Bug 4455 - --reuse should only reuse when X-Spam-Status present
Summary: --reuse should only reuse when X-Spam-Status present
Status: RESOLVED DUPLICATE of bug 4461
Alias: None
Product: Spamassassin
Classification: Unclassified
Component: Masses (show other bugs)
Version: SVN Trunk (Latest Devel Version)
Hardware: Other other
: P5 normal
Target Milestone: Undefined
Assignee: SpamAssassin Developer Mailing List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-07-03 20:11 UTC by Duncan Findlay
Modified: 2005-07-05 13:03 UTC (History)
0 users


Attachment Type Modified Status Actions Submitter/CLA Status
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Duncan Findlay 2005-07-03 20:11:02 UTC 
Currently, --reuse disables all reusable rules by setting the score to 0.
Instead, it should attempt to run those tests when the message has no
X-Spam-Status header  This will skew results for mass-checks, but hopefully not
enough to invalidate them for 3.1.0.
Comment 1 Daniel Quinlan 2005-07-03 21:27:14 UTC 
If people running mass-checks have large sections of mail without X-Spam-Status
headers, perhaps that should be run without --reuse.  This is non-trivial.

Of course, it's much much easier to make sure your corpus has real-time
X-Spam-Status headers.
Comment 2 Rod Begbie 2005-07-03 22:27:47 UTC 
My corpus is a mixture of both.  There's the mail that comes into my "good"
addresses, get's run through SA and sorted based on the score, then there's the
"bad" addresses, which have never been legitimately used and get forwarded into
my corpus without touching SA, thus saving my poor stuttering mail server some
cycles.

Of the 17,450 spams in my June '05 folder, 9,269 have X-Spam-Status.

Which skew is "better"?  Running against the current DNSBL databases, or using
the "real time" values, but ignoring the hits for almost half my spam corpus?
Comment 3 Duncan Findlay 2005-07-05 21:03:12 UTC 

*** This bug has been marked as a duplicate of 4461 ***