From 62d65d20ad1157474a4948c4acbc4be22e40a6fa Mon Sep 17 00:00:00 2001
From: Ivo Oskamp <ivooskamp@oskamp.info>
Date: Fri, 9 Jan 2026 09:35:39 +0100
Subject: [PATCH] Auto-commit local changes before build (2026-01-09 09:35:39)

---
 .last-branch                                  |  2 +-
 .../src/backend/app/parsers/veeam.py          | 34 +++++++++++++++----
 docs/changelog.md                             |  5 +++
 3 files changed, 34 insertions(+), 7 deletions(-)

diff --git a/.last-branch b/.last-branch
index b538399..d5e3c52 100644
--- a/.last-branch
+++ b/.last-branch
@@ -1 +1 @@
-v20260108-39-changelog-0.1.19
+v20260109-01-veeam-m365-overall-message
diff --git a/containers/backupchecks/src/backend/app/parsers/veeam.py b/containers/backupchecks/src/backend/app/parsers/veeam.py
index 2b54e5e..4e3ff6d 100644
--- a/containers/backupchecks/src/backend/app/parsers/veeam.py
+++ b/containers/backupchecks/src/backend/app/parsers/veeam.py
@@ -347,12 +347,34 @@ def _extract_m365_overall_details_message(html: str) -> Optional[str]:
     if not html:
         return None
 
-    # Look for the summary "Details" cell (typically a header_td with rowspan).
-    candidates = re.findall(
-        r'<td[^>]*rowspan\s*=\s*["\']?\s*2\s*["\']?[^>]*>(.*?)</td>',
-        html,
-        flags=re.IGNORECASE | re.DOTALL,
-    )
+    html = _normalize_html(html)
+
+    # Strategy 1 (preferred): locate the "Details" header cell and then scan a small
+    # window after it for a rowspan cell that contains the overall message.
+    #
+    # We intentionally avoid a single giant regex over the entire HTML body to keep
+    # parsing fast and prevent worst-case backtracking on large messages.
+    candidates: List[str] = []
+
+    hdr = re.search(r'(?is)<td[^>]*>\s*<b>\s*Details\s*</b>\s*</td>', html)
+    if hdr:
+        window = html[hdr.end() : hdr.end() + 6000]
+        m = re.search(
+            r'(?is)<td[^>]*rowspan\s*=\s*["\']?\s*(?:2|3|4|5|6|7|8|9|10)\s*["\']?[^>]*>(.*?)</td>',
+            window,
+        )
+        if m:
+            candidates = [m.group(1)]
+
+    # Strategy 2 (fallback): look for rowspan cells with rowspan >= 2.
+    if not candidates:
+        all_rowspans = re.findall(
+            r'(?is)<td[^>]*rowspan\s*=\s*["\']?\s*([2-9]|10)\s*["\']?[^>]*>(.*?)</td>',
+            html,
+        )
+        # re.findall above returns tuples (rowspan, content)
+        candidates = [c[1] for c in all_rowspans] if all_rowspans else []
+
     if not candidates:
         return None
 
diff --git a/docs/changelog.md b/docs/changelog.md
index 3d435e7..b9e897a 100644
--- a/docs/changelog.md
+++ b/docs/changelog.md
@@ -1,3 +1,8 @@
+## v20260109-01-veeam-m365-overall-message
+
+- Fixed Veeam Backup for Microsoft 365 mail parsing where the overall summary message was not stored.
+- Improved extraction of the overall details message from the mail content, ensuring permission and role warnings are correctly captured.
+- Ensured the extracted overall message is consistently available in job details, run checks, and reporting views.
 
 
 ================================================================================================================================================