compat: add pptx.compat sub-package

Author: karthik1024

pptx/compat/__init__.py

@@ -0,0 +1,20 @@
+# encoding: utf-8
+
+"""
+Provides Python 2/3 compatibility objects
+"""
+
+from __future__ import (
+    absolute_import, division, print_function, unicode_literals
+)
+
+import sys
+
+if sys.version_info >= (3, 0):
+    from .python3 import (  # noqa
+        BytesIO, is_integer, is_string, is_unicode, to_unicode, Unicode
+    )
+else:
+    from .python2 import (  # noqa
+        BytesIO, is_integer, is_string, is_unicode, to_unicode, Unicode
+    )

pptx/compat/python2.py

@@ -0,0 +1,48 @@
+# encoding: utf-8
+
+"""
+Provides Python 2 compatibility objects
+"""
+
+from StringIO import StringIO as BytesIO  # noqa
+
+
+def is_integer(obj):
+    """
+    Return True if *obj* is an integer (int, long), False otherwise.
+    """
+    return isinstance(obj, (int, long))
+
+
+def is_string(obj):
+    """
+    Return True if *obj* is a string, False otherwise.
+    """
+    return isinstance(obj, basestring)
+
+
+def is_unicode(obj):
+    """
+    Return True if *obj* is a unicode string, False otherwise.
+    """
+    return isinstance(obj, unicode)
+
+
+def to_unicode(text):
+    """
+    Return *text* as a unicode string. *text* can be a 7-bit ASCII string,
+    a UTF-8 encoded 8-bit string, or unicode. String values are converted to
+    unicode assuming UTF-8 encoding. Unicode values are returned unchanged.
+    """
+    # both str and unicode inherit from basestring
+    if not isinstance(text, basestring):
+        tmpl = 'expected UTF-8 encoded string or unicode, got %s value %s'
+        raise TypeError(tmpl % (type(text), text))
+    # return unicode strings unchanged
+    if isinstance(text, unicode):
+        return text
+    # otherwise assume UTF-8 encoding, which also works for ASCII
+    return unicode(text, 'utf-8')
+
+
+Unicode = unicode

pptx/compat/python3.py

@@ -0,0 +1,42 @@
+# encoding: utf-8
+
+"""
+Provides Python 3 compatibility objects
+"""
+
+from io import BytesIO  # noqa
+
+
+def is_integer(obj):
+    """
+    Return True if *obj* is an int, False otherwise.
+    """
+    return isinstance(obj, int)
+
+
+def is_string(obj):
+    """
+    Return True if *obj* is a string, False otherwise.
+    """
+    return isinstance(obj, str)
+
+
+def is_unicode(obj):
+    """
+    Return True if *obj* is a unicode string, False otherwise.
+    """
+    return isinstance(obj, str)
+
+
+def to_unicode(text):
+    """
+    Return *text* as a unicode string. All text in Python 3 is unicode, so
+    this just returns *text* unchanged.
+    """
+    if not isinstance(text, str):
+        tmpl = 'expected unicode string, got %s value %s'
+        raise TypeError(tmpl % (type(text), text))
+    return text
+
+
+Unicode = str

pptx/oxml/text.py

@@ -7,6 +7,7 @@
 from __future__ import absolute_import
 
 from . import parse_xml
+from ..compat import to_unicode
 from ..enum.text import (
     MSO_AUTO_SIZE, MSO_TEXT_UNDERLINE_TYPE, MSO_VERTICAL_ANCHOR,
     PP_PARAGRAPH_ALIGNMENT
@@ -18,7 +19,7 @@
     ST_TextSpacingPoint, ST_TextTypeface, ST_TextWrappingType, XsdBoolean,
     XsdString
 )
-from ..util import Emu, Length, to_unicode
+from ..util import Emu, Length
 from .xmlchemy import (
     BaseOxmlElement, Choice, OneAndOnlyOne, OneOrMore, OptionalAttribute,
     RequiredAttribute, ZeroOrMore, ZeroOrOne, ZeroOrOneChoice

pptx/shapes/table.py

@@ -9,9 +9,10 @@
 from warnings import warn
 
 from . import Subshape
+from ..compat import to_unicode
 from ..dml.fill import FillFormat
 from ..text.text import TextFrame
-from ..util import lazyproperty, to_unicode
+from ..util import lazyproperty
 
 
 class Table(object):

pptx/text/text.py

@@ -6,6 +6,7 @@
 
 from __future__ import absolute_import, print_function
 
+from ..compat import to_unicode
 from ..dml.fill import FillFormat
 from ..enum.dml import MSO_FILL
 from ..enum.text import MSO_AUTO_SIZE, MSO_UNDERLINE
@@ -14,7 +15,7 @@
 from ..opc.constants import RELATIONSHIP_TYPE as RT
 from ..oxml.simpletypes import ST_TextWrappingType
 from ..shapes import Subshape
-from ..util import Centipoints, Emu, lazyproperty, Pt, to_unicode
+from ..util import Centipoints, Emu, lazyproperty, Pt
 
 
 class TextFrame(Subshape):

pptx/util.py

@@ -177,22 +177,3 @@ def get_prop_value(obj):
             return value
 
     return property(get_prop_value, doc=docstring)
-
-
-def to_unicode(text):
-    """
-    Return *text* as a unicode string.
-
-    *text* can be a 7-bit ASCII string, a UTF-8 encoded 8-bit string, or
-    unicode. String values are converted to unicode assuming UTF-8 encoding.
-    Unicode values are returned unchanged.
-    """
-    # both str and unicode inherit from basestring
-    if not isinstance(text, basestring):
-        tmpl = 'expected UTF-8 encoded string or unicode, got %s value %s'
-        raise TypeError(tmpl % (type(text), text))
-    # return unicode strings unchanged
-    if isinstance(text, unicode):
-        return text
-    # otherwise assume UTF-8 encoding, which also works for ASCII
-    return unicode(text, 'utf-8')

tests/test_util.py

@@ -9,9 +9,8 @@
 import platform
 import pytest
 
-from pptx.util import (
-    Length, Centipoints, Cm, Emu, Inches, Mm, Pt, Px, to_unicode
-)
+from pptx.compat import to_unicode
+from pptx.util import Length, Centipoints, Cm, Emu, Inches, Mm, Pt, Px
 
 
 def test_to_unicode_raises_on_non_string():