Regex with Chinese characters

Question

I am looking for a text_ which: 本周 (3 月 25 日 - 3 月 31 日), 国内 油厂 开机率 继续 下降, 全国 各地 油厂 大豆 压榨 压榨 1456000 吨 (出 粕 1157520 吨, 出 油 262080 吨), 较上周 的 ... [continued]

  crush <- str_extract(string = text_, pattern = perl("(?<=量).*(?=吨（出粕)")) meal <- str_extract(string = text_, pattern = perl("(?<=粕).*(?=吨，出)")) oil <- str_extract(string = text_, pattern = perl("(?<=出油).*(?=吨）)")) 

prints

 [1] "1456000" ## correct [1] "1157520" ## correct [1] NA ## looking for 262080 here 

Why the first two matches, but not the last? I am using the stringr library.

Answer 1

Note that the current version of stringr based on the ICU regex library, and the use of perl() is deprecated.

Note that lookbehind patterns are fixed-width, and there seems to be a problem with the way the ICU parses the first letter in your lookbehind pattern (it cannot calculate its width for some unknown reason).

Since you use stringr , you can simply rely on a capture that can be achieved with str_match to extract part of the template:

 > match <- str_match(s, "出油(\\d+)吨") > match[,2] [1] "262080" 

Thus, you will avoid any possible problems in the future. In addition, these regular expressions are faster because there is no unanchored lookbehind in the template that runs at every place in the search string.

Alternatively, you can simply use the PCRE regular expression with an R base:

 > regmatches(s, regexpr("(?<=出油)\\d+(?=吨)", s, perl=TRUE)) [1] "262080" 

Answer 2

For some reason, I still don't know, I was not able to use the @ WiktorStribiżew commented solution, but it ended up working:

 oil <- str_extract(string = text_, pattern = perl("(?<=吨).*(?=吨)")) # [1] "（出粕1157520吨，出油262080吨），较oil <- str_extract(string = oil, pattern = perl("(?<=油)\\d+(?=吨)")) # [1] 262080 

Answer 3

Try the following:

  oil <- str_extract(string = text_, pattern = perl("(?<=出油).*(?=吨)，较上周的）)")) 

Since a simple 吨 may reappear again in your text, it cannot determine exactly which part may exceed the length of the data or cause a problem with the data type.