使用R中的regex从文件名中提取模式?

你可以用

sub("^.*_(.*)\\.txt$", "\\1", x)

regex demo

sub 将执行单个seasrch和replace操作。如果字符串符合以下条件,它将找到匹配项:

• ^ 字符串开头
• .*_ -任何0+字符,尽可能多,直到最后一个 _
• (.*) \1 从替换模式),尽可能多,直到并包括。。。
• \\.txt$ - .txt . 必须转义以匹配字符串末尾的文本点( $ ).

R demo

x <- "UNKNOWN_{_requestID___b9b6bcc4-c163-45d7-82d9-423a96cf5fe1_,_deviceID___9c84f871-9e95-45d5-9335-12e7d42b96a0_}_2018-08-15-15-43-01-296_529307b7-6316-4cdc-ab53-2e1158c651c6.txt"
sub("^.*_(.*)\\.txt$", "\\1", x)
## => [1] "529307b7-6316-4cdc-ab53-2e1158c651c6"

这里使用的是工具中隐藏的宝石。

x <- "UNKNOWN_{_requestID___b9b6bcc4-c163-45d7-82d9-423a96cf5fe1_,_deviceID___9c84f871-9e95-45d5-9335-12e7d42b96a0_}_2018-08-15-15-43-01-296_529307b7-6316-4cdc-ab53-2e1158c651c6.txt"

out <- strsplit(x, "_")[[1]]
out <- out[length(out)]
tools::file_path_sans_ext(out)

[1] "529307b7-6316-4cdc-ab53-2e1158c651c6"

    text <- c("UNKNOWN_{_requestID___b9b6bcc4-c163-45d7-82d9-423a96cf5fe1_,_deviceID___9c84f871-9e95-45d5-9335-12e7d42b96a0_}_2018-08-15-15-43-01-296_529307b7-6316-4cdc-ab53-2e1158c651c6.txt" )
    sub("\\.txt.*", "", sub(".*\\_", "", text)) 

你能试试下面的吗。

gsub(".*_|\\.txt","",x)

输出如下。

[1] "529307b7-\n6316-4cdc-ab53-2e1158c651c6"

说明: 添加以下内容仅供解释之用。

gsub(     ##Using gsub(Global substitution function of R to perform multiple substitution on variables)
".*_      ##Mentioning REGEX to select everything from starting till _(underscore)
|         ##|(pipe) defines OR so it should match either previous or coing REGEX in varibale's value.
\\.txt"   ##\\. means escaping DOT so that DOT should be treated as a DOT not with its special meaning so it should match string .txt
,""       ##If above mentioned REGEXs any one of them OR both matches then substitute them with "" means NULL.
,x)       ##Mentioning variable named x on which we have to perform gsub.

x 的值如下。

x <- "UNKNOWN_{_requestID___b9b6bcc4-c163-45d7-82d9-423a96cf5fe1_,_deviceID
___9c84f871-9e95-45d5-9335-12e7d42b96a0_}_2018-08-15-15-43-01-296_529307b7-
6316-4cdc-ab53-2e1158c651c6.txt"