If file or path contains non ascii character, tern_for_vim won't work on Windows. about tern_for_vim

[marijnh]

Attached patch fixes two problems I ran into. If you find more, could you file bugs where you describe exactly what you did and what went wrong?

[sys9kdr]

I closed this issue by mistake, sorry.

On Windows I still have this issue.

1. Opening a file contains non-ascii char in file name. (e.g. c:\大根.js)
2. Press in insert mode.
3. Omni completion doesn't work and vim return error code.

Vim return this code.

Error detected while processing function tern#Complete:
line    2:
Traceback (most recent call last):
Error detected while processing function tern#Complete:
line    2:
  File "<string>", line 1, in <module>
Error detected while processing function tern#Complete:
line    2:
  File "C:\Users\username\.vim\bundle\tern_for_vim/script/tern.py", line 231, in tern_ensureCompletionCached
Error detected while processing function tern#Complete:
line    2:
    {"line": curRow - 1, "ch": curCol})
Error detected while processing function tern#Complete:
line    2:
  File "C:\Users\username\.vim\bundle\tern_for_vim/script/tern.py", line 165, in tern_runCommand
Error detected while processing function tern#Complete:
line    2:
    doc["files"].append(tern_fullBuffer())
Error detected while processing function tern#Complete:
line    2:
  File "C:\Users\username\.vim\bundle\tern_for_vim/script/tern.py", line 123, in tern_fullBuffer
Error detected while processing function tern#Complete:
line    2:
    "name": tern_relativeFile(),
Error detected while processing function tern#Complete:
line    2:
  File "C:\Users\username\.vim\bundle\tern_for_vim/script/tern.py", line 111, in tern_relativeFile
Error detected while processing function tern#Complete:
line    2:
    filename = vim.eval("expand('%:p')").decode('utf-8')
Error detected while processing function tern#Complete:
line    2:
  File "C:\Python27\Lib\encodings\utf_8.py", line 16, in decode
Error detected while processing function tern#Complete:
line    2:
    return codecs.utf_8_decode(input, errors, True)
Error detected while processing function tern#Complete:
line    2:
UnicodeDecodeError: 'utf8' codec can't decode byte 0x91 in position 3: invalid start byte

I can avoid to use non-ascii char file, so I keep this issue closing.

[marijnh]

The data that is causing the error is coming from vim.eval. Unfortunately, I have no idea what encoding that is supposed to return. I wouldn't be too surprised if it was different from platform to platform.