FW: [BioC] makecdfenv()

[Lana Schaffer]

Greetings,
Can someone tell me how to use 'pipes' from R to send the output
of my process to a file?
ie. make.cdf.package(.....) | ??

Lana

-----Original Message-----
From: Laurent Gautier [mailto:lgautier at altern.org]
Sent: Wednesday, July 14, 2004 7:25 PM
To: Lana Schaffer
Subject: Re: [BioC] makecdfenv()

>>Now I can add:
>>- go to the C source of the parser, turn the debug/verbose switch on and
>>recompile/install the pack (then the parser will be _very_ verbose).
> 
> 
> I turned on _very_verbose in the C code and did the recompile/install.  Now
> the output goes to the screen, but how to put the "very verbose" output to
> a file??

what about using 'pipes' to send the output of your process to a file ?

> Lana
> 
> 
> 
> 
> -----Original Message-----
> From: Laurent Gautier [mailto:lgautier at altern.org]
> Sent: Tuesday, July 13, 2004 6:30 AM
> To: Robert Gentleman
> Cc: Lana Schaffer
> Subject: Re: [BioC] makecdfenv()
> 
> 
> Robert Gentleman wrote:
> 
>>I think you need to provide a little more detail than that.
>>
>>Which gene is skipped?
>>
>>It could have something to do with the reading of lines with comment 
>>characters in them (especially if there is only one gene skipped).
>>Or it could have to do with there being ' (single quotes) in the gene
>>names. 
>>
>>scan, read.table etc all have notions of comment.char and of what
>>quotes are (and how to interpret them),
> 
> 
> 
> Robert,
> 
> 
> I believe that Lana refers to the function read.cdffile. That function
> calls C level parsers written to parse Affymetrix genuine CDF files.
> Writting a R level level function, using itself R functions such as 
> 'scan' would have been possible (and actually was tried when I first 
> wrote the parsers, now 3 years ago). I quickly (i.e., before it left my 
> hd) moved the code to C to improve speed (CDF file can be quite large).
> 
> What Lana does not specify in her mail is that she uses a 'home-made' 
> CDF file. She already had trouble with her CDF, but apparently managed 
> to overcome this. I included few consistency checks in the parser but
> not enough to detect everything. Lana did not tell us _how_ she built 
> her custom CDF, which limits the amount of help I can provide.
> 
> As before, I would recommend her to:
> - explain what she want to achieve with a custom CDF
> - have a look at the pack makecdfenvs (note that the devel version of 
> altcdfenvs has a vignette with the Affymetrix malaria/anopheles chip, 
> and we have a documentation in a non-vignette form (submitted paper)).
> Now I can add:
> - go to the C source of the parser, turn the debug/verbose switch on and
> recompile/install the pack (then the parser will be _very_ verbose).
> 
> 
> Hopin' it helps,
> 
> 
> 
> L.
> 
> 
> 
> 
> 
> 
>>  Robert
>>
>>
>>
>>On Mon, Jul 12, 2004 at 04:40:06PM -0700, Lana Schaffer wrote:
>>
>>
>>>Hi,
>>>I am using R-1.9.1 and makecdfenv-1.4.7 and 1.4.1, but found that in reading
>>>my CDF file a gene gets skipped.  Does anyone know why this could happen?
>>>Lana
>>>
>>>_______________________________________________
>>>Bioconductor mailing list
>>>Bioconductor at stat.math.ethz.ch
>>>https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor
>>
>>
> 
>