Update filelock and virtualenv

Reviewed-on: #7

.filedust.conf.example

@@ -2,9 +2,12 @@
 # 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.
+# Patterns are matched against paths relative to $HOME
+# Supports:
+#   *   = one path segment
+#   **  = zero or more path segments (recursive)
+# Matching is case-sensitive
 
 [exclude]
 # Add directories or patterns you want filedust to ignore.

.gitea/workflows/lint-and-security.yml

@@ -22,8 +22,15 @@ jobs:
       - name: Run pre-commit hooks
         run: pre-commit run --all-files --color always
 
+      - name: Install Poetry
+        run: |
+          pip install poetry
+          poetry self add poetry-plugin-export
+
       - name: Install pip-audit
         run: pip install pip-audit
 
-      - name: Run pip-audit
-        run: pip-audit
+      - name: Audit dependencies (Poetry lockfile)
+        run: |
+          poetry export -f requirements.txt --without-hashes \
+            | pip-audit -r /dev/stdin

.gitea/workflows/trivy-scan.yml

@@ -0,0 +1,61 @@
+---
+name: Trivy Scan
+on:
+  schedule:
+    - cron: 17 8 * * *
+  workflow_dispatch:
+
+jobs:
+  security-scan:
+    runs-on: running-man
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Trivy scan via Docker
+        id: trivy
+        continue-on-error: true
+        run: |
+          docker run --rm \
+            --volumes-from "$HOSTNAME" \
+            aquasec/trivy:latest \
+            fs /workspace/guardutils/filedust \
+              --scanners vuln \
+              --pkg-types library \
+              --include-dev-deps \
+              --severity MEDIUM,HIGH,CRITICAL \
+              --ignore-unfixed \
+              --format json \
+              --output /workspace/guardutils/filedust/trivy.json \
+              --exit-code 1
+
+      - name: Notify Node-RED on vulnerabilities
+        if: steps.trivy.outcome == 'failure'
+        run: |
+          jq -r '
+          {
+            repo: "guardutils/filedust",
+            summary: (
+              "Total: " +
+              ((.Results[].Vulnerabilities | length) | tostring)
+            ),
+            vulnerabilities: [
+              .Results[].Vulnerabilities[] | {
+                library: .PkgName,
+                cve: .VulnerabilityID,
+                severity: .Severity,
+                installed: .InstalledVersion,
+                fixed: .FixedVersion,
+                title: .Title,
+                url: .PrimaryURL
+              }
+            ]
+          }
+          ' trivy.json \
+          | curl -s -X POST https://nodered.sysmd.uk/trivy-alert \
+              -H "Content-Type: application/json" \
+              --data-binary @-
+
+      - name: Fail workflow if vulnerabilities found
+        if: steps.trivy.outcome == 'failure'
+        run: exit 1

README.md

@@ -4,6 +4,10 @@
 
 # filedust
 
+<div align="center">
+  <img src="filedust.png" alt="filedust logo" width="256" />
+</div>
+
 **filedust** is a small, fast, and safe command-line tool that scans your filesystem for obvious junk — things like Python __pycache__ folders, build artifacts, editor backup files, and leftover temporary files — and cleans them up.
 
 Think of it as “`autoremove` for files.”

poetry.lock

@@ -173,13 +173,13 @@ test = ["pytest (>=6)"]
 
 [[package]]
 name = "filelock"
-version = "3.20.0"
+version = "3.20.3"
 description = "A platform independent file lock."
 optional = false
 python-versions = ">=3.10"
 files = [
-    {file = "filelock-3.20.0-py3-none-any.whl", hash = "sha256:339b4732ffda5cd79b13f4e2711a31b0365ce445d95d243bb996273d072546a2"},
-    {file = "filelock-3.20.0.tar.gz", hash = "sha256:711e943b4ec6be42e1d4e6690b48dc175c822967466bb31c0c293f34334c13f4"},
+    {file = "filelock-3.20.3-py3-none-any.whl", hash = "sha256:4b0dda527ee31078689fc205ec4f1c1bf7d56cf88b6dc9426c4f230e46c2dce1"},
+    {file = "filelock-3.20.3.tar.gz", hash = "sha256:18c57ee915c7ec61cff0ecf7f0f869936c7c30191bb0cf406f1341778d0834e1"},
 ]
 
 [[package]]
@@ -533,18 +533,18 @@ files = [
 
 [[package]]
 name = "virtualenv"
-version = "20.35.4"
+version = "20.36.1"
 description = "Virtual Python Environment builder"
 optional = false
 python-versions = ">=3.8"
 files = [
-    {file = "virtualenv-20.35.4-py3-none-any.whl", hash = "sha256:c21c9cede36c9753eeade68ba7d523529f228a403463376cf821eaae2b650f1b"},
-    {file = "virtualenv-20.35.4.tar.gz", hash = "sha256:643d3914d73d3eeb0c552cbb12d7e82adf0e504dbf86a3182f8771a153a1971c"},
+    {file = "virtualenv-20.36.1-py3-none-any.whl", hash = "sha256:575a8d6b124ef88f6f51d56d656132389f961062a9177016a50e4f507bbcc19f"},
+    {file = "virtualenv-20.36.1.tar.gz", hash = "sha256:8befb5c81842c641f8ee658481e42641c68b5eab3521d8e092d18320902466ba"},
 ]
 
 [package.dependencies]
 distlib = ">=0.3.7,<1"
-filelock = ">=3.12.2,<4"
+filelock = {version = ">=3.20.1,<4", markers = "python_version >= \"3.10\""}
 platformdirs = ">=3.9.1,<5"
 typing-extensions = {version = ">=4.13.2", markers = "python_version < \"3.11\""}
 

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "filedust"
-version = "0.3.1"
+version = "0.4.1"
 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"

src/filedust/junk.py

@@ -20,6 +20,7 @@ def load_user_rules() -> UserRules:
 
     if cfg_path.exists():
         parser = configparser.ConfigParser(allow_no_value=True)
+        parser.optionxform = str
         parser.read(cfg_path)
 
         if parser.has_section("include"):
@@ -32,8 +33,46 @@ def load_user_rules() -> UserRules:
 
 
 def matches_any(patterns: list[str], relpath: Path) -> bool:
-    posix = relpath.as_posix()
-    return any(fnmatch(posix, p) for p in patterns)
+    """
+    True globstar matcher.
+
+    Rules:
+    - *  matches exactly one path segment
+    - ** matches zero or more segments
+    - Patterns are relative to $HOME
+    """
+
+    path_parts = relpath.parts
+
+    for pat in patterns:
+        pat = pat.strip("/")
+
+        pat_parts = tuple(pat.split("/"))
+
+        if _match_parts(pat_parts, path_parts):
+            return True
+
+    return False
+
+
+def _match_parts(pat: tuple[str, ...], path: tuple[str, ...]) -> bool:
+    """Recursive glob matcher with ** support."""
+    if not pat:
+        return not path
+
+    if pat[0] == "**":
+        # ** matches zero or more segments
+        return _match_parts(pat[1:], path) or (
+            bool(path) and _match_parts(pat, path[1:])
+        )
+
+    if not path:
+        return False
+
+    if fnmatch(path[0], pat[0]):
+        return _match_parts(pat[1:], path[1:])
+
+    return False
 
 
 @dataclass
@@ -148,11 +187,11 @@ def iter_junk(root: Path, rules: UserRules | None = None) -> Iterable[Finding]:
     for dirpath, dirnames, filenames in os.walk(root, followlinks=False):
         dirpath_p = Path(dirpath)
 
-        # Fast relative path computation
-        if dirpath == root_str:
-            rel_dir = Path(".")
-        else:
-            rel_dir = Path(dirpath[len(root_str) :].lstrip("/"))
+        try:
+            rel_dir = dirpath_p.resolve().relative_to(HOME)
+        except ValueError:
+            # Should never happen due to earlier checks
+            continue
 
         # USER EXCLUDE → skip entire subtree
         if matches_any(rules.exclude, rel_dir):