From 007efdef5a664744b9f67aa096c7a27b4cedd686 Mon Sep 17 00:00:00 2001
From: worldmozara <worldmozara@163.com>
Date: Wed, 15 Jul 2026 10:15:13 +0800
Subject: [PATCH] core/tools/file: add glob_paths workspace search tool

Stdlib Path.glob with VCS/hidden skip and match cap; paths relative to
workspace root.
---
 src/plyngent/tools/file/glob_paths.py | 101 ++++++++++++++++++++++++++
 tests/test_tools/test_glob_paths.py   |  42 +++++++++++
 2 files changed, 143 insertions(+)
 create mode 100644 src/plyngent/tools/file/glob_paths.py
 create mode 100644 tests/test_tools/test_glob_paths.py

diff --git a/src/plyngent/tools/file/glob_paths.py b/src/plyngent/tools/file/glob_paths.py
new file mode 100644
index 0000000..7f37964
--- /dev/null
+++ b/src/plyngent/tools/file/glob_paths.py
@@ -0,0 +1,101 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from plyngent.agent import tool
+from plyngent.tools.file.tree import VCS_DIR_NAMES
+from plyngent.tools.workspace import WorkspaceError, get_workspace_root, resolve_path
+
+if TYPE_CHECKING:
+    from pathlib import Path
+
+DEFAULT_MAX_MATCHES = 200
+
+
+def _rel(path: Path, root: Path) -> str:
+    try:
+        return str(path.relative_to(root))
+    except ValueError:
+        return str(path)
+
+
+def _hidden_or_vcs(parts: tuple[str, ...]) -> bool:
+    return any(part in VCS_DIR_NAMES or part.startswith(".") for part in parts)
+
+
+def _collect_glob(
+    base: Path,
+    root: Path,
+    pattern: str,
+    *,
+    max_matches: int,
+    skip_hidden_dirs: bool,
+) -> tuple[list[str], bool] | str:
+    matches: list[str] = []
+    try:
+        candidates = sorted(base.glob(pattern), key=lambda p: str(p).casefold())
+    except (OSError, ValueError) as exc:
+        return f"error: glob failed: {exc}"
+
+    for candidate in candidates:
+        try:
+            resolved = candidate.resolve()
+            rel = resolved.relative_to(root)
+        except (OSError, ValueError):
+            continue
+        # Skip anything under (or itself) VCS / hidden path components.
+        if skip_hidden_dirs and _hidden_or_vcs(rel.parts):
+            continue
+        if not resolved.is_file() and not resolved.is_dir():
+            continue
+        matches.append(_rel(resolved, root))
+        if len(matches) >= max_matches:
+            return matches, True
+    return matches, False
+
+
+def _resolve_glob_base(path: str) -> tuple[Path, Path] | str:
+    try:
+        root = get_workspace_root()
+        base = resolve_path(path)
+    except WorkspaceError as exc:
+        return f"error: {exc}"
+    if not base.is_dir():
+        return f"error: not a directory: {path}"
+    return root, base
+
+
+@tool
+def glob_paths(
+    pattern: str,
+    path: str = ".",
+    *,
+    max_matches: int = DEFAULT_MAX_MATCHES,
+    skip_hidden_dirs: bool = True,
+) -> str:
+    """Find files under the workspace matching a glob pattern (e.g. ``**/*.py``).
+
+    Search is relative to ``path`` (default workspace root). Skips VCS dirs
+    (``.git``, …) and hidden directories by default. Returns paths relative to
+    the workspace root, one per line.
+    """
+    if not pattern.strip() or max_matches < 1:
+        return "error: pattern must not be empty and max_matches must be >= 1"
+
+    resolved = _resolve_glob_base(path)
+    if isinstance(resolved, str):
+        return resolved
+    root, base = resolved
+
+    result = _collect_glob(
+        base, root, pattern, max_matches=max_matches, skip_hidden_dirs=skip_hidden_dirs
+    )
+    if isinstance(result, str):
+        return result
+    matches, truncated = result
+    if not matches:
+        return "(no matches)"
+    body = "\n".join(matches)
+    if truncated:
+        body += f"\n...[truncated at {max_matches} matches]"
+    return body
diff --git a/tests/test_tools/test_glob_paths.py b/tests/test_tools/test_glob_paths.py
new file mode 100644
index 0000000..40b8630
--- /dev/null
+++ b/tests/test_tools/test_glob_paths.py
@@ -0,0 +1,42 @@
+from __future__ import annotations
+
+from pathlib import Path
+
+from plyngent.tools.file.glob_paths import glob_paths
+from tests.test_tools.helpers import call_sync
+
+
+def test_glob_paths_basic(workspace: object) -> None:
+    assert isinstance(workspace, Path)
+    _ = (workspace / "a.py").write_text("x", encoding="utf-8")
+    (workspace / "sub").mkdir()
+    _ = (workspace / "sub" / "b.py").write_text("y", encoding="utf-8")
+    _ = (workspace / "c.txt").write_text("z", encoding="utf-8")
+    out = call_sync(glob_paths, "**/*.py")
+    assert "a.py" in out
+    assert "sub/b.py" in out
+    assert "c.txt" not in out
+
+
+def test_glob_paths_skip_git(workspace: object) -> None:
+    assert isinstance(workspace, Path)
+    (workspace / ".git").mkdir()
+    _ = (workspace / ".git" / "config").write_text("x", encoding="utf-8")
+    _ = (workspace / "ok.py").write_text("x", encoding="utf-8")
+    out = call_sync(glob_paths, "**/*")
+    assert "ok.py" in out
+    assert ".git" not in out
+
+
+def test_glob_paths_max_matches(workspace: object) -> None:
+    assert isinstance(workspace, Path)
+    for i in range(5):
+        _ = (workspace / f"f{i}.txt").write_text("x", encoding="utf-8")
+    out = call_sync(glob_paths, "*.txt", max_matches=2)
+    assert "truncated" in out
+    assert out.count("\n") >= 1
+
+
+def test_glob_paths_empty_pattern(workspace: object) -> None:
+    del workspace
+    assert "error" in call_sync(glob_paths, "")