Files
codebase-memory-mcp/internal/parser/parser.go
T
Martin Vogel 02bc3b7352 Release v0.3.0: 25 languages, install/update CLI, skills, benchmark
Language expansion (13 → 25):
  Add Ruby, C, Bash, Zig, Elixir, Haskell, OCaml, HTML, CSS, YAML,
  TOML, HCL with tree-sitter grammars and language specs.
  Restore Erlang and SQL with extraction fixes.

CLI install/update/uninstall commands:
  Auto-detect Claude Code and Codex CLI, register MCP server,
  install task-specific skills, self-update with SHA-256 verification.

Embedded skills (4):
  exploring, tracing, quality, reference — auto-trigger in Claude Code
  for graph-first code discovery.

Pipeline improvements:
  Docstring extraction, enrichment pass for params/returns/complexity,
  graceful context cancellation, improved test detection.

35-language benchmark (BENCHMARK.md):
  12 questions × 35 repos, 91.8% overall score, Linux kernel stress
  test (20K nodes, zero timeouts). Replaces old benchmark artifacts.
2026-03-01 18:01:52 +01:00

172 lines
7.3 KiB
Go

package parser
import (
"fmt"
"sync"
tree_sitter "github.com/tree-sitter/go-tree-sitter"
tree_sitter_bash "github.com/tree-sitter/tree-sitter-bash/bindings/go"
tree_sitter_c_sharp "github.com/tree-sitter/tree-sitter-c-sharp/bindings/go"
tree_sitter_c "github.com/tree-sitter/tree-sitter-c/bindings/go"
tree_sitter_cpp "github.com/tree-sitter/tree-sitter-cpp/bindings/go"
tree_sitter_css "github.com/tree-sitter/tree-sitter-css/bindings/go"
tree_sitter_go "github.com/tree-sitter/tree-sitter-go/bindings/go"
tree_sitter_html "github.com/tree-sitter/tree-sitter-html/bindings/go"
tree_sitter_java "github.com/tree-sitter/tree-sitter-java/bindings/go"
tree_sitter_javascript "github.com/tree-sitter/tree-sitter-javascript/bindings/go"
tree_sitter_ocaml "github.com/tree-sitter/tree-sitter-ocaml/bindings/go"
tree_sitter_php "github.com/tree-sitter/tree-sitter-php/bindings/go"
tree_sitter_python "github.com/tree-sitter/tree-sitter-python/bindings/go"
tree_sitter_ruby "github.com/tree-sitter/tree-sitter-ruby/bindings/go"
tree_sitter_rust "github.com/tree-sitter/tree-sitter-rust/bindings/go"
tree_sitter_scala "github.com/tree-sitter/tree-sitter-scala/bindings/go"
tree_sitter_typescript "github.com/tree-sitter/tree-sitter-typescript/bindings/go"
tree_sitter_hcl "github.com/tree-sitter-grammars/tree-sitter-hcl/bindings/go"
tree_sitter_kotlin "github.com/tree-sitter-grammars/tree-sitter-kotlin/bindings/go"
tree_sitter_lua "github.com/tree-sitter-grammars/tree-sitter-lua/bindings/go"
tree_sitter_toml "github.com/tree-sitter-grammars/tree-sitter-toml/bindings/go"
tree_sitter_yaml "github.com/tree-sitter-grammars/tree-sitter-yaml/bindings/go"
tree_sitter_zig "github.com/tree-sitter-grammars/tree-sitter-zig/bindings/go"
tree_sitter_elixir "github.com/tree-sitter/tree-sitter-elixir/bindings/go"
tree_sitter_haskell "github.com/tree-sitter/tree-sitter-haskell/bindings/go"
tree_sitter_objc "github.com/tree-sitter-grammars/tree-sitter-objc/bindings/go"
tree_sitter_scss "github.com/tree-sitter-grammars/tree-sitter-scss/bindings/go"
tree_sitter_sql "github.com/DerekStride/tree-sitter-sql/bindings/go"
tree_sitter_dart "github.com/UserNobody14/tree-sitter-dart/bindings/go"
tree_sitter_swift "github.com/alex-pinkus/tree-sitter-swift/bindings/go"
tree_sitter_dockerfile "github.com/camdencheek/tree-sitter-dockerfile/bindings/go"
tree_sitter_groovy "github.com/murtaza64/tree-sitter-groovy/bindings/go"
tree_sitter_r "github.com/r-lib/tree-sitter-r/bindings/go"
tree_sitter_erlang "github.com/tree-sitter/tree-sitter-erlang/bindings/go"
tree_sitter_perl "github.com/tree-sitter/tree-sitter-perl/bindings/go"
"github.com/DeusData/codebase-memory-mcp/internal/lang"
)
var (
languagesOnce sync.Once
languages map[lang.Language]*tree_sitter.Language
parserPools map[lang.Language]*sync.Pool
)
func initLanguages() {
languagesOnce.Do(func() {
languages = map[lang.Language]*tree_sitter.Language{
lang.Python: tree_sitter.NewLanguage(tree_sitter_python.Language()),
lang.JavaScript: tree_sitter.NewLanguage(tree_sitter_javascript.Language()),
lang.TypeScript: tree_sitter.NewLanguage(tree_sitter_typescript.LanguageTypescript()),
lang.TSX: tree_sitter.NewLanguage(tree_sitter_typescript.LanguageTSX()),
lang.Go: tree_sitter.NewLanguage(tree_sitter_go.Language()),
lang.Rust: tree_sitter.NewLanguage(tree_sitter_rust.Language()),
lang.Java: tree_sitter.NewLanguage(tree_sitter_java.Language()),
lang.CPP: tree_sitter.NewLanguage(tree_sitter_cpp.Language()),
lang.CSharp: tree_sitter.NewLanguage(tree_sitter_c_sharp.Language()),
lang.PHP: tree_sitter.NewLanguage(tree_sitter_php.LanguagePHPOnly()),
lang.Lua: tree_sitter.NewLanguage(tree_sitter_lua.Language()),
lang.Scala: tree_sitter.NewLanguage(tree_sitter_scala.Language()),
lang.Kotlin: tree_sitter.NewLanguage(tree_sitter_kotlin.Language()),
lang.Ruby: tree_sitter.NewLanguage(tree_sitter_ruby.Language()),
lang.C: tree_sitter.NewLanguage(tree_sitter_c.Language()),
lang.Bash: tree_sitter.NewLanguage(tree_sitter_bash.Language()),
lang.Zig: tree_sitter.NewLanguage(tree_sitter_zig.Language()),
lang.Elixir: tree_sitter.NewLanguage(tree_sitter_elixir.Language()),
lang.Haskell: tree_sitter.NewLanguage(tree_sitter_haskell.Language()),
lang.OCaml: tree_sitter.NewLanguage(tree_sitter_ocaml.LanguageOCaml()),
lang.HTML: tree_sitter.NewLanguage(tree_sitter_html.Language()),
lang.CSS: tree_sitter.NewLanguage(tree_sitter_css.Language()),
lang.YAML: tree_sitter.NewLanguage(tree_sitter_yaml.Language()),
lang.TOML: tree_sitter.NewLanguage(tree_sitter_toml.Language()),
lang.HCL: tree_sitter.NewLanguage(tree_sitter_hcl.Language()),
lang.ObjectiveC: tree_sitter.NewLanguage(tree_sitter_objc.Language()),
lang.Swift: tree_sitter.NewLanguage(tree_sitter_swift.Language()),
lang.Dart: tree_sitter.NewLanguage(tree_sitter_dart.Language()),
lang.Perl: tree_sitter.NewLanguage(tree_sitter_perl.Language()),
lang.Groovy: tree_sitter.NewLanguage(tree_sitter_groovy.Language()),
lang.SCSS: tree_sitter.NewLanguage(tree_sitter_scss.Language()),
lang.R: tree_sitter.NewLanguage(tree_sitter_r.Language()),
lang.Erlang: tree_sitter.NewLanguage(tree_sitter_erlang.Language()),
lang.SQL: tree_sitter.NewLanguage(tree_sitter_sql.Language()),
lang.Dockerfile: tree_sitter.NewLanguage(tree_sitter_dockerfile.Language()),
}
parserPools = make(map[lang.Language]*sync.Pool, len(languages))
for l, tsLang := range languages {
tsLang := tsLang
parserPools[l] = &sync.Pool{
New: func() any {
p := tree_sitter.NewParser()
if err := p.SetLanguage(tsLang); err != nil {
panic(fmt.Sprintf("set language: %v", err))
}
return p
},
}
}
})
}
// GetLanguage returns the tree-sitter Language for a lang.Language.
func GetLanguage(l lang.Language) (*tree_sitter.Language, error) {
initLanguages()
tsLang, ok := languages[l]
if !ok {
return nil, fmt.Errorf("unsupported language: %s", l)
}
return tsLang, nil
}
// Parse parses source code into a tree-sitter AST Tree.
// The caller must call tree.Close() when done.
// Parsers are pooled per language via sync.Pool to avoid per-file allocation.
func Parse(l lang.Language, source []byte) (*tree_sitter.Tree, error) {
initLanguages()
pool, ok := parserPools[l]
if !ok {
return nil, fmt.Errorf("unsupported language: %s", l)
}
p, _ := pool.Get().(*tree_sitter.Parser)
if p == nil {
return nil, fmt.Errorf("failed to get parser for language %s", l)
}
tree := p.Parse(source, nil)
pool.Put(p)
if tree == nil {
return nil, fmt.Errorf("parse failed for language %s", l)
}
return tree, nil
}
// WalkFunc is called for each node during AST traversal.
// Return false to skip children.
type WalkFunc func(node *tree_sitter.Node) bool
// Walk traverses the AST in depth-first order.
func Walk(node *tree_sitter.Node, fn WalkFunc) {
if node == nil {
return
}
if !fn(node) {
return
}
for i := uint(0); i < node.ChildCount(); i++ {
child := node.Child(i)
if child != nil {
Walk(child, fn)
}
}
}
// NodeText returns the text content of a node.
func NodeText(node *tree_sitter.Node, source []byte) string {
return string(source[node.StartByte():node.EndByte()])
}