Guys, specially Serge,
please examine this patch and tell me why I shouldn't commit it.
The symptom: setting headers on a MimeMessageWrapper object appeared to cause the
message-id to change
The problem: MimeMessageWrapper was calling saveChanges() in writeTo(), and
saveChanges() sets a new Message_Id every time, regardless.
The solution: remove saveChanges from writeTo()
Any other solution to the same problem would be OK by me, but its my firm belief that
we should not change message-ID's unless we clone the message.
d.
Index: MimeMessageWrapper.java
===================================================================
RCS file:
/home/cvs/jakarta-james/src/java/org/apache/james/core/MimeMessageWrapper.java,v
retrieving revision 1.19
diff -u -r1.19 MimeMessageWrapper.java
--- MimeMessageWrapper.java 7 Jan 2003 11:06:35 -0000 1.19
+++ MimeMessageWrapper.java 5 Feb 2003 14:44:35 -0000
@@ -200,7 +200,10 @@
}
in.close();
} else {
- message.saveChanges();
+ if(message.getMessageID()!=null){
+
+ }
+ // message.saveChanges();
message.writeTo(os);
}
}
@@ -263,7 +266,7 @@
MimeMessageWrapper wrapper = (MimeMessageWrapper)message;
wrapper.writeTo(headerOs, bodyOs, ignoreList);
} else {
- message.saveChanges();
+ //message.saveChanges();
//Write the headers (minus ignored ones)
Enumeration headers = message.getNonMatchingHeaderLines(ignoreList);
