Hi R programmers
I am reading multiple .xls and .xlsx files from a directory using readxl from
tidyverse. When reading fails, the code should continue on to the next file.
However, when I call the custom function readExcelSheets (in a loop and with
the tryCatch function) I get an error for some files and the code then stops
executing. How can I force my code to continue on to the next files?
Here is the function:
readExcelSheets <- function(curPath) {
out <- tryCatch(
{
message("This is the 'try' part")
dat <- excel_sheets(curPath)
},
error=function(cond) {
message(paste("Error in opening Excel file with readxl read sheets:",
curPath))
message("Here's the original error message:")
message(cond)
},
warning=function(cond) {
message(paste("readxl caused a warning en reading sheets:", curPath))
message("Here's the original warning message:")
message(cond)
},
finally={
message(paste("Processed file for sheets:", curPath))
message("End of processing file for sheets.")
}
)
return(out)
}
The loop looks like this:
listLength <- length(excelList)
for (excel_file in excelList) {
curPath <- excel_file
sheetNames <- NULL
sheetNames <- withTimeout({readExcelSheets(curPath)}, timeout = 5,
onTimeout="silent")
if(is.null(sheetNames)){next}
for (sheetName in sheetNames){
# do something
}
}
The problem is that I get an error:
Error: Evaluation error: zip file '<the path to the file>' cannot be opened.
And then execution of the loop stops without progressing to the next Excel
file. Note that for the first n=+-20 files the code works as expected. I think
that there may be an error in the full path name (such as a text encoding
error), but my point is that it should exit silently and progress to the next
Excel file even if the path is not found.
Best regards
Phillip
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
