Remove non-letters from a string

From CodeCodex

Revision as of 23:22, 8 October 2007 by Tanktop (Talk | contribs)

Related content:

Implementations

Java

This code removes common symbols and numbers from a string, returning letters only. (Note, not all symbols are included.)

static String lettersOnly(String s) {
   return s.replaceAll("[0-9]|[ !@#\\$%\\^&\\*\\(\\)_\\+\\-={}\\|:\"<>\\?\\-=\\[\\];',\\./`~'£€¥]","");
}

OCaml

# let remove_nonalpha = Str.global_replace (Str.regexp "[^a-zA-Z]+") "";;
val remove_nonalpha : string -> string = <fun>

For example:

# remove_nonalpha "133t H4x0r";;
- : string = "tHxr"

Perl

<HIGHLIGHTSYNTAX language="perl">

s{[\W\d_]}{}g; # remove all non-word characters and digits and underscores

</HIGHLIGHTSYNTAX>

Python

from string import letters
s = "hello world! how are you? 0"

# Short version
print filter(lambda c: c in letters, s)

# Faster version for long ASCII strings:
id_tab = "".join(map(chr, xrange(256)))
tostrip = "".join(c for c in id_tab if c not in letters)
print s.translate(id_tab, tostrip)

Tcl

proc remove_non_ascii_letters s {regsub -all {[^a-zA-Z]} $s ""}
remove_non_ascii_letters "hello world! how are you? 0é" ;# -> helloworldhowareyou

Visual Basic

Public Function ReturnAlpha(ByVar sString As String) As String
Dim i As Integer
For i = 1 To Len(sString)
    If Mid(sString, i, 1) Like "[A-Za-z]" Then
        ReturnAlpha = ReturnAlpha + Mid(sString, i, 1)
    End If
Next i
End Function