I have several list objects, each containing 31 dataframes, which I have names 'file1980 through to 'file2010'. These were made by splitting the original (11315 rows) data frame into 31 equal sized (365 rows) data frames using the following:
n <- 31
dataList <- split(MainData, factor(sort(rank(row.names(MainData))%%n)))
names(dataList) <- paste0("file",1980:2010)
The individual data frames look like this:
jDate V1 V2 V3 V4 V5 V6 V7
1 001 -6.83 -5.83 -7.83 0.05 0.8217593 8.101852 100.0
2 002 -6.33 -4.83 -7.83 0.10 2.2453704 9.259259 100.0
3 003 -5.83 -4.83 -6.83 0.30 1.9444444 8.101852 94.7
4 004 -5.83 -4.83 -6.83 0.10 1.0416667 8.101852 97.5
5 005 -6.33 -4.83 -7.83 0.00 1.1226852 9.259259 98.5
6 006 -7.83 -5.83 -9.83 0.03 2.0949074 10.416667 100.0
They will be exported with row names removed into *.txt files for use in another piece of software. However, this software starts by reading the first row (which is the column names incsv or txt formats), as the file name, but for the software to run the first row needs to be the file name, so 'file1980' and so on.
I'm hoping to split the list into 31 equally sized files that look like this (with sequential file names, file1980 to file 2010, in the top row):
file1980
1 001 -6.83 -5.83 -7.83 0.05 0.8217593 8.101852 100.0
2 002 -6.33 -4.83 -7.83 0.10 2.2453704 9.259259 100.0
3 003 -5.83 -4.83 -6.83 0.30 1.9444444 8.101852 94.7
4 004 -5.83 -4.83 -6.83 0.10 1.0416667 8.101852 97.5
5 005 -6.33 -4.83 -7.83 0.00 1.1226852 9.259259 98.5
6 006 -7.83 -5.83 -9.83 0.03 2.0949074 10.416667 100.0
7 007 -5.33 -4.83 -5.83 0.00 1.4930556 8.101852 97.6
8 008 -7.33 -5.83 -8.83 0.00 0.9027778 9.259259 100.0
9 009 -7.33 -6.83 -7.83 0.03 0.8217593 8.101852 90.2
I have now seen two methods which add a new row1 with the file name, but the results here both failed to replace/remove the column names, or replaced them with NA. ON an individual file it is easy to just use names(main file) <- NULL, but this only solves half the problem.
The resulting files all need to be sent along a preset filepath.
I believe it's better not to edit the data frame to do this sort of thing, better first print the required string first to the file, then append the data.
write.table() is quite flexible luckily :
# data prep, taking 2 chunks of iris
MainData <- head(iris, 10)
# optional if you want columns with equal width in the csv :
MainData[] <- lapply(MainData, format)
n <- 2
dataList <- split(
MainData,
factor(sort(rank(row.names(MainData_formatted))%%n))
)
names(dataList) <- paste0("file",seq(n))
dataList
# print to file
for (file in names(dataList)) {
# define the path, here in the working directory
path <- paste0(file, ".csv")
# print(path) # uncomment to see the path
# print string to file
writeLines(file, path)
# append the data, without headers, quotes or row names
write.table(
dataList[[file]], path,
col.names = FALSE, row.names = FALSE, quote = FALSE, append = TRUE
)
}
# in file1.csv :
# file1
# 5.1 3.5 1.4 0.2 setosa
# 4.9 3.0 1.4 0.2 setosa
# 4.7 3.2 1.3 0.2 setosa
# 4.6 3.1 1.5 0.2 setosa
# 5.0 3.6 1.4 0.2 setosa
# cleanup
file.remove(c("file1.csv", "file2.csv"))
write.table() uses " " as a separator, you might use sep = "\t" if you want it tab delimited