qdapRegex::rm_nchar_words returns different results when non English letters involved?

Question

Please help me with the following confusion:

qdapRegex::rm_nchar_words("è ûé", "1,2")
[1] "è ûé"

qdapRegex::rm_nchar_words('k ku ppp d', "1,2")
[1] "ppp"

Why in the first code line it doesn't respond with "" but in the second one it works as expected. What do I miss here? The only thing I can think that in the first line of code the string is built from non English letters.

Any solution?

Comments are not for extended discussion; this conversation has been [moved to chat](https://chat.stackoverflow.com/rooms/194791/discussion-on-question-by-steves-qdapregexrm-nchar-words-returns-different-res). — Samuel Liew, Jun 12 '19 at 01:23

SteveS · Accepted Answer · 2019-06-19T08:59:42.927

As mentioned by the author of the package:

It uses \w to define letters which is defined as [A-Za-z0-9_]. You would need to write your own custom regex to handle the non-ascii letters

UPDATE:

On my Win 7 machine the output is as expected.

One of the possible ways to solve it using pattern "[\\pL_]" (any word in any language)

rm_nchar_words("è ûé", "1,2", pattern = "[\\pL_]")

Locale on Win machine:

locale:
[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252    LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C                           LC_TIME=English_United States.1252

I will keep investigate this and post updates for my answer.

UPDATE 2:

rm_nchar_words("è ûé", "1,2", pattern = "[\\pL_]")
""

works on my Ubuntu 18.04.

qdapRegex::rm_nchar_words returns different results when non English letters involved?

1 Answers1