Skip to content

How to parse text file into a table?

3 messages · Daren Tan, Ben Bolker, jim holtman

Daren Tan
# 
I am given a text file of records to be converted into a table format.
I have searched related topics or packages, but can't find any similar
cases. Please help.

Sample record is given below. Take note the last element doesn't have
a semi colon.

###---------Start of record----------------------

Name : John
Height: 170cm
Weight  : 70kg
Age: 30

Status: Married
Children: 2

Employment  Engineer

###---------End of record-----------------------

Table format should have this header
Name	Height	Weight	Age	Status	Children	Employment
Ben Bolker
# 
Daren Tan <darentan76 <at> gmail.com> writes:
You can put something together with
readLines, grep (to find the appropriate lines),
gsub (to strip off the first field) ... if you need
an explicit example, ask again ...

  Ben Bolker
jim holtman
# 
Here is one way of reading in your data:
+
+ Name: John
+ Height: 170cm
+ Weight: 70kg
+ Age: 30
+
+ Status: Married
+ Children: 2
+
+ Employment:  Engineer
+
+ ###---------End of record-----------------------"))
+     if (nchar(i) == 0) next
+     if (length(grep("Start of record", i)) != 0){
+         # initialize next element of the list
+         result[[recordNo]] <- c(Name=NA, Height=NA, Weight=NA,
+             Age=NA, Status=NA, Children=NA)
+
+     }
+     else if (length(grep("End of record", i)) != 0) recordNo <- recordNo + 1
+     else {
+         # follow assumes you have consistent naming with ":" terminating data
+         # if not, add some error checking code
+         name <- sub("^(.*):.*", "\\1", i)
+         value <- sub(".*:\\s*(.*)", "\\1", i)
+         result[[recordNo]][name] <- value
+     }
+ }
[[1]]
      Name     Height     Weight        Age     Status   Children Employment
    "John"    "170cm"     "70kg"       "30"  "Married"        "2" "Engineer"

        
On Sun, Feb 22, 2009 at 8:45 AM, Daren Tan <darentan76 at gmail.com> wrote: