I am a R user and I have a plain text file which looks like this:

t # 0 * 3
v 0 4
v 1 7
e 0 1 1
x 0 1 2

t # 1 * 2
v 0 4
v 1 7
v 2 10
e 0 1 1
e 1 2 1
x 0 1

t # 2 * 2
v 0 7
v 1 10
e 0 1 1
x 0 2

I need to convert it into three data frames which looks like this:

# First data frame    
object id number
0 0 4
0 1 7
1 0 4
1 1 7
1 2 10
2 0 7
2 1 10

# Second data frame    
object from to number
0 0 1 1
1 0 1 1
1 1 2 1
2 0 1 1

# Third data frame
x t0 t1 t2
0 1 1 1
1 1 1 0
2 1 0 1

Correspondences are illustrated below:

• The first data frame contains information of v in the original file
• The second data frame contains information of e in the original file
• The third data frame contains information of x in the original file
• t is a identifier of each object in new data frames
• In the original text file, we have object t0, t1 and t2, which are indicated in t1, t2 and t3 field in the third new data frame. In the original text file, numbers in x row, i.e., x 0 1 2, indicates that object t0 appears in other set of objects called x 0, 1 and 2. Set of x (0, 1, 2) is given preliminarily.

I appreciate your suggestions to read the original text file and convert it into desired format. I could read text file using some functions such as readLines("sample.txt") and scan("sample.txt", what=character(), sep="\n", blank.lines.skip=T), however I have no knowledge to deal with these text data into data frames.

CodePudding user response：

Somehow I could make it using the following inefficient scripts. Other solutions are appreciated.

#### Load text file
result <- scan("result.txt", what=character(), sep="\n", blank.lines.skip=T)

## Line number of subgraphs
nsub <- grep("t", result)

#### Initial data frame
## Node and edge
node.s <- data.frame(matrix(ncol=3, nrow=0, dimnames=list(NULL, c("object", "id", "func"))))
edge.s <- data.frame(matrix(ncol=4, nrow=0, dimnames=list(NULL, c("object", "from", "to", "connect"))))

## Equivalence table of original graphs and subgraphs
eq.s.1 <- data.frame(t=c(0:(nrow(g)-1)))
eq.s.2 <- data.frame(matrix(ncol=length(nsub), nrow=nrow(g), dimnames=list(NULL, c(paste0("sub",c(0:(length(nsub)-1)))))))
eq.s <- cbind(eq.s.1, eq.s.2)

#### Loop to update data frames
for (i in 1:length(result)) {

  ## Store t number for other data frames
  if (grepl("t", result[i])==1) {  
    t <- word(result[i],3)
  }
  
  #### Update node list (node.s)
  if (grepl("v", result[i])==1) {
    ## Extract node
    tent.v <- data.frame(object=t, id=word(result[i],2), fun=word(result[i],3))
    ## Append
    node.s <- rbind(node.s, tent.v)
  }
  
  #### Update edge list (edge.s)
  if (grepl("e", result[i])==1) {
    ## Extract edge
    tent.e <- data.frame(object=t, from=word(result[i],2), to=word(result[i],3), connect=word(result[i],4))
    ## Append
    edge.s <- rbind(edge.s, tent.e)
  }
  
  #### Update equivalent table (eq.s)
  if (grepl("x", result[i])==1) {
    ## Equivalence between original graphs and subgraphs
    tent.e <- data.frame(t=as.integer(unlist(strsplit(result[i], " "))[-1]), a=1)
    ## Update field name
    names(tent.e) <- c("t",paste0("sub",t))
    ## Overwrite join
    eq.s <- merge_dfs_overwrite_col(eq.s, tent.e, cols=names(tent.e)[2], bycol="t")
  }
}

CodePudding user response：

You could do the following:

dat <- readLines('file.txt')
f <- cumsum(grepl('^t', dat))
dat1 <- paste(dat, f - 1)
df1 <- read.table(text = grep('^v', dat1, value = TRUE))[c(4,2:3)]
df1

  V4 V2 V3
1  0  0  4
2  0  1  7
3  1  0  4
4  1  1  7
5  1  2 10
6  2  0  7
7  2  1 10

df2 <- read.table(text = grep('^e', dat1, value = TRUE))[c(5,2:4)]
df2
  V5 V2 V3 V4
1  0  0  1  1
2  1  0  1  1
3  1  1  2  1
4  2  0  1  1

x <- read.table(text = grep('^x', dat, value = TRUE), fill=T)[-1]
df3 <- data.frame(cbind(x=0:2, apply(x, 1, \(y) 0:2 %in% y)))
df3

  x V2 V3 V4
1 0  1  1  1
2 1  1  1  0
3 2  1  0  1