R help - Apr 2007 - Using R to create pdf's from each file in a directory

The Platform I am using R on is RHEL3.  I run a bash script that collects
data into many CSV files and have been processing them one at a time on my
local machine with an excel macro.  I would like to use R to take data
points from each of the CSV files and create line graphs in PDF format
because it will save me ALOT of time.  I am able to successfully do this
when I call the file name directly...however my script bombs when I try to
do multiple files.  I would like the created pdf's to have the same filename
as the original csv files.  I have looked quite a bit and not found much
help on "batch processing" an entire directory.  My current code is as
follows:

list <- dir("/tmp/data")
for(x in list){
d <- read.table(x, sep="\t", header=TRUE) # read data
pdf("/tmp/graph/x.pdf")                              # file for graph
plot(d$BlockSeqNum, d$MBs,                              # Blocks as x, MB/s
as y     
     type="l",                                     # plot lines, not
points
     xlab="Blocks",                                  # label x axis
     ylab="MB/s",                                  # label y axis
     main=x)          # add title
dev.off()                                          # close file
q()                                                # quit

ERROR: Error in plot.window(xlim, ylim, log, asp, ...) :
        need finite 'xlim' values
In addition: Warning messages:
1: no non-missing arguments to min; returning Inf
2: no non-missing arguments to max; returning -Inf
3: no non-missing arguments to min; returning Inf
4: no non-missing arguments to max; returning -Inf
Execution halted

Thank you for your wisdom!
-- 
View this message in context:
http://www.nabble.com/Using-R-to-create-pdf%27s-from-each-file-in-a-directory-tf3621434.html#a10112832
Sent from the R help mailing list archive at Nabble.com.

On 4/20/2007 9:40 PM, gecko951 wrote:
> The Platform I am using R on is RHEL3.  I run a bash script that collects
> data into many CSV files and have been processing them one at a time on my
> local machine with an excel macro.  I would like to use R to take data
> points from each of the CSV files and create line graphs in PDF format
> because it will save me ALOT of time.  I am able to successfully do this
> when I call the file name directly...however my script bombs when I try to
> do multiple files.  I would like the created pdf's to have the same
filename
> as the original csv files.  I have looked quite a bit and not found much
> help on "batch processing" an entire directory.  My current code
is as
> follows:
> 
> list <- dir("/tmp/data")
> for(x in list){
> d <- read.table(x, sep="\t", header=TRUE) # read data
> pdf("/tmp/graph/x.pdf")                              # file for
graph
> plot(d$BlockSeqNum, d$MBs,                              # Blocks as x, MB/s
> as y     
>      type="l",                                     # plot lines,
not points
>      xlab="Blocks",                                  # label x
axis
>      ylab="MB/s",                                  # label y axis
>      main=x)          # add title
> dev.off()                                          # close file
> q()                                                # quit
> 
> ERROR: Error in plot.window(xlim, ylim, log, asp, ...) :
>         need finite 'xlim' values
> In addition: Warning messages:
> 1: no non-missing arguments to min; returning Inf
> 2: no non-missing arguments to max; returning -Inf
> 3: no non-missing arguments to min; returning Inf
> 4: no non-missing arguments to max; returning -Inf
What you're doing appears to be a reasonable approach; all you need are 
some checks that the data is reasonable before calling plot and possibly 
bombing.

I'd suggest adding a "print(x); print(str(d))" statements after
you read
d; then you'll be able to see when it is that sometimes d doesn't 
contain plotable data.

One other problem you'll find:  you're writing all plots to
"x.pdf".  I
think you really want to construct that name differently for each x. 
I'd suggest using paste() and basename() (and perhaps some regular 
expressions and gsub()) to construct the output filename.

Duncan Murdoch

At 09:40 PM 4/20/2007, gecko951 wrote:
><snip>
>list <- dir("/tmp/data")
>for(x in list){
>d <- read.table(x, sep="\t", header=TRUE) # read data
>pdf("/tmp/graph/x.pdf")                              # file for
graph
><snip>
I'm a tyro at R, but it's obvious here that the line

pdf("/tmp/graph/x.pdf")

has the intended file name string 'x' embedded with the literal 
string enclosed in quotation marks. Obviously he needs "/tmp/graph/" 
string concatenated with x and then ".pdf".

I would suggest a fix, but I am unable to find in the documentation 
how to concatenate two strings into a single string.

So I will amplify gecko951's question to include "How you you 
concatenate two strings in R?". I.e.,

x<-"abc"
y<-"def"

What operator or function returns "abcdef" as a result?

Thanks.

===============================================================Robert A.
LaBudde, PhD, PAS, Dpl. ACAFS  e-mail: ral at lcfltd.com
Least Cost Formulations, Ltd.            URL: http://lcfltd.com/
824 Timberlake Drive                     Tel: 757-467-0954
Virginia Beach, VA 23464-3239            Fax: 757-467-2947

"Vere scire est per causas scire"

gecko951 wrote:
> The Platform I am using R on is RHEL3.  I run a bash script that collects
> data into many CSV files and have been processing them one at a time on my
> local machine with an excel macro.  I would like to use R to take data
> points from each of the CSV files and create line graphs in PDF format
> because it will save me ALOT of time.  I am able to successfully do this
> when I call the file name directly...however my script bombs when I try to
> do multiple files.  I would like the created pdf's to have the same
filename
> as the original csv files.  I have looked quite a bit and not found much
> help on "batch processing" an entire directory.  My current code
is as
> follows:
> 
> list <- dir("/tmp/data")
> for(x in list){
> d <- read.table(x, sep="\t", header=TRUE) # read data
> pdf("/tmp/graph/x.pdf")                              # file for
graph
> plot(d$BlockSeqNum, d$MBs,                              # Blocks as x, MB/s
> as y     
>      type="l",                                     # plot lines,
not points
>      xlab="Blocks",                                  # label x
axis
>      ylab="MB/s",                                  # label y axis
>      main=x)          # add title
> dev.off()                                          # close file
> q()                                                # quit
Below will get you closer to what you want, assuming that your files end 
in .csv, which they should, especially if you'll be creating new files 
in the same directory with a different extension. You certainly don't 
want to re-run your R code and call read.table on a pdf. Another point 
is that your current working directory for R, returned by getwd(), is 
already '/tmp/data'. Otherwise read.table wouldn't work, and a more 
portable solution is to use a variable to hold the directory name:

workdir <- '/tmp/data'
for (x in dir(workdir,pattern='.csv$')){
   d <- read.table(paste(workdir,'/',x,sep=''),
sep="\t", header=TRUE)
  
pdf(paste(workdir,'/',sub('.csv$','.pdf',x),sep=''))
   plot(d$BlockSeqNum, d$MBs,
     type="l",
     xlab="Blocks",
     ylab="MB/s",
     main=x)
   dev.off()
}
q()

Best,

Jeff
---
http://biostat.mc.vanderbilt.edu/JeffreyHorner