Question

当我尝试通过以下方式将下表读入dataframe（data100）时

data100 <- read.table(header=TRUE, text='
                                 verb_object SESSION_ID
1:   BA31C1CC63E5043483FAE25F085E25E5 INSERT   41595370
2: BECE6374D91D47E6285EFDEBA6D65BB9 DATABASE   41595371
3:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE   41595282
4:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE   41595282
5: 2BC5A4199A0DDA16FA17A9CA1AA17C02 DATABASE   41595373
6:   6D944D54C54ED75D487288FE1505BB59 INSERT   41595368
')

I get the following error:
Error in read.table(header = TRUE, text = "\n                               verb_object SESSION_ID\n   BA31C1CC63E5043483FAE25F085E25E5 INSERT   41595370\n                      BECE6374D91D47E6285EFDEBA6D65BB9 DATABASE   41595371\n                         26D695C8CA82CAFFDF985201F3AA44D7 UPDATE   41595282\n                         26D695C8CA82CAFFDF985201F3AA44D7 UPDATE   41595282\n                     2BC5A4199A0DDA16FA17A9CA1AA17C02 DATABASE   41595373\n                         6D944D54C54ED75D487288FE1505BB59 INSERT   41595368\n") : 
  duplicate 'row.names' are not allowed

我怎么读？

使用

后

lines <- readLines(textConnection("       verb_object SESSION_ID



> data100<-read.table(text=gsub('(?<=\\:)\\s+|\\s+(?=\\s[0-9])', " '", lines, perl=TRUE), sep='', fill=TRUE)

结果如下：

> data100
           V1                               V2       V3       V4 V5                                         V6       V7
1 verb_object                       SESSION_ID                NA                                                     NA
2         1:  BA31C1CC63E5043483FAE25F085E25E5   INSERT 41595370 2: BECE6374D91D47E6285EFDEBA6D65BB9 DATABASE  41595371
3         3:  26D695C8CA82CAFFDF985201F3AA44D7   UPDATE 41595282 4:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE  41595282
4         5:  2BC5A4199A0DDA16FA17A9CA1AA17C02 DATABASE 41595373 6:   6D944D54C54ED75D487288FE1505BB59 INSERT  41595368
>

Answer 1

我们可以使用readLines阅读，使用gsub放置引号，然后使用read.table阅读

lines <- readLines(textConnection("verb_object SESSION_ID
1:   BA31C1CC63E5043483FAE25F085E25E5 INSERT   41595370
2: BECE6374D91D47E6285EFDEBA6D65BB9 DATABASE   41595371
3:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE   41595282
4:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE   41595282
5: 2BC5A4199A0DDA16FA17A9CA1AA17C02 DATABASE   41595373
6:   6D944D54C54ED75D487288FE1505BB59 INSERT   41595368"))



read.table(text=gsub('(?<=\\:)\\s+|\\s+(?=\\s[0-9])', " '", lines, perl=TRUE), sep='')
#                                  verb_object SESSION_ID
#1:   BA31C1CC63E5043483FAE25F085E25E5 INSERT    41595370
#2: BECE6374D91D47E6285EFDEBA6D65BB9 DATABASE    41595371
#3:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE    41595282
#4:   26D695C8CA82CAFFDF985201F3AA44D7 UPDATE    41595282
#5: 2BC5A4199A0DDA16FA17A9CA1AA17C02 DATABASE    41595373
#6:   6D944D54C54ED75D487288FE1505BB59 INSERT    41595368

更新

OP的新数据集可以像以前一样用readLines读取，

lines <- readLines(textConnection("items newitem
1: BA31C1CC63E5043483FAE25F085E25E5 INSERT OV1
2: BECE6374D91D47E6285EFDEBA6D65BB9 DATABASE OV2
3: 26D695C8CA82CAFFDF985201F3AA44D7 UPDATE OV3
4: 2BC5A4199A0DDA16FA17A9CA1AA17C02 DATABASE OV4
5: 6D944D54C54ED75D487288FE1505BB59 INSERT OV5"))

我们应该注意，我们在早期数据集（\\s+(?=\\s[0-9])）中匹配的模式在这里不起作用，因为'SESSIONID'中的第一个字符是数字，而在'newitem'中它是一个大写字母。因此，我们会匹配一个或多个不是:的字符，从字符串的开头（^[^:]+）后跟:，后跟一个或多个空格（\\s+），然后我们使用括号()将字符捕获为一组，即一个或多个不是空格的字符，后跟一个或多个空格，而字符不是空格（([^ ]+\\s+[^ ]+)，匹配一个或多个空格（{{ 1}}）后跟一个或多个字符，直到字符串结尾为另一个捕获组（\\s+）。我们通过在第一个捕获组（(.*)$）周围放置引号来替换，后跟空格第二个捕获组。

'\\1'

read.table重复row.names出错

1 个答案:

更新