Text Line Count in R [closed] - r

Question

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, visit the help center.
Closed 9 years ago.
I'm doing some basic text analysis in R and want to count the number of lines for a transcript from a .txt file that I load into R. With the example below to yield a count in which each speaker gets a new line attached to the linecount such that Mr. Smith = 4, Mr. Gordon = 6, Mr. Catalano = 3.
[71] "\"511\"\t\"MR Smith: Mr Speaker, I like the spirit in which we are agreeing on this. The administration of FUFA is present here. FUFA could be used as a conduit, but the intention of what hon. Beti Kamya brought up and what hon. Rose Namayanja has said was okufuwa - just giving a token of appreciation to the players who achieved this.\""
[72] "\"513\"\t\"MR Gordon: Thank you very much, Mr Speaker. FUFA is an organisation and the players are the ones who got the cup for us. To promote motivation in all activities, not only football, you should remunerate people who have done well. In this case, we have heard about FUFA with their problems. They have not paid water bills and they can take this money to pay the water bills. If we agree that this money is supposed to go to the players and the coaches, then when it goes there they would know the amount and they will sit among themselves and distribute according to what we will have given. (Applause) I thank you.\""
[73] "\"515\"\t\"MR Catalano: Mr Speaker, I want to give information to my dear colleagues. The spirit is very good but you must be mindful that the administration of FUFA is what has made this happen. The money to the players. That indicates to you that FUFA is very trustworthy. This is not the old FUFA we are talking about.\""
The function countLine() doesn't work since it requires a connection - these are just .txt imported into R. I realize that the line count is subject to the formatting of what the text is opened up in, but any general help on if this is feasible would help. Thanks.

Answer 1

I didn't think your example was reproducible, so I edited it to get it to contain what you posted but I do not know if the names will match:
txtvec <- structure(list(`'511' ` = "MR Smith: Mr Speaker, I like the spirit in which we are agreeing on this. The administration of FUFA is present here. FUFA could be used as a conduit, but the intention of what hon. Beti Kamya brought up and what hon. Rose Namayanja has said was okufuwa - just giving a token of appreciation to the players who achieved this.\"",
`'513' ` = "MR Gordon: Thank you very much, Mr Speaker. FUFA is an organisation and the players are the ones who got the cup for us. To promote motivation in all activities, not only football, you should remunerate people who have done well. In this case, we have heard about FUFA with their problems. They have not paid water bills and they can take this money to pay the water bills. If we agree that this money is supposed to go to the players and the coaches, then when it goes there they would know the amount and they will sit among themselves and distribute according to what we will have given. (Applause) I thank you.\"",
`'515' ` = "MR Catalano: Mr Speaker, I want to give information to my dear colleagues. The spirit is very good but you must be mindful that the administration of FUFA is what has made this happen. The money to the players. That indicates to you that FUFA is very trustworthy. This is not the old FUFA we are talking about.\""), .Names = c("'511'\t",
"'513'\t", "'515'\t"))
So it's only a matter or running a regex expression across it and tabling the results:
> table( sapply(txtvec, function(x) sub("(^MR.+)\\:.+", "\\1", x) ) )
#MR Catalano MR Gordon MR Smith
1 1 1
There was concern expressed that the names were not in the original structure. This is another version with unnamed vector and a slightly modified regex:
txtvec <- c("\"511\"\t\"\nMR Smith: Mr Speaker, I like the spirit in which we are agreeing on this. The administration of FUFA is present here. FUFA could be used as a conduit, but the intention of what hon. Beti Kamya brought up and what hon. Rose Namayanja has said was okufuwa - just giving a token of appreciation to the players who achieved this.\"",
"\"513\"\t\"\nMR Gordon: Thank you very much, Mr Speaker. FUFA is an organisation and the players are the ones who got the cup for us. To promote motivation in all activities, not only football, you should remunerate people who have done well. In this case, we have heard about FUFA with their problems. They have not paid water bills and they can take this money to pay the water bills. If we agree that this money is supposed to go to the players and the coaches, then when it goes there they would know the amount and they will sit among themselves and distribute according to what we will have given. (Applause) I thank you.\"",
"\"515\"\t\"\nMR Catalano: Mr Speaker, I want to give information to my dear colleagues. The spirit is very good but you must be mindful that the administration of FUFA is what has made this happen. The money to the players. That indicates to you that FUFA is very trustworthy. This is not the old FUFA we are talking about.\""
)
table( sapply(txtvec, function(x) sub(".+\\n(MR.+)\\:.+", "\\1", x) ) )
#MR Catalano MR Gordon MR Smith
# 1 1 1
To count the number of "lines" these would occupy on a wrapping device with 80 characters per line you could use this code (which could easily be converted to a function):
sapply(txtvec, function(tt) 1+nchar(tt) %/% 80)
#[1] 5 8 4

Answer 2

This is raised in the comments, but it really bares being it's own answer:
You cannot "count lines" without defining what a "line" is. A line is a very vague concept and can vary by the program being used.
Unless of course the data contains some indicator of a line break, such as \n. But even then, you would not be counting lines, you would be counting linebreaks. You would then have to ask yourself if the hardcoded line break is in accord with what you are hoping to analyze.
--
If your data does not contain linebreaks, but you still want to count the number of lines, then we're back to the question of "how do you define a line"? The most basic way, is as #flodel suggests, which is to use character length. For example, you can define a line as 76 characters long, and then take
ceiling(nchar(X) / 76))
This of course assumes that you can cut words. (If you need words to remain whole, then you have to get craftier)