Description of problem:
When trying to do a history list with user names rather than command lines I get a UnicodeDecodeError exception.  Some experimentation led me to the conclusion that is because of non-ascii letters in my login and full names.

Version-Release number of selected component (if applicable):
dnf-0.5.2-1.fc20.noarch

How reproducible:
Every time

Steps to Reproduce:
1. LANG=en_US.utf8 dnf --setopt=history_list_view=users history

Actual results:
ID     | Login user               | Date a | Action | Altere
-------------------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/bin/dnf", line 36, in <module>
    main.user_main(sys.argv[1:], exit_code=True)
  File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 206, in user_main
    errcode = main(args)
  File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 82, in main
    return _main(base, args)
  File "/usr/lib/python2.7/site-packages/dnf/cli/main.py", line 132, in _main
    cli.run()
  File "/usr/lib/python2.7/site-packages/dnf/cli/cli.py", line 1163, in run
    return self.command.run(self.base.extcmds)
  File "/usr/lib/python2.7/site-packages/dnf/cli/commands/__init__.py", line 1530, in run
    ret = self.output.historyListCmd(extcmds)
  File "/usr/lib/python2.7/site-packages/dnf/cli/output.py", line 1668, in historyListCmd
    name = self._pwd_ui_username(old.loginuid, 24)
  File "/usr/lib/python2.7/site-packages/dnf/cli/output.py", line 1513, in _pwd_ui_username
    fullname = _safe_split_0(user.pw_gecos, ';', 2)
  File "/usr/lib/python2.7/site-packages/dnf/cli/output.py", line 1506, in _safe_split_0
    ret = text.split(*args)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 1: ordinal not in range(128)


Expected results:
A history list with my name in most of the transactions.

Additional info:
Both my login id and my full name contain non-ascii letters.  Having a non-ascii login id might be stretching the the standards a bit.  But having a non ascii full name (gecos) must be considered quite normal.

I made the following patch which seems to work around the problem:

--- output.py.orig	2014-05-28 10:27:24.000000000 +0200
+++ output.py	2014-07-18 22:49:55.000000000 +0200
@@ -1510,8 +1510,8 @@
 
         try:
             user = pwd.getpwuid(uid)
-            fullname = _safe_split_0(user.pw_gecos, ';', 2)
-            name = "%s <%s>" % (fullname, user.pw_name)
+            fullname = _safe_split_0(ucd(user.pw_gecos), ';', 2)
+            name = "%s <%s>" % (fullname, ucd(user.pw_name))
             if limit is not None and len(name) > limit:
                 name = "%s ... <%s>" % (_safe_split_0(fullname), user.pw_name)
                 if len(name) > limit:

If this is the right thing to do, I don't know.  Encodings in Python are a mystery to me!

Bug 1108908, on its way to be fixed, concerns another unicode issue in dnf.