# HG changeset patch
# User Paul Boddie <paul@boddie.org.uk>
# Date 1481578395 -3600
# Node ID 0538ce057931e1a2de31385c7f8ead0720fe781b
# Parent  f326b1d00b210f347c6e6aaa7f46a4495911478b
Added initial support for Unicode strings based on byte strings.

diff -r f326b1d00b21 -r 0538ce057931 lib/__builtins__/__init__.py
--- a/lib/__builtins__/__init__.py	Mon Dec 12 22:30:09 2016 +0100
+++ b/lib/__builtins__/__init__.py	Mon Dec 12 22:33:15 2016 +0100
@@ -60,7 +60,6 @@
     ValueError
     )
 
-
 # Classes.
 
 from __builtins__.boolean import bool, False, True
@@ -77,8 +76,9 @@
 from __builtins__.none import None, NoneType
 from __builtins__.notimplemented import NotImplemented, NotImplementedType
 from __builtins__.set import frozenset, set
-from __builtins__.str import basestring, str, string, unicode
+from __builtins__.str import basestring, str, string
 from __builtins__.tuple import tuple
+from __builtins__.unicode import unicode, utf8string
 
 # Functions.
 
diff -r f326b1d00b21 -r 0538ce057931 lib/__builtins__/str.py
--- a/lib/__builtins__/str.py	Mon Dec 12 22:30:09 2016 +0100
+++ b/lib/__builtins__/str.py	Mon Dec 12 22:33:15 2016 +0100
@@ -19,7 +19,7 @@
 this program.  If not, see <http://www.gnu.org/licenses/>.
 """
 
-from __builtins__.int import maxint, minint
+from __builtins__.int import maxint
 from __builtins__.operator import _negate
 from __builtins__.sequence import itemaccess
 from __builtins__.types import check_int
@@ -33,21 +33,30 @@
     _p = maxint / 32
     _a = 31
 
-    def __init__(self):
+    def __init__(self, other=None):
 
-        "Initialise the string."
+        "Initialise the string, perhaps from 'other'."
 
         # Note the __data__ member. Since strings are either initialised from
         # literals or converted using routines defined for other types, no form
         # of actual initialisation is performed here.
 
-        self.__data__ = None
+        # NOTE: Cannot perform "other and other.__data__ or None" since the
+        # NOTE: __data__ attribute is not a normal attribute.
+
+        if other:
+            self.__data__ = other.__data__
+        else:
+            self.__data__ = None
 
         # Note the __key__ member. This is also initialised statically. Where
         # a string is the same as an attribute name, the __key__ member contains
         # attribute position and code details.
 
-        self.__key__ = None
+        if other:
+            self.__key__ = other.__key__
+        else:
+            self.__key__ = None
 
     def __hash__(self):
 
@@ -212,10 +221,10 @@
         return str_substr(self.__data__, start, end, step)
 
 class string(basestring):
-    pass
 
-class unicode(basestring):
-    def encode(self, encoding): pass
+    "A plain string of bytes."
+
+    pass
 
 def str(obj):
 
diff -r f326b1d00b21 -r 0538ce057931 lib/__builtins__/types.py
--- a/lib/__builtins__/types.py	Mon Dec 12 22:30:09 2016 +0100
+++ b/lib/__builtins__/types.py	Mon Dec 12 22:33:15 2016 +0100
@@ -32,7 +32,7 @@
 
     "Check the given string 's'."
 
-    if not _isinstance(s, string):
+    if not _isinstance(s, basestring):
         raise ValueError(s)
 
 # vim: tabstop=4 expandtab shiftwidth=4
diff -r f326b1d00b21 -r 0538ce057931 lib/__builtins__/unicode.py
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/lib/__builtins__/unicode.py	Mon Dec 12 22:33:15 2016 +0100
@@ -0,0 +1,60 @@
+#!/usr/bin/env python
+
+"""
+Unicode objects.
+
+Copyright (C) 2015, 2016 Paul Boddie <paul@boddie.org.uk>
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 3 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT
+ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
+details.
+
+You should have received a copy of the GNU General Public License along with
+this program.  If not, see <http://www.gnu.org/licenses/>.
+"""
+
+from __builtins__.str import basestring
+from posix.iconv import Converter
+
+class utf8string(basestring):
+
+    "A character string representation based on UTF-8."
+
+    def encode(self, encoding):
+
+        "Encode the string to the given 'encoding'."
+
+        from_utf8 = Converter("UTF-8", encoding)
+        try:
+            from_utf8.feed(self)
+            return str(from_utf8)
+        finally:
+            from_utf8.close()
+
+def unicode(s, encoding):
+
+    "Convert 's' to a Unicode object, interpreting 's' as using 'encoding'."
+
+    if isinstance(s, utf8string):
+        return s
+
+    # Obtain a string representation.
+
+    s = s.__str__()
+
+    # Convert the string to UTF-8.
+
+    to_utf8 = Converter(encoding, "UTF-8")
+    try:
+        to_utf8.feed(s)
+        return utf8string(str(to_utf8))
+    finally:
+        to_utf8.close()
+
+# vim: tabstop=4 expandtab shiftwidth=4