First, I don't know what it means to say that a server is typed or untyped.
Second, I never put any programs on C: unless their installers give me
no choice. Drive C: is for the operating system, device drivers, etc. I
reserve drive D: for programs. Such granularity makes backing up easier
and safer.
Third, RExcel is in D:\Stat\R\RExcel. That is NOT R_HOME. My R_HOME is
D:\Stat\R.
Erich Neuwirth wrote:
Did you do an uninstall of the previous version of the (D)COM server?
Your problem should not have happened if you had uninstalled.
Were your servers configured as typed or as untyped?
RExcelInstaller now installs RExcel in Program Files\RExcel
One of the reasons is that it can even be used without R on the same
machine, and putting it in a subdirectory of R_HOME would not
make sense in that scenario.
Murray Eisenberg wrote:
I had downloaded (D)COM server 3.0 beta and installed it several days ago.
But I just discovered why, even with that, why things were not working:
It was the path to (D)COM! I had to execute the RExcel
Set R Server > Configure R server(s) command and enter the path to
(D)Com there.
I don't think I had seen anywhere that by default (D)COM was supposed to
be at ..\(D)COM Server\ relative to RExcel.xla. I had installed (D)COM
to one subdirectory of my main R directory (of which R-2.8.0 is a
subdirectory, too) and RExcel to another subdirectory of that directory.
I thought that was a sensible layout where I could neatly keep the two
distributions, (D)COM and RExcel, separate from one another.
I guess I had not encountered this before because, as I recall, RExcel
used to be bundled in with the (D)COM server's installer.
Erich Neuwirth wrote:
(D)COM server 2.5.0 does not work with R 2.8.0
because the inner workings of R have been changed.
We already have a beta version of (D)COM server 3.0 working with R
2.8.0, and Thomas already posted to the list where to get it.
Murray Eisenberg wrote:
The setting for the Excel R Server was Background. When I set it to
Foreground, I could indeed start R from the Check Server button in the R
Server dialog.
So what is the role of (D)COM to be? Do I have to wait until some one
or more of the various components -- (D)COM itself, rscproxy, rcom,
RExcel, perhaps even base R -- gets something fixed?
Erich Neuwirth wrote:
Check if your server is set to foreground server in Excel:
RExcel -> Set R server
In this dialog box you also can check if you can connect to the
foreground (rcom) or background ((D)COM) server.
The foreground server does not need the (D)COM server.
It needs rcom and rscproxy.
A minimalistic version of getting something that works is
Unistall any previous version of the (D)COM server from the control
panel.
Install R
Install the packages rscproxy, rcom, and RExcelInstaller
then run the following commands
library(rcom)
comRegisterRegistry()
library(RExcelInstaller)
installRExcel()
After that, RExcel should work.
Prokaj Vilmos wrote:
The latest rscproxy is 1.0-12. 1.0-11 is buggy. Wait for the compiled
binary or compile yourself from sources. I could download this morning
the source of the new version from the austrian mirror of CRAN.
Murray Eisenberg írta:
I have R 2.8.0. In its library I have the latest rcom (2.0.4) and
rscproxy (1.0.11) from CRAN (as of 15 minutes ago). I installed the
latest RExcelInstaller (3.0.5), and I still have the recently
installed DCOM from R_Scilab_DCOM3.0-1B1beta.exe
When I start R it tells me that rcom and rscproxy are required
packages that have been loaded. Just to make sure, I executed:
library(rcom)
comRegisterRegistry()
Still, when I I try "Server 01- Basic Test," I get a pop-up
"StatConnector Test" window. If I click the Start R button, I see:
Loading StatConnector Server... Done
Initializing R...Function call failed
Code: -2147221485
Text: installation problem: unable to load connector
and I get a separate "Test1" pop-up saying "Method '~' of object '~'
failed.
And still if, from Excel's RExcel menu item I select R Start, I get a
pop-up saying: Could not start RServer.
What's going on here?
------------------------------------------------------------------------
No virus found in this incoming message.
Checked by AVG - http://www.avg.com Version: 8.0.175 / Virus
Database: 270.8.5/1755 - Release Date: 10/29/2008 5:27 PM
------------------------------------------------------------------------
No virus found in this incoming message.
Checked by AVG - http://www.avg.com Version: 8.0.175 / Virus
Database: 270.8.5/1755 - Release Date: 10/29/2008 5:27 PM
------------------------------------------------------------------------
No virus found in this incoming message.
Checked by AVG - http://www.avg.com
Version: 8.0.175 / Virus Database: 270.8.5/1757 - Release Date: 10/30/2008 2:35 PM
--
Murray Eisenberg [EMAIL PROTECTED]
Mathematics & Statistics Dept.
Lederle Graduate Research Tower phone 413 549-1020 (H)
University of Massachusetts 413 545-2859 (W)
710 North Pleasant Street fax 413 545-1801
Amherst, MA 01003-9305
_______________________________________________
Rcom-l mailing list
Rcom-l@mailman.csd.univie.ac.at
http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
More information (including a Wiki) at http://rcom.univie.ac.at
