Transfer text file to table in R with some conditions on it - r

Question

I have one text file like this
DOB
Name
Address
13-03-2003
ABC
xyz.
12-08-2004
dfs
1 infinite loop.
text goes till approx 300 lines. And sometimes Address data exceeds to two second line also i want to convert this text data to either cvs format which will have data like this
DOB, Name, Address
13-03-2003,ABC,xyz.
or at least in one data frame. I tried so many things, when i am giving read.table("file.txt",sep="\n") it makes everything in one column and i also tried first making headers by using
header <- read.table("file.txt",sep= "\n")
and then another data <- read.table("file.txt",skip = 3, sep ="\n") and then combining both but its not working out as my header vector has 3 and data vector has like 300 approx columns, its not working as required. Any help will be really helpful :)

Answer 1

You could try
entries <- unlist(strsplit(text, "\\n")) #separate entries by line breaks
entries <- entries[nchar(entries) > 0] #remove empty lines
as.data.frame(matrix(entries, ncol=3, byrow=TRUE)) #assemble dataframe
# V1 V2 V3
#1 DOB Name Address
#2 13-03-2003 ABC xyz.
#3 12-08-2004 dfs 1 infinite loop.
data
text <-'DOB
Name
Address
13-03-2003
ABC
xyz.
12-08-2004
dfs
1 infinite loop.'
df <- read.table(text = text)

Answer 2

Two assumptions were made, 1 there will not be any blank names or date of births. By "blank" I do not mean "NA", "", or any other marker that the value was missing. Second assumption was that names and DOBs will only occupy one line each.
s1 <- gsub("^\n|\n$", "", strsplit(x, "\n\n+")[[1]])
stars <- gsub("\n", ", ", sub("\n", "*", sub("\n", "*", s1)))
mat <- t(as.data.frame(strsplit(stars, "\\*")))
dimnames(mat) <- c(NULL, NULL)
write.csv(mat,"filename.csv")
We start by splitting the text by the blank lines and eliminating any leading or trailing newline tokens. Then we replace the first and second "\n" symbols with stars. Next we split on those new star markers that we created to always have 3 elements for each row. We create a matrix with the values and transpose it for display. Then write the data to csv.
When opened with Notepad on a test file, I get:
"","V1","V2","V3"
"1","DOB","Name","Address"
"2","13-03-2003","ABC","xyz."
"3","12-08-2004","dfs","1 infinite loop"
"4","01-01-2000","Bob Smith","1234 Main St, Suite 400"
row and column names can be set to FALSE with ?write.csv if desired.
Data
x <- "DOB
Name
Address
13-03-2003
ABC
xyz.
12-08-2004
dfs
1 infinite loop
01-01-2000
Bob Smith
1234 Main St
Suite 400
"