Merge pull request #2 from guardutils/update_filedust_20251129

Improve sefety and add config file
- Add .cache and build to the skip dir list, make filedust run ONLY in the user home directory
- Major rewrite of junk.py, adding user config file for custom rules, don't treat broken symlink as junk
- Add filedust config file, update README, version bump

.filedust.conf.example

@@ -0,0 +1,20 @@
+# filedust configuration file
+# Place at: ~/.filedust.conf
+#
+# Use this file to customize cleanup behavior.
+# Only keys matter (no values). Paths are relative to $HOME.
+#
+# Patterns (globs) are allowed.
+
+[exclude]
+# Add directories or patterns you want filedust to ignore.
+# Examples:
+# Projects/important/*
+
+[include]
+# Add directories or patterns you want filedust to remove.
+# Examples:
+# node_modules
+# dist
+# *.tmp
+# *~

.gitignore

@@ -182,9 +182,9 @@ cython_debug/
 .abstra/
 
 # Visual Studio Code
-#  Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#  Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore
 #  that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
-#  and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#  and can be added to the global gitignore or merged into this file. However, if you prefer,
 #  you could uncomment the following to ignore the entire vscode folder
 # .vscode/
 

README.md

@@ -1,8 +1,7 @@
-![License](https://img.shields.io/github/license/mdaleo404/filedust)
-[![Language](https://img.shields.io/github/languages/top/mdaleo404/filedust.svg)](https://github.com/mdaleo404/filedust/)
-![GitHub Release](https://img.shields.io/github/v/release/mdaleo404/filedust?display_name=release&logo=github)
+[![License](https://img.shields.io/github/license/guardutils/filedust?style=flat)](LICENCE)
+[![Language](https://img.shields.io/github/languages/top/guardutils/filedust.svg)](https://github.com/guardutils/filedust/)
+![GitHub Release](https://img.shields.io/github/v/release/guardutils/filedust?display_name=release&logo=github)
 ![PyPI - Version](https://img.shields.io/pypi/v/filedust?logo=pypi)
-[![Build Status](https://img.shields.io/github/actions/workflow/status/mdaleo404/filedust/.github/workflows/lint-and-security.yml)](https://github.com/mdaleo404/filedust/actions)
 [![PyPI downloads](https://img.shields.io/pypi/dm/filedust.svg)](https://pypi.org/project/filedust/)
 
 # filedust
@@ -35,4 +34,38 @@ One interactive prompt at the end of the run (unless -y is used).
 Shows how much disk space can be freed.
 
 ### Safe by design
-Never touches dotfiles, configs, project files, or anything important.
+* It ONLY runs within user's `$HOME`
+
+* Put user in control by reading `~/.filedust.conf`
+
+* Never touches dotfiles, configs, project files, or anything important unless you want.
+
+## Installation
+
+### From PyPI
+```
+pip install filedust
+```
+
+### From this repository
+```
+git clone https://github.com/guardutils/filedust.git
+cd filedust/
+poetry install
+```
+
+### Custom config
+You can download the example and add your custom rule
+```
+wget -O ~/.filedust.conf https://raw.githubusercontent.com/guardutils/filedust/main/.filedust.conf.example
+```
+
+### TAB completion
+Add this to your `.bashrc`
+```
+eval "$(register-python-argcomplete filedust)"
+```
+And then
+```
+source ~/.bashrc
+```

poetry.lock

@@ -1,5 +1,19 @@
 # This file is automatically @generated by Poetry 1.8.4 and should not be changed by hand.
 
+[[package]]
+name = "argcomplete"
+version = "3.6.3"
+description = "Bash tab completion for argparse"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "argcomplete-3.6.3-py3-none-any.whl", hash = "sha256:f5007b3a600ccac5d25bbce33089211dfd49eab4a7718da3f10e3082525a92ce"},
+    {file = "argcomplete-3.6.3.tar.gz", hash = "sha256:62e8ed4fd6a45864acc8235409461b72c9a28ee785a2011cc5eb78318786c89c"},
+]
+
+[package.extras]
+test = ["coverage", "mypy", "pexpect", "ruff", "wheel"]
+
 [[package]]
 name = "cfgv"
 version = "3.5.0"
@@ -193,6 +207,40 @@ files = [
     {file = "iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730"},
 ]
 
+[[package]]
+name = "markdown-it-py"
+version = "4.0.0"
+description = "Python port of markdown-it. Markdown parsing, done right!"
+optional = false
+python-versions = ">=3.10"
+files = [
+    {file = "markdown_it_py-4.0.0-py3-none-any.whl", hash = "sha256:87327c59b172c5011896038353a81343b6754500a08cd7a4973bb48c6d578147"},
+    {file = "markdown_it_py-4.0.0.tar.gz", hash = "sha256:cb0a2b4aa34f932c007117b194e945bd74e0ec24133ceb5bac59009cda1cb9f3"},
+]
+
+[package.dependencies]
+mdurl = ">=0.1,<1.0"
+
+[package.extras]
+benchmarking = ["psutil", "pytest", "pytest-benchmark"]
+compare = ["commonmark (>=0.9,<1.0)", "markdown (>=3.4,<4.0)", "markdown-it-pyrs", "mistletoe (>=1.0,<2.0)", "mistune (>=3.0,<4.0)", "panflute (>=2.3,<3.0)"]
+linkify = ["linkify-it-py (>=1,<3)"]
+plugins = ["mdit-py-plugins (>=0.5.0)"]
+profiling = ["gprof2dot"]
+rtd = ["ipykernel", "jupyter_sphinx", "mdit-py-plugins (>=0.5.0)", "myst-parser", "pyyaml", "sphinx", "sphinx-book-theme (>=1.0,<2.0)", "sphinx-copybutton", "sphinx-design"]
+testing = ["coverage", "pytest", "pytest-cov", "pytest-regressions", "requests"]
+
+[[package]]
+name = "mdurl"
+version = "0.1.2"
+description = "Markdown URL utilities"
+optional = false
+python-versions = ">=3.7"
+files = [
+    {file = "mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8"},
+    {file = "mdurl-0.1.2.tar.gz", hash = "sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba"},
+]
+
 [[package]]
 name = "nodeenv"
 version = "1.9.1"
@@ -402,6 +450,25 @@ files = [
     {file = "pyyaml-6.0.3.tar.gz", hash = "sha256:d76623373421df22fb4cf8817020cbb7ef15c725b9d5e45f17e189bfc384190f"},
 ]
 
+[[package]]
+name = "rich"
+version = "13.9.4"
+description = "Render rich text, tables, progress bars, syntax highlighting, markdown and more to the terminal"
+optional = false
+python-versions = ">=3.8.0"
+files = [
+    {file = "rich-13.9.4-py3-none-any.whl", hash = "sha256:6049d5e6ec054bf2779ab3358186963bac2ea89175919d699e378b99738c2a90"},
+    {file = "rich-13.9.4.tar.gz", hash = "sha256:439594978a49a09530cff7ebc4b5c7103ef57baf48d5ea3184f21d9a2befa098"},
+]
+
+[package.dependencies]
+markdown-it-py = ">=2.2.0"
+pygments = ">=2.13.0,<3.0.0"
+typing-extensions = {version = ">=4.0.0,<5.0", markers = "python_version < \"3.11\""}
+
+[package.extras]
+jupyter = ["ipywidgets (>=7.5.1,<9)"]
+
 [[package]]
 name = "tomli"
 version = "2.3.0"
@@ -488,4 +555,4 @@ test = ["covdefaults (>=2.3)", "coverage (>=7.2.7)", "coverage-enable-subprocess
 [metadata]
 lock-version = "2.0"
 python-versions = ">=3.10,<4.0"
-content-hash = "98acd9fd57ec90c98a407b83122fd9c8ed432383e095a47d44e201bf187d3107"
+content-hash = "676393f654b241eb2ea6f983d589e83daff16e0b0ca2cb76228c0cd5ed447591"

pyproject.toml

@@ -1,16 +1,18 @@
 [tool.poetry]
 name = "filedust"
-version = "0.1.0"
+version = "0.3.0"
 description = "Opinionated junk cleaner for dev machines (caches, build artifacts, editor backups)."
 authors = ["Marco D'Aleo <marco@marcodaleo.com>"]
 license = "GPL-3.0-or-later"
 readme = "README.md"
-homepage = "https://github.com/mdaleo404/filedust"
-repository = "https://github.com/mdaleo404/filedust"
+homepage = "https://github.com/guardutils/filedust"
+repository = "https://github.com/guardutils/filedust"
 packages = [{ include = "filedust", from = "src" }]
 
 [tool.poetry.dependencies]
 python = ">=3.10,<4.0"
+rich = "^13.0.0"
+argcomplete = "^3.3.0"
 
 [tool.poetry.scripts]
 filedust = "filedust.cli:main"

src/filedust/cli.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import importlib.metadata
 import argparse
+import argcomplete
 import shutil
 from pathlib import Path
 from typing import List
@@ -11,7 +12,7 @@ from rich.table import Table
 from rich.prompt import Confirm
 from rich import box
 
-from .junk import Finding, iter_junk
+from .junk import Finding, iter_junk, load_user_rules
 
 
 console = Console()
@@ -96,6 +97,11 @@ def build_parser() -> argparse.ArgumentParser:
         help="Delete without prompting for confirmation.",
     )
 
+    try:
+        argcomplete.autocomplete(parser)
+    except ImportError:
+        pass
+
     return parser
 
 
@@ -162,10 +168,24 @@ def delete_all(findings: List[Finding]) -> int:
 
 
 def main(argv: list[str] | None = None) -> int:
+    print("Looking for junk ...")
     parser = build_parser()
     args = parser.parse_args(argv)
 
     root = Path(args.path).expanduser()
+    home = Path.home().resolve()
+    root_resolved = root.resolve()
+
+    # Ensure root is inside the user's home directory
+    try:
+        root_resolved.relative_to(home)
+    except ValueError:
+        console.print(
+            f"[red]Error:[/] Refusing to operate outside the user's home directory.\n"
+            f"Requested: {root_resolved}\n"
+            f"Allowed:   {home}"
+        )
+        return 1
 
     if not root.exists():
         console.print(f"[red]Error:[/] Path not found: {root}")
@@ -177,7 +197,8 @@ def main(argv: list[str] | None = None) -> int:
             "This may take a while and may require sudo for deletions.[/]"
         )
 
-    findings = list(iter_junk(root))
+    rules = load_user_rules()
+    findings = list(iter_junk(root, rules=rules))
     total_size = compute_total_size(findings)
 
     if not findings:

src/filedust/junk.py

@@ -1,12 +1,41 @@
 from __future__ import annotations
 
 import os
+import configparser
 from dataclasses import dataclass
 from fnmatch import fnmatch
 from pathlib import Path
 from typing import Iterable, List
 
 
+class UserRules:
+    def __init__(self):
+        self.include: list[str] = []
+        self.exclude: list[str] = []
+
+
+def load_user_rules() -> UserRules:
+    rules = UserRules()
+    cfg_path = Path.home() / ".filedust.conf"
+
+    if cfg_path.exists():
+        parser = configparser.ConfigParser(allow_no_value=True)
+        parser.read(cfg_path)
+
+        if parser.has_section("include"):
+            rules.include = list(parser["include"].keys())
+
+        if parser.has_section("exclude"):
+            rules.exclude = list(parser["exclude"].keys())
+
+    return rules
+
+
+def matches_any(patterns: list[str], relpath: Path) -> bool:
+    posix = relpath.as_posix()
+    return any(fnmatch(posix, p) for p in patterns)
+
+
 @dataclass
 class Finding:
     path: Path
@@ -23,7 +52,6 @@ JUNK_DIR_NAMES = {
     ".nox",
     ".tox",
     ".hypothesis",
-    ".cache",
     ".gradle",
     ".parcel-cache",
     ".turbo",
@@ -31,7 +59,6 @@ JUNK_DIR_NAMES = {
     ".vite",
     ".sass-cache",
     ".sass-cache",
-    "build",
     "dist",
 }
 
@@ -53,6 +80,9 @@ JUNK_FILE_PATTERNS = [
 
 # VCS / system dirs
 SKIP_DIR_NAMES = {
+    ".cache",
+    "build",
+    ".gnupg",
     ".git",
     ".hg",
     ".svn",
@@ -62,6 +92,34 @@ SKIP_DIR_NAMES = {
 }
 
 
+HOME = Path.home().resolve()
+
+
+def safe_exists(path: Path) -> bool | None:
+    """Return True/False if the path exists, or None if permission denied."""
+    try:
+        return path.exists()
+    except Exception:
+        return None
+
+
+def safe_resolve(path: Path, root: Path) -> Path | None:
+    """
+    Resolve symlinks only if safe.
+    Return resolved path if it stays within root.
+    Return None if:
+      - resolution escapes the root
+      - resolution fails
+      - permission denied
+    """
+    try:
+        resolved = path.resolve(strict=False)  # NEVER strict
+        resolved.relative_to(root)  # ensure containment
+        return resolved
+    except Exception:
+        return None
+
+
 def is_junk_dir_name(name: str) -> bool:
     return name in JUNK_DIR_NAMES
 
@@ -70,37 +128,140 @@ def is_junk_file_name(name: str) -> bool:
     return any(fnmatch(name, pattern) for pattern in JUNK_FILE_PATTERNS)
 
 
-def iter_junk(root: Path) -> Iterable[Finding]:
+def iter_junk(root: Path, rules: UserRules | None = None) -> Iterable[Finding]:
     """
-    Walk the tree under `root` and yield junk candidates.
+    Safe, fast junk scanner:
+      - Never follows symlinks.
+      - Broken symlinks are not automatically junk — they follow normal rules.
+      - User include/exclude overrides all.
+      - Built-in junk rules applied only when safe.
+      - SKIP_DIR_NAMES protected unless user includes.
+      - Fully contained in $HOME.
+      - No crashes from PermissionError or unreadable paths.
+    """
+    if rules is None:
+        rules = UserRules()
 
-    filedust:
-      - Skips known critical / config directories (SKIP_DIR_NAMES).
-      - Treats known "junk" directory names as removable as a whole.
-      - Treats known junk file patterns as removable.
-    """
     root = root.resolve()
+    root_str = str(root)
 
-    for dirpath, dirnames, filenames in os.walk(root):
+    for dirpath, dirnames, filenames in os.walk(root, followlinks=False):
         dirpath_p = Path(dirpath)
 
-        # Prune dirs we never touch at all.
-        dirnames[:] = [d for d in dirnames if d not in SKIP_DIR_NAMES]
+        # Fast relative path computation
+        if dirpath == root_str:
+            rel_dir = Path(".")
+        else:
+            rel_dir = Path(dirpath[len(root_str) :].lstrip("/"))
 
-        # Detect junk directories (and skip walking inside them).
+        # USER EXCLUDE → skip entire subtree
+        if matches_any(rules.exclude, rel_dir):
+            dirnames[:] = []
+            continue
+
+        pruned = []
+
+        # Handling dirs
+        for d in dirnames:
+            child = dirpath_p / d
+
+            try:
+                st = child.lstat()
+            except Exception:
+                continue  # unreadable
+
+            is_symlink = (st.st_mode & 0o170000) == 0o120000
+
+            if is_symlink:
+                # If broken symlink dir treat as file later via filenames (skip descent)
+                continue
+
+            rel_child = rel_dir / d
+
+            # User exclude wins
+            if matches_any(rules.exclude, rel_child):
+                continue
+
+            # SKIP_DIR_NAMES unless user includes
+            if d in SKIP_DIR_NAMES and not matches_any(
+                rules.include, rel_child
+            ):
+                continue
+
+            pruned.append(d)
+
+        dirnames[:] = pruned
+
+        # Detect JUNK dirs
         i = 0
         while i < len(dirnames):
             name = dirnames[i]
-            if is_junk_dir_name(name):
-                junk_dir = dirpath_p / name
-                yield Finding(path=junk_dir, kind="dir", reason="junk_dir")
-                # Remove from walk so we don't descend into it.
+            rel_child = rel_dir / name
+
+            # User include directory
+            if matches_any(rules.include, rel_child):
+                yield Finding(dirpath_p / name, "dir", "user_include")
                 del dirnames[i]
                 continue
+
+            # Built-in safe junk dirs
+            if is_junk_dir_name(name):
+                yield Finding(dirpath_p / name, "dir", "junk_dir")
+                del dirnames[i]
+                continue
+
             i += 1
 
-        # Now process files.
+        # Handling files (including symlinks)
         for fname in filenames:
+            fpath = dirpath_p / fname
+            rel_file = rel_dir / fname
+
+            try:
+                st = fpath.lstat()
+            except Exception:
+                continue
+
+            is_symlink = (st.st_mode & 0o170000) == 0o120000
+
+            # Handling broken symlinks
+            if is_symlink:
+                exists = safe_exists(fpath)
+
+                # Permission denied → skip
+                if exists is None:
+                    continue
+
+                # User exclude wins
+                if matches_any(rules.exclude, rel_file):
+                    continue
+
+                # User include wins
+                if matches_any(rules.include, rel_file):
+                    yield Finding(fpath, "file", "user_include")
+                    continue
+
+                # Broken symlink?
+                if exists is False:
+                    # DO NOT auto-delete — classify like regular file
+                    # Only built-in junk patterns apply
+                    if is_junk_file_name(fname):
+                        yield Finding(fpath, "file", "broken_symlink")
+                    continue
+
+                # Valid symlink — NEVER follow; only user-include counts
+                continue
+
+            # Regular files
+            # User exclude wins
+            if matches_any(rules.exclude, rel_file):
+                continue
+
+            # User include wins
+            if matches_any(rules.include, rel_file):
+                yield Finding(fpath, "file", "user_include")
+                continue
+
+            # Built-in junk patterns (safe ones)
             if is_junk_file_name(fname):
-                fpath = dirpath_p / fname
-                yield Finding(path=fpath, kind="file", reason="junk_file")
+                yield Finding(fpath, "file", "junk_file")