Hi Dave,

Thanks for explaining your thoughts on this.  I think I agree.  In your
case, you are doing something a bit unexpected - making excel take an
in-proc control over managed code.  At some point, this won't be
un-conventional - after such point when Excel has been redone in managed
code, for instance.

I wish I had an easy answer for you.  You probably have discovered that
using reflection may let you do the over-ride on a new instance of the
class - but it doesn't update the one that is being used by the CLR to
manage the application.  So the hacking concerns there aren't as serious.

Do you really need to set this on a per application basis?  Why?  Wouldn't
it be better to handle time-differences (if you need message time-out) at
the application level?

Dan
--------------------
Thread-Topic: Need for programmatic config:timeToleranceInSeconds, MaxReques
thread-index: AcTdSpMp89gsN2HJSM2k0qyloyGq7Q==
X-WBNR-Posting-Host: 204.210.55.215
From: "=?Utf-8?B?ZHJrcmF1cw==?=" <drkraus@nospam.nospam>
References:  <30AB465D-A703-4943-B259-C588C7CAE6BC@microsoft.com>
<EnnNGDM3EHA.3984@cpmsftngxa10.phx.gbl>
Subject: RE: Need for programmatic config:timeToleranceInSeconds, MaxReques
Date: Wed, 8 Dec 2004 09:23:03 -0800
Lines: 109
Message-ID: <1244B818-C97A-4AC0-B436-18A6C51C7AB9@microsoft.com>
MIME-Version: 1.0
Content-Type: text/plain;
    charset="Utf-8"
Content-Transfer-Encoding: 7bit
X-Newsreader: Microsoft CDO for Windows 2000
Content-Class: urn:content-classes:message
Importance: normal
Priority: normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.3790.0
Newsgroups: microsoft.public.dotnet.framework.webservices.enhancements
NNTP-Posting-Host: TK2MSFTNGXA03.phx.gbl 10.40.1.29
Path:
cpmsftngxa10.phx.gbl!TK2MSFTFEED01.phx.gbl!TK2MSFTNGP08.phx.gbl!TK2MSFTNGXA0
3.phx.gbl
Xref: cpmsftngxa10.phx.gbl
microsoft.public.dotnet.framework.webservices.enhancements:5086
X-Tomcat-NG: microsoft.public.dotnet.framework.webservices.enhancements

Hi Dan,

Well, it did bother me a bit that I could override a read-only property
using reflection, but only in principle.

A little more background:
In reality, I would have been satisfied if I could have created an
application config file for our client DLL, that is invoked as an add-in to
Excel. Our client group attempted this without success. They ended up
creating a config file (ie. appname.dll.config) that they specifically
query
to pick up their application settings. Any non-application settings that
are
present are ignored at DLL load time, and would have to be programmtically
set as I have described. The next alternative was to create an
Excel.exe.config file, which is unacceptable.

I have been assuming  that when a setting like timeToleranceInSeconds is
configured in a app.config file, that this setting is specific to that
application only. The same goes for any programmatic setting of the same
property by a client. If this is true, I am not sure where the concern
lies.
We are just talking about how the property is set. Now, if it is easier to
compromise/spoof an application program, than it is to compromise/spoof a
config file, then I might see a problem. This would then lead me to believe
that I may have indeed "hacked into" the MaxRequestLengthInBytes property
using reflection.

Finally, even though there is only one of our clients per box, a per-site
config setting would not allow other non-related applications to use other
values.

thanks, Dave

"Dan Rogers" wrote:

(Microsoft.Web.Services2.Configuration.WebServicesConfiguration.MessagingCon