UnicodeDecodeError: 'utf8' codec can't decode byte 0xf4 in position 13: invalid continuation byte

[lmaroncelli]

Hi all,
when I try to rebuild tags I get this error in console

Exception in thread Thread-25:
Traceback (most recent call last):
  File ".\threading.py", line 532, in __bootstrap_inner
  File ".\threading.py", line 484, in run
  File "./ctagsplugin.py", line 118, in run
    result = func(*args, **kwargs)
  File "./ctagsplugin.py", line 862, in build_ctags
    recursive=recursive, cmd=command)
  File "./ctags.py", line 315, in build_ctags
    resort_ctags(tag_file)
  File "./ctags.py", line 348, in resort_ctags
    for line in fh:
  File ".\codecs.py", line 679, in next
  File ".\codecs.py", line 610, in next
  File ".\codecs.py", line 525, in readline
  File ".\codecs.py", line 472, in read
UnicodeDecodeError: 'utf8' codec can't decode byte 0xf4 in position 13: invalid continuation byte

[stephenfin]

This is a highly specific error - likely due to the contents of your files. It's likely you have non-unicode characters in your file. I can't really test for this as I don't use such characters in my files :)

Try to figure out which character(s) are causing issues. If you can suggest a fix, please file pull request. Otherwise, you'll should still be able to use the plugin - only the Show Symbols (Alt + S) command won't work.

PS: I'm going to close this issue for now and it's not something that can be fixed yet. Feel free to reopen if you disagree.

[wangzhen127]

I encountered similar errors. Sometimes it is hard to dig out where the bad character is. One suggestion is to modify 2 lines in function resort_ctags() in ctags.py to ignore or replace the bad character. For example, to replace the errors with some standard python character, add errors='replace' to the following two lines in function resort_ctags()

with codecs.open(tag_file, encoding='utf-8', errors='replace') as fh:

with codecs.open(tag_file+'_sorted_by_file', 'w', encoding='utf-8', errors='replace') as fw:

My UnicodeDecodeError are gone after manually doing the above.

See also:
http://docs.python.org/2/library/codecs.html#codec-base-classes
The default value for errors is 'strict'.