Then you have to establish a reference anyway.
The alternatove to the Auto_Open macro is to use
Tools->References
in the VBA IDE.




Lindsay, Dwayne wrote:
> Sorry, let me provide further information:
> 
> I strictly start R via VBA programming.  As such, I have experienced a
> failure in either R or Excel, (not sure which side this is on), to load
> the appropriate references.  It's no big deal, but has caused to time to
> troubleshoot the issue.
> 
> I hope this clarifies my previous note.
> D-
> 
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Erich
> Neuwirth
> Sent: Wednesday, November 14, 2007 12:30 PM
> To: R (D)COM and RExcel server related issues
> Subject: Re: [Rcom-l] Resources for calling Excel from R
> 
> Does that mean that starting R from the RExcel menu
> and then using Get R Value or Get R dataframe does not work?
> 
> The reference you introduce should not be necessary if you do as
> described above. If you do it this way and it does not work, please give
> more information.
> 
> 
> Lindsay, Dwayne wrote:
>> Dear all-
>>
>> With recent upgrades/version of R, I have found that most of my
> programs
>> now fail to execute because the proper libraries/references fail to
>> load.  As such, I have had to use the following VBA code in the:
>> Auto_Open(), macro to load the appropriate libraries/references: 
>>
>> 'Auto add Rexcel.xla references
>>      strAdd = "C:\Program Files\R\(D)COM Server\xls\RExcel.xla"
>>
>>      On Error GoTo 111
>>              SendKeys "{ENTER}"
>>              ThisWorkbook.VBProject.References.AddFromFile strAdd
>>         
>> 111 Err.Number = 0 
>>
>> Especially, with R 2.6.0, I now have 5-8 libraries to load just to get
>> my Excel based programs to work.  I try to avoid VBA programming like
>> the plaque, so I tend to get everything in to R then, use R as a
> client
>> to obtain my results.  In this manner, I find myself doing a lot less
>> maintenance on my programs.
>>
>> Hope this helps someone out there.
>> Dwayne
>>
>> -----Original Message-----
>> From: [EMAIL PROTECTED]
>> [mailto:[EMAIL PROTECTED] On Behalf Of
> Charilaos
>> Skiadas
>> Sent: Wednesday, November 14, 2007 10:50 AM
>> To: R ((D))COM and RExcel server related issues
>> Subject: Re: [Rcom-l] Resources for calling Excel from R
>>
>> On Nov 14, 2007, at 9:50 AM, Prokaj Vilmos wrote:
>>
>>> I resend this message. This was sent previously in reply to another  
>>> message.
>>>
>>> Hi,
>>>
>>> I use rcom in a similar way as you plan to do. I usually start with  
>>> Visual
>>> Basic of Excel. Then I simply translate the VBA code to R code (macro
>>> recorder could help a lot, and for the contants there is the built in
>>> object browser). If you want to get all data from a spreadsheet the
>>> usedrange property could help you. Example
>>>
>>> XL<-comGetObject("Excel.application")
>>> r<-XL[["activesheet"]][["usedrange"]][["value"]]
>>>
>>> gives a matrix of the values from the activesheet.
>>> Sometimes value2 property is also useful
>>> I hope this helps
>> Thanks Prokaj, this was indeed the connecting link I was looking for.
> 
>> I had not realized that VBA and COM were essentially different sides  
>> of the same coin. So this means I now have to learn a bit of VBA,  
>> which I have always avoided (never had to use Word/Excel, and not a  
>> Windows user for more than 2 years), and I need to locate the library
> 
>> reference for VBA functions (which so far has seemed far from  
>> trivial; any links for a "VBA newbie but long time programmer" would  
>> be welcome).
>>
>> One further question, is the "bracketing" approach recommended, over  
>> the comGetProperty approach? I.e. any advantages over one or the  
>> other of:
>>
>> comGetProperty(comGetProperty(ws,"UsedRange"),"Value")
>> ws[["UsedRange"]][["Value"]]
>>
>> I suppose I don't quite see what the syntax would have been for  
>> getting the equivalent of:
>>
>> comGetProperty(ws,"Range","A1","B4")
>>
>> but it seems that this works:
>>
>> ws[["Range","A1","B4"]]
>>
>> Ok, I just read the documentation for [[.COMObject, and it now makes  
>> a bit more sense. Isn't that a somewhat unusual use of [[ though?  
>> Merely worrying for aesthetic reasons.
>>
>>> Vilmos Prokaj
>>>
>> Haris Skiadas
>> Department of Mathematics and Computer Science
>> Hanover College
>>
>>
>>
>>
>> _______________________________________________
>> Rcom-l mailing list
>> [email protected]
>> http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
>> More information (including a Wiki) at http://rcom.univie.ac.at
>> _______________________________________________
>> Rcom-l mailing list
>> [email protected]
>> http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
>> More information (including a Wiki) at http://rcom.univie.ac.at
>>
>>
> 
> _______________________________________________
> Rcom-l mailing list
> [email protected]
> http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
> More information (including a Wiki) at http://rcom.univie.ac.at
> _______________________________________________
> Rcom-l mailing list
> [email protected]
> http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
> More information (including a Wiki) at http://rcom.univie.ac.at
> 
> 

_______________________________________________
Rcom-l mailing list
[email protected]
http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
More information (including a Wiki) at http://rcom.univie.ac.at

Reply via email to