(tika) branch main updated: TIKA-4798: Don't overwrite dcterms:modified with PR_LAST_MODIFICATION_TIME for MSG files (#2987)

[tallison]

This is an automated email from the ASF dual-hosted git repository.

tballison pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tika.git


The following commit(s) were added to refs/heads/main by this push:
     new 0a8a39ff1e TIKA-4798: Don't overwrite dcterms:modified with 
PR_LAST_MODIFICATION_TIME for MSG files (#2987)
0a8a39ff1e is described below

commit 0a8a39ff1ecec1d8e433e8f3789b54f9f2b94bc3
Author: Nicholas DiPiazza <nicholas.dipia...@gmail.com>
AuthorDate: Fri Sep 18 10:18:36 2026 -0500

    TIKA-4798: Don't overwrite dcterms:modified with PR_LAST_MODIFICATION_TIME 
for MSG files (#2987)
---
 .../org/apache/tika/parser/microsoft/OutlookExtractor.java   | 12 +++++-------
 .../org/apache/tika/parser/microsoft/OutlookParserTest.java  |  7 +++++++
 2 files changed, 12 insertions(+), 7 deletions(-)

diff --git 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OutlookExtractor.java
 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OutlookExtractor.java
index 7c813c5085..13a42c5e13 100644
--- 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OutlookExtractor.java
+++ 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OutlookExtractor.java
@@ -575,13 +575,11 @@ public class OutlookExtractor extends 
AbstractPOIFSExtractor {
                 }
             }
         }
-        //try to overwrite the modified property if the actual 
LAST_MODIFICATION_TIME property exists.
-        List<PropertyValue> timeProp = 
msg.getMainChunks().getProperties().get(MAPIProperty.LAST_MODIFICATION_TIME);
-        if (timeProp != null && ! timeProp.isEmpty()) {
-            Calendar cal = 
((PropertyValue.TimePropertyValue)timeProp.get(0)).getValue();
-            metadata.set(TikaCoreProperties.MODIFIED, cal);
-        }
-
+        //Note: PR_LAST_MODIFICATION_TIME is intentionally not used to 
overwrite dcterms:modified here.
+        //For a .msg file, that MAPI property reflects when the message's 
underlying storage was
+        //last written (e.g. when the item was saved/exported to disk), not 
when its content was
+        //actually edited, and can be long after the message was 
sent/received. It remains available,
+        //unmodified, as mapi:last-modification-time (see 
handleMessageInfo/LITERAL_TIME_PROPERTIES).
     }
 
     private void handleBodyChunks(Chunk htmlChunk, Chunk rtfChunk, Chunk 
textChunk,
diff --git 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/test/java/org/apache/tika/parser/microsoft/OutlookParserTest.java
 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/test/java/org/apache/tika/parser/microsoft/OutlookParserTest.java
index 15088f98fa..d7628d0f32 100644
--- 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/test/java/org/apache/tika/parser/microsoft/OutlookParserTest.java
+++ 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/test/java/org/apache/tika/parser/microsoft/OutlookParserTest.java
@@ -82,6 +82,13 @@ public class OutlookParserTest extends TikaTest {
         // Stored as Thu, 5 Apr 2007 09:26:06 -0700
         assertEquals("2007-04-05T16:26:06Z", 
metadata.get(TikaCoreProperties.CREATED));
 
+        //TIKA-4798: dcterms:modified should reflect the message date, not
+        //PR_LAST_MODIFICATION_TIME, which for a .msg file records when the 
file's
+        //underlying storage was last written (e.g. resaved/exported to disk), 
and
+        //can be much later than when the message was actually sent/received.
+        assertEquals("2007-04-05T16:26:06Z", 
metadata.get(TikaCoreProperties.MODIFIED));
+        assertEquals("2007-10-05T05:37:49Z", 
metadata.get("mapi:last-modification-time"));
+
         String content = handler.toString();
         assertNotContained("Microsoft Outlook Express 6", content);
         assertNotContained("L'\u00C9quipe Microsoft Outlook Express", content);