Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 6728

Summary:	RFC742 "/w" not expanded properly.
Product:	[Retired] Red Hat Linux	Reporter:	Dennis Brylow <dennis.brylow>
Component:	finger	Assignee:	Crutcher Dunnavant <crutcher>
Status:	CLOSED RAWHIDE	QA Contact:
Severity:	low	Docs Contact:
Priority:	low
Version:	6.2	CC:	menthos
Target Milestone:	---
Target Release:	---
Hardware:	i386
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2000-08-04 20:24:32 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	14002

Description Dennis Brylow 1999-11-04 17:38:09 UTC

bsd-finger-0.10-25 has code for compliance with an obscure
feature of RFC742, which, roughly paraphrased, states that
"/w" or "/W" is equivalent to "-l".
The code simply overwrites the "/w" with the "-l" before
passing the args to the finger command, which knows nothing
of "/w".  However, this results in finger requests of the
form "finger /wuser@machine" becoming "-luser" instead of
"-l user".

Comment 1 Jeff Johnson 1999-12-22 15:06:59 UTC

Does this problem persist in the latest finger-0.16-1 package in Raw Hide?
I'm not quite sure how to test, so, if you can supply a working test case,
I'd be grateful.

Comment 2 Jeff Johnson 2000-01-04 16:56:59 UTC

Fixed in finger-0.16-2. Thanks for the patch.

Comment 3 Dennis Brylow 2000-05-18 07:08:59 UTC

The bug was fixed in finger-0.16-2, but it has now reappeared in finger-0.16-5
for RedHat v6.2.

Here, again, is the patch:

This patch fixes the code for compliance with an obscure feature of RFC742,
which, roughly paraphrased, states that "/w" or "/W" is equivalent to "-l".
The previous code simply overwrites the "/w" with the "-l" before passing
the args to the finger command, which knows nothing of "/w".  This patch
adds in a space, so that "/wuser" is expanded to "-l user" instead of
"-luser", which would not be understood by finger.

--- bsd-finger-0.16/fingerd/fingerd.c	Sun Dec 12 13:46:28 1999
+++ bsd-finger-0.16/fingerd/fingerd.c	Thu May 18 01:56:54 2000
@@ -214,7 +214,12 @@
 	av[k++] = "finger";
 	for (s = strtok(line, WS); s && k<ENTRIES; s = strtok(NULL, WS)) {
 		/* RFC742: "/[Ww]" == "-l" */
-		if (!strncasecmp(s, "/w", 2)) memcpy(s, "-l", 2);
+		if (!strncasecmp(s, "/w", 2))
+		  {
+		    if (k < ENTRIES - 1)
+		      av[k++] = "-l";
+		    s += 2;
+		  }
 		if (!forwarding) {
 		    t = strchr(s, '@');
 		    if (t) {

Comment 4 Pekka Savola 2000-07-16 07:40:06 UTC

This is still an issue in Rawhide finger (0.17-0.1).

Comment 5 Jeff Johnson 2000-07-22 19:05:18 UTC

Fixed in finger-0.17-3. Thanks (again) for the patch.

Comment 6 Christian Rose 2000-08-04 18:39:13 UTC

I can't verify that with finger 0.17-3.

finger /W user

and

finger /w user

produce correct results, but not

finger /Wuser

or

finger /wuser

The just return "unknown user".

Comment 7 Dennis Brylow 2000-08-04 18:56:50 UTC

Executing a "finger /wuser" should be expected to return "unknown user".  The
"/w" is really not proper Linux/Unix command line style.  The issue is that the
finger daemon must translate /w's incoming from other systems.  Therefore,
the appropriate test is "finger /wuser@machine".

Comment 8 Jeff Johnson 2000-08-04 20:24:26 UTC

*** Bug 14002 has been marked as a duplicate of this bug. ***

Comment 9 Christian Rose 2000-08-04 20:28:29 UTC

My bad. One always learns something.
Could verify correct /wuser@machine and /Wuser@machine behavior when fingering
from another machine.
Closing this.