如何将制表符分隔的数据(不同格式)解析为data.table/data.frame? [英] How to parse tab-delimited data (of different formats) into a data.table/data.frame?

问题描述

我正在尝试分析制表符分隔的数据，该数据已与其他数据一起另存为文本文件.我希望这是一个R data.table/data.frame.

I am trying to parse tab-delimited data, which has been saved as a text file with extraneous data. I would like this to be an R data.table/data.frame.

制表符分隔的格式如下:

The tab-delimited format is the following:

A   1092    -   1093    +   1X
B   1093    HRDCPMRFYT
A   1093    +   1094    -   1X
B   1094    BSZSDFJRVF
A   1094    +   1095    +   1X
B   1095    SSTFCLEPVV
...

只有两种类型的行，A和B.A始终有5列，例如对于第一行，

There are only two types of rows, A and B. A consistently has 5 columns, e.g. for the first row,

1092    -   1093    +   1X

B始终有两列:

1093    HRDCPMRFYT

问题:如何解析具有交替"行且格式不同的文件?

Question: How do you parse a file with "alternating" rows with different formats?

让我们说这是一个文本文件，仅具有这种格式，将A和B的行交替显示，分别为5列和2列.您如何将其解析为R data.table?我的想法是如何创建以下格式:

Let's say that this was a text file which was only of this format, alternating rows of A and B, with 5 columns and 2 columns respectively. How do you parse this into an R data.table? My idea how be to create the following format:

1092    -    1093    +    1X    1093    HRDCPMRFYT
1093    +    1094    -    1X    1094    BSZSDFJRVF
1094    +    1095    +    1X    1095    SSTFCLEPVV
... 

推荐答案

一种方法是使用readLines读入数据，取出所需的位，然后传递给read.table以形成数据帧.因此，如果行是交替的，则:

One way to go is to read in your data with readLines, pull out the bits you want, and pass to read.table to form the dataframe. So if the rows are alternating then:

txt <- 
'1092    -   1093    +   1X
1093    HRDCPMRFYT
1093    +   1094    -   1X
1094    BSZSDFJRVF
1094    +   1095    +   1X
1095    SSTFCLEPVV'


rd <- readLines(textConnection(txt))
data.frame(read.table(text=rd[c(TRUE, FALSE)]), 
           read.table(text=rd[c(FALSE, TRUE)]))

将textConnection(txt)更改为您的文件路径

另一种方法是只读一次，然后进行后处理

Another way is to read in only once and then post-process

r <- read.table(text=txt, fill=TRUE, stringsAsFactors=FALSE, na.strings = "")
d <- cbind(r[c(TRUE, FALSE),], r[c(FALSE, TRUE),])
d[ colSums(is.na(d)) < nrow(d)]

这篇关于如何将制表符分隔的数据(不同格式)解析为data.table/data.frame?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！