pdf-metaclean 1.0.1__tar.gz

pdf_metaclean-1.0.1/PKG-INFO

@@ -0,0 +1,91 @@
+Metadata-Version: 2.4
+Name: pdf-metaclean
+Version: 1.0.1
+Summary: Strip hidden metadata (author, producer, XMP) from PDFs before sharing — byte-level, no re-encoding.
+Author: Danilo Fortunato
+License-Expression: MIT
+Project-URL: Homepage, https://danyblitz-bit.github.io/pdf-metaclean/
+Project-URL: Repository, https://github.com/danyblitz-bit/pdf-metaclean
+Project-URL: Bug Tracker, https://github.com/danyblitz-bit/pdf-metaclean/issues
+Keywords: pdf,metadata,privacy,xmp,strip,redact,gdpr,cli
+Classifier: Programming Language :: Python :: 3
+Classifier: Environment :: Console
+Classifier: Topic :: Security
+Classifier: Topic :: Text Processing
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: End Users/Desktop
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: pypdf>=6.15.0
+
+# PDF MetaClean
+
+Remove sensitive metadata from PDF files before you share them. One command, three modes.
+
+## Why
+
+A PDF exported from your business software can silently carry your **name, company, software, and even editing timestamps** — freelancers leak their identity this way all the time. PDF MetaClean strips it all.
+
+## Install & Run
+
+```bash
+pip install -r requirements.txt
+
+# Inspect what's hidden in a PDF
+python -m tools.pdf-metaclean report.pdf
+
+# See the full metadata list
+python -m tools.pdf-metaclean report.pdf --verbose
+
+# Write a sanitized copy
+python -m tools.pdf-metaclean report.pdf --clean
+# -> report_clean.pdf
+
+# Overwrite the original
+python -m tools.pdf-metaclean report.pdf --in-place
+
+# Verify the tool works on your machine
+python -m tools.pdf-metaclean --self-check
+```
+
+## Options
+
+| Flag | Description |
+|------|-------------|
+| `--clean` | Write cleaned copy as `<name>_clean.pdf` |
+| `--in-place` | Overwrite the original file after cleaning |
+| `--verbose` | List all metadata found before cleaning |
+| `--self-check` | Run a built-in correctness test |
+
+## Why the byte-level strip
+
+pypdf silently re-injects `/Producer` on every write. PDF MetaClean strips the PDF `/Info` trailer entry at the byte level after writing, so **no metadata can survive** — verified by the built-in self-check.
+
+## Support the Project
+
+PDF MetaClean is free and open source. Like it?
+
+- [Get your site audited — €49](https://danyblitz.gumroad.com/l/jsuyla) — I run the audit and send you the fixes
+- [Buy me a coffee](https://danyblitz.gumroad.com/l/hrvpiu) — one-time support
+
+## Report a bug
+
+Found a bug or something weird? Run:
+
+```bash
+python -m tools.pdf-metaclean --report
+```
+
+This opens a pre-filled email. Send it and I'll get notified automatically.
+
+You can also email **danyblitz@googlemail.com** directly. Use the subject format:
+
+```
+[TOOL-REPORT] pdf-metaclean <what happened>
+```
+
+Attach the PDF or log output if you have one.
+
+## License
+
+MIT

pdf_metaclean-1.0.1/README.md

@@ -0,0 +1,71 @@
+# PDF MetaClean
+
+Remove sensitive metadata from PDF files before you share them. One command, three modes.
+
+## Why
+
+A PDF exported from your business software can silently carry your **name, company, software, and even editing timestamps** — freelancers leak their identity this way all the time. PDF MetaClean strips it all.
+
+## Install & Run
+
+```bash
+pip install -r requirements.txt
+
+# Inspect what's hidden in a PDF
+python -m tools.pdf-metaclean report.pdf
+
+# See the full metadata list
+python -m tools.pdf-metaclean report.pdf --verbose
+
+# Write a sanitized copy
+python -m tools.pdf-metaclean report.pdf --clean
+# -> report_clean.pdf
+
+# Overwrite the original
+python -m tools.pdf-metaclean report.pdf --in-place
+
+# Verify the tool works on your machine
+python -m tools.pdf-metaclean --self-check
+```
+
+## Options
+
+| Flag | Description |
+|------|-------------|
+| `--clean` | Write cleaned copy as `<name>_clean.pdf` |
+| `--in-place` | Overwrite the original file after cleaning |
+| `--verbose` | List all metadata found before cleaning |
+| `--self-check` | Run a built-in correctness test |
+
+## Why the byte-level strip
+
+pypdf silently re-injects `/Producer` on every write. PDF MetaClean strips the PDF `/Info` trailer entry at the byte level after writing, so **no metadata can survive** — verified by the built-in self-check.
+
+## Support the Project
+
+PDF MetaClean is free and open source. Like it?
+
+- [Get your site audited — €49](https://danyblitz.gumroad.com/l/jsuyla) — I run the audit and send you the fixes
+- [Buy me a coffee](https://danyblitz.gumroad.com/l/hrvpiu) — one-time support
+
+## Report a bug
+
+Found a bug or something weird? Run:
+
+```bash
+python -m tools.pdf-metaclean --report
+```
+
+This opens a pre-filled email. Send it and I'll get notified automatically.
+
+You can also email **danyblitz@googlemail.com** directly. Use the subject format:
+
+```
+[TOOL-REPORT] pdf-metaclean <what happened>
+```
+
+Attach the PDF or log output if you have one.
+
+## License
+
+MIT

pdf_metaclean-1.0.1/__init__.py

@@ -0,0 +1,2 @@
+"""PDF MetaClean — strip sensitive metadata from PDF files."""
+__version__ = "1.0.1"

pdf_metaclean-1.0.1/__main__.py

@@ -0,0 +1,151 @@
+"""PDF MetaClean — strip sensitive metadata before sharing PDFs."""
+import io
+import re
+import sys
+import argparse
+from pathlib import Path
+
+from pypdf import PdfReader, PdfWriter
+
+
+def audit(path: Path) -> dict:
+    reader = PdfReader(str(path))
+    meta = reader.metadata
+    return {
+        "pages": len(reader.pages),
+        "metadata": {k: str(v) for k, v in meta.items()} if meta else {},
+        "has_metadata": bool(meta),
+    }
+
+
+def clean(path: Path, out: Path) -> dict:
+    original = path.stat().st_size
+    reader = PdfReader(str(path))
+    writer = PdfWriter()
+    for page in reader.pages:
+        writer.add_page(page)
+
+    buf = io.BytesIO()
+    writer.write(buf)
+    raw = buf.getvalue().decode("latin-1")
+    # pypdf injects /Producer; strip the /Info reference and dict from the trailer outright.
+    # ponytail: flat-key Info dicts only; nested binary streams untouched. Upgrade to qpdf if exotic trailers appear.
+    raw = re.sub(r"/Info \d+ \d+ R", "", raw)
+    raw = re.sub(r"/Info <<[^>]*>>", "", raw)
+    out.write_bytes(raw.encode("latin-1"))
+
+    result = out.stat().st_size
+    return {"cleaned": True, "input_bytes": original, "output_bytes": result, "saved_bytes": original - result}
+
+
+def fmt(v: int) -> str:
+    return f"{v} B" if v < 1024 else f"{v/1024:.1f} KB"
+
+
+def self_check() -> int:
+    import tempfile
+
+    from pypdf import PdfWriter
+
+    buf = io.BytesIO()
+    w = PdfWriter()
+    w.add_blank_page(width=595, height=842)
+    w.add_metadata({"/Author": "Sensitive Name", "/Title": "secret"})
+    w.write(buf)
+
+    with tempfile.TemporaryDirectory() as d:
+        src = Path(d) / "in.pdf"
+        out = Path(d) / "out.pdf"
+        src.write_bytes(buf.getvalue())
+        res = clean(src, out)
+        assert res["cleaned"]
+        assert "/Author" not in out.read_bytes().decode("latin-1"), "Author metadata survived"
+        check = audit(out)
+        assert not check["has_metadata"], f"Metadata remained: {check['metadata']}"
+        assert check["pages"] == 1, "Page count changed"
+
+        # in-place: out is path, so the input size must be read before writing
+        ip = Path(d) / "inplace.pdf"
+        ip.write_bytes(buf.getvalue())
+        res2 = clean(ip, ip)
+        assert res2["saved_bytes"] > 0, f"in-place reported no savings: {res2}"
+    print("self-check OK: metadata stripped, page preserved")
+    return 0
+
+
+def report_to_email(message: str):
+    import webbrowser
+    from urllib.parse import quote
+
+    subject = "[TOOL-REPORT] pdf-metaclean bug or issue"
+    webbrowser.open(f"mailto:danyblitz@googlemail.com?subject={quote(subject)}&body={quote(message)}")
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        prog="pdf-metaclean",
+        description="Remove sensitive metadata from PDF files. Print-audit a file or clean it in place.",
+    )
+    parser.add_argument("file", nargs="*", help="PDF file(s) to audit or clean")
+    parser.add_argument("--self-check", action="store_true", help="Run internal correctness check")
+    output_mode = parser.add_mutually_exclusive_group()
+    output_mode.add_argument("--clean", action="store_true", help="Write a cleaned copy as <name>_clean.pdf")
+    output_mode.add_argument("--in-place", action="store_true", help="Overwrite the original after cleaning")
+    parser.add_argument("--verbose", action="store_true", help="Show full metadata before cleaning")
+    parser.add_argument("--report", action="store_true", help="Open a pre-filled email to report an issue")
+    args = parser.parse_args()
+
+    if args.self_check:
+        sys.exit(self_check())
+
+    if not args.file:
+        parser.error("at least one PDF file is required (or use --self-check)")
+
+    summary_lines = []
+    for file_arg in args.file:
+        p = Path(file_arg)
+        if not p.exists():
+            print(f"[SKIP] {p}: not found")
+            summary_lines.append(f"[SKIP] {p}: not found")
+            continue
+        if p.suffix.lower() != ".pdf":
+            print(f"[SKIP] {p}: not a PDF")
+            summary_lines.append(f"[SKIP] {p}: not a PDF")
+            continue
+
+        info = audit(p)
+        print(f"\n=== {p.name} ===")
+        print(f"  Pages: {info['pages']}  |  Size: {fmt(p.stat().st_size)}")
+        summary_lines.append(f"=== {p.name} === Pages: {info['pages']} Size: {fmt(p.stat().st_size)}")
+
+        if not info["has_metadata"]:
+            print("  Metadata: none found", "  (already clean)" if args.clean else "")
+            if args.clean and not args.in_place:
+                src = p.read_bytes()
+                out = p.with_name(p.stem + "_clean.pdf")
+                out.write_bytes(src)
+                print(f"  Cleaned copy written to {out.name} (no metadata to remove, size unchanged)")
+            continue
+
+        if args.verbose:
+            print("  Metadata found:")
+            for k, v in info["metadata"].items():
+                print(f"    {k}: {v[:80]}")
+                summary_lines.append(f"    {k}: {v[:80]}")
+
+        if args.clean or args.in_place:
+            out_path = p if args.in_place else p.with_name(p.stem + "_clean.pdf")
+            res = clean(p, out_path)
+            print(f"  Cleaned -> {out_path.name} ({fmt(res['output_bytes'])}, {fmt(res['saved_bytes'])} saved)")
+        else:
+            fields = ", ".join(info["metadata"].keys()) or "(none printable)"
+            print(f"  Metadata present: {fields}")
+            print("  Use --clean to write a sanitized copy, or --in-place to overwrite.")
+
+    if args.report:
+        report_to_email("\n".join(summary_lines))
+        print("\n  Email draft opened — send it and I'll get notified automatically.")
+
+
+if __name__ == "__main__":
+    main()

pdf_metaclean-1.0.1/pdf_metaclean.egg-info/PKG-INFO

@@ -0,0 +1,91 @@
+Metadata-Version: 2.4
+Name: pdf-metaclean
+Version: 1.0.1
+Summary: Strip hidden metadata (author, producer, XMP) from PDFs before sharing — byte-level, no re-encoding.
+Author: Danilo Fortunato
+License-Expression: MIT
+Project-URL: Homepage, https://danyblitz-bit.github.io/pdf-metaclean/
+Project-URL: Repository, https://github.com/danyblitz-bit/pdf-metaclean
+Project-URL: Bug Tracker, https://github.com/danyblitz-bit/pdf-metaclean/issues
+Keywords: pdf,metadata,privacy,xmp,strip,redact,gdpr,cli
+Classifier: Programming Language :: Python :: 3
+Classifier: Environment :: Console
+Classifier: Topic :: Security
+Classifier: Topic :: Text Processing
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: End Users/Desktop
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: pypdf>=6.15.0
+
+# PDF MetaClean
+
+Remove sensitive metadata from PDF files before you share them. One command, three modes.
+
+## Why
+
+A PDF exported from your business software can silently carry your **name, company, software, and even editing timestamps** — freelancers leak their identity this way all the time. PDF MetaClean strips it all.
+
+## Install & Run
+
+```bash
+pip install -r requirements.txt
+
+# Inspect what's hidden in a PDF
+python -m tools.pdf-metaclean report.pdf
+
+# See the full metadata list
+python -m tools.pdf-metaclean report.pdf --verbose
+
+# Write a sanitized copy
+python -m tools.pdf-metaclean report.pdf --clean
+# -> report_clean.pdf
+
+# Overwrite the original
+python -m tools.pdf-metaclean report.pdf --in-place
+
+# Verify the tool works on your machine
+python -m tools.pdf-metaclean --self-check
+```
+
+## Options
+
+| Flag | Description |
+|------|-------------|
+| `--clean` | Write cleaned copy as `<name>_clean.pdf` |
+| `--in-place` | Overwrite the original file after cleaning |
+| `--verbose` | List all metadata found before cleaning |
+| `--self-check` | Run a built-in correctness test |
+
+## Why the byte-level strip
+
+pypdf silently re-injects `/Producer` on every write. PDF MetaClean strips the PDF `/Info` trailer entry at the byte level after writing, so **no metadata can survive** — verified by the built-in self-check.
+
+## Support the Project
+
+PDF MetaClean is free and open source. Like it?
+
+- [Get your site audited — €49](https://danyblitz.gumroad.com/l/jsuyla) — I run the audit and send you the fixes
+- [Buy me a coffee](https://danyblitz.gumroad.com/l/hrvpiu) — one-time support
+
+## Report a bug
+
+Found a bug or something weird? Run:
+
+```bash
+python -m tools.pdf-metaclean --report
+```
+
+This opens a pre-filled email. Send it and I'll get notified automatically.
+
+You can also email **danyblitz@googlemail.com** directly. Use the subject format:
+
+```
+[TOOL-REPORT] pdf-metaclean <what happened>
+```
+
+Attach the PDF or log output if you have one.
+
+## License
+
+MIT

pdf_metaclean-1.0.1/pdf_metaclean.egg-info/SOURCES.txt

@@ -0,0 +1,12 @@
+README.md
+__init__.py
+__main__.py
+pyproject.toml
+./__init__.py
+./__main__.py
+pdf_metaclean.egg-info/PKG-INFO
+pdf_metaclean.egg-info/SOURCES.txt
+pdf_metaclean.egg-info/dependency_links.txt
+pdf_metaclean.egg-info/entry_points.txt
+pdf_metaclean.egg-info/requires.txt
+pdf_metaclean.egg-info/top_level.txt

pdf_metaclean-1.0.1/pdf_metaclean.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

pdf_metaclean-1.0.1/pdf_metaclean.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+pdfmetaclean = pdfmetaclean.__main__:main

pdf_metaclean-1.0.1/pdf_metaclean.egg-info/requires.txt

@@ -0,0 +1 @@
+pypdf>=6.15.0

pdf_metaclean-1.0.1/pdf_metaclean.egg-info/top_level.txt

@@ -0,0 +1 @@
+pdfmetaclean

pdf_metaclean-1.0.1/pyproject.toml

@@ -0,0 +1,45 @@
+[build-system]
+requires = ["setuptools>=68"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "pdf-metaclean"
+version = "1.0.1"
+description = "Strip hidden metadata (author, producer, XMP) from PDFs before sharing — byte-level, no re-encoding."
+readme = "README.md"
+requires-python = ">=3.9"
+license = "MIT"
+authors = [{ name = "Danilo Fortunato" }]
+keywords = [
+    "pdf",
+    "metadata",
+    "privacy",
+    "xmp",
+    "strip",
+    "redact",
+    "gdpr",
+    "cli",
+]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Environment :: Console",
+    "Topic :: Security",
+    "Topic :: Text Processing",
+    "Intended Audience :: Developers",
+    "Intended Audience :: End Users/Desktop",
+]
+dependencies = ["pypdf>=6.15.0"]
+
+[project.urls]
+Homepage = "https://danyblitz-bit.github.io/pdf-metaclean/"
+Repository = "https://github.com/danyblitz-bit/pdf-metaclean"
+"Bug Tracker" = "https://github.com/danyblitz-bit/pdf-metaclean/issues"
+
+[project.scripts]
+pdfmetaclean = "pdfmetaclean.__main__:main"
+
+[tool.setuptools]
+packages = ["pdfmetaclean"]
+
+[tool.setuptools.package-dir]
+pdfmetaclean = "."

pdf_metaclean-1.0.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+