MRAB python at mrabarnett.plus.com
Sun Mar 19 23:36:45 EDT 2017

• Previous message (by thread): python script Non-ASCII character
• Next message (by thread): how to corss-compile azure-iot-sdk-python
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

---

On 2017-03-20 02:50, eryk sun wrote:
> On Sun, Mar 19, 2017 at 11:06 PM, MRAB <python at mrabarnett.plus.com> wrote:
>>
>> If you're using Unicode string literals, your choices are:
>>
>> 1. Raw string literals:
>>
>>     var1 = ur"C:\Users\username\Desktop\η γλωσσα μου\mylanguage\myfile"
>
> Raw unicode literals are practically useless in Python 2. They're not
> actually raw because \u and \U are still parsed as Unicode escapes,
> e.g. ur"C:\Users" and ur"C:\users" are syntax errors.
>
Thanks for the correction.

I haven't used Python 2 for a long time (apart from running the unit 
tests when I make a new release of the regex module)!

>> 2. Slashes:
>>
>>     var1 = u"C:/Users/username/Desktop/η γλωσσα μου/mylanguage/myfile"
>
> I prefer to normalize a path with os.path.normpath, or pathlib.Path in
> Python 3. This converts slashes to backslashes to get a canonical
> Windows path. IMO, it's also visually better in a text representation
> or error messages. Otherwise at runtime joined paths often end up with
> mixed slashes, which to me is ugly.
>

---

• Previous message (by thread): python script Non-ASCII character
• Next message (by thread): how to corss-compile azure-iot-sdk-python
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

---

More information about the Python-list mailing list