# String#chop chops last byte, not char

**URL:** <https://www.ruby-forum.com/t/string-chop-chops-last-byte-not-char/135656>\
**Category:** Ruby\
**Created:** [April 23, 2008, 10:35pm UTC](https://www.ruby-forum.com/t/string-chop-chops-last-byte-not-char/135656 "2008-04-23T22:35:08Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![ebelin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/e/e19b73/32.png) [@ebelin](https://www.ruby-forum.com/u/ebelin)\
**Post date:** [April 23, 2008, 10:35pm UTC](https://www.ruby-forum.com/t/string-chop-chops-last-byte-not-char/135656/1 "2008-04-23T22:35:08Z")

</div>

This is happening in ruby 1.8.6:

% ruby --version  
ruby 1.8.6 (2007-09-23 patchlevel 110) [i686-darwin9.1.0]

Documentation says:

% ri String#chop  
…Returns a new +String+ with the last character removed. If…

This is irb and Rails script/console sessions:

% ./script/console  
Loading development environment (Rails 2.0.2)

> > “абвгд”.chop  
> > =\> “абвг\320”
> 
> > “абвгд”.chop.chop  
> > =\> “абвг”

% irb

> > “абвгд”.chop  
> > =\> “\320\260\320\261\320\262\320\263\320”
> 
> > “д”.chop  
> > =\> “\320”
> 
> > “д”.chop.chop  
> > =\> “”

As you can see chop removes last byte vs last char. Btw, problem  
happened while stuffing strings into legacy (ms sql server) db,  
varchar(255) column. I came up with the following method (see below).  
Maybe there is an easier alternative?

def truncate(text, size = 254, suffix = “…”)  
if text.nil? then return end  
l = size - suffix.size  
if text.size \> size  
truncated\_text = “”  
text.each\_char do |c|  
if truncated\_text.size + c.size \< l  
truncated\_text \<\< c  
else  
break  
end  
end  
truncated\_text += suffix  
else  
text  
end  
end

I did not report this as a bug, since I am not sure who is supposed to  
be right - ruby or documentation.

---

<div class="post-metadata">

**Author:** ![ebelin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/e/e19b73/32.png) [@ebelin](https://www.ruby-forum.com/u/ebelin)\
**Post date:** [April 23, 2008, 11:10pm UTC](https://www.ruby-forum.com/t/string-chop-chops-last-byte-not-char/135656/2 "2008-04-23T23:10:55Z")

</div>

On Wed, 23 Apr 2008 15:35:11 -0500, Evgeni Belin wrote:

> This is happening in ruby 1.8.6:
> 
> % ruby --version  
> ruby 1.8.6 (2007-09-23 patchlevel 110) [i686-darwin9.1.0]

> > > “д”.chop  
> > > =\> “\320”
> > 
> > > “д”.chop.chop  
> > > =\> “”
> 
> As you can see chop removes last byte vs last char. Btw, problem

Ruby 1.8 does not have multi-byte character support built in, so it  
assumes each character is one byte. If you would like unicode support,  
include in your scripts

$KCODE=‘u’  
require ‘jcode’

Then chop will work properly. (Though I’m not sure everything will be  
perfect)

Ruby 1.9 has unicode support built in and handles this properly out of  
the box.

---

<div class="post-metadata">

**Author:** ![ebelin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/e/e19b73/32.png) [@ebelin](https://www.ruby-forum.com/u/ebelin)\
**Post date:** [April 23, 2008, 11:12pm UTC](https://www.ruby-forum.com/t/string-chop-chops-last-byte-not-char/135656/3 "2008-04-23T23:12:41Z")

</div>

Evgeni Belin wrote:

> This is happening in ruby 1.8.6:

In Ruby 1.8.6, character == byte basically everywhere. You’ll need to  
use a regex to remove the last character safely (UTF-8).

- Charlie