"
+
+# PHP grammar .so must exist
+if [ ! -f "${TREE_SITTER_DIR}/php.so" ]; then
+    echo "FAIL: PHP grammar not found at ${TREE_SITTER_DIR}/php.so"
+    exit 1
+fi
+echo "PASS: PHP grammar installed at ${TREE_SITTER_DIR}/php.so"

test/tree-sitter/scenarios.json

@@ -0,0 +1,30 @@
+{
+  "install_tree_sitter_latest": {
+    "image": "mcr.microsoft.com/devcontainers/base:ubuntu",
+    "features": {
+      "tree-sitter": {}
+    }
+  },
+  "install_tree_sitter_with_php": {
+    "image": "mcr.microsoft.com/devcontainers/base:ubuntu",
+    "features": {
+      "tree-sitter": {
+        "grammarPhp": true
+      }
+    }
+  },
+  "install_tree_sitter_with_all_grammars": {
+    "image": "mcr.microsoft.com/devcontainers/base:ubuntu",
+    "features": {
+      "tree-sitter": {
+        "grammarPhp": true,
+        "grammarTypescript": true,
+        "grammarJavascript": true,
+        "grammarGo": true,
+        "grammarDart": true,
+        "grammarPython": true,
+        "grammarYaml": true
+      }
+    }
+  }
+}

test/tree-sitter/test.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+set -e
+
+# Devcontainer feature test — run by devcontainers/action in CI
+# Verifies Tree-sitter CLI is correctly installed and on PATH.
+
+echo "==> Testing Tree-sitter feature..."
+
+# tree-sitter binary must be on PATH
+if ! command -v tree-sitter &>/dev/null; then
+    echo "FAIL: tree-sitter not found on PATH"
+    exit 1
+fi
+echo "PASS: tree-sitter on PATH ($(command -v tree-sitter))"
+
+# tree-sitter --version must succeed
+TS_VERSION=$(tree-sitter --version 2>&1)
+if [ -z "$TS_VERSION" ]; then
+    echo "FAIL: tree-sitter --version returned empty output"
+    exit 1
+fi
+echo "PASS: tree-sitter --version => ${TS_VERSION}"
+
+# TREE_SITTER_DIR env var must be set
+if [ -z "$TREE_SITTER_DIR" ]; then
+    echo "FAIL: TREE_SITTER_DIR not set"
+    exit 1
+fi
+echo "PASS: TREE_SITTER_DIR=${TREE_SITTER_DIR}"
+
+echo "==> All Tree-sitter base tests passed"