R语言学习 day_4

• 字符串处理

  1. 常用字符串函数：nchar(),substring(),paste().

     nchar(),可以显示字符串里面的字符数量

     substring(),可以截取字符串

     paste（）可以拼贴字符串

     > charvector=c("1970 1,1003.2|4.11|6.21 Mio","1975|21,034.6|5.31|7.11 Mio")
     > charvector
     [1] "1970 1,1003.2|4.11|6.21 Mio" "1975|21,034.6|5.31|7.11 Mio"
     > nchar(charvector)
     [1] 27 27
     > years=substring(charvector,first = 1,last=4) #结构中first = 1,last=4，可以简写为1，4
     > years
     [1] "1970" "1975"
     > years=substring(charvector,1,4)
     > years
     [1] "1970" "1975"
     > paste(years,"12","31",sep="-") #sep表示以何种符号分割
     [1] "1970-12-31" "1975-12-31"
     > paste(years,"31","12",sep="-")
     [1] "1970-31-12" "1975-31-12"
     > #按输入字符的顺序paste
     

  2. 搜索字符串

     grep(),regexpr()

     grep(),可以搜索指定内容或者含有指定内容的字符串。

     #搜索字符串中的模式
     > #grep()
     > data(longley)
     > names=names(longley)
     > names
     [1] "GNP.deflator" "GNP"          "Unemployed"   "Armed.Forces" "Population"  
     [6] "Year"         "Employed"    
     > index=grep("GNP",names) #搜索含GNP字样的字符串
     > index
     [1] 1 2
     > names[index]
     [1] "GNP.deflator" "GNP"         
     > index=grep("^G",names) #搜索以G字母开头的字符串
     > index
     [1] 1 2
     #regexpr()判断向量names中的各元素是否含有字符串,有则返回位次，无则返回值-1，返回结果还有一些其他内容
     > gnpMatch=regexpr("GNP",names)
     > gnpMatch
     [1]  1  1 -1 -1 -1 -1 -1
     attr(,"match.length")
     [1]  3  3 -1 -1 -1 -1 -1
     attr(,"index.type")
     [1] "chars"
     attr(,"useBytes")
     [1] TRUE
     > index=1:length(names)
     > index[gnpMatch>0]
     [1] 1 2
     

  3. 字符串替换

     利用gsub()可以进行字符串替换

     #字符串替换
     > #gsub()
     > gsub(".Forces","_Forces",names)
     [1] "GNP.deflator" "GNP"          "Unemployed"   "Armed_Forces" "Population"  
     [6] "Year"         "Employed"
     

• 因子处理

  1. cut()函数，对数据进行切片分组，cut返回的结果类型为 因子

     > data("longley")
     > gnp=longley$GNP
     > regimes=(2:6)*100;regimes
     [1] 200 300 400 500 600
     > cut(x=gnp,regimes) #不指定分组时，函数会默认分组
      [1] (200,300] (200,300] (200,300] (200,300] (300,400] (300,400] (300,400] (300,400]
      [9] (300,400] (400,500] (400,500] (400,500] (400,500] (500,600] (500,600] (500,600]
     Levels: (200,300] (300,400] (400,500] (500,600]
     > gnp
      [1] 234.289 259.426 258.054 284.599 328.975 346.999 365.385 363.112 397.469 419.180
     [11] 442.769 444.546 482.704 502.601 518.173 554.894
     > cut(x=gnp,breaks=3)
      [1] (234,341] (234,341] (234,341] (234,341] (234,341] (341,448] (341,448] (341,448]
      [9] (341,448] (341,448] (341,448] (341,448] (448,555] (448,555] (448,555] (448,555]
     Levels: (234,341] (341,448] (448,555]
     > cut(x=gnp,breaks=5)
      [1] (234,298] (234,298] (234,298] (234,298] (298,363] (298,363] (363,427] (363,427]
      [9] (363,427] (363,427] (427,491] (427,491] (427,491] (491,555] (491,555] (491,555]
     Levels: (234,298] (298,363] (363,427] (427,491] (491,555]
     > groups=cut(x=gnp,breaks=3,labels=c("low","median","high"));groups
      [1] low    low    low    low    low    median median median median median median
     [12] median high   high   high   high  
     Levels: low median high
     > groups=cut(x=gnp,breaks=3,labels=c("high","median","low"));groups #注意标签顺序，与排序方式有关。建议指定方式
      [1] high   high   high   high   high   median median median median median median
     [12] median low    low    low    low   
     Levels: high median low
     > 
     

  2. 利用factor()函数可以实现类型转换

     > #注意标签顺序，与排序方式有关。建议指定方式
     > #函数factor
     > x=1:6;x
     [1] 1 2 3 4 5 6
     > class(x)
     [1] "integer"
     > y=factor(x)
     > y
     [1] 1 2 3 4 5 6
     Levels: 1 2 3 4 5 6
     > class(y)
     [1] "factor"
     > x+y
     [1] NA NA NA NA NA NA
     Warning message:
     In Ops.factor(x, y) : ‘+’ not meaningful for factors
     

• 日期序列对象的类型

  1. date类型

     包含日期数据，不包含时间和时区信息

     常用获取时间的方法就是Sys.Date()函数，返回当前系统的时间

     as.numeric(),unclass()可以将日期对象转换为它的内部形式

  2. POSIX类型

     名称来源于“UNIX可移植操作系统接口”（Portable Operation System Interface of UNIX）的缩写。

     两种类型：POSIXct（日历时间，calendar time），POSIXlt（列表时间，list time）

     POSIXct只需要一个双精度数值，而POSIXlt需要一个列表，前者更节省空间。

  ###date()
  > today<-Sys.Date()
  > today
  [1] "2022-07-27"
  > class(today)
  [1] "Date"
  > as.numeric(today)
  [1] 19200
  > ndate<-unclass(today)
  > ndate
  [1] 19200
  > as.Date(ndate,origin = '1970-01-01')
  #在R语言中，是以距离1970-01-01的天数储存数据
  [1] "2022-07-27"
  > class(ndate)='Date'
  #引号不可省
  > ndate
  [1] "2022-07-27"
  
  > ###POSIX类型
  > time_ct<-Sys.time()
  > time_ct
  [1] "2022-07-27 09:32:13 CST"
  > class(time_ct)
  [1] "POSIXct" "POSIXt" 
  > as.numeric(time_ct)
  [1] 1658885533
  > unclass(time_ct)
  [1] 1658885533
  > time_lt<-as.POSIXlt(time_ct)
  > time_ct
  [1] "2022-07-27 09:32:13 CST"
  > time_lt
  [1] "2022-07-27 09:32:13 CST"
  > class(time_ct)
  [1] "POSIXct" "POSIXt" 
  > class(time_lt)
  [1] "POSIXlt" "POSIXt" 
  > as.numeric(time_lt)
  [1] 1658885533
  > unclass(time_lt)
  $sec
  [1] 13.02043
  
  $min
  [1] 32
  
  $hour
  [1] 9
  
  $mday
  [1] 27
  
  $mon
  [1] 6
  
  $year
  [1] 122
  
  $wday
  [1] 3
  
  $yday
  [1] 207
  
  $isdst
  [1] 0
  
  $zone
  [1] "CST"
  
  $gmtoff
  [1] 28800
  
  attr(,"tzone")
  [1] ""    "CST" "CDT"
  > n_sec<-as.numeric(time_lt)
  > class(time_lt)
  [1] "POSIXlt" "POSIXt" 
  > class(n_sec)=c("POSIXct","POSIXt");n_sec;class(n_sec)
  [1] "2022-07-27 09:32:13 CST"
  [1] "POSIXct" "POSIXt"