Remove hashtags from beginning and end of tweets in R

I am trying to remove hashtags from beginning of strings in R. For example:

 x<- "I didn't know it could be #boring. guess I need some fun #movie #lateNightThoughts"

I want to remove the hashtags at the end of string which are #lateNightThoughts and #movie. Result:

 - "I didn't know it could be #boring. guess I need some fun"

I tried :

stringi::stri_replace_last_regex(x,'#\\S+',"")

but it removes only the very last hashtag.

- "I didn't know it could be #boring. guess I need some fun #movie "

Any idea how to get the expected result?

Edit:

How about removing hashtag from beginning of text ? eg:

x<- "#Thomas20 I didn't know it could be #boring. guess I need some fun #movie #lateNightThoughts"

Solution

You may use

>  x<- "I didn't know it could be #boring. guess I need some fun #movie #lateNightThoughts"
> sub("\\s*\\B#\\w+(?:\\s*#\\w+)*\\s*$", "", x)
[1] "I didn't know it could be #boring. guess I need some fun"

Or, if you do not care about the context of the first # you want to start matching from, you may even use

sub("(?:\\s*#\\w+)+\\s*$", "", x)

See the regex demo.

Details

\s* - zero or more whitespaces
\B - right before the current location, there can be start of string or a non-word char (this is usually used to ensure you do not match # inside a "word", so if you do not need it, you may remove this non-word boundary)
# - a # char
\w+ - 1 or more word chars (letters, digits or _)
(?:\s*#\w+)* - zero or more occurrences of:
- \s* - zero or more whitespaces
- # - a # char
- \w+ - 1+ word chars
\s* - zero or more whitespaces
$ - end of string.