R: Row numbers unmatched for Sentence to word-table

Question

From my previous problem, I have some texts in different rows, and from the text I am trying to generating word-table for each of the words. But problem is occurring when the row number of the text column, and row number of the word-table unlike. It has been found for some text, two or more rows are being created. So finally I cannot cbind these two together. Code is here. I just want the outcome will be exactly same row number of the text that I can bind them together to show for which text is which word-table.

texts <- c("concratulations    successfully signed   company  please find attached quick guide   can  t figure   immediately ", " conversation   laughing services  sweden", "p please find attached drafted budget   p ", "p please finad attached  agenda  today s board meeting  p ", "p hi   nbsp   p    p please find attached  darft meeting minutes  today s meeting   p ", "p please find attached  final version   minutes  updated action log  please let  know  actions   done   ll update  excel  nbsp   p ", "p hi    p    p please find attached  draft meeting minutes   action log  please provide comments   end  next week   p    p   nice spring party  saturday    p    p   tuija  p ", " p welcome team priority   hope   enjoy yo  p ", "p please find attached  flyer   can study  share   p ", "p attached new version  voice   receiver   p    p minor change request  invitation code       mentioned    invitation code may       tell  check  code  invitation email    end    alarm bell  example telling  new comments   ", "comment  etc     front page  now  seemed  end without warning    p ", "p memo attached  actions    p ", "p please find attached  updated board roles  responsibilities    made  changes   red   document   please review  especially   role  relevant contact info   prepare  comment   meeting  wednesday  nbsp   p ", "p attached documents  review  please comment  soonest   p ")
texts <- cbind(texts)
## to remove multi-white spaces
MyDf <- gsub("\\s+"," ",texts)
MyDf <- gsub("\r?\n|\r", " ", MyDf)
MyDf <- cbind(MyDf)
colnames(MyDf) <- c("Introduction")

## this way, extra rows are being generated
word_table <- read.table(text = paste(gsub('\n', ' ', MyDf), collapse = '\n'), fill = TRUE)

## this way, the words are being repeated to match with the largest text
word_table <- do.call(rbind, strsplit(as.character(MyDf), " "))

More details: the texts had multiple whitespaces, or tab. Initial assumption was, may be that additional spaces creating the problem, but after removing the additional white spaces, still it is in the same problem.

Please Help

score 0 · Accepted Answer · answered Mar 10 '16 at 20:24

Solution: honorable mention for breaker and cSplit function.

texts <- c("concratulations    successfully signed   company  please find attached quick guide   can  t figure   immediately ", " conversation   laughing services  sweden", "p please find attached drafted budget   p ", "p please finad attached  agenda  today s board meeting  p ", "p hi   nbsp   p    p please find attached  darft meeting minutes  today s meeting   p ", "p please find attached  final version   minutes  updated action log  please let  know  actions   done   ll update  excel  nbsp   p ", "p hi    p    p please find attached  draft meeting minutes   action log  please provide comments   end  next week   p    p   nice spring party  saturday    p    p   tuija  p ", " p welcome team priority   hope   enjoy yo  p ", "p please find attached  flyer   can study  share   p ", "p attached new version  voice   receiver   p    p minor change request  invitation code       mentioned    invitation code may       tell  check  code  invitation email    end    alarm bell  example telling  new comments   ", "comment  etc     front page  now  seemed  end without warning    p ", "p memo attached  actions    p ", "p please find attached  updated board roles  responsibilities    made  changes   red   document   please review  especially   role  relevant contact info   prepare  comment   meeting  wednesday  nbsp   p ", "p attached documents  review  please comment  soonest   p ")
texts <- cbind(texts)
## to remove multi-white spaces
MyDf <- gsub("\\s+"," ",texts)
MyDf <- gsub("\r?\n|\r", " ", MyDf)
MyDf <- cbind(MyDf)
colnames(MyDf) <- c("Introduction")
n <- matrix(texts[ ,1 ], nrow = nrow(texts), ncol = ncol(texts))
library(splitstackshape)
library(data.table)
breaker <- function(X) {
        strsplit(X, "[[:space:]]|(?=[.!?])", perl=TRUE)
}
aaa <- breaker(n)
aaa <- cbind(aaa)
#############################################################################################
cSplit <- function(indt, splitCols, sep = ",", direction = "wide", 
                   makeEqual = NULL, fixed = TRUE, drop = TRUE, 
                   stripWhite = FALSE) {
        message("`cSplit` is now part of the 'splitstackshape' package (V1.4.0)")
        ## requires data.table >= 1.8.11
        require(data.table)
        if (!is.data.table(indt)) setDT(indt)
        if (is.numeric(splitCols)) splitCols <- names(indt)[splitCols]
        if (any(!vapply(indt[, splitCols, with = FALSE],
                        is.character, logical(1L)))) {
                indt[, eval(splitCols) := lapply(.SD, as.character),
                     .SDcols = splitCols]
        }

        if (length(sep) == 1) 
                sep <- rep(sep, length(splitCols))
        if (length(sep) != length(splitCols)) {
                stop("Verify you have entered the correct number of sep")
        }

        if (isTRUE(stripWhite)) {
                indt[, eval(splitCols) := mapply(function(x, y) 
                        gsub(sprintf("\\s+%s\\s+|\\s+%s|%s\\s+", 
                                     x, x, x), x, y), 
                        sep, indt[, splitCols, with = FALSE], 
                        SIMPLIFY = FALSE)]
        }  

        X <- lapply(seq_along(splitCols), function(x) {
                strsplit(indt[[splitCols[x]]], split = sep[x], fixed = fixed)
        })

        if (direction == "long") {
                if (is.null(makeEqual)) {
                        IV <- function(x,y) if (identical(x,y)) TRUE else FALSE
                        makeEqual <- ifelse(Reduce(IV, rapply(X, length, how = "list")),
                                            FALSE, TRUE)
                }
        } else if (direction == "wide") {
                if (!is.null(makeEqual)) {
                        if (!isTRUE(makeEqual)) {
                                message("makeEqual specified as FALSE but set to TRUE")
                                makeEqual <- TRUE
                        }
                        makeEqual <- TRUE
                } else {
                        makeEqual <- TRUE
                }
        }
        if (isTRUE(makeEqual)) {
                SetUp <- lapply(seq_along(X), function(y) {
                        A <- vapply(X[[y]], length, 1L)
                        list(Mat = cbind(rep(seq_along(A), A), sequence(A)),
                             Val = unlist(X[[y]]))
                })    
                Ncol <- max(unlist(lapply(SetUp, function(y) y[["Mat"]][, 2]), 
                                   use.names = FALSE))
                X <- lapply(seq_along(SetUp), function(y) {
                        M <- matrix(NA_character_, nrow = nrow(indt), ncol = Ncol)
                        M[SetUp[[y]][["Mat"]]] <- SetUp[[y]][["Val"]]
                        M
                })
                if (direction == "wide") {
                        X <- lapply(seq_along(X), function(x) {
                                colnames(X[[x]]) <- paste(splitCols[x], 
                                                          sequence(ncol(X[[x]])), 
                                                          sep = "_")
                                X[[x]]
                        })
                        if (isTRUE(drop)) {
                                cbind(indt, do.call(cbind, X))[, eval(splitCols) := NULL][]
                        } else {
                                cbind(indt, do.call(cbind, X))
                        }
                } else {
                        indt <- indt[rep(sequence(nrow(indt)), each = Ncol)]
                        X <- lapply(X, function(y) as.vector(t(y)))
                        indt[, eval(splitCols) := lapply(X, unlist, use.names = FALSE)][]
                }  
        } else {
                Rep <- vapply(X[[1]], length, integer(1L))
                indt <- indt[rep(sequence(nrow(indt)), Rep)]
                indt[, eval(splitCols) := lapply(X, unlist, use.names = FALSE)][]
        }
}

df <- cSplit(as.data.frame(aaa), "aaa", ",")
df <- data.frame(cbind(texts, df))

######################################################################################
## Heading
Heading <- df[ ,1]

## Word Table
df <- df[ ,2:ncol(df)]

## first column
aaa_first <- df[,1]
aaa_first <- cbind(aaa_first)
c <- substring(aaa_first, 3)

## last column
aaa_end <- df[ ,ncol(df)]
aaa_end <- cbind(aaa_end)
e <- substr(aaa_end, 1, nchar(aaa_end)-1)

## Middole columns
d <- df[ ,3:ncol(df)-1]

cc <- cbind(Heading, c, d, e )
## cc <- cbind( c, d, e )

cc <- data.frame(lapply(cc, as.character), stringsAsFactors = FALSE)

df2 <- as.data.frame(sapply(cc,gsub,pattern= ")",replacement=""))
# df2 <- as.data.frame(sapply(df2,gsub,pattern="(",replacement=""))
df3 <- as.data.frame(sapply(df2, function(x) gsub("\"", "", x)))

R: Row numbers unmatched for Sentence to word-table

1 Answers1