utf-8 and python files about fhir-parser HOT 5 CLOSED

Hmmm, I have # -*- coding: utf-8 -*- specified in all executable Python files, is there another way to specify the encoding so Windows doesn't fail?

from fhir-parser.

Unfortunately, that coding refers to the python file's text encoding, not to the default encoding used to read or write text files. So, that allows unicode variable names, unicode comments, etc.

Sorry I wasn't clear enough above. You need to specify encoding when you open text files, by specifying the encoding kw argument.
example:
with file as open(path, 'r', encoding='utf-8'):
...

It's kind of idiomatic to do so when you want your code to run without fail under windows. So, I do it always, just as a matter of course. It's an unfortunate platform cross compatibility issue that bites a lot of folks and should be mentioned in any text the describes the use of os.path. But it often isn't.

from fhir-parser.

Ah I see. I use io.open with specified encoding for writing the classes, will apply to reading as well. I have no access to a Windows machine so I'll let you re-open the issue when I close it and it's still giving you headaches.

from fhir-parser.

One last comment on the opening of files: io.open() is an alias for the builtin open() There's no need to import io for most file access and processing.

from fhir-parser.

Yes. I like to use io.open to contrast with codecs.open when I specify the encoding (i.e. write more thorough code). More a useless habit than anything really.

from fhir-parser.