Fixes to the controller text model (#17527)

* Fixes to the controller text model

* Change command line args to text type
* Make display replace undecodable bytes with replacement chars.  This
  is only a problem on pyhton3 where surrogates can enter into the msg
  but sys.stdout doesn't know how to handle them.
* Remove a deprecated playbook syntax in unicode.yml

* Fix up run_cmd to change its parameters to byte string at appropriate times.

bin/ansible

@@ -91,9 +91,16 @@ if __name__ == '__main__':
             else:
                 raise
 
-        cli = mycli(sys.argv)
-        cli.parse()
-        exit_code = cli.run()
+        try:
+            args = [to_text(a, errors='surrogate_or_strict') for a in sys.argv]
+        except UnicodeError:
+            display.error('Command line args are not in utf-8, unable to continue.  Ansible currently only understands utf-8')
+            display.display(u"The full traceback was:\n\n%s" % to_text(traceback.format_exc()))
+            exit_code = 6
+        else:
+            cli = mycli(args)
+            cli.parse()
+            exit_code = cli.run()
 
     except AnsibleOptionsError as e:
         cli.parser.print_help()

lib/ansible/utils/cmd_functions.py

@@ -24,15 +24,26 @@ import shlex
 import subprocess
 import select
 
+from ansible.compat.six import PY2
+from ansible.module_utils._text import to_bytes
+
 def run_cmd(cmd, live=False, readsize=10):
 
     #readsize = 10
 
+    # On python2, shlex needs byte strings
+    if PY2:
+        cmd = to_bytes(cmd, errors='surrogate_or_strict')
     cmdargs = shlex.split(cmd)
+
+    # subprocess should be passed byte strings.  (on python2.6 it must be
+    # passed byte strtings)
+    cmdargs = [to_bytes(a, errors='surrogate_or_strict') for a in cmdargs]
+
     p = subprocess.Popen(cmdargs, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
 
-    stdout = ''
-    stderr = ''
+    stdout = b''
+    stderr = b''
     rpipes = [p.stdout, p.stderr]
     while True:
         rfd, wfd, efd = select.select(rpipes, [], rpipes, 1)
@@ -42,14 +53,14 @@ def run_cmd(cmd, live=False, readsize=10):
             if live:
                 sys.stdout.write(dat)
             stdout += dat
-            if dat == '':
+            if dat == b'':
                 rpipes.remove(p.stdout)
         if p.stderr in rfd:
             dat = os.read(p.stderr.fileno(), readsize)
             stderr += dat
             if live:
                 sys.stdout.write(dat)
-            if dat == '':
+            if dat == b'':
                 rpipes.remove(p.stderr)
         # only break out if we've emptied the pipes, or there is nothing to
         # read from and the process has finished.

lib/ansible/utils/display.py

@@ -124,7 +124,7 @@ class Display:
                 # Convert back to text string on python3
                 # We first convert to a byte string so that we get rid of
                 # characters that are invalid in the user's locale
-                msg2 = to_text(msg2, self._output_encoding(stderr=stderr))
+                msg2 = to_text(msg2, self._output_encoding(stderr=stderr), errors='replace')
 
             if not stderr:
                 fileobj = sys.stdout

test/integration/unicode.yml

@@ -32,7 +32,7 @@
         groups: 'ĪīĬĭ'
         ansible_ssh_host: 127.0.0.1
         ansible_connection: local
-      with_items: hostnames
+      with_items: "{{ hostnames }}"
 
     - name: 'A task with unicode extra vars'
       debug: var=extra_var