Support terminal types, such as kermit and avatar, that use bytes 127-255 in their escape sequences. Close #47.1.6

3 files changed, 15 insertions, 7 deletions

diff --git a/README.rst b/README.rst
index 7336ebd..b7f2401 100644
--- a/README.rst
+++ b/README.rst
@@ -441,6 +441,8 @@ Version History
   * Add ``fullscreen()`` and ``hidden_cursor()`` to the auto-generated docs.
   * Fall back to ``LINES`` and ``COLUMNS`` environment vars to find height and
     width. (jquast)
+  * Support terminal types, such as kermit and avatar, that use bytes 127-255
+    in their escape sequences. (jquast)
 
 1.5.1
   * Clean up fabfile, removing the redundant ``test`` command.
diff --git a/blessings/__init__.py b/blessings/__init__.py
index 4506869..04b10a5 100644
--- a/blessings/__init__.py
+++ b/blessings/__init__.py
@@ -376,10 +376,8 @@ class Terminal(object):
         """
         code = tigetstr(self._sugar.get(atom, atom))
         if code:
-            # We can encode escape sequences as UTF-8 because they never
-            # contain chars > 127, and UTF-8 never changes anything within that
-            # range..
-            return code.decode('utf-8')
+            # See the comment in ParametrizingString for why this is latin1.
+            return code.decode('latin1')
         return u''
 
     def _resolve_color(self, color):
@@ -448,7 +446,15 @@ class ParametrizingString(unicode):
             # Re-encode the cap, because tparm() takes a bytestring in Python
             # 3. However, appear to be a plain Unicode string otherwise so
             # concats work.
-            parametrized = tparm(self.encode('utf-8'), *args).decode('utf-8')
+            #
+            # We use *latin1* encoding so that bytes emitted by tparm are
+            # encoded to their native value: some terminal kinds, such as
+            # 'avatar' or 'kermit', emit 8-bit bytes in range 0x7f to 0xff.
+            # latin1 leaves these values unmodified in their conversion to
+            # unicode byte values. The terminal emulator will "catch" and
+            # handle these values, even if emitting utf8-encoded text, where
+            # these bytes would otherwise be illegal utf8 start bytes.
+            parametrized = tparm(self.encode('latin1'), *args).decode('latin1')
             return (parametrized if self._normal is None else
                     FormattingString(parametrized, self._normal))
         except curses.error:
diff --git a/blessings/tests.py b/blessings/tests.py
index 168c396..aff3a2f 100644
--- a/blessings/tests.py
+++ b/blessings/tests.py
@@ -28,12 +28,12 @@ TestTerminal = partial(Terminal, kind='xterm-256color')
 
 def unicode_cap(cap):
     """Return the result of ``tigetstr`` except as Unicode."""
-    return tigetstr(cap).decode('utf-8')
+    return tigetstr(cap).decode('latin1')
 
 
 def unicode_parm(cap, *parms):
     """Return the result of ``tparm(tigetstr())`` except as Unicode."""
-    return tparm(tigetstr(cap), *parms).decode('utf-8')
+    return tparm(tigetstr(cap), *parms).decode('latin1')
 
 
 def test_capability():