From 7c8db7999a9192d50ad92f8f6322a53828e698bd Mon Sep 17 00:00:00 2001
From: Maks Snegov <snegov@spqr.link>
Date: Wed, 7 Oct 2026 13:04:52 -0700
Subject: [PATCH] Treat a known Message-ID as a duplicate in the mail importer

Outlook for Mac rewrites headers on every export, so a re-exported
message never hashes the same twice. Hash-only matching stored each
re-export as a variant beside the original.
---
 README.md              |  6 ++++--
 import-outlook-mail.py | 20 +++++++++-----------
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/README.md b/README.md
index 123301a..57115bb 100644
--- a/README.md
+++ b/README.md
@@ -21,8 +21,10 @@ filename. Unknown or invalid dates are stored under `unknown-date/`.
 The importer regenerates `index.csv` from the archive with the full SHA-256, selected normalized
 header values, and archive path. Its duplicate rules are:
 
-- An existing SHA-256 is an exact duplicate and is skipped.
-- A repeated `Message-ID` with different bytes is preserved and reported as a variant.
+- An existing SHA-256 is a duplicate and is skipped.
+- An existing `Message-ID` is a duplicate and is skipped. Outlook rewrites headers on every
+  export, so the same message rarely hashes the same twice.
+- A message without a `Message-ID` is checked by SHA-256 only.
 - A repeated subject alone is not a duplicate.
 
 Each new message is copied to a temporary file beside its final destination, SHA-256 verified, and
diff --git a/import-outlook-mail.py b/import-outlook-mail.py
index f66cd55..0056548 100755
--- a/import-outlook-mail.py
+++ b/import-outlook-mail.py
@@ -150,7 +150,7 @@ def main():
         parser.error(f"could not scan source directory: {error}")
     messages = [path for path in files if path.suffix.lower() == ".eml"]
     unsupported = [path for path in files if path.suffix.lower() != ".eml"]
-    imported = duplicates = variants = failures = 0
+    imported = duplicates = failures = 0
 
     for path in sorted(messages):
         try:
@@ -161,12 +161,13 @@ def main():
                 continue
 
             headers = read_headers(path)
+            if headers["message_id"] and headers["message_id"] in message_ids:
+                print(f"DUPLICATE {path.relative_to(source)}")
+                duplicates += 1
+                continue
+
             target = destination(archive, headers, digest)
-            is_variant = bool(
-                headers["message_id"] and headers["message_id"] in message_ids
-            )
-            label = "VARIANT" if is_variant else "IMPORT"
-            print(f"{label} {path.relative_to(source)} -> {target.relative_to(archive)}")
+            print(f"IMPORT {path.relative_to(source)} -> {target.relative_to(archive)}")
 
             target.parent.mkdir(parents=True, exist_ok=True)
             temporary = target.with_name(f".{target.name}.{os.getpid()}.tmp")
@@ -191,10 +192,7 @@ def main():
             hashes.add(digest)
             if headers["message_id"]:
                 message_ids.add(headers["message_id"])
-            if is_variant:
-                variants += 1
-            else:
-                imported += 1
+            imported += 1
         except (OSError, ValueError) as error:
             print(f"FAILED {path.relative_to(source)}: {error}")
             failures += 1
@@ -205,7 +203,7 @@ def main():
     write_index(archive, rows)
 
     print(
-        f"{imported} imported, {variants} variants, {duplicates} duplicates, "
+        f"{imported} imported, {duplicates} duplicates, "
         f"{len(unsupported)} unsupported, {failures} failures"
     )
     return 1 if failures or unsupported else 0