Encode differences between idle python and python

[pretoriano_2001]

Hello:
Under win32 XP y select python command line and execute next code with
results indicated:

Python 2.5 (r25:51908, Sep 19 2006, 09:52:17) [MSC v.1310 32 bit
(Intel)] on
Type "help", "copyright" , "credits" or "license" for more information.

>>u=u'áéíóú'
u

u'\xe1\xe9\xed\ xf3\xfa'

>>print u

áéíóú

>>a=u.encode('l atin-1')
a

'\xe1\xe9\xed\x f3\xfa'

>>print a

ßÚݾ·

>>type(a)

<type 'str'>

>>type(u)

<type 'unicode'>

>>>

using python IDLE I repeat the code, but get next differen result:
IDLE 1.2

>>u=u'áéíóú'
u

u'\xe1\xe9\xed\ xf3\xfa'

>>print u

áéíóú

>>a=u.encode('l atin-1')
a

'\xe1\xe9\xed\x f3\xfa'

>>print a

áéíóú

>>type(a)

<type 'str'>

>>type(u)

<type 'unicode'>

>>>

What do you think is happending and how can I solve this ? The IDLE
looks fine but command line has problems.

[Peter Otten]

pr************* @hotmail.com wrote:

>u=u'áéíóú'
u

u'\xe1\xe9\xed\ xf3\xfa'

>print u

áéíóú

>a=u.encode('la tin-1')
a

'\xe1\xe9\xed\x f3\xfa'

>print a

ßÚݾ·

That means that Python is better at guessing the correct encoding than you
are. Here's how you can make it share its secrets:

>>import sys
sys.stdout.en coding

'UTF-8' # something else on your machine (cp850, maybe)

Then you can use that encoding to print:

>>your_encodi ng = sys.stdout.enco ding
print u"áéíóú".encode (your_encoding)

áéíóú

On the other hand: why not always print the unicode string directly?

Peter

[Gabriel Genellina]

At Tuesday 10/10/2006 02:44, pr************* @hotmail.com wrote:

>Hello:
Under win32 XP y select python command line and execute next code with
results indicated:

Python 2.5 (r25:51908, Sep 19 2006, 09:52:17) [MSC v.1310 32 bit
(Intel)] on
Type "help", "copyright" , "credits" or "license" for more information.

>u=u'áéíóú'
u

u'\xe1\xe9\xed \xf3\xfa'

>print u

áéíóú

>a=u.encode('la tin-1')
a

'\xe1\xe9\xed\ xf3\xfa'

>print a

ßÚݾ·

>type(a)

<type 'str'>

>type(u)

<type 'unicode'>

>>

Because the console code page != windows code page.
Exit Python. At the console prompt, type:

>chcp

If it says 850 - your console is using codepage 850.
Enter Python again, and replace 'latin-1' with
'cp850'. You should get the right representation.
--
Gabriel Genellina
Softlab SRL


_______________ _______________ _______________ _____
Preguntá. Respondé. Descubrí.
Todo lo que querías saber, y lo que ni imaginabas,
está en Yahoo! Respuestas (Beta).
¡Probalo ya!
http://www.yahoo.com.ar/respuestas

[pretoriano_2001]

Gabriel, Peter:
Many thanks for your clear answers!!
Best regards.

Vizcayno

[Neil Cerutti]

On 2006-10-10, pr************* @hotmail.com <pr************ *@hotmail.comwr ote:

>
Gabriel, Peter:
Many thanks for your clear answers!! Best regards.

Something I've been working on is currently using the following
trick:

# Create some string of non-ASCII text in ISO 8859-1.
some_string = ''.join(chr(a) for a in range(0xc0, 0xdf)).decode(' ISO 8859-1')
# Print it to stdout, converting to the terminal's encoding, replacing
# unprintable characters with '?'.
print some_string.enc ode(sys.stdout. encoding, 'replace')

--
Neil Cerutti
That's the biggest laughingstock I've ever heard of in my life.
--Trot Nixon