Python: ascii codec can't encode en-dash

Question

I'm trying to print a poem from the Poetry Foundation's daily poem RSS feed with a thermal printer that supports an encoding of CP437. This means I need to translate some characters; in this case an en-dash to a hyphen. But python won't even encode the en dash to begin with. When I try to decode the string and replace the en-dash with a hyphen I get the following error:

Traceback (most recent call last):
  File "pftest.py", line 46, in <module>
    str = str.decode('utf-8')
  File "/usr/lib/python2.7/encodings/utf_8.py", line 16, in decode
    return codecs.utf_8_decode(input, errors, True)
UnicodeEncodeError: 'ascii' codec can't encode character u'\u2013' in position 140: ordinal not in range(128)

And here is my code:

#!/usr/bin/python
#-*- coding: utf-8 -*-

# This string is actually a variable entitled d['entries'][1].summary_detail.value
str = """Love brought by night a vision to my bed,
One that still wore the vesture of a child
But eighteen years of age – who sweetly smiled"""

str = str.decode('utf-8')
str = str.replace("\u2013", "-") #en dash
str = str.replace("\u2014", "--") #em dash
print (str)

I can actually print the output using the following code without errors in my terminal window (Mac), but my printer spits out sets of 3 CP437 characters:

str = u''.str.encode('utf-8')

I'm using Sublime Text as my editor, and I've saved the page with UTF-8 encoding, but I'm not sure that will help things. I would greatly appreciate any help with this code. Thank you!

Answer 1

I don't fully understand what's happening in your code, but I've also been trying to replace en-dashes with hyphens in a string I got from the Web, and here's what's working for me. My code is just this:

txt = re.sub(u"\u2013", "-", txt)

I'm using Python 2.7 and Sublime Text 2, but I don't bother setting -*- coding: utf-8 -*- in my script, as I'm trying not to introduce any new encoding issues. (Even though my variables may contain Unicode I like to keep my code pure ASCII.) Do you need to include Unicode in your .py file, or was that just to help with debugging?

I'll note that my txt variable is already a unicode string, i.e.

print type(txt)

produces

<type 'unicode'>

I'd be curious to know what type(str) would produce in your case.

---

One thing I noticed in your code is

str = str.replace("\u2013", "-") #en dash

Are you sure that does anything? My understanding is that \u only means "unicode character' inside a u"" string, and what you've created there is a string with 5 characters, a "u", a "2", a "0", etc. (The first character is because you can escape any character and if there's no special meaning, like in the case of '\n' or '\t', it just ignores the backslash.)

Also, the fact that you get 3 CP437 characters from your printer makes me suspect that you still have an en-dash in your string. The UTF-8 encoding of an en-dash is 3 bytes: 0xe2 0x80 0x93. When you call str.encode('utf-8') on a unicode string that contains an en-dash you get those three bytes in the returned string. I'm guessing that your terminal knows how to interpret that as an en-dash and that's what you're seeing.

If you can't get my first method to work, I'll mention that I also had success with this:

txt = txt.encode('utf-8')
txt = re.sub("\xe2\x80\x93", "-", txt)

Maybe that re.sub() would work for you if you put it after your call to encode(). And in that case you might not even need that call to decode() at all. I'll confess that I really don't understand why it's there.