Trim Characters from the Left and/or Right Side of a String
These functions may be used, e.g., to remove unnecessary
white-spaces from strings. Trimming ends at the first or
starts at the last pattern
match.
stri_trim_both(str, pattern = "\\P{Wspace}") stri_trim_left(str, pattern = "\\P{Wspace}") stri_trim_right(str, pattern = "\\P{Wspace}") stri_trim(str, side = c("both", "left", "right"), pattern = "\\P{Wspace}")
str |
a character vector of strings to be trimmed |
pattern |
a single pattern, specifying character
classes that should be preserved (see stringi-search-charclass).
Defaults to ' |
side |
character [ |
Vectorized over str
and pattern
.
stri_trim
is a convenience wrapper over stri_trim_left
and stri_trim_right
.
Contrary to many other string processing libraries,
our trimming functions are universal. A character class,
given by pattern
, may be adjusted to suit your needs
(yet, most often you stick to the default value).
For replacing pattern matches with
arbitrary replacement string, see stri_replace
.
Trimming can also be used where you would normally rely on
regular expressions. For instance, you may get
'23.5'
out of 'total of 23.5 bitcoins'
.
For trimming white-spaces, please note the difference
between Unicode binary property '\p{Wspace}
' (more universal)
and general character category '\p{Z}
',
see stringi-search-charclass.
All these functions return a character vector.
Other search_replace:
about_search
,
stri_replace_all()
Other search_charclass:
about_search_charclass
,
about_search
stri_trim_left(' aaa') stri_trim_right('r-project.org/', '\\p{P}') stri_trim_both(' Total of 23.5 bitcoins. ', '\\p{N}') stri_trim_both(' Total of 23.5 bitcoins. ', '\\p{L}')
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.