From maristbrandi at gmail.com  Fri Jun  1 12:23:10 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Fri Jun  1 12:23:34 2007
Subject: [Rt-devel] Re: logo change
In-Reply-To: <a3a53f9f0705301253o78e80408o1c32deb6f293fb03@mail.gmail.com>
References: <270ACA5D2DB53248809CCD6885A16CA40283C889@PTEEXB02.mtsallstream.com>
	<a3a53f9f0705301253o78e80408o1c32deb6f293fb03@mail.gmail.com>
Message-ID: <a3a53f9f0706010923q76424c2eu44fe752fb5bfceee@mail.gmail.com>

Sean,

I've followed your directions but I'm still having a funny problem and I'm
wondering if you saw something similar. When I do

Set( $LogoImageURL, $WebImagesURL . " foxylogo.gif");

the image won't display. I'm positive that the image name is correct. But,
when I do

Set( $LogoImageURL, $WebImagesURL . " bplogo.gif");

it works just fine.

Any suggestions?

-Brandi L



On 5/30/07, Brandi L <maristbrandi@gmail.com> wrote:
>
> Hi Sean,
>
> Thanks so much for the advice! I'd tried following the directions on the
> Wiki and was having problems. I'll give it a try.
>
> Warmest regards,
> Brandi L
>
>
>  On 5/30/07, Conway, Sean <sean.conway@mtsallstream.com > wrote:
> >
> >  G'Day,
> >     I saw your posting and knowing I had the same problem. It is not as
> > straight forward as you are lead to believe on the wiki and the 3.6.3fix is not the same for
> > 3.6.4. I got it working for 3.6.4 but only with some help from BP
> > developers.
> >
> > stop http daemon
> >
> > In the RT_SiteConfig.pm add the following. Change the details for your
> > site
> > Set( $LogoLinkURL, 'http://www.mtsallstream.com');
> > Set( $LogoImageURL, $WebImagesURL . " mtslogo.gif");
> > Set( $LogoAltText, "MTS Allstream Corporate Logo");
> > Set( $LogoWidth, '177');
> > Set( $LogoHeight, '33');
> >
> > put your logo file in this location
> >
> >  /opt/rt3/share/html/NoAuth/images/mtslogo.gif
> >
> >
> >
> >
> >
> > modify /opt/rt3/share/html/Elements/Logo with the following line. I
> > don't have a copy of the original so I can't tell you what was there
> > originally.
> >
> > <a href="<%$RT::LogoLinkURL%>"><img src="<%$RT::LogoImageURL%>" alt="<%
> > loc($RT::LogoAltText) %>" width="<%$RT::LogoWidth%>"
> > height="<%$RT::LogoHeight%>" /></a>
> >
> >
> > if httpd was started stop the service (i.e. CentOS example) remove a
> > directory created everytime by mason then restart http with the modified
> > entries.
> >
> > service httpd stop
> >
> > rm -rf /opt/rt3/var/mason_data/obj
> >
> > service httpd start
> >
> >
> >
> >
> >
> > Sean Conway
> > Network System Specialist
> > Enhanced Systems
> > MTS Allstream Inc.
> > 640 Corydon Ave
> > Winnipeg, Manitoba
> >
> > Phone      204-941-7382
> > Cell          204-795-1497
> > Email       sean.conway@mtsallstream.com
> >
> > CONFIDENTIALITY WARNING:  THIS E-MAIL MESSAGE IS INTENDED ONLY FOR THE
> > PERSONAL AND CONFIDENTIAL USE  OF THE RECIPIENT(S) NAMED ABOVE.  IF THE
> > READER OF THIS MESSAGE IS NOT THE INTENDED RECIPIENT OR AN AGENT RESPONSIBLE
> > FOR DELIVERING IT TO THE INTENDED RECIPIENT, OR IF AFTER RECEIPT YOU ARE
> > ADVISED THAT THIS MESSAGE HAS BEEN SENT TO YOU IN ERROR, YOU ARE HEREBY
> > NOTIFIED THAT ANY REVIEW, DISSEMINATION, DISTRIBUTION, OR COPYING OF THIS
> > MESSAGE IS STRICTLY PROHIBITED.  IF YOU HAVE RECEIVED THIS MESSAGE IN ERROR,
> > PLEASE NOTIFY US IMMEDIATELY BY TELEPHONE AT THE NUMBER SET OUT ABOVE AND
> > DELETE THE ORIGINAL MESSAGE.
> >
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070601/d1f6f99f/attachment.htm
From joseph85750 at yahoo.com  Fri Jun  1 12:46:00 2007
From: joseph85750 at yahoo.com (Joseph Spenner)
Date: Fri Jun  1 12:53:32 2007
Subject: [Rt-devel] Re: logo change
In-Reply-To: <a3a53f9f0706010923q76424c2eu44fe752fb5bfceee@mail.gmail.com>
Message-ID: <603751.52377.qm@web30305.mail.mud.yahoo.com>

--- Brandi L <maristbrandi@gmail.com> wrote:
> 
> I've followed your directions but I'm still having a
> funny problem and I'm
> wondering if you saw something similar. When I do
> 
> Set( $LogoImageURL, $WebImagesURL . "
> foxylogo.gif");
> 
> the image won't display. I'm positive that the image
> name is correct. But,
> when I do
> 
> Set( $LogoImageURL, $WebImagesURL . " bplogo.gif");
> 
> it works just fine.
> 
> Any suggestions?
> 
> -Brandi L
> 

Try to right click on the 'broken link image', and see
what the URL is that it is trying to get.  That might
give you some idea where it is looking and failing.

Also, you could be experiencing an image cache
somewhere.  Try to SHIFT reload the image.



       
____________________________________________________________________________________
Get the Yahoo! toolbar and be alerted to new email wherever you're surfing.
http://new.toolbar.yahoo.com/toolbar/features/mail/index.php
From mikef at ack.berkeley.edu  Fri Jun  1 13:59:00 2007
From: mikef at ack.berkeley.edu (Mike Friedman)
Date: Fri Jun  1 14:01:01 2007
Subject: [Rt-devel] Re: logo change
In-Reply-To: <603751.52377.qm@web30305.mail.mud.yahoo.com>
References: <603751.52377.qm@web30305.mail.mud.yahoo.com>
Message-ID: <20070601104741.X63996@malcolm.berkeley.edu>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>> I've followed your directions but I'm still having a funny problem and 
>> I'm wondering if you saw something similar. When I do
>>
>> Set( $LogoImageURL, $WebImagesURL . " foxylogo.gif");
>>
>> the image won't display. I'm positive that the image name is correct. 
>> But, when I do
>>
>> Set( $LogoImageURL, $WebImagesURL . " bplogo.gif");
>>
>> it works just fine.

(That should be $LogoURL).

Although Brandi may be having a different problem, I should point out that 
even as recently as 3.6.3, RT doesn't use $LogoURL at all, even though you 
can set it in RT_SiteConfig.pm.

I've been carrying forward a local customization since 3.4.2, that changes 
the code in html/Elements/Logo to use $RT::LogoURL intead of the hardcoded 
'bplogo.gif' that appears in the RT distribution.  (I have also customized 
the 'alt' text and the link that is associated with the image).

Mike

_________________________________________________________________________
Mike Friedman                        Information Services & Technology
mikef@ack.Berkeley.EDU               2484 Shattuck Avenue
1-510-642-1410                       University of California at Berkeley
http://socrates.berkeley.edu/~mikef  http://ist.berkeley.edu
_________________________________________________________________________

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8

iQA/AwUBRmBeZ60bf1iNr4mCEQJLDACcDadDyj1AG1YjBocHT7uCEBjbapMAn1qk
3yGS48q9z1N2buBFOANJORjy
=Jecc
-----END PGP SIGNATURE-----
From sworley at chkno.net  Sun Jun  3 18:56:20 2007
From: sworley at chkno.net (Scott Worley)
Date: Sun Jun  3 19:02:40 2007
Subject: [Rt-devel] DefaultOwner
Message-ID: <46634714.3070509@chkno.net>

A previous round a discussion on this topic:
http://lists.fsck.com/pipermail/rt-devel/2005-June/007291.html

I'm looking for something with a little less duct-tape nature.

Attached is a patch to add a per-Queue DefaultOwner property that can be
manipulated through the UI in the normal way ( Admin/Queues/Modify.html ).

Note: I am not familiar with the RT codebase.  Critique is welcome.

-------------- next part --------------
diff -Naur rt-3.6.3/lib/RT/Queue.pm rt-3.6.3.with-default-owner/lib/RT/Queue.pm
--- rt-3.6.3/lib/RT/Queue.pm	2006-06-19 22:44:04.000000000 +0000
+++ rt-3.6.3.with-default-owner/lib/RT/Queue.pm	2007-06-03 19:23:37.000000000 +0000
@@ -94,6 +94,7 @@
   int(11) 'InitialPriority'.
   int(11) 'FinalPriority'.
   int(11) 'DefaultDueIn'.
+  int(11) 'DefaultOwner'.
   smallint(6) 'Disabled'.
 
 =cut
@@ -111,6 +112,7 @@
                 InitialPriority => '0',
                 FinalPriority => '0',
                 DefaultDueIn => '0',
+                DefaultOwner => $RT::Nobody->Id,
                 Disabled => '0',
 
 		  @_);
@@ -122,6 +124,7 @@
                          InitialPriority => $args{'InitialPriority'},
                          FinalPriority => $args{'FinalPriority'},
                          DefaultDueIn => $args{'DefaultDueIn'},
+                         DefaultOwner => $args{'DefaultOwner'},
                          Disabled => $args{'Disabled'},
 );
 
@@ -264,6 +267,24 @@
 =cut
 
 
+=head2 DefaultOwner
+
+Returns the current value of DefaultOwner. 
+(In the database, DefaultOwner is stored as int(11).)
+
+
+
+=head2 SetDefaultOwner VALUE
+
+
+Set DefaultOwner to VALUE. 
+Returns (1, 'Status message') on success and (0, 'Error Message') on failure.
+(In the database, DefaultOwner will be stored as a int(11).)
+
+
+=cut
+
+
 =head2 Creator
 
 Returns the current value of Creator. 
@@ -338,6 +359,8 @@
 		{read => 1, write => 1, sql_type => 4, length => 11,  is_blob => 0,  is_numeric => 1,  type => 'int(11)', default => '0'},
         DefaultDueIn => 
 		{read => 1, write => 1, sql_type => 4, length => 11,  is_blob => 0,  is_numeric => 1,  type => 'int(11)', default => '0'},
+        DefaultOwner => 
+		{read => 1, write => 1, sql_type => 4, length => 11,  is_blob => 0,  is_numeric => 1,  type => 'int(11)', default => '0'},
         Creator => 
 		{read => 1, auto => 1, sql_type => 4, length => 11,  is_blob => 0,  is_numeric => 1,  type => 'int(11)', default => '0'},
         Created => 
diff -Naur rt-3.6.3/lib/RT/Queue_Overlay.pm rt-3.6.3.with-default-owner/lib/RT/Queue_Overlay.pm
--- rt-3.6.3/lib/RT/Queue_Overlay.pm	2006-09-26 15:06:31.000000000 +0000
+++ rt-3.6.3.with-default-owner/lib/RT/Queue_Overlay.pm	2007-06-03 19:23:37.000000000 +0000
@@ -327,6 +327,7 @@
   InitialPriority
   FinalPriority
   DefaultDueIn
+  DefaultOwner
  
 If you pass the ACL check, it creates the queue and returns its queue id.
 
@@ -353,6 +354,7 @@
         InitialPriority   => "0",
         FinalPriority     => "0",
         DefaultDueIn      => "0",
+        DefaultOwner      => $RT::Nobody->Id,
         @_
     );
 
diff -Naur rt-3.6.3/lib/RT/Ticket_Overlay.pm rt-3.6.3.with-default-owner/lib/RT/Ticket_Overlay.pm
--- rt-3.6.3/lib/RT/Ticket_Overlay.pm	2006-12-01 19:00:23.000000000 +0000
+++ rt-3.6.3.with-default-owner/lib/RT/Ticket_Overlay.pm	2007-06-03 19:39:57.000000000 +0000
@@ -507,6 +507,10 @@
           )
           unless ( $Owner->Id );
     }
+    else {
+        $Owner = RT::User->new( $self->CurrentUser );
+        $Owner->Load( $QueueObj->DefaultOwner );
+    }
 
     #If we have a proposed owner and they don't have the right
     #to own a ticket, scream about it and make them not the owner
diff -Naur rt-3.6.3/html/Admin/Queues/Modify.html rt-3.6.3.with-default-owner/html/Admin/Queues/Modify.html
--- rt-3.6.3/html/Admin/Queues/Modify.html	2006-06-19 22:44:04.000000000 +0000
+++ rt-3.6.3.with-default-owner/html/Admin/Queues/Modify.html	2007-06-03 19:26:29.000000000 +0000
@@ -97,6 +97,11 @@
 </td><td>
 <input name="DefaultDueIn" value="<% ($Create) ? "" : $QueueObj->DefaultDueIn%>" /> <&|/l&>days</&>.
 </td>
+<td align="right">
+<&|/l&>Default owner</&>:
+</td><td>
+<& /Elements/SelectOwner, Name => "DefaultOwner", QueueObj => $QueueObj, Default => ($Create) ? $RT::Nobody->Id : $QueueObj->DefaultOwner, DefaultValue => 0 &>
+</td>
 </tr>
 <tr>
 <td>
@@ -144,7 +149,7 @@
 if ($QueueObj->Id()) {
     delete $session{'create_in_queues'};
 my @attribs= qw(Description CorrespondAddress CommentAddress Name 
-                InitialPriority FinalPriority DefaultDueIn);
+                InitialPriority FinalPriority DefaultDueIn DefaultOwner);
 
   @results = UpdateRecordObject( AttributesRef => \@attribs, 
 				    Object => $QueueObj, 
@@ -186,6 +191,7 @@
 $InitialPriority => undef
 $FinalPriority => undef
 $DefaultDueIn => undef
+$DefaultOwner => undef
 $SetEnabled => undef
 $Enabled => undef
 </%ARGS>
diff -Naur rt-3.6.3/html/REST/1.0/Forms/queue/default rt-3.6.3.with-default-owner/html/REST/1.0/Forms/queue/default
--- rt-3.6.3/html/REST/1.0/Forms/queue/default	2006-06-19 22:44:04.000000000 +0000
+++ rt-3.6.3.with-default-owner/html/REST/1.0/Forms/queue/default	2007-06-03 19:26:29.000000000 +0000
@@ -56,7 +56,7 @@
 my %data = %$changes;
 my $queue = new RT::Queue $session{CurrentUser};
 my @fields = qw(Name Description CorrespondAddress CommentAddress
-                InitialPriority FinalPriority DefaultDueIn);
+                InitialPriority FinalPriority DefaultDueIn DefaultOwner);
 my %fields = map { lc $_ => $_ } @fields;
 
 if ($id ne 'new') {
@@ -79,6 +79,7 @@
                 InitialPriority => "",
                 FinalPriority => "",
                 DefaultDueIn => "",
+                DefaultOwner => "",
             },
             0
         ];
diff -Naur rt-3.6.3/html/Ticket/Create.html rt-3.6.3.with-default-owner/html/Ticket/Create.html
--- rt-3.6.3/html/Ticket/Create.html	2006-09-26 15:06:31.000000000 +0000
+++ rt-3.6.3.with-default-owner/html/Ticket/Create.html	2007-06-03 19:26:29.000000000 +0000
@@ -72,7 +72,7 @@
 <&|/l&>Owner</&>:
 </td>
 <td class="value">
-<& /Elements/SelectOwner, Name => "Owner", QueueObj => $QueueObj, Default => $ARGS{Owner}||$RT::Nobody->Id, DefaultValue => 0 &>
+<& /Elements/SelectOwner, Name => "Owner", QueueObj => $QueueObj, Default => $ARGS{Owner}||$QueueObj->DefaultOwner||$RT::Nobody->Id, DefaultValue => 0 &>
 </td>
 </tr>
 <tr>
diff -Naur rt-3.6.3/etc/schema.Informix rt-3.6.3.with-default-owner/etc/schema.Informix
--- rt-3.6.3/etc/schema.Informix	2004-04-02 13:43:44.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/schema.Informix	2007-06-03 19:23:37.000000000 +0000
@@ -32,6 +32,7 @@
 	InitialPriority 	INTEGER DEFAULT 0 NOT NULL,
 	FinalPriority 		INTEGER DEFAULT 0 NOT NULL,
 	DefaultDueIn 		INTEGER DEFAULT 0 NOT NULL,
+	DefaultOwner 		INTEGER DEFAULT 0 NOT NULL,
 	Creator 		INTEGER DEFAULT 0 NOT NULL,
 	Created 		DATETIME YEAR TO SECOND,
 	LastUpdatedBy 		INTEGER DEFAULT 0 NOT NULL,
diff -Naur rt-3.6.3/etc/schema.Oracle rt-3.6.3.with-default-owner/etc/schema.Oracle
--- rt-3.6.3/etc/schema.Oracle	2006-12-15 04:39:19.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/schema.Oracle	2007-06-03 19:23:37.000000000 +0000
@@ -30,6 +30,7 @@
 	InitialPriority 	NUMBER(11,0) DEFAULT 0 NOT NULL,
 	FinalPriority 		NUMBER(11,0) DEFAULT 0 NOT NULL,
 	DefaultDueIn 		NUMBER(11,0) DEFAULT 0 NOT NULL,
+	DefaultOwner 		NUMBER(11,0) DEFAULT 0 NOT NULL,
 	Creator 		NUMBER(11,0) DEFAULT 0 NOT NULL,
 	Created 		DATE,
 	LastUpdatedBy 		NUMBER(11,0) DEFAULT 0 NOT NULL,
diff -Naur rt-3.6.3/etc/schema.Pg rt-3.6.3.with-default-owner/etc/schema.Pg
--- rt-3.6.3/etc/schema.Pg	2004-11-11 08:46:17.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/schema.Pg	2007-06-03 19:23:37.000000000 +0000
@@ -53,6 +53,7 @@
   InitialPriority integer NOT NULL DEFAULT 0  ,
   FinalPriority integer NOT NULL DEFAULT 0  ,
   DefaultDueIn integer NOT NULL DEFAULT 0  ,
+  DefaultOwner integer NOT NULL DEFAULT 0  ,
   Creator integer NOT NULL DEFAULT 0  ,
   Created TIMESTAMP NULL  ,
   LastUpdatedBy integer NOT NULL DEFAULT 0  ,
diff -Naur rt-3.6.3/etc/schema.SQLite rt-3.6.3.with-default-owner/etc/schema.SQLite
--- rt-3.6.3/etc/schema.SQLite	2004-11-11 08:46:17.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/schema.SQLite	2007-06-03 19:23:37.000000000 +0000
@@ -31,6 +31,7 @@
   InitialPriority integer NULL  ,
   FinalPriority integer NULL  ,
   DefaultDueIn integer NULL  ,
+  DefaultOwner integer NULL  ,
   Creator integer NULL  ,
   Created DATETIME NULL  ,
   LastUpdatedBy integer NULL  ,
diff -Naur rt-3.6.3/etc/schema.Sybase rt-3.6.3.with-default-owner/etc/schema.Sybase
--- rt-3.6.3/etc/schema.Sybase	2004-11-11 08:03:05.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/schema.Sybase	2007-06-03 19:23:37.000000000 +0000
@@ -31,6 +31,7 @@
   InitialPriority integer NOT NULL   ,
   FinalPriority integer NOT NULL   ,
   DefaultDueIn integer NOT NULL   ,
+  DefaultOwner integer NOT NULL   ,
   Creator integer NOT NULL   ,
   Created DATETIME NULL  ,
   LastUpdatedBy integer NOT NULL   ,
diff -Naur rt-3.6.3/etc/schema.mysql rt-3.6.3.with-default-owner/etc/schema.mysql
--- rt-3.6.3/etc/schema.mysql	2006-01-19 15:11:21.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/schema.mysql	2007-06-03 19:23:37.000000000 +0000
@@ -30,6 +30,7 @@
   InitialPriority integer NOT NULL DEFAULT 0  ,
   FinalPriority integer NOT NULL DEFAULT 0  ,
   DefaultDueIn integer NOT NULL DEFAULT 0  ,
+  DefaultOwner integer NOT NULL DEFAULT 0  ,
   Creator integer NOT NULL DEFAULT 0  ,
   Created DATETIME NULL  ,
   LastUpdatedBy integer NOT NULL DEFAULT 0  ,
diff -Naur rt-3.6.3/etc/upgrade/3.6.4/schema.Oracle rt-3.6.3.with-default-owner/etc/upgrade/3.6.4/schema.Oracle
--- rt-3.6.3/etc/upgrade/3.6.4/schema.Oracle	1970-01-01 00:00:00.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/upgrade/3.6.4/schema.Oracle	2007-06-03 19:46:54.000000000 +0000
@@ -0,0 +1 @@
+ALTER TABLE Queues ADD DefaultOwner NUMBER(11,0) NULL;
diff -Naur rt-3.6.3/etc/upgrade/3.6.4/schema.Pg rt-3.6.3.with-default-owner/etc/upgrade/3.6.4/schema.Pg
--- rt-3.6.3/etc/upgrade/3.6.4/schema.Pg	1970-01-01 00:00:00.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/upgrade/3.6.4/schema.Pg	2007-06-03 19:46:54.000000000 +0000
@@ -0,0 +1 @@
+alter table Queues add column DefaultOwner integer NULL;
diff -Naur rt-3.6.3/etc/upgrade/3.6.4/schema.mysql rt-3.6.3.with-default-owner/etc/upgrade/3.6.4/schema.mysql
--- rt-3.6.3/etc/upgrade/3.6.4/schema.mysql	1970-01-01 00:00:00.000000000 +0000
+++ rt-3.6.3.with-default-owner/etc/upgrade/3.6.4/schema.mysql	2007-06-03 19:46:54.000000000 +0000
@@ -0,0 +1 @@
+alter table Queues add column DefaultOwner integer NULL;

From vivek at khera.org  Sun Jun  3 19:12:41 2007
From: vivek at khera.org (Vivek Khera)
Date: Sun Jun  3 19:20:08 2007
Subject: [Rt-devel] DefaultOwner
In-Reply-To: <46634714.3070509@chkno.net>
References: <46634714.3070509@chkno.net>
Message-ID: <89D36AF4-60A1-41E3-BF47-DB91FBC4611F@khera.org>


On Jun 3, 2007, at 6:56 PM, Scott Worley wrote:

> I'm looking for something with a little less duct-tape nature.

Does setting the owner on the OnCreate scrip for the queue not work?

From jesse at bestpractical.com  Tue Jun  5 19:54:19 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Tue Jun  5 19:54:54 2007
Subject: [Rt-devel] RT Wiki - now running wifty
Message-ID: <EFE38DE4-2830-45E9-8E5A-0CCEB1CA1B03@bestpractical.com>

I've just finally converted it over from Kwiki to Wifty. I'm sure  
there will be hiccups, but please report them to me personally,  
rather than cluttering the lists.

Best,

Jesse

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070605/05577eab/PGP.pgp
From ruz at bestpractical.com  Tue Jun  5 22:13:31 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Tue Jun  5 22:13:38 2007
Subject: [Rt-devel] 3.6.4rc1 performance regression on a fairly complex
	search (CROSS JOIN???)
In-Reply-To: <C910AA47C468581A16DC3679@192.168.1.7>
References: <C910AA47C468581A16DC3679@192.168.1.7>
Message-ID: <589c94400706051913we052d67q8fa7b7881067f164@mail.gmail.com>

Hi, Dirk.

Can you try attached patch?

On 5/29/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
> Halle Jesse, hello Ruslan,
>
> the following ticket search query (captured from the advanced tab) leads to
> very different Searchbuilder compilations in 3.6.3 an 3.6.4rc1 (captured
> from mysql-slow.log):
>
>  Queue = 'trash' AND  (  Requestor.EmailAddress LIKE 'staff@' OR
> Requestor.EmailAddress LIKE 'math.fu' OR Requestor.EmailAddress LIKE
> 'inf.fu' OR Requestor.EmailAddress LIKE 'mi.fu' OR
> Requestor.EmailAddress LIKE 'cedis.fu' OR Requestor.EmailAddress LIKE
> 'zuv.fu' OR Requestor.EmailAddress LIKE 'pcpool.mi' ) AND  (  Status =
> 'new' OR Status = 'open' )
>
> is very straight forward and reasonably fast with 3.6.3:
>
[snip]

-- 
Best regards, Ruslan.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 3.6-try_to_speed_searches_by_watchers.patch
Type: text/x-patch
Size: 2822 bytes
Desc: not available
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070606/5e8eebb4/3.6-try_to_speed_searches_by_watchers.bin
From pape-rt at inf.fu-berlin.de  Wed Jun  6 03:17:07 2007
From: pape-rt at inf.fu-berlin.de (Dirk Pape)
Date: Wed Jun  6 03:17:49 2007
Subject: [Rt-devel] 3.6.4rc1 performance regression on a fairly complex
	search (CROSS JOIN???)
In-Reply-To: <589c94400706051913we052d67q8fa7b7881067f164@mail.gmail.com>
References: <C910AA47C468581A16DC3679@192.168.1.7>
	<589c94400706051913we052d67q8fa7b7881067f164@mail.gmail.com>
Message-ID: <4D93C6AF16BDBE36DFE05E66@[192.168.1.7]>

Hello Ruslan,

--Am 6. Juni 2007 06:13:31 +0400 schrieb Ruslan Zakirov 
<ruz@bestpractical.com>:

> Can you try attached patch?

yes, I did. It looks better now and answer is identical to that produced by 
3.6.3. Though the search is faster in 3.6.3 (it does not appear in 
mysql-slow.log)

Dirk.

# Query_time: 7  Lock_time: 0  Rows_sent: 1  Rows_examined: 259476
SELECT COUNT(DISTINCT main.id) FROM Tickets main CROSS JOIN Users Users_3 
JOIN Groups Groups_1  ON ( Groups_1.Domain = '
RT::Ticket-Role' ) AND ( Groups_1.Type = 'Requestor' ) AND ( 
Groups_1.Instance = main.id ) JOIN CachedGroupMembers Cache
dGroupMembers_2  ON ( CachedGroupMembers_2.MemberId = Users_3.id ) AND ( 
CachedGroupMembers_2.GroupId = Groups_1.id )  W
HERE (main.Status != 'deleted') AND (main.Queue = '57' AND  (  ( 
Users_3.EmailAddress LIKE '%staff@%' AND CachedGroupMem
bers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE '%math.fu%' AND 
CachedGroupMembers_2.id IS NOT NULL )  OR  ( Us
ers_3.EmailAddress LIKE '%inf.fu%' AND CachedGroupMembers_2.id IS NOT NULL 
)  OR  ( Users_3.EmailAddress LIKE '%mi.fu%'
AND CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE 
'%cedis.fu%' AND CachedGroupMembers_2.id IS N
OT NULL )  OR  ( Users_3.EmailAddress LIKE '%zuv.fu%' AND 
CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddr
ess LIKE '%pcpool.mi%' AND CachedGroupMembers_2.id IS NOT NULL )  )  AND  ( 
main.Status = 'new' OR main.Status = 'open'
) ) AND (main.Type = 'ticket') AND (main.EffectiveId = main.id);


-- 
Dr. Dirk Pape (eAS - Projektleitung Campus Management)
Freie Universitaet Berlin
Grunewaldstr. 34a, 12165 Berlin
Tel. +49 (30) 838 75143, Fax. +49 (30) 838 54654

From ruz at bestpractical.com  Wed Jun  6 07:17:09 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Wed Jun  6 07:17:57 2007
Subject: [Rt-devel] 3.6.4rc1 performance regression on a fairly complex
	search (CROSS JOIN???)
In-Reply-To: <4D93C6AF16BDBE36DFE05E66@192.168.1.7>
References: <C910AA47C468581A16DC3679@192.168.1.7>
	<589c94400706051913we052d67q8fa7b7881067f164@mail.gmail.com>
	<4D93C6AF16BDBE36DFE05E66@192.168.1.7>
Message-ID: <589c94400706060417j2e55f706vdf9f9222874c994b@mail.gmail.com>

Can you send EXPLAINs for both they are slightly different? Also, I
don't remember what version of mysql you're using.

On 6/6/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
> Hello Ruslan,
>
> --Am 6. Juni 2007 06:13:31 +0400 schrieb Ruslan Zakirov
> <ruz@bestpractical.com>:
>
> > Can you try attached patch?
>
> yes, I did. It looks better now and answer is identical to that produced by
> 3.6.3. Though the search is faster in 3.6.3 (it does not appear in
> mysql-slow.log)
>
> Dirk.
>
> # Query_time: 7  Lock_time: 0  Rows_sent: 1  Rows_examined: 259476
> SELECT COUNT(DISTINCT main.id) FROM Tickets main CROSS JOIN Users Users_3
> JOIN Groups Groups_1  ON ( Groups_1.Domain = '
> RT::Ticket-Role' ) AND ( Groups_1.Type = 'Requestor' ) AND (
> Groups_1.Instance = main.id ) JOIN CachedGroupMembers Cache
> dGroupMembers_2  ON ( CachedGroupMembers_2.MemberId = Users_3.id ) AND (
> CachedGroupMembers_2.GroupId = Groups_1.id )  W
> HERE (main.Status != 'deleted') AND (main.Queue = '57' AND  (  (
> Users_3.EmailAddress LIKE '%staff@%' AND CachedGroupMem
> bers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE '%math.fu%' AND
> CachedGroupMembers_2.id IS NOT NULL )  OR  ( Us
> ers_3.EmailAddress LIKE '%inf.fu%' AND CachedGroupMembers_2.id IS NOT NULL
> )  OR  ( Users_3.EmailAddress LIKE '%mi.fu%'
> AND CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE
> '%cedis.fu%' AND CachedGroupMembers_2.id IS N
> OT NULL )  OR  ( Users_3.EmailAddress LIKE '%zuv.fu%' AND
> CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddr
> ess LIKE '%pcpool.mi%' AND CachedGroupMembers_2.id IS NOT NULL )  )  AND  (
> main.Status = 'new' OR main.Status = 'open'
> ) ) AND (main.Type = 'ticket') AND (main.EffectiveId = main.id);
>
>
> --
> Dr. Dirk Pape (eAS - Projektleitung Campus Management)
> Freie Universitaet Berlin
> Grunewaldstr. 34a, 12165 Berlin
> Tel. +49 (30) 838 75143, Fax. +49 (30) 838 54654
>
>


-- 
Best regards, Ruslan.
From nick at ccl4.org  Wed Jun  6 12:39:47 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Wed Jun  6 12:40:18 2007
Subject: [Rt-devel] reliably serving images from inside the standalone server
Message-ID: <20070606163947.GI65611@plum.flirble.org>

I'm using the standalone server for development. It's attempting to server
image files from inside html/NoAuth/images verbatim by using an autohandler.
But Mason always wants to pass the files first as Mason components, so if they
happen to have '<&' (or a couple of other strings) Mason chokes.

I can't see a great way to override Mason. The best I can manage is to subvert
Mason as appended, by sniffing the file on the way in, and if it appears to
be anything recognisable, wrap the Mason source object to return a faked-up
Mason component that has the real content as a large literal '' string.

There is rather too much copying of the literal data here for my liking.
But I can't see a way to get the real absolute filename of the component
early enough to feed into File::Type, I can't see an official way to override
the Mason component compiler (which would avoid the need for the <%'$source'%>
hack, and I can't see a way to cleanly wrap the code callback used in the
upstream comp_source_ref() method.

Yes, it works.
Yes, it's small.
Yes, it's non-invasive.

That's about all it has going for it.

Nicholas Clark

--- lib/RT/Interface/Web/Standalone.pm~	Mon May 21 14:38:41 2007
+++ lib/RT/Interface/Web/Standalone.pm	Wed Jun  6 14:39:56 2007
@@ -49,6 +49,7 @@
 use base 'HTTP::Server::Simple::Mason';
 use RT::Interface::Web::Handler;
 use RT::Interface::Web;
+use MasonX::SimplisticServer;
 
 sub handler_class { "RT::Interface::Web::Handler" }
 
@@ -62,7 +63,7 @@
 } 
 
 sub default_mason_config {
-    return @RT::MasonParameters;
+    return interp_class => 'MasonX::SimplisticServer', @RT::MasonParameters;
 } 
 
 sub handle_request {
--- /dev/null	Wed Jun  6 16:33:00 2007
+++ lib/MasonX/SimplisticServer.pm	Wed Jun  6 16:24:34 2007
@@ -0,0 +1,58 @@
+package MasonX::SimplisticServer;
+use strict;
+use warnings;
+use vars qw($VERSION @ISA);
+$VERSION = '0.01';
+
+require HTML::Mason::Interp;
+
+@ISA = 'HTML::Mason::Interp';
+
+require File::Type;
+
+sub resolve_comp_path_to_source {
+    my $self = shift;
+    my $source_obj = $self->SUPER::resolve_comp_path_to_source(@_);
+
+    return unless $source_obj;
+
+    if ($source_obj) {
+	my $source = $source_obj->comp_source();
+	my $type = File::Type->new()->checktype_contents($source);
+	if (defined $type and $type !~ m!^text/!
+	    and $type ne 'application/octet-stream') {
+	    # application/octet-stream is the fallback default, and rather a
+	    # lot of Mason components end up there, as File::Type is looking
+	    # for an <html> tag before it says text/html
+	    # In fact, File::Type is 100% reluctant to say text/*
+	    bless $source_obj, 'MasonX::SimplisticServer::LiteralSource';
+	}
+    }
+
+    $source_obj;
+}
+
+package MasonX::SimplisticServer::LiteralSource;
+
+use strict;
+use warnings;
+use vars qw($VERSION @ISA);
+$VERSION = '0.01';
+
+require HTML::Mason::ComponentSource;
+
+@ISA = 'HTML::Mason::ComponentSource';
+
+sub comp_source_ref {
+    my $self = shift;
+    my $ref = $self->SUPER::comp_source_ref();
+    my $source = $$ref;
+    # Escape all the single quotes and backslashes
+    $source =~ s/([\\\'])/\\$1/g;
+    # So that we can make this a valid '' string literal that encodes the
+    # file contents, and return a reference to a page that has that '' string
+    \"<%'$source'%>";
+}
+
+1;
From js at oxado.com  Thu Jun  7 04:32:09 2007
From: js at oxado.com (Jonathan Scher)
Date: Thu Jun  7 04:37:26 2007
Subject: [Rt-devel] Search on custom fields
Message-ID: <4667C289.1090206@oxado.com>

Hello,

It is easy to get a list of ticket for a given requestor. Is it possible 
to search tickets based on a given custom field?
I've searched in the FAQ, but I have only found this page that allow us 
to display custom fields in result's searches...

Regards,
Jonathan Scher
From pape-rt at inf.fu-berlin.de  Thu Jun  7 05:32:28 2007
From: pape-rt at inf.fu-berlin.de (Dirk Pape)
Date: Thu Jun  7 05:32:41 2007
Subject: [Rt-devel] 3.6.4rc1 performance regression on a fairly complex
	search (CROSS JOIN???)
In-Reply-To: <589c94400706060417j2e55f706vdf9f9222874c994b@mail.gmail.com>
References: <C910AA47C468581A16DC3679@192.168.1.7>	
	<589c94400706051913we052d67q8fa7b7881067f164@mail.gmail.com>	
	<4D93C6AF16BDBE36DFE05E66@192.168.1.7>
	<589c94400706060417j2e55f706vdf9f9222874c994b@mail.gmail.com>
Message-ID: <39767ACA63FEA36BA7AA9C7E@idefix.mi.fu-berlin.de>

Hello Ruslan,

--Am 6. Juni 2007 15:17:09 +0400 schrieb Ruslan Zakirov 
<ruz@bestpractical.com>:

> Can you send EXPLAINs for both they are slightly different? Also, I
> don't remember what version of mysql you're using.

The version is: mysql  Ver 12.22 Distrib 4.0.24, for pc-linux-gnu (i386)

The following explain is for 3.6.3:

mysql> explain SELECT COUNT(DISTINCT main.id) FROM Tickets main JOIN Groups 
Groups_1  ON ( Groups_1.Instance = main.id ) JOIN CachedGroupMembers 
CachedGroupMembers_2 ON ( CachedGroupMembers_2.GroupId != 
CachedGroupMembers_2.MemberId ) AND ( CachedGroupMembers_2.GroupId = 
Groups_1.id ) JOIN Users Users_3  ON ( Users_3.id = 
CachedGroupMembers_2.MemberId )  WHERE (Groups_1.Domain = 
'RT::Ticket-Role') AND (main.Status != 'deleted') AND (Groups_1.Type = 
'Requestor') AND (main.Queue = '57' AND  (  ( Users_3.EmailAddress LIKE 
'%staff@%' OR Users_3.EmailAddress LIKE '%math.fu%' OR Users_3.EmailAddress 
LIKE '%inf.fu%' OR Users_3.EmailAddress LIKE '%mi.fu%' OR 
Users_3.EmailAddress LIKE '%cedis.fu%' OR Users_3.EmailAddress LIKE 
'%zuv.fu%' OR Users_3.EmailAddress LIKE '%pcpool.mi%' )  )  AND  ( 
main.Status = 'new' OR main.Status = 'open' ) ) AND (main.Type = 'ticket') 
AND (main.EffectiveId = 
main.id);+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
| table                | type   | possible_keys                      | key 
| key_len | ref                          | rows  | Extra 
|
+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
| Users_3              | index  | PRIMARY,Users3                     | 
Users4        |     121 | NULL                         | 20436 | Using 
where; Using index |
| CachedGroupMembers_2 | ref    | DisGrouMem,SHREDDER_CGM1           | 
SHREDDER_CGM1 |       5 | Users_3.id                   |     1 | Using 
where; Using index |
| Groups_1             | eq_ref | PRIMARY,Groups1,Groups2            | 
PRIMARY       |       4 | CachedGroupMembers_2.GroupId |     1 | Using 
where              |
| main                 | eq_ref | PRIMARY,Tickets1,Tickets5,Tickets4 | 
PRIMARY       |       4 | Groups_1.Instance            |     1 | Using 
where              |
+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
4 rows in set (0.00 sec)


and the next is for 3.6.4rc1+Ruslan-patch:


mysql> explain SELECT COUNT(DISTINCT main.id) FROM Tickets main CROSS JOIN 
Users Users_3 JOIN Groups Groups_1  ON ( Groups_1.Domain = 
'RT::Ticket-Role' ) AND ( Groups_1.Type = 'Requestor' ) AND ( 
Groups_1.Instance = main.id ) JOIN CachedGroupMembers CachedGroupMembers_2 
ON ( CachedGroupMembers_2.MemberId = Users_3.id ) AND ( 
CachedGroupMembers_2.GroupId = Groups_1.id )  WHERE (main.Status != 
'deleted') AND (main.Queue = '57' AND  (  ( Users_3.EmailAddress LIKE 
'%staff@%' AND CachedGroupMembers_2.id IS NOT NULL )  OR  ( 
Users_3.EmailAddress LIKE '%math.fu%' AND CachedGroupMembers_2.id IS NOT 
NULL )  OR  ( Users_3.EmailAddress LIKE '%inf.fu%' AND 
CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE 
'%mi.fu%'AND CachedGroupMembers_2.id IS NOT NULL )  OR  ( 
Users_3.EmailAddress LIKE '%cedis.fu%' AND CachedGroupMembers_2.id IS NOT 
NULL )  OR  ( Users_3.EmailAddress LIKE '%zuv.fu%' AND 
CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE 
'%pcpool.mi%' AND CachedGroupMembers_2.id IS NOT NULL )  )  AND  ( 
main.Status = 'new' OR main.Status = 'open') ) AND (main.Type = 'ticket') 
AND (main.EffectiveId = main.id);
+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
| table                | type   | possible_keys                      | key 
| key_len | ref                          | rows  | Extra 
|
+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
| Users_3              | index  | PRIMARY,Users3                     | 
Users4        |     121 | NULL                         | 20436 | Using 
where; Using index |
| CachedGroupMembers_2 | ref    | PRIMARY,DisGrouMem,SHREDDER_CGM1   | 
SHREDDER_CGM1 |       5 | Users_3.id                   |     1 | Using 
where; Using index |
| Groups_1             | eq_ref | PRIMARY,Groups1,Groups2            | 
PRIMARY       |       4 | CachedGroupMembers_2.GroupId |     1 | Using 
where              |
| main                 | eq_ref | PRIMARY,Tickets1,Tickets5,Tickets4 | 
PRIMARY       |       4 | Groups_1.Instance            |     1 | Using 
where              |
+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
4 rows in set (0.00 sec)

Regards, Dirk.

-- 
Dr. Dirk Pape (eAS - Projektleitung Campus Management)
Freie Universitaet Berlin
Grunewaldstr. 34a, 12165 Berlin
Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
From gentgeen at linuxmail.org  Thu Jun  7 09:09:52 2007
From: gentgeen at linuxmail.org (Kevin Squire)
Date: Thu Jun  7 09:13:24 2007
Subject: [Rt-devel] Search on custom fields
In-Reply-To: <4667C289.1090206@oxado.com>
References: <4667C289.1090206@oxado.com>
Message-ID: <20070607090952.046a1060@localhost.localdomain>

On Thu, 07 Jun 2007 10:32:09 +0200
Jonathan Scher <js@oxado.com> wrote:

> Hello,
> 
> It is easy to get a list of ticket for a given requestor. Is it possible 
> to search tickets based on a given custom field?
> I've searched in the FAQ, but I have only found this page that allow us 
> to display custom fields in result's searches...
> 
> Regards,
> Jonathan Scher
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
> 

You must first set the Queue in your search.  Then you will be given the options to search on the custom fields related to that Queue.

-- 
http://gentgeen.homelinux.org

#############################################################
 Associate yourself with men of good quality if you esteem    
 your own reputation; for 'tis better to be alone then in bad 
 company.        - George Washington, Rules of Civility
From ruz at bestpractical.com  Thu Jun  7 10:54:46 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Thu Jun  7 10:55:25 2007
Subject: [Rt-devel] 3.6.4rc1 performance regression on a fairly complex
	search (CROSS JOIN???)
In-Reply-To: <39767ACA63FEA36BA7AA9C7E@idefix.mi.fu-berlin.de>
References: <C910AA47C468581A16DC3679@192.168.1.7>
	<589c94400706051913we052d67q8fa7b7881067f164@mail.gmail.com>
	<4D93C6AF16BDBE36DFE05E66@192.168.1.7>
	<589c94400706060417j2e55f706vdf9f9222874c994b@mail.gmail.com>
	<39767ACA63FEA36BA7AA9C7E@idefix.mi.fu-berlin.de>
Message-ID: <589c94400706070754y1d835ab9va63abfee5c0ade50@mail.gmail.com>

Ok, equiv patch is in the repo. Thanks, Dirk, for reporting,
investigating and testing :)

On 6/7/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
> Hello Ruslan,
>
> --Am 6. Juni 2007 15:17:09 +0400 schrieb Ruslan Zakirov
> <ruz@bestpractical.com>:
>
> > Can you send EXPLAINs for both they are slightly different? Also, I
> > don't remember what version of mysql you're using.
>
> The version is: mysql  Ver 12.22 Distrib 4.0.24, for pc-linux-gnu (i386)
>
> The following explain is for 3.6.3:
>
> mysql> explain SELECT COUNT(DISTINCT main.id) FROM Tickets main JOIN Groups
> Groups_1  ON ( Groups_1.Instance = main.id ) JOIN CachedGroupMembers
> CachedGroupMembers_2 ON ( CachedGroupMembers_2.GroupId !=
> CachedGroupMembers_2.MemberId ) AND ( CachedGroupMembers_2.GroupId =
> Groups_1.id ) JOIN Users Users_3  ON ( Users_3.id =
> CachedGroupMembers_2.MemberId )  WHERE (Groups_1.Domain =
> 'RT::Ticket-Role') AND (main.Status != 'deleted') AND (Groups_1.Type =
> 'Requestor') AND (main.Queue = '57' AND  (  ( Users_3.EmailAddress LIKE
> '%staff@%' OR Users_3.EmailAddress LIKE '%math.fu%' OR Users_3.EmailAddress
> LIKE '%inf.fu%' OR Users_3.EmailAddress LIKE '%mi.fu%' OR
> Users_3.EmailAddress LIKE '%cedis.fu%' OR Users_3.EmailAddress LIKE
> '%zuv.fu%' OR Users_3.EmailAddress LIKE '%pcpool.mi%' )  )  AND  (
> main.Status = 'new' OR main.Status = 'open' ) ) AND (main.Type = 'ticket')
> AND (main.EffectiveId =
> main.id);+----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
> | table                | type   | possible_keys                      | key
> | key_len | ref                          | rows  | Extra
> |
> +----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
> | Users_3              | index  | PRIMARY,Users3                     |
> Users4        |     121 | NULL                         | 20436 | Using
> where; Using index |
> | CachedGroupMembers_2 | ref    | DisGrouMem,SHREDDER_CGM1           |
> SHREDDER_CGM1 |       5 | Users_3.id                   |     1 | Using
> where; Using index |
> | Groups_1             | eq_ref | PRIMARY,Groups1,Groups2            |
> PRIMARY       |       4 | CachedGroupMembers_2.GroupId |     1 | Using
> where              |
> | main                 | eq_ref | PRIMARY,Tickets1,Tickets5,Tickets4 |
> PRIMARY       |       4 | Groups_1.Instance            |     1 | Using
> where              |
> +----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
> 4 rows in set (0.00 sec)
>
>
> and the next is for 3.6.4rc1+Ruslan-patch:
>
>
> mysql> explain SELECT COUNT(DISTINCT main.id) FROM Tickets main CROSS JOIN
> Users Users_3 JOIN Groups Groups_1  ON ( Groups_1.Domain =
> 'RT::Ticket-Role' ) AND ( Groups_1.Type = 'Requestor' ) AND (
> Groups_1.Instance = main.id ) JOIN CachedGroupMembers CachedGroupMembers_2
> ON ( CachedGroupMembers_2.MemberId = Users_3.id ) AND (
> CachedGroupMembers_2.GroupId = Groups_1.id )  WHERE (main.Status !=
> 'deleted') AND (main.Queue = '57' AND  (  ( Users_3.EmailAddress LIKE
> '%staff@%' AND CachedGroupMembers_2.id IS NOT NULL )  OR  (
> Users_3.EmailAddress LIKE '%math.fu%' AND CachedGroupMembers_2.id IS NOT
> NULL )  OR  ( Users_3.EmailAddress LIKE '%inf.fu%' AND
> CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE
> '%mi.fu%'AND CachedGroupMembers_2.id IS NOT NULL )  OR  (
> Users_3.EmailAddress LIKE '%cedis.fu%' AND CachedGroupMembers_2.id IS NOT
> NULL )  OR  ( Users_3.EmailAddress LIKE '%zuv.fu%' AND
> CachedGroupMembers_2.id IS NOT NULL )  OR  ( Users_3.EmailAddress LIKE
> '%pcpool.mi%' AND CachedGroupMembers_2.id IS NOT NULL )  )  AND  (
> main.Status = 'new' OR main.Status = 'open') ) AND (main.Type = 'ticket')
> AND (main.EffectiveId = main.id);
> +----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
> | table                | type   | possible_keys                      | key
> | key_len | ref                          | rows  | Extra
> |
> +----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
> | Users_3              | index  | PRIMARY,Users3                     |
> Users4        |     121 | NULL                         | 20436 | Using
> where; Using index |
> | CachedGroupMembers_2 | ref    | PRIMARY,DisGrouMem,SHREDDER_CGM1   |
> SHREDDER_CGM1 |       5 | Users_3.id                   |     1 | Using
> where; Using index |
> | Groups_1             | eq_ref | PRIMARY,Groups1,Groups2            |
> PRIMARY       |       4 | CachedGroupMembers_2.GroupId |     1 | Using
> where              |
> | main                 | eq_ref | PRIMARY,Tickets1,Tickets5,Tickets4 |
> PRIMARY       |       4 | Groups_1.Instance            |     1 | Using
> where              |
> +----------------------+--------+------------------------------------+---------------+---------+------------------------------+-------+--------------------------+
> 4 rows in set (0.00 sec)
>
> Regards, Dirk.
>
> --
> Dr. Dirk Pape (eAS - Projektleitung Campus Management)
> Freie Universitaet Berlin
> Grunewaldstr. 34a, 12165 Berlin
> Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
>


-- 
Best regards, Ruslan.
From jesse at bestpractical.com  Thu Jun  7 13:32:06 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Thu Jun  7 13:32:24 2007
Subject: [Rt-devel] RT 3.6.4rc2
Message-ID: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>

I'm pleased to announce the second release candidate of RT 3.6.4:

	http://download.bestpractical.com/pub/rt/devel/rt-3.6.4rc2.tar.gz
	http://download.bestpractical.com/pub/rt/devel/rt-3.6.4rc2.tar.gz.sig

	MD5 (rt-3.6.4rc2.tar.gz) = 12578c66dde1a90067857a834b9edbc2
	MD5 (rt-3.6.4rc2.tar.gz.sig) = a0fbc02145766e0b4b9c34cdd405e556



Changes:
	several fixes from Nicholas,
	back to fast searches by watchers (regression in rc1),
	updated deps
	"Log a notice when we choose not to redistribute autogenerated  
messages"
	$RT::Timezone should now actually work.


I'd like to hear a couple success reports and confirmation that we've  
fixed the regression in rc1 before rolling the final next week.

(As a heads up, I'll be pretty-much 100% offline from 20 June to 4  
July. So if it doesn't happen by 15 June or so, expect it not to  
happen until July)

	Jesse

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070607/9c914dd8/PGP.pgp
From jesse at bestpractical.com  Thu Jun  7 13:32:06 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Thu Jun  7 13:40:34 2007
Subject: [Rt-devel] RT 3.6.4rc2
Message-ID: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>

I'm pleased to announce the second release candidate of RT 3.6.4:

	http://download.bestpractical.com/pub/rt/devel/rt-3.6.4rc2.tar.gz
	http://download.bestpractical.com/pub/rt/devel/rt-3.6.4rc2.tar.gz.sig

	MD5 (rt-3.6.4rc2.tar.gz) = 12578c66dde1a90067857a834b9edbc2
	MD5 (rt-3.6.4rc2.tar.gz.sig) = a0fbc02145766e0b4b9c34cdd405e556



Changes:
	several fixes from Nicholas,
	back to fast searches by watchers (regression in rc1),
	updated deps
	"Log a notice when we choose not to redistribute autogenerated  
messages"
	$RT::Timezone should now actually work.


I'd like to hear a couple success reports and confirmation that we've  
fixed the regression in rc1 before rolling the final next week.

(As a heads up, I'll be pretty-much 100% offline from 20 June to 4  
July. So if it doesn't happen by 15 June or so, expect it not to  
happen until July)

	Jesse

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070607/9c914dd8/PGP-0001.pgp
From jesse at bestpractical.com  Thu Jun  7 19:17:59 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Thu Jun  7 19:18:18 2007
Subject: [Rt-devel] RT 3.4.7rc1
Message-ID: <5F47DB20-6C59-4592-AF63-BE88748EE9E5@bestpractical.com>

I'm pleased to announce a new release candidate for the previous  
stable series of RT.


* Fixed compatibility with DBIx::SearchBuilder 1.48 or greater on
   RT instances with the RTIR extension installed.

* Fixed generation of SQL for searches by attachments, so searches
   like 'Subject LIKE "xxx" OR Content LIKE "yyy"' should work
   much better, but they are still may require many resourses.

* Fixed display of the SelfService "Display" page after create on the  
web.
   Thanks to Wilson Chow for finding the issue.

* Added missing UNIVERSAL::require dependency.

* Bumped version requirements of Text::Quoted module as there are fixes
   for several bugs related to RT.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070607/d3f34e68/PGP.pgp
From Sean.Edge at mtvnmix.com  Mon Jun 11 15:37:33 2007
From: Sean.Edge at mtvnmix.com (Edge, Sean)
Date: Mon Jun 11 15:48:10 2007
Subject: [Rt-devel] Display scrip transactions in "Results:"
Message-ID: <9DB5A62CCC96094EBA95F593C589CE1638B757@LAZYTOWN.mtvn.ad.viacom.com>

I first sent this to RT-Users but realized this is probably the better
place to send it....

> I created a Scrip to make sure that "Time Worked" is not zero when a
> ticket is 'resolved'.  If there is no "Time Worked" defined then I set
> the ticket status back to 'open' and log a comment.  Here's the code:
> 
> *****************
> if ((! defined($self->TicketObj->TimeWorked) ||
>    ($self->TicketObj->TimeWorked == 0)) { 
>     my ($id, $msg) = $self->TicketObj->SetStatus("open");
>     $self->TicketObj->Comment(Content => "You must set the Time
> Worked!");
> }
> return 1;
> *****************
> 
> I would like to display the transaction ('resolved' => 'open') and
> possibly the message when it goes back to display the changes made to
> the ticket.  Currently here's all it displays:
> 
> *****************
> Results 
> 
> *	Ticket 32: Status changed from 'open' to 'resolved'
> *****************
> 
> Does anyone have any idea how to do this?  Otherwise they will
> probably be confused when the result shows that the ticket was
> resolved but it's still open and they may not notice the comment at
> the end of the thread.
> 
> Thanks for any insight/help.
> 
> -Sean



This email (including any attachments) is for its intended-recipient's
use only. This email may contain information that is confidential or
privileged.  If you received this email in error, please immediately
advise the sender by replying to this email and then delete this message
from your system.  For further information and disclaimers that apply to
this email, see [http://www.viacom.com/email_disclaimer.jhtml].
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070611/b583e7e3/attachment.htm
From Sean.Edge at mtvnmix.com  Mon Jun 11 15:48:36 2007
From: Sean.Edge at mtvnmix.com (Edge, Sean)
Date: Mon Jun 11 15:49:20 2007
Subject: [Rt-devel] RE: Display scrip transactions in "Results:"
In-Reply-To: <9DB5A62CCC96094EBA95F593C589CE1638B757@LAZYTOWN.mtvn.ad.viacom.com>
References: <9DB5A62CCC96094EBA95F593C589CE1638B757@LAZYTOWN.mtvn.ad.viacom.com>
Message-ID: <9DB5A62CCC96094EBA95F593C589CE1638B758@LAZYTOWN.mtvn.ad.viacom.com>

Again, forgot to mention... RT 3.6.3.

> _____________________________________________ 
> From: 	Edge, Sean  
> Sent:	Monday, June 11, 2007 3:38 PM
> To:	'rt-devel@lists.bestpractical.com'
> Subject:	Display scrip transactions in "Results:"
> 
> I first sent this to RT-Users but realized this is probably the better
> place to send it....
> 
> I created a Scrip to make sure that "Time Worked" is not zero when a
> ticket is 'resolved'.  If there is no "Time Worked" defined then I set
> the ticket status back to 'open' and log a comment.  Here's the code:
> 
> *****************
> if ((! defined($self->TicketObj->TimeWorked) ||
>    ($self->TicketObj->TimeWorked == 0)) { 
>     my ($id, $msg) = $self->TicketObj->SetStatus("open");
>     $self->TicketObj->Comment(Content => "You must set the Time
> Worked!");
> }
> return 1;
> *****************
> 
> I would like to display the transaction ('resolved' => 'open') and
> possibly the message when it goes back to display the changes made to
> the ticket.  Currently here's all it displays:
> 
> *****************
> Results 
> 
> *	Ticket 32: Status changed from 'open' to 'resolved'
> *****************
> 
> Does anyone have any idea how to do this?  Otherwise they will
> probably be confused when the result shows that the ticket was
> resolved but it's still open and they may not notice the comment at
> the end of the thread.
> 
> Thanks for any insight/help.
> 
> -Sean



This email (including any attachments) is for its intended-recipient's
use only. This email may contain information that is confidential or
privileged.  If you received this email in error, please immediately
advise the sender by replying to this email and then delete this message
from your system.  For further information and disclaimers that apply to
this email, see [http://www.viacom.com/email_disclaimer.jhtml].
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070611/14c70ee2/attachment.htm
From JStark at sumtotalsystems.com  Tue Jun 12 15:07:22 2007
From: JStark at sumtotalsystems.com (Jeff Stark)
Date: Tue Jun 12 15:07:48 2007
Subject: [Rt-devel] RT and Attachments
Message-ID: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>

Hello, it seems that attachments are killing our system, even if they
are not being accessed.

Is that the attachments (standard and custom field) are returned on
every query of a ticket?  
 
Every ticket that has an attachment greater than (roughly) 6MB is
causing our system to hand and I must then clear the mason obj cache and
restart httpd.
 
Is this because the attachments are stored in the database and not
published to the file system in some fashion?

Has anyone possibly created some performance patches for attachments or
created a patch that copies the files to the file system instead?
 
Thanks,
 
 
- Stark

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070612/15dade39/attachment.htm
From ruz at bestpractical.com  Wed Jun 13 04:12:50 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Wed Jun 13 04:12:57 2007
Subject: [Rt-devel] RT and Attachments
In-Reply-To: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>
References: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>
Message-ID: <589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>

Jeff, as you're writing to devel list you should know how to report
things in the right way.
We need to know versions of perl, RT, apache, FCGI/MP and DB at least.
We also need to know each your step after server restart. So you have
to stop the server, turn on debug log, prune logs or at least know
where it ends at this point, then start the server, reproduce the
problem and send us the log and shiny report.

On 6/12/07, Jeff Stark <JStark@sumtotalsystems.com> wrote:
>
>
> Hello, it seems that attachments are killing our system, even if they are
> not being accessed.
>
> Is that the attachments (standard and custom field) are returned on every
> query of a ticket?
>
> Every ticket that has an attachment greater than (roughly) 6MB is causing
> our system to hand and I must then clear the mason obj cache and restart
> httpd.
>
> Is this because the attachments are stored in the database and not published
> to the file system in some fashion?
>
> Has anyone possibly created some performance patches for attachments or
> created a patch that copies the files to the file system instead?
>
> Thanks,
>
>
>
>
> - Stark
> _______________________________________________
> List info:
> http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>


-- 
Best regards, Ruslan.
From JStark at sumtotalsystems.com  Wed Jun 13 10:20:25 2007
From: JStark at sumtotalsystems.com (Jeff Stark)
Date: Wed Jun 13 10:23:19 2007
Subject: [Rt-devel] RT and Attachments
In-Reply-To: <589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>
References: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>
	<589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>
Message-ID: <1AC37FBEF7856646AC61A9F285BC14CC03A0AAC1@mtn-exch1.sumtotalsystems.com>

Sorry Ruslan.

RT: 3.6.3
Perl: v5.8.5 under linux RHEL 4
Apache: 2.0.52
FCGI: v0.67;
DB:  10g Enterprise Edition Release 10.2.0.1.0 

Steps are quite simple at this point....someone accesses a ticket with
an attachment, after churning for several minutes, the receive a 500
error, the entire system is pretty much hosed at this point.  I connect
to the box, stop httpd, clear mason obj cache, restart httpd and system
is fine, until someone accesses a ticket with a large attachment again.

Our Oracle DBA has also identified that the joins created are not
allowing Oracle to utilize indexes which is causing 13 to 15 seconds (on
average) for each page load.  After changing custom fields from
attachment to text box, the system response time increased slightly to 7
to 9 seconds (on average) for each page load.  Is there something I can
do to boost the performance of RT on Oracle so that it can utilize
indexes to speed up queries?  

Thanks,
 
Jeff Stark, Sr Systems Administrator      
SumTotal Systems, Inc. (Nasdaq; SUMT)                       


OFFICE    +1 919 326 7548
MOBILE  +1 919 622 0418

EMAIL     jstark@sumtotalsystems.com

 
The contents of this communication are considered SumTotal Confidential,
and should not be shared with anyone outside of SumTotal, either
electronically or verbally, without the express permission of the
author(s).

-----Original Message-----
From: ruslan.zakirov@gmail.com [mailto:ruslan.zakirov@gmail.com] On
Behalf Of Ruslan Zakirov
Sent: Wednesday, June 13, 2007 4:13 AM
To: Jeff Stark
Cc: rt-devel@lists.bestpractical.com
Subject: Re: [Rt-devel] RT and Attachments

Jeff, as you're writing to devel list you should know how to report
things in the right way.
We need to know versions of perl, RT, apache, FCGI/MP and DB at least.
We also need to know each your step after server restart. So you have to
stop the server, turn on debug log, prune logs or at least know where it
ends at this point, then start the server, reproduce the problem and
send us the log and shiny report.

On 6/12/07, Jeff Stark <JStark@sumtotalsystems.com> wrote:
>
>
> Hello, it seems that attachments are killing our system, even if they 
> are not being accessed.
>
> Is that the attachments (standard and custom field) are returned on 
> every query of a ticket?
>
> Every ticket that has an attachment greater than (roughly) 6MB is 
> causing our system to hand and I must then clear the mason obj cache 
> and restart httpd.
>
> Is this because the attachments are stored in the database and not 
> published to the file system in some fashion?
>
> Has anyone possibly created some performance patches for attachments 
> or created a patch that copies the files to the file system instead?
>
> Thanks,
>
>
>
>
> - Stark
> _______________________________________________
> List info:
> http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>


--
Best regards, Ruslan.

From JoopvandeWege at mococo.nl  Wed Jun 13 10:41:09 2007
From: JoopvandeWege at mococo.nl (Joop)
Date: Wed Jun 13 10:49:31 2007
Subject: [Rt-devel] RT and Attachments
In-Reply-To: <1AC37FBEF7856646AC61A9F285BC14CC03A0AAC1@mtn-exch1.sumtotalsystems.com>
References: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>	<589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>
	<1AC37FBEF7856646AC61A9F285BC14CC03A0AAC1@mtn-exch1.sumtotalsystems.com>
Message-ID: <46700205.8050200@mococo.nl>

Jeff Stark wrote:
> Sorry Ruslan.
> 
> RT: 3.6.3
> Perl: v5.8.5 under linux RHEL 4
> Apache: 2.0.52
> FCGI: v0.67;
> DB:  10g Enterprise Edition Release 10.2.0.1.0 
> 
> Steps are quite simple at this point....someone accesses a ticket with
> an attachment, after churning for several minutes, the receive a 500
> error, the entire system is pretty much hosed at this point.  I connect
> to the box, stop httpd, clear mason obj cache, restart httpd and system
> is fine, until someone accesses a ticket with a large attachment again.
The 500 error is due to a timeout on your FCGI process (probably) and I 
know that when we used FastCGI I set it to a higher value. I think the 
default is 180sec and we changed it to atleast 300 but maybe even 600 
because of searching in the content of email message took sometime.

> Our Oracle DBA has also identified that the joins created are not
> allowing Oracle to utilize indexes which is causing 13 to 15 seconds (on
> average) for each page load.  After changing custom fields from
> attachment to text box, the system response time increased slightly to 7
> to 9 seconds (on average) for each page load.  Is there something I can
> do to boost the performance of RT on Oracle so that it can utilize
> indexes to speed up queries?  
Please post query and explain plan.

Yes I have a patch specifically for searching content using OracleText 
but not seeing your problem queries I can't say if you'll benefit from 
those patches.

Joop

PS:
My config is:
VirtualMachine:Ubuntu-5.10 Svr, rt-3.6.3, apache2, mod_perl2, 
perl-5.8.7, Oracle XE
Performance is OK, frontpage loading time is: ~2-3sec with 7 portlets.
(30 newest, 30 I own, quick ticket creation, Calendar, Reminders, Quick 
Search (35 rows,3 columns), Don't refresh this page)
From ruz at bestpractical.com  Wed Jun 13 13:24:49 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Wed Jun 13 13:25:00 2007
Subject: [Rt-devel] RT and Attachments
In-Reply-To: <46700205.8050200@mococo.nl>
References: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>
	<589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>
	<1AC37FBEF7856646AC61A9F285BC14CC03A0AAC1@mtn-exch1.sumtotalsystems.com>
	<46700205.8050200@mococo.nl>
Message-ID: <589c94400706131024k1216ff21n3b7166b950a051bd@mail.gmail.com>

On 6/13/07, Joop <JoopvandeWege@mococo.nl> wrote:
> Jeff Stark wrote:
> > Sorry Ruslan.
> >
> > RT: 3.6.3
> > Perl: v5.8.5 under linux RHEL 4
> > Apache: 2.0.52
> > FCGI: v0.67;
> > DB:  10g Enterprise Edition Release 10.2.0.1.0
> >
> > Steps are quite simple at this point....someone accesses a ticket with
> > an attachment, after churning for several minutes, the receive a 500
> > error, the entire system is pretty much hosed at this point.  I connect
> > to the box, stop httpd, clear mason obj cache, restart httpd and system
> > is fine, until someone accesses a ticket with a large attachment again.
> The 500 error is due to a timeout on your FCGI process (probably) and I
> know that when we used FastCGI I set it to a higher value. I think the
> default is 180sec and we changed it to atleast 300 but maybe even 600
> because of searching in the content of email message took sometime.
As I understand Jeff access to a ticket, not searching for tickets,
but it's not absolutely clear from description. Also, Jeff, can you
block access to RT to all users and try to reproduce the problem in
"single user mode"? As really it could be not big attachments, but a
search by content of attachments which terribly slow down system so
other requests to RT timeout.

We fixed an issue with searches by content of attachments in 3.6.4rc2.
Can you try it? Things should be better, still not ideal, but better.

>
> > Our Oracle DBA has also identified that the joins created are not
> > allowing Oracle to utilize indexes which is causing 13 to 15 seconds (on
> > average) for each page load.  After changing custom fields from
> > attachment to text box, the system response time increased slightly to 7
> > to 9 seconds (on average) for each page load.  Is there something I can
> > do to boost the performance of RT on Oracle so that it can utilize
> > indexes to speed up queries?
> Please post query and explain plan.
>
> Yes I have a patch specifically for searching content using OracleText
> but not seeing your problem queries I can't say if you'll benefit from
> those patches.
>
> Joop
>
> PS:
> My config is:
> VirtualMachine:Ubuntu-5.10 Svr, rt-3.6.3, apache2, mod_perl2,
> perl-5.8.7, Oracle XE
> Performance is OK, frontpage loading time is: ~2-3sec with 7 portlets.
> (30 newest, 30 I own, quick ticket creation, Calendar, Reminders, Quick
> Search (35 rows,3 columns), Don't refresh this page)
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>


-- 
Best regards, Ruslan.
From JoopvandeWege at mococo.nl  Wed Jun 13 14:16:54 2007
From: JoopvandeWege at mococo.nl (Joop van de Wege)
Date: Wed Jun 13 14:26:06 2007
Subject: [Rt-devel] RT and Attachments
In-Reply-To: <589c94400706131024k1216ff21n3b7166b950a051bd@mail.gmail.com>
References: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>	
	<589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>	
	<1AC37FBEF7856646AC61A9F285BC14CC03A0AAC1@mtn-exch1.sumtotalsystems.com>	
	<46700205.8050200@mococo.nl>
	<589c94400706131024k1216ff21n3b7166b950a051bd@mail.gmail.com>
Message-ID: <46703496.2090105@mococo.nl>

Ruslan Zakirov wrote:

> 
> We fixed an issue with searches by content of attachments in 3.6.4rc2.
> Can you try it? Things should be better, still not ideal, but better.
Could you give us an idea to what is changed?

Joop

From ruz at bestpractical.com  Wed Jun 13 14:57:39 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Wed Jun 13 14:57:46 2007
Subject: [Rt-devel] RT and Attachments
In-Reply-To: <46703496.2090105@mococo.nl>
References: <1AC37FBEF7856646AC61A9F285BC14CC03985146@mtn-exch1.sumtotalsystems.com>
	<589c94400706130112m37e85a0ao6037a73806ba822b@mail.gmail.com>
	<1AC37FBEF7856646AC61A9F285BC14CC03A0AAC1@mtn-exch1.sumtotalsystems.com>
	<46700205.8050200@mococo.nl>
	<589c94400706131024k1216ff21n3b7166b950a051bd@mail.gmail.com>
	<46703496.2090105@mococo.nl>
Message-ID: <589c94400706131157s169c7e34wf218b586d2e5e620@mail.gmail.com>

We changed SQL query we build, but this only affects queries where
"Content like x" joined with another condition, for example "Content
LIKE X OR Subject LIKE X", but we still don't use full text indexes
(ideal solution).

On 6/13/07, Joop van de Wege <JoopvandeWege@mococo.nl> wrote:
> Ruslan Zakirov wrote:
>
> >
> > We fixed an issue with searches by content of attachments in 3.6.4rc2.
> > Can you try it? Things should be better, still not ideal, but better.
> Could you give us an idea to what is changed?
>
> Joop
>
>


-- 
Best regards, Ruslan.
From mattea at rescomp.berkeley.edu  Thu Jun 14 00:58:55 2007
From: mattea at rescomp.berkeley.edu (Matthew Ekstrand-Abueg)
Date: Thu Jun 14 01:29:22 2007
Subject: [Rt-devel] Filtering Spam
Message-ID: <20070614045855.GK23047@rescomp.berkeley.edu>

Hello,

I'm sure this question has been asked before, so please feel free to
direct me to a place where I can find an answer.

I have SpamAssassin running to flag all emails before they reach RT, and
would like to be able to filter the Spam messages (i.e. matching the
header "X-Spam-Flag: Yes") to a Queue called "Spam" from a Scrip within RT.

My problem is simply that I can not figure out how to extract that
portion of the header to place it within my scrip.  I am using RT
version 3.6.3 should this make any difference.

Thank you very much for your time.

-- 
Matthew Ekstrand-Abueg
Systems Administrator
Network Infrastructure, RSSP-IT
UC Berkeley

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070613/675ad981/attachment.pgp
From falcone at bestpractical.com  Thu Jun 14 11:01:32 2007
From: falcone at bestpractical.com (Kevin Falcone)
Date: Thu Jun 14 11:02:59 2007
Subject: [Rt-devel] Filtering Spam
In-Reply-To: <20070614045855.GK23047@rescomp.berkeley.edu>
References: <20070614045855.GK23047@rescomp.berkeley.edu>
Message-ID: <C07B2510-7DE4-4A1B-B3DF-EEA6B2998BE8@bestpractical.com>


On Jun 14, 2007, at 12:58 AM, Matthew Ekstrand-Abueg wrote:
> I have SpamAssassin running to flag all emails before they reach  
> RT, and
> would like to be able to filter the Spam messages (i.e. matching the
> header "X-Spam-Flag: Yes") to a Queue called "Spam" from a Scrip  
> within RT.
>
> My problem is simply that I can not figure out how to extract that
> portion of the header to place it within my scrip.  I am using RT
> version 3.6.3 should this make any difference.

Attachment objects have a GetHeader method so you can say
$attachment->GetHeader('X-Spam-Flag') and it will return the value.

You may want to take a look at the RT-Extension-ExtractCustomFieldValues
code on CPAN or on the wiki.  It has code you can steal.

-kevin
From i.norton at lancaster.ac.uk  Thu Jun 14 12:35:28 2007
From: i.norton at lancaster.ac.uk (Norton, Ian)
Date: Thu Jun 14 13:05:41 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>
References: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>
Message-ID: <BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local>

Hi all,

> 	$RT::Timezone should now actually work.
> 
> I'd like to hear a couple success reports and confirmation 
> that we've fixed the regression in rc1 before rolling the 
> final next week.

The timezone fix doesn't appear to be working for a test installation.

The system timezone is set to GMT, with the RT Timezone set to
"US/Eastern".  Recording comments in a ticket shows them logged at the
GMT time rather than the local timezone.  By this we know that the
database is correctly logging them at GMT but that the UI is not
translating them for display.

I *think* that the following sample code should illustrate the problem:

--SNIP--

#!/usr/bin/perl -w

use strict;
use lib "/usr/local/rt3.6.4rc2/lib";
use RT;

# Load config and init
RT::LoadConfig();
RT::Init();

require "RT/Date.pm";

# Create a date object and set it to now
my $date = RT::Date->new(RT::SystemUser);
$date->SetToNow();

# timezone as defined by RT
print("LocalTimezone - " . $date->LocalTimezone() . "\n");

# timezone environment variable
print("TZ Env var    - " . $ENV{TZ} . "\n");

# time AsString (Returns the object's time as a string with the current
timezone)
print("AsString      - " . $date->AsString() . "\n");

# localtime
print("Localtime     - " . localtime() . "\n");

--SNIP--

The results are this:

LocalTimezone - US/Eastern
TZ Env var    - US/Eastern
AsString      - Thu Jun 14 16:14:15 2007
Localtime     - Thu Jun 14 16:14:15 2007

Whilst I expect localtime to return the value it does, AsString returns
the same value.

Assuming this is a valid test which I accept it may not be, are there
any suggestions on how to fix this please?

We're using ubuntu 6.06 with perl 5.8.7 and My SQL 5.0.22.

Thanks, Ian.
From nick at ccl4.org  Thu Jun 14 13:01:02 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Thu Jun 14 13:05:46 2007
Subject: [Rt-devel] obscure stricture explosion
Message-ID: <20070614170102.GX18077@plum.flirble.org>

If I'm getting this error message:

Can't locate object method "EmailAddress" via package "No object mapping for field" (perhaps you forgot to load "No object mapping for field"?) at lib/RT/Ticket_Overlay.pm line 1351.

where the offending lines in 3.6.3's RT/Ticket_Overlay.pm are:

    if ( $self->CurrentUser->PrincipalId == ($args{'PrincipalId'} || 0)
       or    lc( $self->CurrentUser->UserObj->EmailAddress )
          eq lc( RT::User::CanonicalizeEmailAddress(undef, $args{'Email'}) || '' ) )
    {

and my calling code is

$t->AddWatcher(Type => 'Cc', PrincipalId => $g->Id);

where $g->Id is 55, and $t is a valid RT::Ticket object


then, where do I start looking to work out what my bug is?


My script (it's trying to be a regression test) started with:

use_ok('RT');
use_ok('RT::Ticket');
use_ok('RT::Queue');
RT::LoadConfig();
RT::Init();


And I'd not set CurrentUser. Is that my problem?

Nicholas Clark
From jesse at bestpractical.com  Thu Jun 14 13:12:32 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Thu Jun 14 13:12:47 2007
Subject: [Rt-devel] obscure stricture explosion
In-Reply-To: <20070614170102.GX18077@plum.flirble.org>
References: <20070614170102.GX18077@plum.flirble.org>
Message-ID: <AD27826B-2F02-4EF5-BB54-E16B71A7DB8D@bestpractical.com>


On Jun 14, 2007, at 1:01 PM, Nicholas Clark wrote:

> If I'm getting this error message:
>
> Can't locate object method "EmailAddress" via package "No object  
> mapping for field" (perhaps you forgot to load "No object mapping  
> for field"?) at lib/RT/Ticket_Overlay.pm line 1351.
>
> where the offending lines in 3.6.3's RT/Ticket_Overlay.pm are:
>
>     if ( $self->CurrentUser->PrincipalId == ($args{'PrincipalId'}  
> || 0)
>        or    lc( $self->CurrentUser->UserObj->EmailAddress )
>           eq lc( RT::User::CanonicalizeEmailAddress(undef, $args 
> {'Email'}) || '' ) )
>     {
>
> and my calling code is
>
> $t->AddWatcher(Type => 'Cc', PrincipalId => $g->Id);
>

Can I see the initialization of $t?  In RT, you should ~always be  
handing a currentuser to ->new().



> where $g->Id is 55, and $t is a valid RT::Ticket object
>
>
> then, where do I start looking to work out what my bug is?
>
>
> My script (it's trying to be a regression test) started with:
>
> use_ok('RT');
> use_ok('RT::Ticket');
> use_ok('RT::Queue');
> RT::LoadConfig();
> RT::Init();
>
>
> And I'd not set CurrentUser. Is that my problem?
>
> Nicholas Clark
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/ 
> rt-devel
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070614/83bba180/PGP.pgp
From jesse at bestpractical.com  Thu Jun 14 13:47:06 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Thu Jun 14 13:47:59 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local>
References: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>
	<BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local>
Message-ID: <7F856382-DB0C-498D-BC64-E5D50B7CBF59@bestpractical.com>

Can you give me (from the shell):

TZ=US/Eastern date
TZ=America/New_York date
perl -e'$ENV{TZ} = "US/Eastern";  print localtime()."\n";'
perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";  print  
localtime()."\n";'



On Jun 14, 2007, at 12:35 PM, Norton, Ian wrote:

> Hi all,
>
>> 	$RT::Timezone should now actually work.
>>
>> I'd like to hear a couple success reports and confirmation
>> that we've fixed the regression in rc1 before rolling the
>> final next week.
>
> The timezone fix doesn't appear to be working for a test installation.
>
> The system timezone is set to GMT, with the RT Timezone set to
> "US/Eastern".  Recording comments in a ticket shows them logged at the
> GMT time rather than the local timezone.  By this we know that the
> database is correctly logging them at GMT but that the UI is not
> translating them for display.
>
> I *think* that the following sample code should illustrate the  
> problem:
>
> --SNIP--
>
> #!/usr/bin/perl -w
>
> use strict;
> use lib "/usr/local/rt3.6.4rc2/lib";
> use RT;
>
> # Load config and init
> RT::LoadConfig();
> RT::Init();
>
> require "RT/Date.pm";
>
> # Create a date object and set it to now
> my $date = RT::Date->new(RT::SystemUser);
> $date->SetToNow();
>
> # timezone as defined by RT
> print("LocalTimezone - " . $date->LocalTimezone() . "\n");
>
> # timezone environment variable
> print("TZ Env var    - " . $ENV{TZ} . "\n");
>
> # time AsString (Returns the object's time as a string with the  
> current
> timezone)
> print("AsString      - " . $date->AsString() . "\n");
>
> # localtime
> print("Localtime     - " . localtime() . "\n");
>
> --SNIP--
>
> The results are this:
>
> LocalTimezone - US/Eastern
> TZ Env var    - US/Eastern
> AsString      - Thu Jun 14 16:14:15 2007
> Localtime     - Thu Jun 14 16:14:15 2007
>
> Whilst I expect localtime to return the value it does, AsString  
> returns
> the same value.
>
> Assuming this is a valid test which I accept it may not be, are there
> any suggestions on how to fix this please?
>
> We're using ubuntu 6.06 with perl 5.8.7 and My SQL 5.0.22.
>
> Thanks, Ian.
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/ 
> rt-devel
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070614/818a850e/PGP.pgp
From Sean.Edge at mtvnmix.com  Thu Jun 14 13:47:53 2007
From: Sean.Edge at mtvnmix.com (Edge, Sean)
Date: Thu Jun 14 13:51:07 2007
Subject: [Rt-devel] Bug: ObjectCustomFieldValues.Content not being updated
Message-ID: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>

RT Version: 3.6.3
Apache : 2.0.59
Perl: 5.8.5
MySQL: 4.0.20

We have a CustomField called Department.  We use it do reporting on
which departments had work done for them and how much time so that later
we can charge them for that time.  The problem is when the Names are
changed via the RT UI -- the ObjectCustomFieldValues.Content values are
not modified.  If the names were changed on the first of the month and
someone tries to pull a report from last month (or any previous month
for that matter) they will get only the Departments that haven't had
their names changed, obviously.

I have made the updates via the MySQL CLI using the same user/pass as RT
is configured to use, so it's not a database permissions issue.

I've looked through the Apache error logs and they are pretty bare...
infact there isn't even an entry from today.  If more information is
needed please let me know and I will provide it promptly.

Thank you.

Sean Edge
Consultant
MTV Networks



This email (including any attachments) is for its intended-recipient's
use only. This email may contain information that is confidential or
privileged.  If you received this email in error, please immediately
advise the sender by replying to this email and then delete this message
from your system.  For further information and disclaimers that apply to
this email, see [http://www.viacom.com/email_disclaimer.jhtml].
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070614/f7ffe65e/attachment.htm
From JStark at sumtotalsystems.com  Thu Jun 14 13:57:06 2007
From: JStark at sumtotalsystems.com (Jeff Stark)
Date: Thu Jun 14 14:09:09 2007
Subject: [Rt-devel] Bug: ObjectCustomFieldValues.Content not being updated
In-Reply-To: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>
References: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>
Message-ID: <1AC37FBEF7856646AC61A9F285BC14CC03A0B2FA@mtn-exch1.sumtotalsystems.com>

Yeah we had to put a process in place to manually update
"ObjectCustomFieldValues" with that in "CustomFieldValues".
 
 
- Stark


________________________________

From: rt-devel-bounces@lists.bestpractical.com
[mailto:rt-devel-bounces@lists.bestpractical.com] On Behalf Of Edge,
Sean
Sent: Thursday, June 14, 2007 1:48 PM
To: rt-devel@lists.bestpractical.com
Subject: [Rt-devel] Bug: ObjectCustomFieldValues.Content not being
updated



RT Version: 3.6.3 
Apache : 2.0.59 
Perl: 5.8.5 
MySQL: 4.0.20 

We have a CustomField called Department.  We use it do reporting on
which departments had work done for them and how much time so that later
we can charge them for that time.  The problem is when the Names are
changed via the RT UI -- the ObjectCustomFieldValues.Content values are
not modified.  If the names were changed on the first of the month and
someone tries to pull a report from last month (or any previous month
for that matter) they will get only the Departments that haven't had
their names changed, obviously.

I have made the updates via the MySQL CLI using the same user/pass as RT
is configured to use, so it's not a database permissions issue.

I've looked through the Apache error logs and they are pretty bare...
infact there isn't even an entry from today.  If more information is
needed please let me know and I will provide it promptly.

Thank you. 

Sean Edge 
Consultant 
MTV Networks 



This email (including any attachments) is for its intended-recipient's
use only. This email may contain information that is confidential or
privileged.  If you received this email in error, please immediately
advise the sender by replying to this email and then delete this message
from your system.  For further information and disclaimers that apply to
this email, see [http://www.viacom.com/email_disclaimer.jhtml].
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070614/4cf11a81/attachment.htm
From jesse at bestpractical.com  Thu Jun 14 14:14:53 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Thu Jun 14 14:15:06 2007
Subject: [Rt-devel] Bug: ObjectCustomFieldValues.Content not being updated
In-Reply-To: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>
References: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>
Message-ID: <241BD248-4D4A-4B65-B0E1-035A3A73F7B6@bestpractical.com>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070614/4984cf8e/PGP-0001.pgp
From joby at u.washington.edu  Thu Jun 14 14:15:39 2007
From: joby at u.washington.edu (Joby Walker)
Date: Thu Jun 14 14:20:06 2007
Subject: [Rt-devel] Bug: ObjectCustomFieldValues.Content not being updated
In-Reply-To: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>
References: <9DB5A62CCC96094EBA95F593C589CE1638B765@LAZYTOWN.mtvn.ad.viacom.com>
Message-ID: <467185CB.90406@u.washington.edu>

This is design feature.  There is no code keep ObjectCustomFieldValues
in sync with the current CustomFieldValues.  This allows you to retire
old CustomFieldValues without impacting Tickets.  This is necessary for us.

I do see the value of on modifying a specific CFV changing the
corresponding OCFVs though.

Joby Walker
C&C SSG, University of Washington


Edge, Sean wrote:
> 
> 
> RT Version: 3.6.3
> Apache : 2.0.59
> Perl: 5.8.5
> MySQL: 4.0.20
> 
> We have a CustomField called Department.  We use it do reporting on
> which departments had work done for them and how much time so that later
> we can charge them for that time.  The problem is when the Names are
> changed via the RT UI -- the ObjectCustomFieldValues.Content values are
> not modified.  If the names were changed on the first of the month and
> someone tries to pull a report from last month (or any previous month
> for that matter) they will get only the Departments that haven't had
> their names changed, obviously.
> 
> I have made the updates via the MySQL CLI using the same user/pass as RT
> is configured to use, so it's not a database permissions issue.
> 
> I've looked through the Apache error logs and they are pretty bare?
> infact there isn't even an entry from today.  If more information is
> needed please let me know and I will provide it promptly.
> 
> Thank you.
> 
> Sean Edge
> Consultant
> MTV Networks
> 
> 
> This email (including any attachments) is for its intended-recipient's
> use only. This email may contain information that is confidential or
> privileged.  If you received this email in error, please immediately
> advise the sender by replying to this email and then delete this message
> from your system.  For further information and disclaimers that apply to
> this email, see [http://www.viacom.com/email_disclaimer.jhtml].
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
From nick at ccl4.org  Thu Jun 14 17:30:03 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Thu Jun 14 17:30:13 2007
Subject: [Rt-devel] obscure stricture explosion
In-Reply-To: <AD27826B-2F02-4EF5-BB54-E16B71A7DB8D@bestpractical.com>
References: <20070614170102.GX18077@plum.flirble.org>
	<AD27826B-2F02-4EF5-BB54-E16B71A7DB8D@bestpractical.com>
Message-ID: <20070614213003.GY18077@plum.flirble.org>

On Thu, Jun 14, 2007 at 01:12:32PM -0400, Jesse Vincent wrote:
> 
> On Jun 14, 2007, at 1:01 PM, Nicholas Clark wrote:
> 
> >If I'm getting this error message:
> >
> >Can't locate object method "EmailAddress" via package "No object  
> >mapping for field" (perhaps you forgot to load "No object mapping  
> >for field"?) at lib/RT/Ticket_Overlay.pm line 1351.
> >
> >where the offending lines in 3.6.3's RT/Ticket_Overlay.pm are:
> >
> >    if ( $self->CurrentUser->PrincipalId == ($args{'PrincipalId'}  
> >|| 0)
> >       or    lc( $self->CurrentUser->UserObj->EmailAddress )
> >          eq lc( RT::User::CanonicalizeEmailAddress(undef, $args 
> >{'Email'}) || '' ) )
> >    {
> >
> >and my calling code is
> >
> >$t->AddWatcher(Type => 'Cc', PrincipalId => $g->Id);
> >
> 
> Can I see the initialization of $t?  In RT, you should ~always be  
> handing a currentuser to ->new().

Code is at work, so I can't get to it from here, but (from memory) it was

$t = RT::Ticket->new($co_user);
# where $co_user was a user retrieved by name

$t->Create(...)

where a ticket is successfully created. (because it shows up)

I can give you the verbatim code by my tomorrow (by which time you'll be
asleep)

Nicholas Clark
From kihong.lee at hypermediasystems.com  Thu Jun 14 18:01:15 2007
From: kihong.lee at hypermediasystems.com (Kihong Lee)
Date: Thu Jun 14 18:01:56 2007
Subject: [Rt-devel] Hi, have a question about outgoing email encoding.
Message-ID: <4671BAAB.3080707@hypermediasystems.com>

Hello.

About a month ago, I have upgraded a RT system from 3.0.9 to 3.6.3. But 
after that some people started to complain about broken Japaneses. So 
I've investigated this issue and what I've found so far is,

If we reply the ticket which is created by email which has 
"Content-Transfer-Encoding: 7bit" header, it works fine. But if we reply 
the ticket which is create by email which doesn't have 
"Content-Transfer-Encoding: 7bit", the requester complains that they get 
broken Japanese.

For example,
This is the part of header of requestor's email who gets broken Japanese.
content-type: text/plain; charset="utf-8"; format="flowed"
X-RT-Original-Encoding: iso-2022-jp

And this is the part of header of requestor's email who gets clear Japanese.
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
X-RT-Original-Encoding: iso-2022-jp

Whatever the header is, incoming emails to RT are fine. But, again, if 
the email of requestor to RT system has " Content-Transfer-Encoding: 
7bit" header in it, they don't get broken Japanses when we reply it. And 
if the email of requetor to RT system DOESN't have 
"Content-Transfer-Encoding: 7bit" header, they GET BROKEN Japanses.

It looks like 3.6.3 use 8 bit encoding. So can I change the outgoing 
encoding options to make it not use 8 bit encoding? I'd like to try with 
7 bit encoding. Which file should I modify to try 7 bit encoding for 
outgoing email?

Or any other comment, advice, please?

Thanks,
Kihong Lee
From mouse at Rodents.Montreal.QC.CA  Thu Jun 14 22:31:16 2007
From: mouse at Rodents.Montreal.QC.CA (der Mouse)
Date: Thu Jun 14 22:57:57 2007
Subject: [Rt-devel] Hi, have a question about outgoing email encoding.
In-Reply-To: <4671BAAB.3080707@hypermediasystems.com>
References: <4671BAAB.3080707@hypermediasystems.com>
Message-ID: <200706150243.WAA05747@Sparkle.Rodents.Montreal.QC.CA>

> For example,
> This is the part of header of requestor's email who gets broken Japanese.
> content-type: text/plain; charset="utf-8"; format="flowed"
> X-RT-Original-Encoding: iso-2022-jp

> And this is the part of header of requestor's email who gets clear Japanese.
> Content-Type: text/plain; charset="utf-8"
> Content-Transfer-Encoding: 7bit
> X-RT-Original-Encoding: iso-2022-jp

> Whatever the header is, incoming emails to RT are fine.

Perhaps, but nto as quoted above, not if they contain Japanese.  (Well,
unless it's written in roumaji, a quibble which does not appear
relevant here.)  Those two header sets are semantically the same (a
missing Content-Transfer-Encoding: is defined to be equivalent to 7bit;
see RFC 2045 section 6.1), and neither is acceptable if the message
contains any non-ASCII characters, since UTF-8 generates non-7bit octet
sequences for them.  It's possible the mail on the wire was OK, if the
X-RT-Original-Encoding: is actually the Content-Transfer-Encoding: from
the incoming message.  I don't know iso-2022-jp; provided it sticks to
the 7bit restrictions (2045 2.7) it's fine - eg, if that's the encoding
that uses X3.64ish escape sequences.  But recoding it to UTF-8 without
changing the Content-Transfer-Encoding: to match is..broken.

That said, why RT treats them differently when they are defined to be
semantically identical is an interesting question, and arguably a bug
in RT.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
From pape-rt at inf.fu-berlin.de  Fri Jun 15 03:58:44 2007
From: pape-rt at inf.fu-berlin.de (Dirk Pape)
Date: Fri Jun 15 03:59:14 2007
Subject: [Rt-devel] patch: Optimize recipients for Notifications Mails
Message-ID: <0ADF91A1AB8F1D246B25F71B@idefix.mi.fu-berlin.de>

Hello,

we found, that if we have ScripActions that notify more than one watcher 
group (e.g. "Notify AdminCC and Cc as Comment"), people get more than one 
message for the same notification, if they have more than one role for the 
ticket (e.g. beeing AdminCc and Cc).

This is not necessary at all but only annoying, so I propose following 
general patch for Action/SendEmail.pm which removes unnecessary recipient 
entries: It removes every Cc already in To and every Bcc if already in To 
or Cc.

The patch is tested here and behaves like expected. Will this make it into 
the core?

<http://page.mi.fu-berlin.de/~pape/rt3/patches/rt/optimize_recipients.patch>

Regards, Dirk.
-- 
Dr. Dirk Pape (eAS - Projektleitung Campus Management)
Freie Universitaet Berlin
Grunewaldstr. 34a, 12165 Berlin
Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
From ruz at bestpractical.com  Fri Jun 15 04:20:28 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Fri Jun 15 04:20:34 2007
Subject: [Rt-devel] patch: Optimize recipients for Notifications Mails
In-Reply-To: <0ADF91A1AB8F1D246B25F71B@idefix.mi.fu-berlin.de>
References: <0ADF91A1AB8F1D246B25F71B@idefix.mi.fu-berlin.de>
Message-ID: <589c94400706150120i6e7f7fd4i23ad46a259f31d9a@mail.gmail.com>

good patch, vote for applying

On 6/15/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
> Hello,
>
> we found, that if we have ScripActions that notify more than one watcher
> group (e.g. "Notify AdminCC and Cc as Comment"), people get more than one
> message for the same notification, if they have more than one role for the
> ticket (e.g. beeing AdminCc and Cc).
>
> This is not necessary at all but only annoying, so I propose following
> general patch for Action/SendEmail.pm which removes unnecessary recipient
> entries: It removes every Cc already in To and every Bcc if already in To
> or Cc.
>
> The patch is tested here and behaves like expected. Will this make it into
> the core?
>
> <http://page.mi.fu-berlin.de/~pape/rt3/patches/rt/optimize_recipients.patch>
>
> Regards, Dirk.
> --
> Dr. Dirk Pape (eAS - Projektleitung Campus Management)
> Freie Universitaet Berlin
> Grunewaldstr. 34a, 12165 Berlin
> Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>


-- 
Best regards, Ruslan.
From nick at ccl4.org  Fri Jun 15 05:31:10 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Fri Jun 15 05:31:24 2007
Subject: [Rt-devel] obscure stricture explosion
In-Reply-To: <AD27826B-2F02-4EF5-BB54-E16B71A7DB8D@bestpractical.com>
References: <20070614170102.GX18077@plum.flirble.org>
	<AD27826B-2F02-4EF5-BB54-E16B71A7DB8D@bestpractical.com>
Message-ID: <20070615093110.GZ18077@plum.flirble.org>

On Thu, Jun 14, 2007 at 01:12:32PM -0400, Jesse Vincent wrote:
> 
> On Jun 14, 2007, at 1:01 PM, Nicholas Clark wrote:
> 
> >If I'm getting this error message:
> >
> >Can't locate object method "EmailAddress" via package "No object  
> >mapping for field" (perhaps you forgot to load "No object mapping  
> >for field"?) at lib/RT/Ticket_Overlay.pm line 1351.
> >
> >where the offending lines in 3.6.3's RT/Ticket_Overlay.pm are:
> >
> >    if ( $self->CurrentUser->PrincipalId == ($args{'PrincipalId'}  
> >|| 0)
> >       or    lc( $self->CurrentUser->UserObj->EmailAddress )
> >          eq lc( RT::User::CanonicalizeEmailAddress(undef, $args 
> >{'Email'}) || '' ) )
> >    {
> >
> >and my calling code is
> >
> >$t->AddWatcher(Type => 'Cc', PrincipalId => $g->Id);
> >
> 
> Can I see the initialization of $t?  In RT, you should ~always be  
> handing a currentuser to ->new().

The whole thing is appended.

Nicholas Clark

#!perl -w
# $Id$

use strict;

use Test::More 'no_plan';

use_ok('RT');
use_ok('RT::Ticket');
use_ok('RT::Queue');
RT::LoadConfig();
RT::Init();

use vars qw ($CO $CO_USER $V_USER $QUEUE_E $QUEUE_I);

$CO_USER='B1';
$V_USER='root';
$QUEUE_E='Systems Requests';
$QUEUE_I='Systems';

ok(my $co_user = RT::User->new($RT::SystemUser));
ok($co_user->Load($CO_USER), "Loaded user '$CO_USER'");

my $cf = $co_user->FirstCustomFieldValue('company group name');
ok(defined $cf);
my $g;

if (defined $cf) {
    $g = RT::Group->new($co_user);
    $g->LoadUserDefinedGroup($cf);
    ok($g->Id);
}

my $queue_e = RT::Queue->new($co_user);
$queue_e->Load($QUEUE_E);
ok ($queue_e, "Loaded queue '$QUEUE_E'");


ok(my $t = RT::Ticket->new($co_user));

my ($id,$t_o,$msg) = $t->Create(Queue => $queue_e,
 Subject => "External request (test $$)",
 Owner => $co_user);
ok($id, "Created ticket");
diag($msg);
diag($t);
diag($g->Id);
$t->AddWatcher(Type => 'Cc', PrincipalId => $g->Id);
From CLoos at netcologne.de  Fri Jun 15 05:47:42 2007
From: CLoos at netcologne.de (Loos, Christian)
Date: Fri Jun 15 06:39:53 2007
Subject: [Rt-devel] Re: Bug: ObjectCustomFieldValues.Content not being
	updated
Message-ID: <657C603258FA594DB9A8C49EB86631190BD651A2@ncmail.netcologne.intern>

Hi,

i think from user point of view this is a bug.
For a user it is very confusing if you have value a, b, c, d in custom field dropdown list and ticket custom field value is z.
As suggest bei jesse you could make an bulk update. But if you have script actions on custom field value change you get problems with this.

So custom field values should be stored by reference.

As mentioned by Joby Walker maybe the function of the object custom field value checkbox (check box to delete) should be changed from delete to disable, so disabled values are no longer visible in the dropdown list.

Chris

From i.norton at lancaster.ac.uk  Fri Jun 15 07:08:53 2007
From: i.norton at lancaster.ac.uk (Norton, Ian)
Date: Fri Jun 15 07:10:50 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <7F856382-DB0C-498D-BC64-E5D50B7CBF59@bestpractical.com>
References: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>
	<BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local>
	<7F856382-DB0C-498D-BC64-E5D50B7CBF59@bestpractical.com>
Message-ID: <BD039FFA11612F4AAF68A9C0C568752F014753BE@exchange-be5.lancs.local>

Hi Jesse,
 
> Can you give me (from the shell):
> 
> TZ=US/Eastern date
> TZ=America/New_York date
> perl -e'$ENV{TZ} = "US/Eastern";  print localtime()."\n";'
> perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";  
> print localtime()."\n";'


iann@elk:/usr/local/rt3.6.4rc2/lib/RT$ TZ=US/Eastern date
Fri Jun 15 07:07:34 EDT 2007
iann@elk:/usr/local/rt3.6.4rc2/lib/RT$ TZ=America/New_York date
Fri Jun 15 07:07:34 EDT 2007
iann@elk:/usr/local/rt3.6.4rc2/lib/RT$ perl -e'$ENV{TZ} = "US/Eastern";
print localtime()."\n";'
Fri Jun 15 07:07:36 2007
iann@elk:/usr/local/rt3.6.4rc2/lib/RT$ perl -e' print localtime()."\n";
$ENV{TZ} = "US/Eastern";  print localtime()."\n";'
Fri Jun 15 11:07:36 2007
Fri Jun 15 11:07:36 2007

Regards, Ian.
From jesse at bestpractical.com  Fri Jun 15 08:46:53 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Fri Jun 15 08:47:17 2007
Subject: [Rt-devel] patch: Optimize recipients for Notifications Mails
In-Reply-To: <589c94400706150120i6e7f7fd4i23ad46a259f31d9a@mail.gmail.com>
References: <0ADF91A1AB8F1D246B25F71B@idefix.mi.fu-berlin.de>
	<589c94400706150120i6e7f7fd4i23ad46a259f31d9a@mail.gmail.com>
Message-ID: <967A8F84-D717-4D63-B966-36C8AC32E164@bestpractical.com>


On Jun 15, 2007, at 4:20 AM, Ruslan Zakirov wrote:

> good patch, vote for applying

Ok. Let's aim for 3.6.5?



>
> On 6/15/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
>> Hello,
>>
>> we found, that if we have ScripActions that notify more than one  
>> watcher
>> group (e.g. "Notify AdminCC and Cc as Comment"), people get more  
>> than one
>> message for the same notification, if they have more than one role  
>> for the
>> ticket (e.g. beeing AdminCc and Cc).
>>
>> This is not necessary at all but only annoying, so I propose  
>> following
>> general patch for Action/SendEmail.pm which removes unnecessary  
>> recipient
>> entries: It removes every Cc already in To and every Bcc if  
>> already in To
>> or Cc.
>>
>> The patch is tested here and behaves like expected. Will this make  
>> it into
>> the core?
>>
>> <http://page.mi.fu-berlin.de/~pape/rt3/patches/rt/ 
>> optimize_recipients.patch>
>>
>> Regards, Dirk.
>> --
>> Dr. Dirk Pape (eAS - Projektleitung Campus Management)
>> Freie Universitaet Berlin
>> Grunewaldstr. 34a, 12165 Berlin
>> Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
>> _______________________________________________
>> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/ 
>> rt-devel
>>
>
>
> -- 
> Best regards, Ruslan.
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/ 
> rt-devel
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070615/2df933fc/PGP.pgp
From jesse at bestpractical.com  Fri Jun 15 08:47:26 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Fri Jun 15 08:47:50 2007
Subject: [Rt-devel] RT 3.6.4?
Message-ID: <586D2E1B-AF77-45E1-9BC0-5C5AC4B3A484@bestpractical.com>

I've heard no reports of _regressions_ in RT 3.6.4rc2. Anyone got  
anything to say before we release?

-j
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070615/5fc1c4e6/PGP.pgp
From ruz at bestpractical.com  Fri Jun 15 08:52:26 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Fri Jun 15 08:52:35 2007
Subject: [Rt-devel] patch: Optimize recipients for Notifications Mails
In-Reply-To: <967A8F84-D717-4D63-B966-36C8AC32E164@bestpractical.com>
References: <0ADF91A1AB8F1D246B25F71B@idefix.mi.fu-berlin.de>
	<589c94400706150120i6e7f7fd4i23ad46a259f31d9a@mail.gmail.com>
	<967A8F84-D717-4D63-B966-36C8AC32E164@bestpractical.com>
Message-ID: <589c94400706150552s4f34fb8bvd011de37aee309e5@mail.gmail.com>

ok.

On 6/15/07, Jesse Vincent <jesse@bestpractical.com> wrote:
>
> On Jun 15, 2007, at 4:20 AM, Ruslan Zakirov wrote:
>
> > good patch, vote for applying
>
> Ok. Let's aim for 3.6.5?
>
>
>
> >
> > On 6/15/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
> >> Hello,
> >>
> >> we found, that if we have ScripActions that notify more than one
> >> watcher
> >> group (e.g. "Notify AdminCC and Cc as Comment"), people get more
> >> than one
> >> message for the same notification, if they have more than one role
> >> for the
> >> ticket (e.g. beeing AdminCc and Cc).
> >>
> >> This is not necessary at all but only annoying, so I propose
> >> following
> >> general patch for Action/SendEmail.pm which removes unnecessary
> >> recipient
> >> entries: It removes every Cc already in To and every Bcc if
> >> already in To
> >> or Cc.
> >>
> >> The patch is tested here and behaves like expected. Will this make
> >> it into
> >> the core?
> >>
> >> <http://page.mi.fu-berlin.de/~pape/rt3/patches/rt/
> >> optimize_recipients.patch>
> >>
> >> Regards, Dirk.
> >> --
> >> Dr. Dirk Pape (eAS - Projektleitung Campus Management)
> >> Freie Universitaet Berlin
> >> Grunewaldstr. 34a, 12165 Berlin
> >> Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
> >> _______________________________________________
> >> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/
> >> rt-devel
> >>
> >
> >
> > --
> > Best regards, Ruslan.
> > _______________________________________________
> > List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/
> > rt-devel
> >
>
>
>


-- 
Best regards, Ruslan.
From jesse at bestpractical.com  Fri Jun 15 08:52:34 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Fri Jun 15 08:52:54 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <BD039FFA11612F4AAF68A9C0C568752F014753BE@exchange-be5.lancs.local>
References: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>
	<BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local>
	<7F856382-DB0C-498D-BC64-E5D50B7CBF59@bestpractical.com>
	<BD039FFA11612F4AAF68A9C0C568752F014753BE@exchange-be5.lancs.local>
Message-ID: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>


On Jun 15, 2007, at 7:08 AM, Norton, Ian wrote:

> perl -e' print localtime()."\n";
> $ENV{TZ} = "US/Eastern";  print localtime()."\n";'


Fascinating.

TZ=GMT perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";   
print localtime()."\n";'

Mac OS X:

Fri Jun 15 12:50:55 2007
Fri Jun 15 08:50:55 2007


Linux:

Fri Jun 15 12:51:51 2007
Fri Jun 15 12:51:51 2007

Both 5.8.8.

I wonder what bit of cleverness I'm missing.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070615/b4f09d97/PGP.pgp
From ruz at bestpractical.com  Fri Jun 15 08:53:21 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Fri Jun 15 08:53:47 2007
Subject: [Rt-devel] RT 3.6.4?
In-Reply-To: <586D2E1B-AF77-45E1-9BC0-5C5AC4B3A484@bestpractical.com>
References: <586D2E1B-AF77-45E1-9BC0-5C5AC4B3A484@bestpractical.com>
Message-ID: <589c94400706150553t7fc64c76x64cd732e8a67728d@mail.gmail.com>

I have nothing.

On 6/15/07, Jesse Vincent <jesse@bestpractical.com> wrote:
> I've heard no reports of _regressions_ in RT 3.6.4rc2. Anyone got
> anything to say before we release?
>
> -j
>
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>
>


-- 
Best regards, Ruslan.
From torsten.brumm at Kuehne-Nagel.com  Fri Jun 15 09:11:26 2007
From: torsten.brumm at Kuehne-Nagel.com (Ham MI-ID, Torsten Brumm)
Date: Fri Jun 15 09:11:41 2007
Subject: AW: [Rt-devel] RT 3.6.4?
Message-ID: <16426EA38D57E74CB1DE5A6AE1DB039448556F@w3hamboex11.ger.win.int.kn>

Yes, I tried it and it running fine here since one week.

Torsten

-----Original Message-----
From: rt-devel-bounces@lists.bestpractical.com <rt-devel-bounces@lists.bestpractical.com>
To: rt-devel@lists.bestpractical.com <rt-devel@lists.bestpractical.com>
Sent: Fri Jun 15 14:47:26 2007
Subject: [Rt-devel] RT 3.6.4?

I've heard no reports of _regressions_ in RT 3.6.4rc2. Anyone got  
anything to say before we release?

-j
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070615/8f540b43/attachment.htm
From kfh at mqsoftware.com  Fri Jun 15 08:59:56 2007
From: kfh at mqsoftware.com (Kelly F. Hickel)
Date: Fri Jun 15 09:12:28 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
References: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com><BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local><7F856382-DB0C-498D-BC64-E5D50B7CBF59@bestpractical.com><BD039FFA11612F4AAF68A9C0C568752F014753BE@exchange-be5.lancs.local>
	<17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
Message-ID: <63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>

Hmm, the obvious combinations work for me on centos 5.5.....

[root@starbug2 ~]# TZ=GMT perl -e' print localtime()."\n";'
Fri Jun 15 12:58:10 2007
[root@starbug2 ~]# TZ="US/Eastern" perl -e' print localtime()."\n";'
Fri Jun 15 08:58:21 2007
[root@starbug2 ~]# perl  -e' $ENV{TZ} = "GMT"; print localtime()."\n";'
Fri Jun 15 12:59:08 2007
[root@starbug2 ~]# perl  -e' $ENV{TZ} = "US/Eastern"; print
localtime()."\n";'
Fri Jun 15 08:59:15 2007
[
-- 
 
Kelly F. Hickel
Senior Software Architect
MQSoftware, Inc
952.345.8677
kfh@mqsoftware.com


> -----Original Message-----
> From: rt-devel-bounces@lists.bestpractical.com [mailto:rt-devel-
> bounces@lists.bestpractical.com] On Behalf Of Jesse Vincent
> Sent: Friday, June 15, 2007 7:53 AM
> To: Norton, Ian
> Cc: rt-devel@lists.bestpractical.com
> Subject: Re: [Rt-devel] RT 3.6.4rc2
> 
> 
> On Jun 15, 2007, at 7:08 AM, Norton, Ian wrote:
> 
> > perl -e' print localtime()."\n";
> > $ENV{TZ} = "US/Eastern";  print localtime()."\n";'
> 
> 
> Fascinating.
> 
> TZ=GMT perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";
> print localtime()."\n";'
> 
> Mac OS X:
> 
> Fri Jun 15 12:50:55 2007
> Fri Jun 15 08:50:55 2007
> 
> 
> Linux:
> 
> Fri Jun 15 12:51:51 2007
> Fri Jun 15 12:51:51 2007
> 
> Both 5.8.8.
> 
> I wonder what bit of cleverness I'm missing.
From ruz at bestpractical.com  Fri Jun 15 09:30:24 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Fri Jun 15 09:30:40 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
References: <0CC5D695-4278-497A-8D3E-4D767652E191@bestpractical.com>
	<BD039FFA11612F4AAF68A9C0C568752F014752FA@exchange-be5.lancs.local>
	<7F856382-DB0C-498D-BC64-E5D50B7CBF59@bestpractical.com>
	<BD039FFA11612F4AAF68A9C0C568752F014753BE@exchange-be5.lancs.local>
	<17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
Message-ID: <589c94400706150630h62b42110pfdcc97948160b4e8@mail.gmail.com>

works on my linux box:
TZ=GMT perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern"; print
localtime()."\n";'
Fri Jun 15 13:21:20 2007
Fri Jun 15 09:21:20 2007

I think this all depends on distribution and how good zone info package is.

On 6/15/07, Jesse Vincent <jesse@bestpractical.com> wrote:
>
> On Jun 15, 2007, at 7:08 AM, Norton, Ian wrote:
>
> > perl -e' print localtime()."\n";
> > $ENV{TZ} = "US/Eastern";  print localtime()."\n";'
>
>
> Fascinating.
>
> TZ=GMT perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";
> print localtime()."\n";'
>
> Mac OS X:
>
> Fri Jun 15 12:50:55 2007
> Fri Jun 15 08:50:55 2007
>
>
> Linux:
>
> Fri Jun 15 12:51:51 2007
> Fri Jun 15 12:51:51 2007
>
> Both 5.8.8.
>
> I wonder what bit of cleverness I'm missing.
>
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>
>


-- 
Best regards, Ruslan.
From nick at ccl4.org  Fri Jun 15 09:38:30 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Fri Jun 15 09:38:40 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>
References: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
	<63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>
Message-ID: <20070615133830.GE18077@plum.flirble.org>

> > -----Original Message-----
> > From: rt-devel-bounces@lists.bestpractical.com [mailto:rt-devel-
> > bounces@lists.bestpractical.com] On Behalf Of Jesse Vincent
> > Sent: Friday, June 15, 2007 7:53 AM

> > Fascinating.
> > 
> > TZ=GMT perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";
> > print localtime()."\n";'
> > 
> > Mac OS X:
> > 
> > Fri Jun 15 12:50:55 2007
> > Fri Jun 15 08:50:55 2007
> > 
> > 
> > Linux:
> > 
> > Fri Jun 15 12:51:51 2007
> > Fri Jun 15 12:51:51 2007
> > 
> > Both 5.8.8.
> > 
> > I wonder what bit of cleverness I'm missing.

On Fri, Jun 15, 2007 at 07:59:56AM -0500, Kelly F. Hickel wrote:
> Hmm, the obvious combinations work for me on centos 5.5.....
> 
> [root@starbug2 ~]# TZ=GMT perl -e' print localtime()."\n";'
> Fri Jun 15 12:58:10 2007
> [root@starbug2 ~]# TZ="US/Eastern" perl -e' print localtime()."\n";'
> Fri Jun 15 08:58:21 2007
> [root@starbug2 ~]# perl  -e' $ENV{TZ} = "GMT"; print localtime()."\n";'
> Fri Jun 15 12:59:08 2007
> [root@starbug2 ~]# perl  -e' $ENV{TZ} = "US/Eastern"; print
> localtime()."\n";'
> Fri Jun 15 08:59:15 2007

There appear to be two things interacting here (at least in Jesse's example)

On a FreeBSD machine that's in GMT I see:

$ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = "US/Eastern"; print localtime()."\n";'
Fri Jun 15 13:31:51 2007
Fri Jun 15 13:31:51 2007
$ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
Fri Jun 15 13:31:54 2007
Fri Jun 15 13:31:54 2007


so TZ is getting ignored.

Whereas an Ubuntu machine recognises the TZ strings:

$ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
Fri Jun 15 09:32:56 2007
Fri Jun 15 09:32:56 2007
$ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = "US/Eastern"; print localtime()."\n";'
Fri Jun 15 07:33:00 2007
Fri Jun 15 07:33:00 2007

then both print out that timezone at initialisation time, and the subsequent
change is ignored.

So OS X must be doing something "special" and resetting the timezone for every
call to localtime. Note that GMT isn't a great one to test with, as the
default for timezones you don't understand is 0, which is GMT.

$ TZ=Europe/London perl -we'print gmtime() . "\n" . localtime()."\n";'
Fri Jun 15 13:36:45 2007
Fri Jun 15 14:36:45 2007
$ TZ=US/Dubya perl -we'print gmtime() . "\n" . localtime()."\n";'
Fri Jun 15 13:37:05 2007
Fri Jun 15 13:37:05 2007


And no, I don't know more than this.

Nicholas Clark
From ruz at bestpractical.com  Fri Jun 15 09:54:24 2007
From: ruz at bestpractical.com (Ruslan Zakirov)
Date: Fri Jun 15 09:54:32 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <20070615133830.GE18077@plum.flirble.org>
References: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
	<63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>
	<20070615133830.GE18077@plum.flirble.org>
Message-ID: <589c94400706150654j42738a85gc55d9f5ea85f78d@mail.gmail.com>

On 6/15/07, Nicholas Clark <nick@ccl4.org> wrote:
> > > -----Original Message-----
> > > From: rt-devel-bounces@lists.bestpractical.com [mailto:rt-devel-
> > > bounces@lists.bestpractical.com] On Behalf Of Jesse Vincent
> > > Sent: Friday, June 15, 2007 7:53 AM
>
> > > Fascinating.
> > >
> > > TZ=GMT perl -e' print localtime()."\n"; $ENV{TZ} = "US/Eastern";
> > > print localtime()."\n";'
> > >
> > > Mac OS X:
> > >
> > > Fri Jun 15 12:50:55 2007
> > > Fri Jun 15 08:50:55 2007
> > >
> > >
> > > Linux:
> > >
> > > Fri Jun 15 12:51:51 2007
> > > Fri Jun 15 12:51:51 2007
> > >
> > > Both 5.8.8.
> > >
> > > I wonder what bit of cleverness I'm missing.
>
> On Fri, Jun 15, 2007 at 07:59:56AM -0500, Kelly F. Hickel wrote:
> > Hmm, the obvious combinations work for me on centos 5.5.....
> >
> > [root@starbug2 ~]# TZ=GMT perl -e' print localtime()."\n";'
> > Fri Jun 15 12:58:10 2007
> > [root@starbug2 ~]# TZ="US/Eastern" perl -e' print localtime()."\n";'
> > Fri Jun 15 08:58:21 2007
> > [root@starbug2 ~]# perl  -e' $ENV{TZ} = "GMT"; print localtime()."\n";'
> > Fri Jun 15 12:59:08 2007
> > [root@starbug2 ~]# perl  -e' $ENV{TZ} = "US/Eastern"; print
> > localtime()."\n";'
> > Fri Jun 15 08:59:15 2007
>
> There appear to be two things interacting here (at least in Jesse's example)
[snip]

> so TZ is getting ignored.
>
> Whereas an Ubuntu machine recognises the TZ strings:
>
> $ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
> Fri Jun 15 09:32:56 2007
> Fri Jun 15 09:32:56 2007
> $ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = "US/Eastern"; print localtime()."\n";'
> Fri Jun 15 07:33:00 2007
> Fri Jun 15 07:33:00 2007
>
> then both print out that timezone at initialisation time, and the subsequent
> change is ignored.
Gentoo linux (Europe/Moscow is default):
$ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} =
"US/Eastern"; print localtime()."\n";'
Fri Jun 15 07:44:55 2007
Fri Jun 15 09:44:55 2007
$ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} =
"US/Mountain"; print localtime()."\n";'
Fri Jun 15 09:45:09 2007
Fri Jun 15 07:45:09 2007

So not only MacOS is doing something special.

>
> So OS X must be doing something "special" and resetting the timezone for every
> call to localtime. Note that GMT isn't a great one to test with, as the
> default for timezones you don't understand is 0, which is GMT.
>
[snip]

>
> And no, I don't know more than this.
>
> Nicholas Clark


-- 
Best regards, Ruslan.
From rg at progtech.net  Fri Jun 15 10:22:59 2007
From: rg at progtech.net (Rolf Grossmann)
Date: Fri Jun 15 10:39:12 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <20070615133830.GE18077@plum.flirble.org>
References: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>	<63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>
	<20070615133830.GE18077@plum.flirble.org>
Message-ID: <4672A0C3.3080306@PROGTECH.net>

Hi,

Nicholas Clark wrote:

> There appear to be two things interacting here (at least in Jesse's example)
> 
> On a FreeBSD machine that's in GMT I see:
> 
> $ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = "US/Eastern"; print localtime()."\n";'
> Fri Jun 15 13:31:51 2007
> Fri Jun 15 13:31:51 2007
> $ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
> Fri Jun 15 13:31:54 2007
> Fri Jun 15 13:31:54 2007
> 
> 
> so TZ is getting ignored.

That is because FreeBSD doesn't have definitions for US/* timezones. If 
you use e.g. perl -e' print localtime()."\n"; $ENV{TZ} = 
"America/New_York"; print localtime()."\n";'
I (being in Germany) get:

Fri Jun 15 16:13:30 2007
Fri Jun 15 10:13:30 2007

Note there is no caching of the TZ variable (I have perl 5.8.7).

> Whereas an Ubuntu machine recognises the TZ strings:
> 
> $ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
> Fri Jun 15 09:32:56 2007
> Fri Jun 15 09:32:56 2007
 > $ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = 
"US/Eastern"; print localtime()."\n";'
 > Fri Jun 15 07:33:00 2007
 > Fri Jun 15 07:33:00 2007
 >
 > then both print out that timezone at initialisation time, and the 
subsequent
 > change is ignored.

Those results (like the original requestor seems to be getting) on the 
other hand look very strange to me. Apparently the TZ is being cached.

> So OS X must be doing something "special" and resetting the timezone for every
> call to localtime. Note that GMT isn't a great one to test with, as the
> default for timezones you don't understand is 0, which is GMT.

I'm inclined to say Linux is doing something special/weird by caching 
the TZ value.

> And no, I don't know more than this.

Me neither. Nor do I have a Linux box available to look for the cause.

Bye, Rolf.
From nick at ccl4.org  Fri Jun 15 12:23:31 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Fri Jun 15 12:23:47 2007
Subject: [Rt-devel] Errors generate warnings
Message-ID: <20070615162331.GF18077@plum.flirble.org>

html/Element/Errors generates uninitialized value warnings if the parameter
$Details is not passed in, and if the SessionType is undefined. The following
patch quietens it.

Nicholas Clark

Index: html/Elements/Error
===================================================================
RCS file: /export/cvsroot/rt-external/html/Elements/Error,v
retrieving revision 1.1.1.1
diff -p -u -r1.1.1.1 Error
--- html/Elements/Error	21 May 2007 14:38:40 -0000	1.1.1.1
+++ html/Elements/Error	15 Jun 2007 16:17:14 -0000
@@ -49,7 +49,7 @@
 <div class="error">
 <%$Why%>
 <br />
-<%$Details%>
+<% defined $Details ? $Details : '' %>
 </div>
 
 <%cleanup>
@@ -65,7 +65,8 @@ $Why => loc("the calling component did n
 </%args>
 
 <%INIT>
-my $error = "WebRT: $Why ($Details)";
+my $error = "WebRT: $Why ";
+$error .= " ($Details)" if defined $Details;
 
 # TODO: Log::Dispatch isn't UTF-8 safe. Autrijus needs to talk to dave rolsky about getting this fixed
 if ($] >= 5.007001) {
@@ -75,10 +76,10 @@ if ($] >= 5.007001) {
 
 $RT::Logger->error($error);
 
-if ( $session{'SessionType'} eq 'REST' ) {
+if ( defined $session{'SessionType'} && $session{'SessionType'} eq 'REST' ) {
     $r->content_type('text/plain');
     $m->out( "Error: " . $Why . "\n" );
-    $m->out( $Details . "\n" );
+    $m->out( $Details . "\n" ) if defined $Details;
     $m->abort();
 }
 </%INIT>
From jesse at bestpractical.com  Fri Jun 15 12:31:08 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Fri Jun 15 12:31:49 2007
Subject: [Rt-devel] Errors generate warnings
In-Reply-To: <20070615162331.GF18077@plum.flirble.org>
References: <20070615162331.GF18077@plum.flirble.org>
Message-ID: <0E85D75F-4518-4CC9-B405-EE51850420E4@bestpractical.com>

About to apply this equivalent, slightly more Masonic patch. Thanks!

=== html/Elements/Error
==================================================================
--- html/Elements/Error (revision 58163)
+++ html/Elements/Error (local)
@@ -61,7 +61,7 @@

  <%args>
  $Code => undef
-$Details => undef
+$Details =>''
  $Title => loc("RT Error")
  $Why => loc("the calling component did not specify why")
  </%args>
@@ -77,7 +77,7 @@

  $RT::Logger->error($error);

-if ( $session{'SessionType'} eq 'REST' ) {
+if ( defined ($session{'SessionType'} && $session{'SessionType'} eq  
'REST' ) {
      $r->content_type('text/plain');
      $m->out( "Error: " . $Why . "\n" );
      $m->out( $Details . "\n" );

On Jun 15, 2007, at 12:23 PM, Nicholas Clark wrote:

> html/Element/Errors generates uninitialized value warnings if the  
> parameter
> $Details is not passed in, and if the SessionType is undefined. The  
> following
> patch quietens it.
>
> Nicholas Clark
>
> Index: html/Elements/Error
> ===================================================================
> RCS file: /export/cvsroot/rt-external/html/Elements/Error,v
> retrieving revision 1.1.1.1
> diff -p -u -r1.1.1.1 Error
> --- html/Elements/Error	21 May 2007 14:38:40 -0000	1.1.1.1
> +++ html/Elements/Error	15 Jun 2007 16:17:14 -0000
> @@ -49,7 +49,7 @@
>  <div class="error">
>  <%$Why%>
>  <br />
> -<%$Details%>
> +<% defined $Details ? $Details : '' %>
>  </div>
>
>  <%cleanup>
> @@ -65,7 +65,8 @@ $Why => loc("the calling component did n
>  </%args>
>
>  <%INIT>
> -my $error = "WebRT: $Why ($Details)";
> +my $error = "WebRT: $Why ";
> +$error .= " ($Details)" if defined $Details;
>
>  # TODO: Log::Dispatch isn't UTF-8 safe. Autrijus needs to talk to  
> dave rolsky about getting this fixed
>  if ($] >= 5.007001) {
> @@ -75,10 +76,10 @@ if ($] >= 5.007001) {
>
>  $RT::Logger->error($error);
>
> -if ( $session{'SessionType'} eq 'REST' ) {
> +if ( defined $session{'SessionType'} && $session{'SessionType'} eq  
> 'REST' ) {
>      $r->content_type('text/plain');
>      $m->out( "Error: " . $Why . "\n" );
> -    $m->out( $Details . "\n" );
> +    $m->out( $Details . "\n" ) if defined $Details;
>      $m->abort();
>  }
>  </%INIT>
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/ 
> rt-devel
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070615/59fe11e9/PGP.pgp
From cbs at cts.ucla.edu  Fri Jun 15 16:20:58 2007
From: cbs at cts.ucla.edu (Chris Stromsoe)
Date: Fri Jun 15 16:29:01 2007
Subject: [Rt-devel] RT::Interface::Web::Redirect() broken (for 3.6) in split
 front/back proxying setup
Message-ID: <Pine.LNX.4.64.0706151311020.22134@potato.cts.ucla.edu>

RT::Interface::Web::Redirect() is broken for a split front-end proxy / 
back-end RT setup.


I'm running multiple instances of RT spread over several backend servers. 
They're all accessible through a single set of front-end proxies that use 
a combination of different ports and paths (though for these purposes they 
could also use different hostnames) to direct access to the correct 
backend.

As an example, the url https://rt-front.domain:8888/rt99/ might proxy 
through to https://rt-back.otherdoman:8080/rt-customer-x/

I'm finding that some times when Redirect() is called, it is sending a 
redirect back out to the client trying to redirect to the back-end 
hostname and port.

Redirect looks like this:

sub Redirect {
     my $redir_to = shift;
     untie $HTML::Mason::Commands::session;
     my $uri = URI->new($redir_to);
     my $server_uri = URI->new($RT::WebURL);

     # If the user is coming in via a non-canonical
     # hostname, don't redirect them to the canonical host,
     # it will just upset them (and invalidate their credentials)
     if ($uri->host  eq $server_uri->host &&
         $uri->port eq $server_uri->port) {
             $uri->host($ENV{'HTTP_HOST'});
             $uri->port($ENV{'SERVER_PORT'});
         }

     $HTML::Mason::Commands::m->redirect($uri->canonical);
     $HTML::Mason::Commands::m->abort;
}



If I comment out the calls to $uri->host() and $uri->port() everything 
works fine.  I'm using 3.6.1 as shipped with Debian stable.  Based on at 
least one comment in the archives from back in April or May, it sounds 
like this also happens with 3.6.3.  I have not looked at 3.6.4 release 
candidates to see if the same code exists.


Can the code block be changed to refer to set variables in 
RT_SiteConfig.pm instead of the environment hash?  Or to check for 
something that was set intentionally?  Or to check if the redirect should 
always be to the canonical name?



-Chris
From nick at ccl4.org  Fri Jun 15 17:41:16 2007
From: nick at ccl4.org (Nicholas Clark)
Date: Fri Jun 15 17:41:22 2007
Subject: [Rt-devel] Errors generate warnings
In-Reply-To: <0E85D75F-4518-4CC9-B405-EE51850420E4@bestpractical.com>
References: <20070615162331.GF18077@plum.flirble.org>
	<0E85D75F-4518-4CC9-B405-EE51850420E4@bestpractical.com>
Message-ID: <20070615214116.GG18077@plum.flirble.org>

On Fri, Jun 15, 2007 at 12:31:08PM -0400, Jesse Vincent wrote:
> About to apply this equivalent, slightly more Masonic patch. Thanks!
> 
> === html/Elements/Error
> ==================================================================
> --- html/Elements/Error (revision 58163)
> +++ html/Elements/Error (local)
> @@ -61,7 +61,7 @@
> 
>  <%args>
>  $Code => undef
> -$Details => undef
> +$Details =>''
>  $Title => loc("RT Error")
>  $Why => loc("the calling component did not specify why")
>  </%args>

I thought about that, but I wasn't sure whether $Details being undef was
intended to be a significant difference from $Details being an empty string.
Thinking about it now, it would seem pretty silly to have an empty string as
the supplied text detailing the error. So, yes, that approach is better.

Nicholas Clark
From pape-rt at inf.fu-berlin.de  Sat Jun 16 03:24:42 2007
From: pape-rt at inf.fu-berlin.de (Dirk Pape)
Date: Sat Jun 16 03:24:53 2007
Subject: [Rt-devel] patch: Optimize recipients for Notifications Mails
In-Reply-To: <589c94400706150552s4f34fb8bvd011de37aee309e5@mail.gmail.com>
References: <0ADF91A1AB8F1D246B25F71B@idefix.mi.fu-berlin.de>	
	<589c94400706150120i6e7f7fd4i23ad46a259f31d9a@mail.gmail.com>	
	<967A8F84-D717-4D63-B966-36C8AC32E164@bestpractical.com>
	<589c94400706150552s4f34fb8bvd011de37aee309e5@mail.gmail.com>
Message-ID: <E371CF8F65174D62392A13D8@[192.168.1.7]>

Hello,

Thanks, but I see that it does not help with the issue I had (more than one 
notification scrip is fireing for the same person being AdminCc, Owner and 
OtherRecipient).

The optimization I did with this patch is already done by the MTA, which 
apparently does not send a message twice to one person within the same 
session.

Nevertheless the patch (or a slightly better version of it, that also 
removes duplicates in one Header field - I will send later) is useful, 
because cleaning the recipient header will help in a better "preview" on 
the Ticket/Update.html who will receive a message, and in the "recorded 
email" transactions.

For my original intention I have to cleanup my scrip logic.

Thanks, Dirk.


--Am 15. Juni 2007 16:52:26 +0400 schrieb Ruslan Zakirov 
<ruz@bestpractical.com>:

> ok.
>
> On 6/15/07, Jesse Vincent <jesse@bestpractical.com> wrote:
>>
>> On Jun 15, 2007, at 4:20 AM, Ruslan Zakirov wrote:
>>
>> > good patch, vote for applying
>>
>> Ok. Let's aim for 3.6.5?
>>
>>
>>
>> >
>> > On 6/15/07, Dirk Pape <pape-rt@inf.fu-berlin.de> wrote:
>> >> Hello,
>> >>
>> >> we found, that if we have ScripActions that notify more than one
>> >> watcher
>> >> group (e.g. "Notify AdminCC and Cc as Comment"), people get more
>> >> than one
>> >> message for the same notification, if they have more than one role
>> >> for the
>> >> ticket (e.g. beeing AdminCc and Cc).
>> >>
>> >> This is not necessary at all but only annoying, so I propose
>> >> following
>> >> general patch for Action/SendEmail.pm which removes unnecessary
>> >> recipient
>> >> entries: It removes every Cc already in To and every Bcc if
>> >> already in To
>> >> or Cc.
>> >>
>> >> The patch is tested here and behaves like expected. Will this make
>> >> it into
>> >> the core?
>> >>
>> >> <http://page.mi.fu-berlin.de/~pape/rt3/patches/rt/
>> >> optimize_recipients.patch>
>> >>
>> >> Regards, Dirk.
>> >> --
>> >> Dr. Dirk Pape (eAS - Projektleitung Campus Management)
>> >> Freie Universitaet Berlin
>> >> Grunewaldstr. 34a, 12165 Berlin
>> >> Tel. +49 (0)30 838 75143, Fax. +49 (0)30 838 54654
>> >> _______________________________________________
>> >> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/
>> >> rt-devel
>> >>
>> >
>> >
>> > --
>> > Best regards, Ruslan.
>> > _______________________________________________
>> > List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/
>> > rt-devel
>> >
>>
>>



-- 
Dr. Dirk Pape (eAS - Projektleitung Campus Management)
Freie Universitaet Berlin
Grunewaldstr. 34a, 12165 Berlin
Tel. +49 (30) 838 75143, Fax. +49 (30) 838 54654

From sven.sternberger at desy.de  Sat Jun 16 06:44:29 2007
From: sven.sternberger at desy.de (Sven Sternberger)
Date: Sat Jun 16 06:44:52 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <4672A0C3.3080306@PROGTECH.net>
References: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
	<63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>
	<20070615133830.GE18077@plum.flirble.org>
	<4672A0C3.3080306@PROGTECH.net>
Message-ID: <1181990669.5670.7.camel@pcx4546.desy.de>

Hello!

here my 2c

my test: 

perl -we'print localtime()."\n";$ENV{TZ} = "Europe/London"; print
localtime()."\n";'

Solaris 10: different
Ubuntu 6.04: equal
SL4.4 (RHEL4): different
Suse8.2: different
Debian Sarge: equal

So because I find it funy I googled and found this one:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=48184
http://www.gnu.org/software/libc/FAQ.html#s-4.3

So for me it looks like TZ change is ignored in an running process 
context (you can see this also with strace).

regards!

sven

On Fri, 2007-06-15 at 16:22 +0200, Rolf Grossmann wrote:
> Hi,
> 
> Nicholas Clark wrote:
> 
> > There appear to be two things interacting here (at least in Jesse's example)
> > 
> > On a FreeBSD machine that's in GMT I see:
> > 
> > $ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = "US/Eastern"; print localtime()."\n";'
> > Fri Jun 15 13:31:51 2007
> > Fri Jun 15 13:31:51 2007
> > $ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
> > Fri Jun 15 13:31:54 2007
> > Fri Jun 15 13:31:54 2007
> > 
> > 
> > so TZ is getting ignored.
> 
> That is because FreeBSD doesn't have definitions for US/* timezones. If 
> you use e.g. perl -e' print localtime()."\n"; $ENV{TZ} = 
> "America/New_York"; print localtime()."\n";'
> I (being in Germany) get:
> 
> Fri Jun 15 16:13:30 2007
> Fri Jun 15 10:13:30 2007
> 
> Note there is no caching of the TZ variable (I have perl 5.8.7).
> 
> > Whereas an Ubuntu machine recognises the TZ strings:
> > 
> > $ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/Mountain"; print localtime()."\n";'
> > Fri Jun 15 09:32:56 2007
> > Fri Jun 15 09:32:56 2007
>  > $ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = 
> "US/Eastern"; print localtime()."\n";'
>  > Fri Jun 15 07:33:00 2007
>  > Fri Jun 15 07:33:00 2007
>  >
>  > then both print out that timezone at initialisation time, and the 
> subsequent
>  > change is ignored.
> 
> Those results (like the original requestor seems to be getting) on the 
> other hand look very strange to me. Apparently the TZ is being cached.
> 
> > So OS X must be doing something "special" and resetting the timezone for every
> > call to localtime. Note that GMT isn't a great one to test with, as the
> > default for timezones you don't understand is 0, which is GMT.
> 
> I'm inclined to say Linux is doing something special/weird by caching 
> the TZ value.
> 
> > And no, I don't know more than this.
> 
> Me neither. Nor do I have a Linux box available to look for the cause.
> 
> Bye, Rolf.
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel

From vivek at khera.org  Sat Jun 16 12:43:05 2007
From: vivek at khera.org (Vivek Khera)
Date: Sat Jun 16 12:43:17 2007
Subject: [Rt-devel] RT 3.6.4rc2
In-Reply-To: <20070615133830.GE18077@plum.flirble.org>
References: <17699F8C-F47E-4CF1-A623-84C68710E07B@bestpractical.com>
	<63BEA5E623E09F4D92233FB12A9F7943F461D0@emailmn.mqsoftware.com>
	<20070615133830.GE18077@plum.flirble.org>
Message-ID: <4CAA87C0-7D17-4CDF-A38C-3BC5E6C25A04@khera.org>


On Jun 15, 2007, at 9:38 AM, Nicholas Clark wrote:

> On a FreeBSD machine that's in GMT I see:
>
> $ TZ=US/Mountain perl -we' print localtime()."\n"; $ENV{TZ} = "US/ 
> Eastern"; print localtime()."\n";'
> Fri Jun 15 13:31:51 2007
> Fri Jun 15 13:31:51 2007
> $ TZ=US/Eastern perl -we' print localtime()."\n"; $ENV{TZ} = "US/ 
> Mountain"; print localtime()."\n";'
> Fri Jun 15 13:31:54 2007
> Fri Jun 15 13:31:54 2007
>
>
> so TZ is getting ignored.
>

It is ignored because there is not US/Eastern timezone file on  
FreeBSD.  Peek in /usr/share/zoneinfo.

On the Mac, there is a US/Eastern file within /usr/share/zoneinfo.   
I'm sure ubuntu has it as well.

This is why you have differences.  The systems don't know the names  
of the zones you're giving them.

From jesse at bestpractical.com  Mon Jun 18 12:07:32 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Mon Jun 18 12:07:58 2007
Subject: [Rt-devel] RT::Interface::Web::Redirect() broken (for 3.6) in
	split front/back proxying setup
In-Reply-To: <Pine.LNX.4.64.0706151311020.22134@potato.cts.ucla.edu>
References: <Pine.LNX.4.64.0706151311020.22134@potato.cts.ucla.edu>
Message-ID: <98D41D5A-CE45-405B-874C-FA7DA7704176@bestpractical.com>


On Jun 15, 2007, at 4:20 PM, Chris Stromsoe wrote:

> RT::Interface::Web::Redirect() is broken for a split front-end  
> proxy / back-end RT setup.
>
> I'm running multiple instances of RT spread over several backend  
> servers. They're all accessible through a single set of front-end  
> proxies that use a combination of different ports and paths (though  
> for these purposes they could also use different hostnames) to  
> direct access to the correct backend.
>
> As an example, the url https://rt-front.domain:8888/rt99/ might  
> proxy through to https://rt-back.otherdoman:8080/rt-customer-x/
>
> I'm finding that some times when Redirect() is called, it is  
> sending a redirect back out to the client trying to redirect to the  
> back-end hostname and port.
>
> Redirect looks like this:
>
> sub Redirect {
>     my $redir_to = shift;
>     untie $HTML::Mason::Commands::session;
>     my $uri = URI->new($redir_to);
>     my $server_uri = URI->new($RT::WebURL);
>
>     # If the user is coming in via a non-canonical
>     # hostname, don't redirect them to the canonical host,
>     # it will just upset them (and invalidate their credentials)
>     if ($uri->host  eq $server_uri->host &&
>         $uri->port eq $server_uri->port) {
>             $uri->host($ENV{'HTTP_HOST'});
>             $uri->port($ENV{'SERVER_PORT'});
>         }
>
>     $HTML::Mason::Commands::m->redirect($uri->canonical);
>     $HTML::Mason::Commands::m->abort;
> }
>
>
>
> If I comment out the calls to $uri->host() and $uri->port()  
> everything works fine.  I'm using 3.6.1 as shipped with Debian  
> stable.  Based on at least one comment in the archives from back in  
> April or May, it sounds like this also happens with 3.6.3.  I have  
> not looked at 3.6.4 release candidates to see if the same code exists.

The behaviour is the same with 3.6.4. (And we're too late to change  
it now. We're only looking at things that are regressions from 3.6.3  
at this point.

I agree that the current behaviour is broken for split frontend/ 
backend setups. That's not ever been something we've explicitly  
supported, but it's not something I want to not support. That is to  
say "I want it to work right for everybody."

So, let's get a solution worked out.

The current code has a special case of "If we are redirecting to a  
URL that appears to be a 'canonical' RT URL and hit RT through  
something that appears to be non-canonical, for the love of god,  
don't redirect them. (When we do, they get asked to log in again  
because their cookie is for another domain)  Unfortunately, that  
directly conflicts with the backend-server case, where RT is running  
on a server and port that shouldn't be advertised.

> Can the code block be changed to refer to set variables in  
> RT_SiteConfig.pm instead of the environment hash?  Or to check for  
> something that was set intentionally?  Or to check if the redirect  
> should always be to the canonical name?
>
So yeah, I think that a config file flag for "always canonicalize on  
redirct" would be a welcome patch for 3.6.5. Is that something you  
can whip up?

Best,
Jesse

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070618/8c69bea6/PGP.pgp
From diablo at roedu.net  Tue Jun 19 06:08:10 2007
From: diablo at roedu.net (=?UTF-8?B?TWFudWVsIMWeVUJSRURV?=)
Date: Tue Jun 19 06:27:15 2007
Subject: [Rt-devel] CustomRights
Message-ID: <4677AB0A.70105@roedu.net>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I'm trying to develop a custom reports tool. Since I want to properly to
the job, I need to add a custom right to RT, to be able to 'filter' the
users who are allowed to see the reports.
 I searched on the wiki and on the web, but I didn't find any reference
on who to add custom rights to RT. Can someone please give me some
pointers ?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGd6sKjGXbUSvc3AsRAqUBAJ9o3o123/VSBDWiEZDLZEpjvYkpeACaA2kc
5Q/gqhx7nT7QLfQDkyEV6q4=
=L7TL
-----END PGP SIGNATURE-----
From cbs at cts.ucla.edu  Tue Jun 19 13:15:45 2007
From: cbs at cts.ucla.edu (Chris Stromsoe)
Date: Tue Jun 19 13:15:54 2007
Subject: [Rt-devel] RT::Interface::Web::Redirect() broken (for 3.6) in
	split front/back proxying setup
In-Reply-To: <98D41D5A-CE45-405B-874C-FA7DA7704176@bestpractical.com>
References: <Pine.LNX.4.64.0706151311020.22134@potato.cts.ucla.edu>
	<98D41D5A-CE45-405B-874C-FA7DA7704176@bestpractical.com>
Message-ID: <Pine.LNX.4.64.0706191011160.11508@potato.cts.ucla.edu>

On Mon, 18 Jun 2007, Jesse Vincent wrote:
> On Jun 15, 2007, at 4:20 PM, Chris Stromsoe wrote:
>
>> Can the code block be changed to refer to set variables in 
>> RT_SiteConfig.pm instead of the environment hash?  Or to check for 
>> something that was set intentionally?  Or to check if the redirect 
>> should always be to the canonical name?
>> 
> So yeah, I think that a config file flag for "always canonicalize on 
> redirct" would be a welcome patch for 3.6.5. Is that something you can 
> whip up?

Nothing fancy, but I've added a call in RT_SiteConfig.pm to:

     Set($CanonicalRedirect, 1);

and then modified Redirect() with a one-liner to this:

     sub Redirect {
         my $redir_to = shift;
         untie $HTML::Mason::Commands::session;
         my $uri = URI->new($redir_to);
         my $server_uri = URI->new($RT::WebURL);

         # If the user is coming in via a non-canonical
         # hostname, don't redirect them to the canonical host,
         # it will just upset them (and invalidate their credentials)
         if ((!defined $RT::CanonicalRedirect || $RT::CanonicalRedirect == 0) &&
             $uri->host  eq $server_uri->host &&
             $uri->port eq $server_uri->port) {
                 $uri->host($ENV{'HTTP_HOST'});
                 $uri->port($ENV{'SERVER_PORT'});
             }

         $HTML::Mason::Commands::m->redirect($uri->canonical);
         $HTML::Mason::Commands::m->abort;
     }


That preserves the current behavior by default, only doing a canonical 
redirect if CanonicalRedirect is set to some non-zero value.



-Chris
From jesse at bestpractical.com  Tue Jun 19 13:57:55 2007
From: jesse at bestpractical.com (Jesse Vincent)
Date: Tue Jun 19 13:58:24 2007
Subject: [Rt-devel] RT::Interface::Web::Redirect() broken (for 3.6) in
	split front/back proxying setup
In-Reply-To: <Pine.LNX.4.64.0706191011160.11508@potato.cts.ucla.edu>
References: <Pine.LNX.4.64.0706151311020.22134@potato.cts.ucla.edu>
	<98D41D5A-CE45-405B-874C-FA7DA7704176@bestpractical.com>
	<Pine.LNX.4.64.0706191011160.11508@potato.cts.ucla.edu>
Message-ID: <F442B491-E985-4A2D-AD2B-CFF8DB68801C@bestpractical.com>

I've forwarded this to rt-bugs so we can address it for 3.6.5. Thanks

On Jun 19, 2007, at 1:15 PM, Chris Stromsoe wrote:

> On Mon, 18 Jun 2007, Jesse Vincent wrote:
>> On Jun 15, 2007, at 4:20 PM, Chris Stromsoe wrote:
>>
>>> Can the code block be changed to refer to set variables in  
>>> RT_SiteConfig.pm instead of the environment hash?  Or to check  
>>> for something that was set intentionally?  Or to check if the  
>>> redirect should always be to the canonical name?
>> So yeah, I think that a config file flag for "always canonicalize  
>> on redirct" would be a welcome patch for 3.6.5. Is that something  
>> you can whip up?
>
> Nothing fancy, but I've added a call in RT_SiteConfig.pm to:
>
>     Set($CanonicalRedirect, 1);
>
> and then modified Redirect() with a one-liner to this:
>
>     sub Redirect {
>         my $redir_to = shift;
>         untie $HTML::Mason::Commands::session;
>         my $uri = URI->new($redir_to);
>         my $server_uri = URI->new($RT::WebURL);
>
>         # If the user is coming in via a non-canonical
>         # hostname, don't redirect them to the canonical host,
>         # it will just upset them (and invalidate their credentials)
>         if ((!defined $RT::CanonicalRedirect ||  
> $RT::CanonicalRedirect == 0) &&
>             $uri->host  eq $server_uri->host &&
>             $uri->port eq $server_uri->port) {
>                 $uri->host($ENV{'HTTP_HOST'});
>                 $uri->port($ENV{'SERVER_PORT'});
>             }
>
>         $HTML::Mason::Commands::m->redirect($uri->canonical);
>         $HTML::Mason::Commands::m->abort;
>     }
>
>
> That preserves the current behavior by default, only doing a  
> canonical redirect if CanonicalRedirect is set to some non-zero value.
>
>
>
> -Chris
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070619/7f63eb54/PGP.pgp
From mattea at rescomp.berkeley.edu  Tue Jun 19 17:03:15 2007
From: mattea at rescomp.berkeley.edu (Matthew Ekstrand-Abueg)
Date: Tue Jun 19 17:03:22 2007
Subject: [Rt-devel] Option for Identical Names
Message-ID: <20070619210315.GL18982@rescomp.berkeley.edu>

Hello all,

I have noticed that the code blocks creation of multiple users with the
same name.  This is not desirable for a large-community where people may
have the same name, but everyone still has difference email addresses.
So, as a workaround, I modified the code in the check so that if it is
trying to auto create a user with the same name, it instead takes the
email address and puts it into the name field, obviously still failing
after that if two of the same email addresses exist.

I honestly don't remember if this was a part of the LDAP_Overlay I
added, so I am mentioning this just in case that is true.

In etc/RT_SiteConfig.pm:

# Allows for multiple users to be created with the same Name
Set($AllowDuplicateNames, 1);

In lib/RT/User_Overlay.pm Create function:

#SANITY CHECK THE NAME AND ABORT IF IT'S TAKEN
if ($RT::SystemUser) {   #This only works if RT::SystemUser has been defined
    my $TempUser = RT::User->new($RT::SystemUser);
    $TempUser->Load( $args{'Name'} );
    if ( $TempUser->Id ) {
        if ( $RT::AllowDuplicateNames ) {
            $args{'Name'} = $args{'EmailAddress'};
        } else {
            return ( 0, $self->loc('Name in use') );
    }
     }

    return ( 0, $self->loc('Email address in use') )
      unless ( $self->ValidateEmailAddress( $args{'EmailAddress'} ) );
}
else {
    $RT::Logger->warning( "$self couldn't check for pre-existing users");
}

Notice the added internal if statements.  I hope my changes are at least
marginally helpful in resolving this issue.  Thank you very much.

-- 
Matthew Ekstrand-Abueg
Systems Administrator
Network Infrastructure, RSSP-IT
UC Berkeley

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070619/d72b10bb/attachment.pgp
From joby at u.washington.edu  Tue Jun 19 17:07:52 2007
From: joby at u.washington.edu (Joby Walker)
Date: Tue Jun 19 17:08:01 2007
Subject: [Rt-devel] Option for Identical Names
In-Reply-To: <20070619210315.GL18982@rescomp.berkeley.edu>
References: <20070619210315.GL18982@rescomp.berkeley.edu>
Message-ID: <467845A8.2040303@u.washington.edu>

Name cannot be identical for this is the login identity of the user!
Real Name is the name of the user and can be non-unique.

Joby Walker
C&C SSG, University of Washington


Matthew Ekstrand-Abueg wrote:
> Hello all,
> 
> I have noticed that the code blocks creation of multiple users with the
> same name.  This is not desirable for a large-community where people may
> have the same name, but everyone still has difference email addresses.
> So, as a workaround, I modified the code in the check so that if it is
> trying to auto create a user with the same name, it instead takes the
> email address and puts it into the name field, obviously still failing
> after that if two of the same email addresses exist.
> 
> I honestly don't remember if this was a part of the LDAP_Overlay I
> added, so I am mentioning this just in case that is true.
> 
> In etc/RT_SiteConfig.pm:
> 
> # Allows for multiple users to be created with the same Name
> Set($AllowDuplicateNames, 1);
> 
> In lib/RT/User_Overlay.pm Create function:
> 
> #SANITY CHECK THE NAME AND ABORT IF IT'S TAKEN
> if ($RT::SystemUser) {   #This only works if RT::SystemUser has been defined
>     my $TempUser = RT::User->new($RT::SystemUser);
>     $TempUser->Load( $args{'Name'} );
>     if ( $TempUser->Id ) {
>         if ( $RT::AllowDuplicateNames ) {
>             $args{'Name'} = $args{'EmailAddress'};
>         } else {
>             return ( 0, $self->loc('Name in use') );
>     }
>      }
> 
>     return ( 0, $self->loc('Email address in use') )
>       unless ( $self->ValidateEmailAddress( $args{'EmailAddress'} ) );
> }
> else {
>     $RT::Logger->warning( "$self couldn't check for pre-existing users");
> }
> 
> Notice the added internal if statements.  I hope my changes are at least
> marginally helpful in resolving this issue.  Thank you very much.
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
From maristbrandi at gmail.com  Wed Jun 20 10:21:24 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Wed Jun 20 10:21:43 2007
Subject: [Rt-devel] Adding an image, tables cut off
Message-ID: <a3a53f9f0706200721i4cd7cdeek4c1bea32c05efe1e@mail.gmail.com>

Hi everyone!

I have two issues here that I'm trying to solve.

FIrstly, I'm trying to add an image to the login page. Its just a logo,
nothing fancy, just for the look of the thing. Here is what I've done...

-First, I put the image in html/NoAuth/images

-To my RT_SiteConfig.pm file I've added the following code:
 Set( $WebImagesURL , $WebPath . "/NoAuth/images/");
 Set( $LoginImageURL, $WebImagesURL . "Helpdesk_Fox.gif");
 Set( $LoginAltText, "Help Desk Fox");

-Then, I added the following code to html/Elements/Login:
 <img src="<%$RT::LoginImageURL%>" alt="<% loc($RT::LoginAltText) %>" />

The image doesn't show up, though. All I get is the box with the x in it and
the LoginAltText. After I made the changes I stopped httpd, cleared the
mason cache, and started httpd again.


My other problem is the fact that in Internet Explorer (but not in Firefox)
the right side of my tables is getting cut off when using the
3.4-compatstylesheets (see screen print link below). Has anyone else
seen this
problem?
Screen print of cut off
tables<http://foxweb.marist.edu/users/kbr4b/screenPrint.png>


Does anyone have any suggestions? Help is, as always, greatly appreciated.

Thanks!

Regards,
Brandi L
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070620/ffd6eae7/attachment.htm
From maristbrandi at gmail.com  Wed Jun 20 12:44:59 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Wed Jun 20 12:45:06 2007
Subject: [Rt-devel] Re: R: [rt-users] Adding an image, tables cut off
In-Reply-To: <A6EA8BA3E2776540BBC22A6C4D259D73017C6548@MAILSERVER.pevianimail.local>
References: <A6EA8BA3E2776540BBC22A6C4D259D73017C6548@MAILSERVER.pevianimail.local>
Message-ID: <a3a53f9f0706200944v7dc7aedev5578afb97ef6fc73@mail.gmail.com>

Hi Gianluca,

In your RT_SiteConfig.pm file you just need to add the following line:
Set( $WebDefaultStylesheet, '3.4-compat');

Hope that helps!
-Brandi L


On 6/20/07, Gianluca Cecchi <gcecchi@peviani.it> wrote:
>
>  As a side question: how did you manage to have the Navigation Bar (Home,
> Simple Search, Tickets, etc) placed vertically instead of the default
> horizontal way?
> From the gui or tricking the code?
>
> thanks
> gianluca
>
>
>
> My other problem is the fact that in Internet Explorer (but not in
> Firefox) the right side of my tables is getting cut off when using the
> 3.4-compat stylesheets (see screen print link below). Has anyone else seen
> this problem?
> Screen print of cut off tables<http://foxweb.marist.edu/users/kbr4b/screenPrint.png>
>
>
>
> _______________________________________________
> http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users
>
> Community help: http://wiki.bestpractical.com
> Commercial support: sales@bestpractical.com
>
>
> Discover RT's hidden secrets with RT Essentials from O'Reilly Media.
> Buy a copy at http://rtbook.bestpractical.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070620/c3834bf5/attachment.htm
From maristbrandi at gmail.com  Thu Jun 21 09:00:06 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Thu Jun 21 09:00:22 2007
Subject: [Rt-devel] Re: R: R: [rt-users] Adding an image, tables cut off
In-Reply-To: <A6EA8BA3E2776540BBC22A6C4D259D73017C659C@MAILSERVER.pevianimail.local>
References: <A6EA8BA3E2776540BBC22A6C4D259D73017C659C@MAILSERVER.pevianimail.local>
Message-ID: <a3a53f9f0706210600g131f2a7bidd891feffcf3d550@mail.gmail.com>

Hi Gianluca,

Yep, I changed a few things. All I did was go into the folder that has the
stylesheets (\usr\share\rt3\html\NoAuth\css\3.4-compat) and changed some
values. I played around with them to my liking. I also changed my logo using
directions from the wiki (http://wiki.bestpractical.com/view/ChangeLogo) and
with some help from people on this list :). Let me know if you'd like
directions!

Best Regards,
-Brandi L


On 6/21/07, Gianluca Cecchi <gcecchi@peviani.it> wrote:
>
>  ok, thanks.
> but apart from this I think you also modified other layout parameters in
> your CSS such as colour, rendering, etc...
> Your printout looks quite better than default '3.4-compat' if I try
> it.....
>
>
>  ------------------------------
> *Da:* Brandi L [mailto:maristbrandi@gmail.com]
> *Inviato:* mercoled? 20 giugno 2007 18.45
> *A:* Gianluca Cecchi
> *Cc:* rt-devel@lists.bestpractical.com
> *Oggetto:* Re: R: [rt-users] Adding an image, tables cut off
>
>
>  Hi Gianluca,
>
> In your RT_SiteConfig.pm file you just need to add the following line:
> Set( $WebDefaultStylesheet, '3.4-compat');
>
> Hope that helps!
> -Brandi L
>
>
> On 6/20/07, Gianluca Cecchi <gcecchi@peviani.it> wrote:
> >
> >  As a side question: how did you manage to have the Navigation Bar
> > (Home, Simple Search, Tickets, etc) placed vertically instead of the default
> > horizontal way?
> > From the gui or tricking the code?
> >
> > thanks
> > gianluca
> >
> >
> >
> > My other problem is the fact that in Internet Explorer (but not in
> > Firefox) the right side of my tables is getting cut off when using the
> > 3.4-compat stylesheets (see screen print link below). Has anyone else
> > seen this problem?
> > Screen print of cut off tables<http://foxweb.marist.edu/users/kbr4b/screenPrint.png>
> >
> >
> >
> > _______________________________________________
> > http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users
> >
> > Community help: http://wiki.bestpractical.com
> > Commercial support: sales@bestpractical.com
> >
> >
> > Discover RT's hidden secrets with RT Essentials from O'Reilly Media.
> > Buy a copy at http://rtbook.bestpractical.com
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070621/12933fa4/attachment.htm
From maristbrandi at gmail.com  Thu Jun 21 11:37:17 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Thu Jun 21 11:37:26 2007
Subject: [Rt-devel] Re: R: R: [rt-users] Adding an image, tables cut off
In-Reply-To: <467A817A.6010208@voskamp.ca>
References: <A6EA8BA3E2776540BBC22A6C4D259D73017C659C@MAILSERVER.pevianimail.local>
	<a3a53f9f0706210600g131f2a7bidd891feffcf3d550@mail.gmail.com>
	<467A817A.6010208@voskamp.ca>
Message-ID: <a3a53f9f0706210837o218e641co2cc5f85e8a15bd08@mail.gmail.com>

Jeff,

Sure!
Here's the changes I've made...

forms.css:

.submit {
%#  background: #069;
    background: #A61E1E;
%#  These borders are needed so the container actually surrounds the floats
inside it
%#  border-top: 1px solid white;
%#  border-bottom: 1px solid white;
    border: 3px solid #A61E1E;
    border: 1px solid black;
%#  color: #ffdb00;
    color: white;
    font-weight: bold;
}


header.css

#header {
%#  background: #4282b5;
    background: #A61E1E;
%#  margin-top: 0;
%#  padding-bottom: 0.2em;
    color: white;
    border-left: 2px solid white;
    float: left;
%#  width: 82%;
    width: 81%;
}

#header h1 {
%#  background: #4282b5;
    background: #A61E1E;
    color: white;
    font-size: 1.7em;
%#  margin: 0;
%#  padding: 0;
    border: 4px solid #A61E1E;
}


main.css
I just added at the end...

A:visited {color: #732B2B}


misc.css

.oddline { background: white; }
%#.evenline { background: #cecfef; }
.evenline { background: #EEE9E9; }


nav.css

#nav {
    clear: left;
    float: left;
%#  width: 18%;
    width: 16%;
    font-size: 1.4em;
%#  color: #eee;
    color: white;
    margin: 0;
%#  background: #4282b5 url(<%$RT::WebImagesURL%>/css/cbr.gif) no-repeat
bottom right;
    background: #A61E1E url(<%$RT::WebImagesURL%>/css/cbr.gif) no-repeat
bottom right;
}

#nav :visited.selected
{
#   color: #ff6;
    color: white;
}


titlebox.css

.titlebox .titlebox-title {
%#  background: #069;
    background: #A61E1E;
    padding: 0.2em 0.5em;
    color: white;
%#  border-top: 1px solid black;
%#  border-bottom: 1px solid black;
    border: 1px solid black;
    font-weight: bold;
    position: relative;
}
.titlebox .titlebox-title .right {
    position: absolute;
%#  right: 1.5em;
    right: 0.5em;
    font-size: 0.9em;
}


transactions.css

.ticket-transaction.even {
%#  background: #cecfef;
    background: #EEE9E9;
}

%#.ticket-transaction.basics { border-color: #9c3031; }
.ticket-transaction.basics { border-color: #000000; }
%#.ticket-transaction.basics .type { background: #9c3031; }
.ticket-transaction.basics .type { background: #363636; }
.ticket-summary .ticket-info-basics .titlebox-content { border-left: none; }
%#.ticket-summary .ticket-info-basics .titlebox-title { background: #9c3031;
}
.ticket-summary .ticket-info-basics .titlebox-title { background: #585252;
color: white; }

%#.ticket-transaction.people { border-color: #31309c; }
.ticket-transaction.people { border-color: #000000; }
%#.ticket-transaction.people .type { background: #31309c; }
.ticket-transaction.people .type { background: #494949; }
.ticket-summary .ticket-info-people .titlebox-content { border-left: none; }
%#.ticket-summary .ticket-info-people .titlebox-title { background: #31309c;
}
.ticket-summary .ticket-info-people .titlebox-title { background: #585252;
color: white; }

%#.ticket-transaction.links { border-color: #316531; }
.ticket-transaction.links { border-color: #000000; }
%#.ticket-transaction.links .type { background: #316531; }
.ticket-transaction.links .type { background: #606060; }
.ticket-summary .ticket-info-links .titlebox-content  { border-left: none; }
%#.ticket-summary .ticket-info-links .titlebox-title  { background: #316531;
}
.ticket-summary .ticket-info-links .titlebox-title  { background: #A61E1E;
color: white; }

%#.ticket-transaction.dates { border-color: #633063; }
.ticket-transaction.dates { border-color: #000000; }
%#.ticket-transaction.dates .type { background: #633063; }
.ticket-transaction.dates .type { background: #707070; }
.ticket-summary .ticket-info-dates .titlebox-content  { border-left: none; }
%#.ticket-summary .ticket-info-dates .titlebox-title  { background: #633063;
}
.ticket-summary .ticket-info-dates .titlebox-title  { background: #585252;
color: white; }

%#.ticket-transaction.message { border-color: #069; }
.ticket-transaction.message { border-color: #000000; }
%#.ticket-transaction.message .type { background: #069; }
.ticket-transaction.message .type { background: #848484; }

Enjoy!

Best Regards,
Brandi L








On 6/21/07, Jeff Voskamp <jeff@voskamp.ca> wrote:
>
> Brandi L wrote:
> > Hi Gianluca,
> >
> > Yep, I changed a few things. All I did was go into the folder that has
> > the stylesheets (\usr\share\rt3\html\NoAuth\css\3.4-compat) and
> > changed some values. I played around with them to my liking. I also
> > changed my logo using directions from the wiki (
> > http://wiki.bestpractical.com/view/ChangeLogo) and with some help from
> > people on this list :). Let me know if you'd like directions!
> >
> > Best Regards,
> > -Brandi L
> I'd be interested in comparing your tweaks with my tweaks.
>
> Jeff Voskamp
>
> diff -rb /var/rt/html/NoAuth/css/3.5-default/titlebox.css ./titlebox.css
> 55c55
> <     /*margin: 1em 2em 0.5em 2em;*/
> ---
> >     /*margin: 1em 1em 0.0em 2em;*/
> 59c59
> <     margin: 1em 2em 0.5em 2em;
> ---
> >     margin: 1em 0em 0em 1em;
> 63c63,67
> <
> ---
> >     margin: 1em 0em 0em 1em;
> > }
> > .titlebox td {
> >     font-size: 1.0em;
> >     margin: 1em 0em 0em 1em;
> 67c71
> <
> ---
> >     margin: 1em -1em 0em 0em;
> 134c138
> <   background: url(<%$RT::WebImagesURL%>/css/rollup-arrow.gif) no-repeat
> center center;
> ---
> >   background: url(<%$RT::WebImagesURL%>/css/rollup- arrow.gif) no-repeat
> center top;
> 166c170
> <     background-position: center 0.3em;
> ---
> >     background-position: center top;
> 171c175
> <     background-image: url(<%$RT::WebImagesURL%>/css/rolldown- arrow.gif
> );
> ---
> >     background-image: url(<%$RT::WebImagesURL%>/css/rolldown-arrow.gif)
> no-repeat center top;
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070621/39ba5d98/attachment-0001.htm
From joby at u.washington.edu  Fri Jun 22 15:03:30 2007
From: joby at u.washington.edu (Joby Walker)
Date: Fri Jun 22 15:03:38 2007
Subject: [Rt-devel] New index on groupmembers table
Message-ID: <467C1D02.5090202@u.washington.edu>

I've noticed some significant improvement by adding the following index:

CREATE INDEX groupmembers1 on groupmembers (groupid, memberid);

Currently the groupmembers table is not indexed (apart from the PK), and
there are several operations that make use of the groupmembers table and
not the cachedgroupmembers table.  For example (in the stock RT) on
*every* page load RT checks to see if the currentuser is Privileged.


-- 

Joby Walker
C&C SSG, University of Washington
From svitter at list.ru  Fri Jun 22 16:05:28 2007
From: svitter at list.ru (Viktor)
Date: Fri Jun 22 16:05:41 2007
Subject: =?koi8-r?Q?Re=3A_[Rt-devel]_New_index_on_groupmembers_table?=
In-Reply-To: <467C1D02.5090202@u.washington.edu>
References: <467C1D02.5090202@u.washington.edu>
Message-ID: <E1I1pNo-0006GZ-00.svitter-list-ru@f37.mail.ru>

Hello Joby,

do you use RTx::Shredder extension? It adds some useful indexes, like
    "shredder_gm1" UNIQUE, btree (memberid, groupid)
in PostgreSQL.
RTx::Shredder is a must, please be sure to try it!

-----Original Message-----
From: Joby Walker <joby@u.washington.edu>
To: rt-devel@lists.bestpractical.com
Date: Fri, 22 Jun 2007 12:03:30 -0700
Subject: [Rt-devel] New index on groupmembers table

> 
> I've noticed some significant improvement by adding the following index:
> 
> CREATE INDEX groupmembers1 on groupmembers (groupid, memberid);
> 
> Currently the groupmembers table is not indexed (apart from the PK), and
> there are several operations that make use of the groupmembers table and
> not the cachedgroupmembers table.  For example (in the stock RT) on
> *every* page load RT checks to see if the currentuser is Privileged.
> 
> 
> -- 
> 
> Joby Walker
> C&C SSG, University of Washington
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
> 
From joby at u.washington.edu  Fri Jun 22 16:08:33 2007
From: joby at u.washington.edu (Joby Walker)
Date: Fri Jun 22 16:08:42 2007
Subject: [Rt-devel] New index on groupmembers table
In-Reply-To: <E1I1pNo-0006GZ-00.svitter-list-ru@f37.mail.ru>
References: <467C1D02.5090202@u.washington.edu>
	<E1I1pNo-0006GZ-00.svitter-list-ru@f37.mail.ru>
Message-ID: <467C2C41.8090506@u.washington.edu>

We haven't updated RTx::Shredder in quite a while...I'll check out the
most recent version.

Joby Walker
C&C SSG, University of Washington


Viktor wrote:
> Hello Joby,
> 
> do you use RTx::Shredder extension? It adds some useful indexes, like
>     "shredder_gm1" UNIQUE, btree (memberid, groupid)
> in PostgreSQL.
> RTx::Shredder is a must, please be sure to try it!
> 
> -----Original Message-----
> From: Joby Walker <joby@u.washington.edu>
> To: rt-devel@lists.bestpractical.com
> Date: Fri, 22 Jun 2007 12:03:30 -0700
> Subject: [Rt-devel] New index on groupmembers table
> 
>> I've noticed some significant improvement by adding the following index:
>>
>> CREATE INDEX groupmembers1 on groupmembers (groupid, memberid);
>>
>> Currently the groupmembers table is not indexed (apart from the PK), and
>> there are several operations that make use of the groupmembers table and
>> not the cachedgroupmembers table.  For example (in the stock RT) on
>> *every* page load RT checks to see if the currentuser is Privileged.
>>
>>
>> -- 
>>
>> Joby Walker
>> C&C SSG, University of Washington
>> _______________________________________________
>> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
>>
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
From alvaro.munoz.sanchez at hp.com  Mon Jun 25 06:54:25 2007
From: alvaro.munoz.sanchez at hp.com (Munoz, Alvaro)
Date: Mon Jun 25 06:55:43 2007
Subject: [Rt-devel] Hide custom fields 
Message-ID: <5230D735379C53489604251BE1B6DB6CD0ED3E@esmexc02.emea.cpqcorp.net>

 
Hi all,
 
Is it possible to show some custom fields only if certain selection has been made in another customfield (combobox)? Any of you have an example of this?
 
Thanks in advance!
 
Alvaro Mu?oz S?nchez
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070625/fef74281/attachment.htm
From maristbrandi at gmail.com  Mon Jun 25 09:41:44 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Mon Jun 25 09:41:51 2007
Subject: [Rt-devel] set table width
Message-ID: <a3a53f9f0706250641k29046413u133d13e8e5f98906@mail.gmail.com>

Hi all!

Does anyone know where in the code I can set the width of the tables
displayed on the 'RT at a Glance' page? I am using the 3.4 stylesheets, if
that makes a difference.

Thanks, in advance, for any help. :)

-Brandi L
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070625/252be450/attachment.htm
From Uli.Staerk at globalways.net  Mon Jun 25 11:34:43 2007
From: Uli.Staerk at globalways.net (Uli Staerk)
Date: Mon Jun 25 11:41:34 2007
Subject: [Rt-devel] REST Interface
Message-ID: <F91A18B8B41BEA4496988E928B3AAE6D45A0C2@EX01.hosted.globalways.net>

Hi,

I created a form on our homepage that can create tickets using the REST
Interface. With the form the user is able to add further information to
his tickets, or mark them as resolved.
But every time the user adds information via the form, the creator is
wrong. The creator is always the user that is authenticated - which is
basically correct. But as the customer-login has no access to the REST
API (gets always redirected to /SelfService), I cant use his login to
create the history entry.
How can I create a correspondence/comment that has the same creator as
the tickets requestor with the REST Interface?
From alvaro.munoz.sanchez at hp.com  Mon Jun 25 18:00:12 2007
From: alvaro.munoz.sanchez at hp.com (Munoz, Alvaro)
Date: Mon Jun 25 18:00:27 2007
Subject: [Rt-devel] Current time
Message-ID: <5230D735379C53489604251BE1B6DB6CD40DC2@esmexc02.emea.cpqcorp.net>

 
Hi all,
 
is there any variable i can use to check current time.
 
I would like to make a query like:
 
Query   => " Owner = '__CurrentUser__' AND ( started  > '__CurrentTime__' ) AND ( Status = 'open') AND (queue ='test_queue')",
 
But I dont know the right variable.
Actually what Im really interested in obtain is all tickets opened and not resolved for more than 20 days.
 
Any hints?
 
Thanks!
 
Alvaro Mu?oz S?nchez
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070626/1988725e/attachment.htm
From svitter at list.ru  Tue Jun 26 01:37:00 2007
From: svitter at list.ru (Viktor)
Date: Tue Jun 26 01:37:18 2007
Subject: =?koi8-r?Q?Re=3A_[Rt-devel]_Current_time?=
In-Reply-To: <5230D735379C53489604251BE1B6DB6CD40DC2@esmexc02.emea.cpqcorp.net>
References: <5230D735379C53489604251BE1B6DB6CD40DC2@esmexc02.emea.cpqcorp.net>
Message-ID: <E1I33jY-0001Uj-00.svitter-list-ru@f109.mail.ru>



-----Original Message-----
From: "Munoz, Alvaro" <alvaro.munoz.sanchez@hp.com>
To: <rt-devel@lists.bestpractical.com>
Date: Tue, 26 Jun 2007 00:00:12 +0200
Subject: [Rt-devel] Current time

>  
> Hi all,
>  
> is there any variable i can use to check current time.
>  
> I would like to make a query like:
>  
> Query   => " Owner = '__CurrentUser__' AND ( started  > '__CurrentTime__' ) AND ( Status = 'open') AND (queue ='test_queue')",
>  
> But I dont know the right variable.
> Actually what Im really interested in obtain is all tickets opened and not resolved for more than 20 days.

Please try

started > '20 days ago'

condition. Actually Date::Manip is used for date/time parsing, it recognizes almost any imaginable formats.

> Any hints?
>  
> Thanks!
>  
> Alvaro Mu?oz S?nchez
> 
> 
> _______________________________________________
> List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
> 
From diablo at roedu.net  Tue Jun 26 03:31:40 2007
From: diablo at roedu.net (=?UTF-8?B?TWFudWVsIMWeVUJSRURV?=)
Date: Tue Jun 26 03:32:21 2007
Subject: [Rt-devel] CustomRights - SOLVED
In-Reply-To: <4677AB0A.70105@roedu.net>
References: <4677AB0A.70105@roedu.net>
Message-ID: <4680C0DC.2090807@roedu.net>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I needed a global custom right to be granted to invidual users (or
groups) so that they can see some custom statistics I have done.
 To add a new custom right (global) to RT 3.6.x you need to create in
/opt/rt3/lib/RT directory a file named System_Vendor.pm with the
following content:

<begin of code>
$CustomRights = {
  SeeCustomStats => 'See the custom statistics',
};

foreach my $custom_right ( keys %{ $CustomRights } ) {
  $RT::ACE::LOWERCASERIGHTNAMES{lc $custom_right} = $custom_right;
  $RIGHTS->{$custom_right} = $CustomRights->{$custom_right};
}

1;
</end of code>

That code snippet will add a new right named SeeCustomStats to global
user and group rights . Ofc, more custom rights can be added.

 That code works for me ! I don't know if it will work on other version
of RT or if I broke something ;) At this moment all works as expected


Manuel ?UBREDU wrote:
> Hi,
> 
> I'm trying to develop a custom reports tool. Since I want to properly to
> the job, I need to add a custom right to RT, to be able to 'filter' the
> users who are allowed to see the reports.
>  I searched on the wiki and on the web, but I didn't find any reference
> on who to add custom rights to RT. Can someone please give me some
> pointers ?
_______________________________________________
List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGgMDcjGXbUSvc3AsRAsj1AJ4sTREHFB+zemS6QFT8XQrK0h0cNQCfaQ78
nYcyZ8Ef+gmZdFQ5ntKx7k4=
=XbCU
-----END PGP SIGNATURE-----
From andrew at mammoth.com.au  Tue Jun 26 02:14:04 2007
From: andrew at mammoth.com.au (Andrew Armstrong)
Date: Tue Jun 26 04:11:58 2007
Subject: [Rt-devel] RT Large Scale Performance Issues: Resolved
Message-ID: <002501c7b7b9$3288fe20$b6f3a305@plasma>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: SearchBuilder_Local.pm
Type: application/octet-stream
Size: 1368 bytes
Desc: not available
Url : http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070626/eaa6a0b0/SearchBuilder_Local.obj
From maristbrandi at gmail.com  Wed Jun 27 09:20:09 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Wed Jun 27 09:20:21 2007
Subject: [Rt-devel] Won't take images of a certain size?
Message-ID: <a3a53f9f0706270620t44ac29f9gc616d3121abb26df@mail.gmail.com>

Hey all,

I'm trying to add an image to my login page. It appears that RT will only
accept images of a certain size (the image I'm using for my logo will appear
just fine but the larger image, in the same place, will not. Anyone know how
I can change that so the image will show on the login page?

-Brandi L
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070627/4f95d789/attachment.htm
From alvaro.munoz.sanchez at hp.com  Wed Jun 27 09:37:28 2007
From: alvaro.munoz.sanchez at hp.com (Munoz, Alvaro)
Date: Wed Jun 27 09:37:46 2007
Subject: [Rt-devel] Current time SOLVED
In-Reply-To: <E1I33jY-0001Uj-00.svitter-list-ru@f109.mail.ru>
References: <5230D735379C53489604251BE1B6DB6CD40DC2@esmexc02.emea.cpqcorp.net>
	<E1I33jY-0001Uj-00.svitter-list-ru@f109.mail.ru>
Message-ID: <5230D735379C53489604251BE1B6DB6CD982DA@esmexc02.emea.cpqcorp.net>


Thanks a lot, that did it! 


-----Original Message-----
From: rt-devel-bounces@lists.bestpractical.com [mailto:rt-devel-bounces@lists.bestpractical.com] On Behalf Of Viktor
Sent: martes, 26 de junio de 2007 7:37
To: rt-devel@lists.bestpractical.com
Subject: Re: [Rt-devel] Current time



-----Original Message-----
From: "Munoz, Alvaro" <alvaro.munoz.sanchez@hp.com>
To: <rt-devel@lists.bestpractical.com>
Date: Tue, 26 Jun 2007 00:00:12 +0200
Subject: [Rt-devel] Current time

>  
> Hi all,
>  
> is there any variable i can use to check current time.
>  
> I would like to make a query like:
>  
> Query   => " Owner = '__CurrentUser__' AND ( started  > '__CurrentTime__' ) AND ( Status = 'open') AND (queue ='test_queue')",
>  
> But I dont know the right variable.
> Actually what Im really interested in obtain is all tickets opened and not resolved for more than 20 days.

Please try

started > '20 days ago'

condition. Actually Date::Manip is used for date/time parsing, it recognizes almost any imaginable formats.

> Any hints?
>  
> Thanks!
>  
> Alvaro Mu?oz S?nchez
> 
> 
> _______________________________________________
> List info: 
> http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
> 
_______________________________________________
List info: http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel
From alvaro.munoz.sanchez at hp.com  Wed Jun 27 18:08:44 2007
From: alvaro.munoz.sanchez at hp.com (Munoz, Alvaro)
Date: Wed Jun 27 18:09:17 2007
Subject: [Rt-devel] Dinamic Combo box
Message-ID: <5230D735379C53489604251BE1B6DB6CDC1CCF@esmexc02.emea.cpqcorp.net>

 
Hi listers,
 
Is there any way to change the contents of a combo box field depending
on the value selected in another combobox field?
Ive seen Jesse doing it but can remember how!
 
Thanks again!
 
Alvaro
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070628/a9803892/attachment.htm
From yanfat.liu at kuehne-nagel.com  Wed Jun 27 21:22:20 2007
From: yanfat.liu at kuehne-nagel.com (HKG RI-CI/LIU, YANFAT)
Date: Wed Jun 27 21:32:44 2007
Subject: [Rt-devel] Passing arguments from scrip action to template
In-Reply-To: <20070627160008.CF3534D80C9@diesel.bestpractical.com>
Message-ID: <000401c7b922$c6a88cf0$06fe460a@KNHKG.COM>

Hi all,

I'm dealing with scrip,different email will be sent to different users
according to the result output from my 'scrip-action'. And, contents of
emails can be changed accordingly.
Is it possible to pass arguments from 'scrip-action' to the template?
Or, is it easier to send email inside 'scrip-action' part with all different
email cases, with blank template?


From chris.fuka at gmail.com  Thu Jun 28 11:54:47 2007
From: chris.fuka at gmail.com (chris fuka)
Date: Thu Jun 28 11:55:00 2007
Subject: [Rt-devel] Scrip condition: "On RTFM Article Create"
Message-ID: <94c2f64d0706280854t407bfe17u753b2db4cdb1bb5f@mail.gmail.com>

I am trying to come up with a Scrip that will run a (mySQL) stored
procedure when an Article in RTFM is created.

The place where I work really likes RT, but we would like to have a
Knowledge Base that is publicly accessible and user friendly. With any
luck having this KB will cut down on the number of tickets that end up
in our support queue. RTFM does a good job of handling the extraction
from an RT ticket and if possible I would like to continue using it
for that purpose. I have written a stored procedure which extracts the
proper question and answer from the RTFM Article id it is passed and
sends it over to our KB.

I am using RT 3.6.1 and RTFM 2.2.0RC5. I gave the RT package (3.6.3 I
believe) a shot on Fedora Core 6 and it worked fine. As soon as RTFM
was installed however, I was knee deep in perl taint and I stopped
trying to debug it. I am on a Debian VM right now using the versions
previously stated.

I have a basic understanding of scrips from the RT Essentials book,
but I do not know how to apply them to RTFM.

Chris
From otto.hirr at olabinc.com  Thu Jun 28 11:28:54 2007
From: otto.hirr at olabinc.com (Otto Hirr)
Date: Thu Jun 28 12:15:46 2007
Subject: [Rt-devel] generator script
Message-ID: <E077297E18B24372BFEA8777AB0BD341@olabinc.com>

Greetings,

I am looking for the script which 'Autogenerates' the files in lib/RT like
User.pm. What data file is being used for its input?

Also what script is used to generate the various schemas in etc? At the top
of schema.Pg, I see 'My2Pg 1.23 translated dump' (for version rt-3.6.3).
What are the others using? or is it by hand? Has anyone tried
SQL::Translator?


Regards,

..Otto


From maristbrandi at gmail.com  Thu Jun 28 12:55:11 2007
From: maristbrandi at gmail.com (Brandi L)
Date: Thu Jun 28 12:55:21 2007
Subject: [Rt-devel] Code to change 'mandatory' message?
Message-ID: <a3a53f9f0706280955v5b035512u339262022a84185d@mail.gmail.com>

Hi All!

I made a couple of custom fields for my tickets. I made them mandatory. I
just want to change the message below the field that says 'Input must match
[Mandatory]' to something simpler, like 'Required.' Does anyone know where
that code can be found? I've been looking and haven't been able to find it.

Thanks!
-Brandi L
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070628/3af292d7/attachment.htm
From JStark at sumtotalsystems.com  Thu Jun 28 13:20:32 2007
From: JStark at sumtotalsystems.com (Jeff Stark)
Date: Thu Jun 28 13:20:49 2007
Subject: [Rt-devel] Code to change 'mandatory' message?
In-Reply-To: <a3a53f9f0706280955v5b035512u339262022a84185d@mail.gmail.com>
References: <a3a53f9f0706280955v5b035512u339262022a84185d@mail.gmail.com>
Message-ID: <1AC37FBEF7856646AC61A9F285BC14CC03C0ADBB@mtn-exch1.sumtotalsystems.com>

Elements/ValidateCustomFields
 
 
-Stark


________________________________

From: rt-devel-bounces@lists.bestpractical.com
[mailto:rt-devel-bounces@lists.bestpractical.com] On Behalf Of Brandi L
Sent: Thursday, June 28, 2007 12:55 PM
To: rt-devel@lists.bestpractical.com
Subject: [Rt-devel] Code to change 'mandatory' message?


Hi All!
 
I made a couple of custom fields for my tickets. I made them mandatory.
I just want to change the message below the field that says 'Input must
match [Mandatory]' to something simpler, like 'Required.' Does anyone
know where that code can be found? I've been looking and haven't been
able to find it. 
 
Thanks!
-Brandi L
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070628/ddbf5f1a/attachment.htm
From falcone at bestpractical.com  Thu Jun 28 16:40:31 2007
From: falcone at bestpractical.com (Kevin Falcone)
Date: Thu Jun 28 19:17:14 2007
Subject: [Rt-devel] generator script
In-Reply-To: <E077297E18B24372BFEA8777AB0BD341@olabinc.com>
References: <E077297E18B24372BFEA8777AB0BD341@olabinc.com>
Message-ID: <70CCEF90-B0F2-450F-8B9F-93F3BE4B8275@bestpractical.com>


On Jun 28, 2007, at 11:28 AM, Otto Hirr wrote:
>
> I am looking for the script which 'Autogenerates' the files in lib/ 
> RT like
> User.pm. What data file is being used for its input?

you want sbin/factory in your RT distribution

> Also what script is used to generate the various schemas in etc? At  
> the top
> of schema.Pg, I see 'My2Pg 1.23 translated dump' (for version  
> rt-3.6.3).

I believe that refers to a program called My2Pg version 1.23 that
translates MySQL schemas to Pg schemas.  It has since been tweaked.

> What are the others using? or is it by hand? Has anyone tried
> SQL::Translator?

using for what?

-kevin

From alvaro.munoz.sanchez at hp.com  Fri Jun 29 12:16:16 2007
From: alvaro.munoz.sanchez at hp.com (Munoz, Alvaro)
Date: Fri Jun 29 12:16:58 2007
Subject: [Rt-devel] Redirect After creating a ticket
Message-ID: <5230D735379C53489604251BE1B6DB6CE12AD2@esmexc02.emea.cpqcorp.net>

Hi all,
 
I was wondering if it is possible to create a ticket and instead of
getting forwaded to the new ticket page, go back to the page where you
clicked "new ticket". For example, if you create a ticket from another
ticket in another queue, you shouldnt get redirected to the new ticket
page cause it is not your business ... 
 
Thanks in advance,
 
Regards,
Alvaro
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070629/ab1739da/attachment.htm
From torsten.brumm at googlemail.com  Fri Jun 29 13:41:23 2007
From: torsten.brumm at googlemail.com (Torsten Brumm)
Date: Fri Jun 29 13:41:35 2007
Subject: [Rt-devel] Re: [rt-users] RT Large Scale Performance Issues:
	Resolved
In-Reply-To: <002501c7b7b9$3288fe20$b6f3a305@plasma>
References: <002501c7b7b9$3288fe20$b6f3a305@plasma>
Message-ID: <bdbd084b0706291041t6d497dd8vcd56df5926324ff8@mail.gmail.com>

Hi Andrew,

the first part of your mail was fine, its working like a charme, but this
part:

*Use another locking scheme*

By default, it appears the user settings locking scheme (for read/write
locks) by default is done via MySQL (using the GET_LOCK() method). When a
user browses RT.

While this sounds fine, unfortunately its because there is no 'read/shared
lock' concept using GET_LOCK(), and so every lock the user makes (eg,
locking their preferences for READ during a page load) is promoted to an
exclusive write lock. This slows performance when loading multiple pages at
once as every page is competing for this exclusive lock, instead of using a
shared read lock.
I dont understand, do you have an idea how to change this like you described
it???

Thanks

Torsten

2007/6/26, Andrew Armstrong <andrew@mammoth.com.au>:
>
>  Hi Everyone,
>
>
>
> We have been making use of RT for almost a year now, however the
> performance began to quickly degrade (tickets would take 30+ seconds to open
> at times) over time.
>
>
>
> We're using RT v3.6.3
>
>
>
> After we checked both this list and other resources for a solution, we
> were unable to find a solid fix which appeared to correct the issue.
>
>
>
> We have managed to change a few core queries to our installation which
> appears to have made a massive difference (tickets load in less than one
> second most of the time).
>
>
>
> We thought others would benefit from these changes and so I am passing on
> the findings we made to fix up our RT install.
>
>
>
> *Force a JOIN order of important tables*
>
> From Nathan at mammoth:
>
> --
>
> This is lib/RT/SearchBuilder_Local.pm - its a new file which is
> automatically used by RT's extension system. We're using 3.6.3
>
>
>
> It works by changing the SQL to do a straight join of tables in a specific
> order. This appears to improve things because MySQL appears unable to choose
> the optimal join order on its own.
>
>
>
> Regards
>
> Nathan
>
> --
>
> * *
>
> This changed file is attached.
>
> * *
>
> *Use another locking scheme*
>
> By default, it appears the user settings locking scheme (for read/write
> locks) by default is done via MySQL (using the GET_LOCK() method). When a
> user browses RT.
>
> While this sounds fine, unfortunately its because there is no 'read/shared
> lock' concept using GET_LOCK(), and so every lock the user makes (eg,
> locking their preferences for READ during a page load) is promoted to an
> exclusive write lock. This slows performance when loading multiple pages at
> once as every page is competing for this exclusive lock, instead of using a
> shared read lock.
>
>
>
> Regards,
>
> Andrew
>
> _______________________________________________
> http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users
>
> Community help: http://wiki.bestpractical.com
> Commercial support: sales@bestpractical.com
>
>
> Discover RT's hidden secrets with RT Essentials from O'Reilly Media.
> Buy a copy at http://rtbook.bestpractical.com
>
>


-- 
MFG

Torsten Brumm

http://www.torsten-brumm.de
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.bestpractical.com/pipermail/rt-devel/attachments/20070629/f674f0c7/attachment.htm