R 数据处理 —— stringr

前言

之前我们介绍 R 数据处理的时候，对字符串的操作都是用自带的函数。

虽然 R 的字符串并不是它的强项，看起来也不是那么的优雅，但是字符串在数据处理和清洗过程中还是扮演者较为重要的角色。

所以，今天我就讲下 R 字符串处理的第三方包 —— stringr

stringr 包提供了一组内聚函数，尽可能使字符串的操作简单化。

stringr 中主要包括四个函数族：

• 1.字符操作：这些函数允许操作字符串向量中每个字符串的字符
• 2.提供添加、删除和操作空白字符的工具
• 3.提供对语境敏感的操作
• 4.模式匹配函数。总共包含 4 中模式，其中最常用的就是正则表达式

安装

# install 
install.packages("stringr")

# 从 GitHub 中安装最新的开发版本:
# 需要用到 devtools 包，如果没有这个包，需要安装一下
if (!require("devtools")) 
  install.packages("devtools")
devtools::install_github("tidyverse/stringr")

使用

导入

library(stringr)

stringr 中的所有函数都以 str_ 开头，并接受一个字符串向量作为第一个参数

1.单字符串操作

1. str_length: 字符串长度

str_length(string)

参数：

• string: 字符串或字符串向量

示例：

> s <- c("why", "video", "cross", "extra", "deal", "authority")
> str_length(s)
# [1] 3 5 5 5 4 9
> str_length("love")
# [1] 4
> str_length(c("love", NA))
# [1]  4 NA

2. str_sub: 提取和修改字符串

str_sub(string, start = 1L, end = -1L)

参数：

• string: 字符串或字符串向量
• start: 开始位置
• end: 结束位置

示例：

x <- c("abcdef", "ghifjk")

# The 3rd letter
> str_sub(x, 3, 3)
# [1] "c" "i"

# The 2nd to 2nd-to-last character
> str_sub(x, 2, -2)
# [1] "bcde" "hifj"

修改字符串

> str_sub(x, 3, 3) <- "X"
> x
# [1] "abXdef" "ghXfjk"

3. str_dup: 复制字符串

str_dup(string, times)

参数：

• string: 字符串或字符串向量
• times: 复制数量

示例：

> str_dup(x, c(2, 3))
# [1] "abcdefabcdef"       "ghifjkghifjkghifjk"
> str_dup(x, c(2))
# [1] "abcdefabcdef" "ghifjkghifjk"
> str_dup(x, 2)
# [1] "abcdefabcdef" "ghifjkghifjk"

4. str_c: 字符串拼接

str_c(..., sep = "", collapse = NULL)

参数：

• ...: 多参数的输入
• sep: 把多个字符串参数拼接为一个字符串的连接符号。
• collapse: 把多个向量参数拼接为一个字符串的连接符号。

> x <- c("why", "video", "cross", "extra", "deal", "authority")
> str_c(x, collapse = ", ")
# [1] "why, video, cross, extra, deal, authority"
> str_c("why", "not", "my", sep = "-")
[1] "why-not-my"

2.空白字符

1.str_pad: 补充字符串的长度

str_pad(string, width, side = c("left", "right", "both"), pad = " ")

参数：

• string:
• width: 字符串填充后的长度
• side: 填充方向
  • left: 左边填充
  • right: 右边填充
  • both: 两边都填充
• pad: 用于填充的字符

> x <- c("abc", "defghi")
> str_pad(x, 10) # 默认在左边
# [1] "       abc" "    defghi"
> str_pad(x, 10, "both")
# [1] "   abc    " "  defghi  "
> str_pad(x, 10, pad = "=")
# [1] "====abcdef" "====ghifjk"

如果你设置的字符串填充长度小于其本身，那么不会产生任何作用

str_pad(x, 4)
#> [1] " abc"   "defghi"

2.str_trunc: 省略过长字符

str_trunc(string, width, side = c("right", "left", "center"), ellipsis = "...")

参数：

• string: 字符串或字符串向量
• width: 最大字符串长度
• side： 省略方式
  • center: 中间省略
  • left: 左边省略
  • right: 右边省略

示例

> x <- "This string is moderately long"
> rbind(
+     str_trunc(x, 20, "right"),
+     str_trunc(x, 20, "left"),
+     str_trunc(x, 20, "center")
+ )
#      [,1]                  
# [1,] "This string is mo..."
# [2,] "...s moderately long"
# [3,] "This stri...ely long"

3. str_trim: 去掉字符串的空格和TAB

str_trim(string, side = c("both", "left", "right"))

参数：

• string: 字符串或字符串向量
• side： 过滤方式
  • both: 两边都过滤
  • left: 左边过滤
  • right: 右边过滤

示例

x <- c("Short", "This is a long string")

x <- c("  a   ", "b   ",  "   c")
str_trim(x)
#> [1] "a" "b" "c"
str_trim(x, "left")
#> [1] "a   " "b   " "c"

4. str_wrap: 控制字符串输出格式

str_wrap(string, width = 80, indent = 0, exdent = 0)

参数：

• string: 字符串或字符串向量。
• width: 设置一行所占的宽度。
• indent: 段落首行的缩进值
• exdent: 段落非首行的缩进值

示例

jabberwocky <- str_c(
  "`Twas brillig, and the slithy toves ",
  "did gyre and gimble in the wabe: ",
  "All mimsy were the borogoves, ",
  "and the mome raths outgrabe. "
)
cat(str_wrap(jabberwocky, width = 40))
# `Twas brillig, and the slithy toves did
# gyre and gimble in the wabe: All mimsy
# were the borogoves, and the mome raths
# outgrabe.

3.语境

1.大小写转换

str_to_upper(string, locale = "")
str_to_lower(string, locale = "")
str_to_title(string, locale = "")

参数：

• string: 字符串。
• locale: 按哪种语言习惯

示例

x <- "I like horses."
str_to_upper(x)
#> [1] "I LIKE HORSES."
str_to_title(x)
#> [1] "I Like Horses."

str_to_lower(x)
#> [1] "i like horses."
# Turkish has two sorts of i: with and without the dot
str_to_lower(x, "tr")
#> [1] "ı like horses."

2. 排序

# 对值排序
str_sort(x, decreasing = FALSE, na_last = TRUE, locale = "", ...)
str_order(x, decreasing = FALSE, na_last = TRUE, locale = "", ...)

参数：

• x: 字符串或字符串向量。
• decreasing: 排序方向。
• na_last: NA 放置的位置：TRUE 放到最后，FALSE 放到最前，NA 过滤处理
• locale: 按哪种语言习惯排序

> x <- c("y", "i", "k")
> str_order(x)
# [1] 2 3 1

> str_sort(x)
# [1] "i" "k" "y"
# In Lithuanian, y comes between i and k
> str_sort(x, locale = "lt")
# [1] "i" "y" "k"

3. 编码方式

str_conv(string, encoding)

参数：

• string: 字符串或字符串向量。
• encoding: 编码名。

示例：

# 把中文字符字节化
> x <- charToRaw('你好')
> x
[1] c4 e3 ba c3

# 默认 win 系统字符集为 GBK，GB2312 为 GBK 字集，转码正常
> str_conv(x, "GBK")
[1] "你好"
> str_conv(x, "GB2312")
[1] "你好"

# 在 mac 系统下
> x <- charToRaw('你好')
> str_conv(x, "GBK")
[1] "浣犲ソ"
> str_conv(x, "GB2312")
# [1] "浣\032濂\032"
# Warning messages:
# 1: In stri_conv(string, encoding, "UTF-8") :
# mac 
> str_conv(x, "UTF-8")
# [1] "你好"

4.模式匹配

每个模式匹配函数的前两个参数都相同

• string： 字符串或字符串向量
• pattern： 匹配模式

字符串和匹配模式

strings <- c(
  "apple", 
  "219 733 8965", 
  "329-293-8753", 
  "Work: 579-499-7527; Home: 543.355.3679"
)
phone <- "([2-9][0-9]{2})[- .]([0-9]{3})[- .]([0-9]{4})"

1.匹配字符串的字符

str_detect(string, pattern)

示例：

# 匹配字符串的字符
> str_detect(strings, phone)
# [1] FALSE  TRUE  TRUE  TRUE

2. 模式在字符串中的位置

str_locate(string, pattern)
str_locate_all(string, pattern)

示例：

> (loc <- str_locate(strings, phone))
#>      start end
#> [1,]    NA  NA
#> [2,]     1  12
#> [3,]     1  12
#> [4,]     7  18
> str_locate_all(strings, phone)
#> [[1]]
#>      start end
#> 
#> [[2]]
#>      start end
#> [1,]     1  12
#> 
#> [[3]]
#>      start end
#> [1,]     1  12
#> 
#> [[4]]
#>      start end
#> [1,]     7  18
#> [2,]    27  38

3. 从字符串中提取匹配模式

str_extract(string, pattern)
str_extract_all(string, pattern, simplify = FALSE)

示例：

> val <- c("a1", 467, "ab2")

# 返回匹配的数字
> str_extract(val, "\\d")
# [1] "1" "4" "2"
> str_extract_all(val, "\\d")
# [[1]]
# [1] "1"

# [[2]]
# [1] "4" "6" "7"

# [[3]]
# [1] "2"

# 返回匹配的字符
> str_extract(val, "[a-z]+")
# [1] "a" NA     "ab"
> str_extract_all(val, "\\w+")
# [[1]]
# [1] "a1"

# [[2]]
# [1] "467"

# [[3]]
# [1] "ab2"

4. 从字符串中提取匹配组

str_match(string, pattern)
str_match_all(string, pattern)

示例

> str_match(strings, phone)
#>      [,1]           [,2]  [,3]  [,4]  
#> [1,] NA             NA    NA    NA    
#> [2,] "219 733 8965" "219" "733" "8965"
#> [3,] "329-293-8753" "329" "293" "8753"
#> [4,] "579-499-7527" "579" "499" "7527"
> str_match_all(strings, phone)
#> [[1]]
#>      [,1] [,2] [,3] [,4]
#> 
#> [[2]]
#>      [,1]           [,2]  [,3]  [,4]  
#> [1,] "219 733 8965" "219" "733" "8965"
#> 
#> [[3]]
#>      [,1]           [,2]  [,3]  [,4]  
#> [1,] "329-293-8753" "329" "293" "8753"
#> 
#> [[4]]
#>      [,1]           [,2]  [,3]  [,4]  
#> [1,] "579-499-7527" "579" "499" "7527"
#> [2,] "543.355.3679" "543" "355" "3679"

5. 替换匹配的字符串

str_replace(string, pattern, replacement)
str_replace_all(string, pattern, replacement)
str_replace_na(string, replacement = "NA")

replacement 为替字符串

示例：

# 替换第一个匹配
> str_replace(strings, phone, "XXX-XXX-XXXX")
#> [1] "apple"                                 
#> [2] "XXX-XXX-XXXX"                          
#> [3] "XXX-XXX-XXXX"                          
#> [4] "Work: XXX-XXX-XXXX; Home: 543.355.3679"
# 替换所有匹配
> str_replace_all(strings, phone, "XXX-XXX-XXXX")
#> [1] "apple"                                 
#> [2] "XXX-XXX-XXXX"                          
#> [3] "XXX-XXX-XXXX"                          
#> [4] "Work: XXX-XXX-XXXX; Home: XXX-XXX-XXXX"
> str_replace_na(c(NA, "abc", "def"))
# [1] "NA"  "abc" "def"

6. 字符串分割

str_split(string, pattern, n = Inf)  # 返回 list
str_split_fixed(string, pattern, n)  # 返回 matrix

n 为分割次数

示例：

str_split("a-b-c", "-")
#> [[1]]
#> [1] "a" "b" "c"
str_split_fixed("a-b-c", "-", n = 2)
#>      [,1] [,2] 
#> [1,] "a"  "b-c"

7. 字符串计数

str_count(string, pattern = "")

示例：

> str_count(strings, phone)
# [1] 0 1 1 2

8. 返回的匹配字符串

str_subset(string, pattern)

示例：

> str_subset(strings, phone)
# [1] "219 733 8965"                          
# [2] "329-293-8753"                          
# [3] "Work: 579-499-7527; Home: 543.355.3679"

9. 提取单词

word(string, start = 1L, end = start, sep = fixed(" "))

• string: 字符串，字符串向量。
• start: 开始位置。
• end: 结束位置。
• sep: 匹配字符。

示例

> val <- c("hello world", "this is my, girl")

# 默认以空格分割，取第一个位置的字符串
> word(val, 1)
# [1] "hello" "this"
> word(val, -1)
# [1] "world" "girl"   
> word(val, 2, -1)
# [1] "world"       "is my, girl"  

# 以,分割，取第一个位置的字符串 
> val <- '111,222,333,444'
> word(val, 1, sep = fixed(','))
[1] "111"
> word(val, 3, sep = fixed(','))
[1] "333"

四种模式

1. boundary: 匹配字符、行、句子或单词之间的边界

x <- "This is a sentence."
str_split(x, boundary("word"))
#> [[1]]
#> [1] "This"     "is"       "a"        "sentence"
str_count(x, boundary("word"))
#> [1] 4
str_extract_all(x, boundary("word"))
#> [[1]]
#> [1] "This"     "is"       "a"        "sentence"

2. coll: 定义字符串标准排序规则。

i <- c("I", "İ", "i", "ı")
i
#> [1] "I" "İ" "i" "ı"

str_subset(i, coll("i", ignore_case = TRUE))
#> [1] "I" "i"
str_subset(i, coll("i", ignore_case = TRUE, locale = "tr"))
#> [1] "İ" "i"

3. fixed: 定义用于匹配的字符，包括正则表达式中的转义符

> str_count(c("a.", ".", ".a.",NA), ".")
[1]  2  1  3 NA

# 用fixed匹配字符
> str_count(c("a.", ".", ".a.",NA), fixed("."))
[1]  1  1  2 NA

4. regex: 定义正则表达式，默认就是正则表达式

> val
[1] "a1"  "467" "ab2"
> str_extract(val, regex("\\w+"))
[1] "a1"  "467" "ab2"