try and fix unicode filenames

Author: kouk

BigStash/upload.py

@@ -26,6 +26,8 @@
 
 from __future__ import print_function
 import boto3
+import six
+import sys
 import os
 import errno
 import sys
@@ -44,9 +46,15 @@
 log = logging.getLogger('bigstash.upload')
 
 
+def smart_str(s):
+    if isinstance(s, six.text_type):
+        return s
+    return s.decode('utf-8')
+
+
 class ProgressPercentage(object):
     def __init__(self, filename):
-        self._filename = filename
+        self._filename = smart_str(filename)
         self._size = os.path.getsize(filename)
         self._seen_so_far = 0
         self._lock = threading.Lock()
@@ -55,7 +63,7 @@ def _write_progress(self, wrote, total):
         percentage = 100
         if total > 0:
             percentage = (wrote / float(total)) * 100
-        sys.stdout.write("\r{} {} / {} ({:.2f}%)".format(
+        sys.stdout.write(u"\r{} {} / {} ({:.2f}%)".format(
             self._filename, wrote, total, percentage))
         sys.stdout.flush()
 
@@ -72,7 +80,12 @@ def __call__(self, bytes_amount):
 
 
 def main():
-    args = docopt(__doc__, version=__version__)
+    argv = sys.argv
+    if not six.PY3 and os.name == 'nt':
+        from BigStash.winargvfix import fix_argv_on_windows, fix_env_on_windows
+        argv = fix_argv_on_windows()
+        fix_env_on_windows()
+    args = docopt(__doc__, argv=argv[1:], version=__version__)
 
     settings = BigStashAPISettings.load_settings()
     BigStashAPI.setup_logging(settings)
@@ -145,15 +158,12 @@ def bgst_settings(args, settings):
 
 def bgst_put(args, settings):
     try:
-
         title = args['--title'] if args['--title'] else None
         opt_silent = False if not args['--silent'] else True
         opt_dont_wait = False if not args['--dont-wait'] else True
         upload = None
-        manifest, errors = Manifest.from_paths(
-            paths=[f.decode('utf-8') for f in args['FILES']],
-            title=title
-            )
+        filepaths = map(smart_str, args['FILES'])
+        manifest, errors = Manifest.from_paths(paths=filepaths, title=title)
         if errors:
             errtext = [": ".join(e) for e in errors]
             print("\n".join(["There were errors:"] + errtext))

BigStash/winargvfix.py

@@ -0,0 +1,52 @@
+import os
+import sys
+from ctypes import (WINFUNCTYPE, windll, POINTER, byref, c_int,
+                    create_unicode_buffer, c_wchar_p)
+from ctypes.wintypes import LPWSTR, LPCWSTR
+
+
+def fix_env_on_windows():
+    def getEnvironmentVariable(name):
+        name= unicode(name) # make sure string argument is unicode
+        n= windll.kernel32.GetEnvironmentVariableW(name, None, 0)
+        if n==0:
+            return None
+        buf= create_unicode_buffer(u'\0'*n)
+        windll.kernel32.GetEnvironmentVariableW(name, buf, n)
+        return buf.value
+    for k in os.environ.keys():
+        os.environ[k] = getEnvironmentVariable(k).encode('utf-8')
+
+
+def fix_argv_on_windows():
+    # This works around <http://bugs.python.org/issue2128>.
+    GetCommandLineW = windll.kernel32.GetCommandLineW
+    GetCommandLineW.restype = c_wchar_p
+    CommandLineToArgvW = windll.shell32.CommandLineToArgvW
+    CommandLineToArgvW.restype = POINTER(c_wchar_p)
+
+    argc = c_int(0)
+    argv_unicode = CommandLineToArgvW(GetCommandLineW(), byref(argc))
+
+    argv = [argv_unicode[i].encode('utf-8') for i in range(0, argc.value)]
+
+    if not hasattr(sys, 'frozen'):
+        # If this is an executable produced by py2exe or bbfreeze, then it will
+        # have been invoked directly. Otherwise, unicode_argv[0] is the Python
+        # interpreter, so skip that.
+        argv = argv[1:]
+
+        # Also skip option arguments to the Python interpreter.
+        while len(argv) > 0:
+            arg = argv[0]
+            if not arg.startswith("-") or arg == "-":
+                break
+            argv = argv[1:]
+            if arg == '-m':
+                # sys.argv[0] should really be the absolute path of the module source,
+                # but never mind
+                break
+            if arg == '-c':
+                argv[0] = '-c'
+                break
+    return argv