Fix CSV quoting for last_comment

CHANGELOG.md

@@ -1,8 +1,10 @@
 # 📝 Changelog
 
+- 2.10.1
+  - Always quote only the `last_comment` CSV field to avoid delimiter-related parsing issues (semicolon-separated).
 - 2.10.0
   - Add CSV reporting for full datastore scans (`--all-guests`), written atomically at the end with an ISO 8601 timestamp filename.
-  - Include namespace path with VM/CT ID, latest snapshot comment, and unique chunk size in GiB (fixed precision) in the CSV output.
+  - Include namespace path with VM/CT ID, latest snapshot comment (always quoted), and unique chunk size in GiB (fixed precision) in the CSV output.
   - Allow selecting the CSV output directory via `--csv-dir` and the interactive Options menu.
 - 2.9.0
   - Add an optional `--show-comments` flag and corresponding interactive option to show a short label derived from the latest PBS snapshot comment next to each VM/CT in the per‑guest summary and interactive search‑path selector (best‑effort, using `proxmox-backup-debug api get /admin/datastore/<DATASTORE>/snapshots`).
@@ -21,4 +23,3 @@
   - Keep `--version` as pure version output again.
 - 2.7.1
   - Split version and update logic: `--version` prints version; `--check-updates` checks and offers updates.
-+

README.md

@@ -9,7 +9,7 @@ It calculates the **real disk space usage** of a specific **namespace**, **VM**,
 
 This allows accurate insights into space consumption per tenant or object — useful for chargeback, reporting, and storage optimization.
 
-**Current version:** 2.10.0 (`./pbs_chunk_checker.py --version`)
+**Current version:** 2.10.1 (`./pbs_chunk_checker.py --version`)
 
 See full changes in `CHANGELOG.md`.
 
@@ -153,6 +153,7 @@ When running with `--all-guests`, the script writes a CSV report **after** the s
 - Output directory: current working directory by default, or via `--csv-dir` / the Options overlay
 - Separator: `;` (semicolon)
 - Columns: `namespace_path`, `last_comment`, `unique_size_gib`
+- The `last_comment` field is always quoted (other fields are not).
 - Size unit: GiB (1024^3 bytes), fixed 3-decimal precision
 - The CSV is written automically at the end so no partial file is left behind on errors
 

pbs_chunk_checker.py

@@ -18,10 +18,9 @@
 - Repository: https://github.com/VoltKraft/PBS_Chunk_Checker
 """
 
-__version__ = "2.10.0"
+__version__ = "2.10.1"
 
 import argparse
-import csv
 import concurrent.futures as futures
 import hashlib
 import json
@@ -2236,10 +2235,10 @@ def _write_full_scan_csv(
     tmp_path = Path(tmp_file.name)
     try:
         with tmp_file as handle:
-            writer = csv.writer(handle, delimiter=";")
-            writer.writerow(["namespace_path", "last_comment", "unique_size_gib"])
+            handle.write("namespace_path;last_comment;unique_size_gib\n")
             for path_label, comment, unique_bytes in rows:
-                writer.writerow([path_label, comment, _bytes_to_gib(unique_bytes)])
+                escaped_comment = (comment or "").replace('"', '""')
+                handle.write(f'{path_label};"{escaped_comment}";{_bytes_to_gib(unique_bytes)}\n')
         os.replace(tmp_path, final_path)
     except Exception:
         try: