RF: add (partial?) python 2 compatibility

Make code somewhat more ugly by adding Python 2 compatibility.

I wasn't able to get the citeproc-test set up properly to test this
well.

Author: matthew-brett

.gitignore

@@ -1 +1,2 @@
 __pycache__
+*.pyc

citeproc/__init__.py

@@ -1,3 +1,4 @@
+from __future__ import absolute_import
 
 import os
 

citeproc/formatter/html.py

@@ -1,12 +1,15 @@
+try:
+    from html import escape
+except ImportError:
+    from cgi import escape
 
-from html import escape
-
+from ..py2compat import text_type
 
 def preformat(text):
-    return escape(str(text))
+    return escape(text_type(text))
 
 
-class TagWrapper(str):
+class TagWrapper(text_type):
     tag = None
     attributes = None
 
@@ -57,7 +60,7 @@ class SmallCaps(TagWrapper):
     attributes = {'style': 'font-variant:small-caps;'}
 
 
-class Bibliography(str):
+class Bibliography(text_type):
     bib_prefix = '<div class="csl-bib-body">'
     bib_suffix = '</div>'
     item_prefix = '  <div class="csl-entry">'
@@ -66,7 +69,7 @@ class Bibliography(str):
     def __new__(cls, items):
         output = [cls.bib_prefix]
         for text in items:
-            text = cls.item_prefix + str(text) + cls.item_suffix
+            text = cls.item_prefix + text_type(text) + cls.item_suffix
             output.append(text)
         output.append(cls.bib_suffix)
         return super().__new__(cls, '\n'.join(output))

citeproc/formatter/plain.py

@@ -1,18 +1,19 @@
+from ..py2compat import text_type
 
 def preformat(text):
     return text
 
 
-Italic = str
-Oblique = str
-Bold = str
-Light = str
-Underline = str
-Superscript = str
-Subscript = str
-SmallCaps = str
+Italic = text_type
+Oblique = text_type
+Bold = text_type
+Light = text_type
+Underline = text_type
+Superscript = text_type
+Subscript = text_type
+SmallCaps = text_type
 
-class Bibliography(str):
+class Bibliography(text_type):
     def __new__(cls, items):
-        items = map(str, items)
-        return super().__new__(cls, '\n'.join(items))
+        items = map(text_type, items)
+        return super(Bibliography, cls).__new__(cls, '\n'.join(items))

citeproc/frontend.py

@@ -35,7 +35,7 @@ class CitationStylesLocale(CitationStylesXML):
     def __init__(self, locale):
         locale_path = os.path.join(LOCALES_PATH, 'locales-{}.xml'.format(locale))
         try:
-            super().__init__(locale_path)
+            super(CitationStylesLocale, self).__init__(locale_path)
         except IOError:
             raise ValueError("'{}' is not a known locale".format(locale))
 
@@ -48,7 +48,7 @@ def __init__(self, style, locale=None):
         except TypeError:
             pass
         try:
-            super().__init__(style)
+            super(CitationStylesStyle, self).__init__(style)
         except IOError:
             raise ValueError("'{}' is not a known style".format(style))
         if locale is None:

citeproc/model.py

@@ -68,7 +68,7 @@ def loc(self):
         print('\n'.join(tree))
 
     def get_option(self, name):
-        return self.get(name, __class__._default_options[name])
+        return self.get(name, self._default_options[name])
 
     def get_macro(self, name):
         expression = "cs:macro[@name='{}'][1]".format(name)
@@ -195,7 +195,7 @@ def get_option(self, name):
         options = self.find('cs:style-options', self.nsmap)
         if options is None:
             raise IndexError
-        return options.get(name, __class__._default_options[name])
+        return options.get(name, self._default_options[name])
 
     def get_formatter(self):
         return self.style.formatter

citeproc/py2compat.py

@@ -0,0 +1,10 @@
+import sys
+
+PY2 = sys.version_info[0] < 3
+
+if PY2:
+    string_types = basestring,
+    text_type = unicode
+else:
+    string_types = str,
+    text_type = str

citeproc/source/__init__.py

@@ -36,7 +36,7 @@ def __getattr__(self, name):
 
     def __getitem__(self, key):
         try:
-            return super().__getitem__(key)
+            return super(CustomDict, self).__getitem__(key)
         except KeyError:
             raise VariableError
 
@@ -48,7 +48,7 @@ def __init__(self, key, type, **args):
         #required_or = [set(csl.VARIABLES)]
         optional = ({'uri', 'container_uri', 'contributor', 'date'} |
                     set(VARIABLES))
-        super().__init__(args, optional=optional)
+        super(Reference, self).__init__(args, optional=optional)
 
     def __repr__(self):
         return '{}({})'.format(self.__class__.__name__, self.key)
@@ -67,7 +67,7 @@ def __init__(self, **args):
             required = {'family'}
             optional = {'given', 'dropping-particle', 'non-dropping-particle',
                         'suffix'}
-        super().__init__(args, required, optional)
+        super(Name, self).__init__(args, required, optional)
 
     def parts(self):
         return (self.get('given'), self.get('family'),
@@ -78,7 +78,7 @@ def parts(self):
 class DateBase(CustomDict):
     def __init__(self, args, required=set(), optional=set()):
         optional = {'circa'} | optional
-        super().__init__(args, required, optional)
+        super(DateBase, self).__init__(args, required, optional)
         # defaults
         if 'circa' not in self:
             self['circa'] = False
@@ -92,7 +92,7 @@ def __init__(self, **args):
             raise TypeError('When specifying the day, you should also specify '
                             'the month')
         args = {key: int(value) for key, value in args.items()}
-        super().__init__(args, required, optional)
+        super(Date, self).__init__(args, required, optional)
 
     def sort_key(self):
         year = self.year
@@ -108,7 +108,7 @@ def is_nil(self):
 class LiteralDate(DateBase):
     def __init__(self, text, **args):
         self.text = text
-        super().__init__(args)
+        super(LiteralDate, self).__init__(args)
 
     def sort_key(self):
         return self.text
@@ -118,7 +118,7 @@ class DateRange(DateBase):
     def __init__(self, **args):
         required = {'begin'}
         optional = {'end'}
-        super().__init__(args, required, optional)
+        super(DateRange, self).__init__(args, required, optional)
 
     def sort_key(self):
         begin = self.begin.sort_key()
@@ -139,15 +139,15 @@ def __init__(self, **args):
                 args[key] = int(value)
             except ValueError:
                 pass
-        super().__init__(args, required, optional)
+        super(Pages, self).__init__(args, required, optional)
 
 
 class Citation(CustomDict):
     def __init__(self, cites, **kwargs):
         for cite in cites:
             cite.citation = self
         self.cites = cites
-        super().__init__(kwargs)
+        super(Citation, self).__init__(kwargs)
 
     def __repr__(self):
         cites = ', '.join([cite.key for cite in self.cites])
@@ -158,7 +158,7 @@ class CitationItem(CustomDict):
     def __init__(self, key, bibliography=None, **args):
         self.key = key.lower()
         optional = {'locator', 'prefix', 'suffix'}
-        super().__init__(args, optional=optional)
+        super(CitationItem, self).__init__(args, optional=optional)
 
     def __repr__(self):
         return '{}({})'.format(self.__class__.__name__, self.key)

citeproc/source/bibtex/bibparse.py

@@ -1,3 +1,6 @@
+from __future__ import print_function, unicode_literals
+
+from ...py2compat import string_types
 
 import unicodedata
 
@@ -6,7 +9,7 @@
 
 class BibTeXEntry(dict):
     def __init__(self, document_type, attributes):
-        super().__init__(attributes)
+        super(BibTeXEntry, self).__init__(attributes)
         self.document_type = document_type
 
 
@@ -26,7 +29,7 @@ class BibTeXParser(dict):
 
     def __init__(self, file_or_filename):
         try:
-            self.file = open(file_or_filename, 'r')
+            self.file = open(file_or_filename, 'rt')
         except TypeError:
             self.file = file_or_filename
         self.variables = {}
@@ -47,7 +50,7 @@ def _parse(self, file):
         self._parse_preamble(self._preamble)
         for key, entry in self.items():
             for attribute, value in entry.items():
-                if isinstance(value, str):
+                if isinstance(value, string_types):
                     entry[attribute] = self._expand_macros(value)
 
     def _parse_entry(self, file):
@@ -164,7 +167,7 @@ def _parse_variable(self, file, char):
     def _parse_integer(self, file, char):
         integer = ''
         restore_point = file.tell()
-        while char.isnumeric():
+        while char.isdigit():
             integer += char
             restore_point = file.tell()
             char = file.read(1)

citeproc/source/bibtex/bibtex.py

@@ -1,4 +1,5 @@
-
+from __future__ import print_function, unicode_literals
+
 import re
 
 from warnings import warn
@@ -195,10 +196,10 @@ def _parse_author(self, authors):
         csl_authors = []
         for author in authors.split(' and '):
             if ',' in author:
-                family, given = map(str.strip, author.split(',', 1))
+                family, given = [a.strip() for a in  author.split(',', 1)]
                 name = Name(family=family, given=given)
             elif ' ' in author:
-                given, family = map(str.strip, author.rsplit(' ', 1))
+                given, family = [a.strip() for a in author.rsplit(' ', 1)]
                 name = Name(family=family, given=given)
             else:
                 # TODO: handle 'others'

citeproc/string.py

@@ -1,6 +1,8 @@
+from __future__ import print_function, unicode_literals
 
+from .py2compat import text_type
 
-class String(str):
+class String(text_type):
     def __radd__(self, other):
         return MixedString([other]).__add__(self)
 
@@ -11,16 +13,16 @@ def __iadd__(self, other):
         return self.__add__(other)
 
     def replace(self, *args, **kwargs):
-        return self.__class__(super().replace(*args, **kwargs))
+        return self.__class__(super(String, self).replace(*args, **kwargs))
 
     def rstrip(self, *args, **kwargs):
-        return self.__class__(super().rstrip(*args, **kwargs))
+        return self.__class__(super(String, self).rstrip(*args, **kwargs))
 
     def lower(self):
-        return self.__class__(super().lower())
+        return self.__class__(super(String, self).lower())
 
     def upper(self):
-        return self.__class__(super().upper())
+        return self.__class__(super(String, self).upper())
 
     def soft_lower(self):
         return self.lower()
@@ -39,9 +41,10 @@ def words(self):
 class MixedString(list):
     def __add__(self, other):
         try:
-            return self.__class__(super().__add__(other))
+            return self.__class__(super(MixedString, self).__add__(other))
         except TypeError:
-            return self.__class__(super().__add__(__class__([other])))
+            return self.__class__(super(MixedString, self).__add__(
+                self.__class__([other])))
 
     def __radd__(self, other):
         return self.__class__([other]).__add__(self)
@@ -50,7 +53,7 @@ def __iadd__(self, other):
         return self.__add__(other)
 
     def __str__(self):
-        return ''.join(map(str, self))
+        return ''.join(map(text_type, self))
 
     def __getitem__(self, index):
         return str(self)[index]
@@ -80,7 +83,7 @@ def isupper(self):
         return all(string.isupper() for string in self)
 
     def split(self, *args, **kwargs):
-        return str(self).split(*args, **kwargs)
+        return text_type(self).split(*args, **kwargs)
 
     def rstrip(self, *args, **kwargs):
         rev_iter = reversed(self)
@@ -96,7 +99,7 @@ def words(self):
 
 class NoCase(String):
     def __repr__(self):
-        return '{}({})'.format(self.__class__.__name__, str(self))
+        return '{}({})'.format(self.__class__.__name__, text_type(self))
 
     def soft_lower(self):
         return self

examples/bibtex.py

@@ -1,3 +1,4 @@
+from __future__ import print_function, unicode_literals
 
 # The references are parsed from a BibTeX database, so we import the
 # corresponding parser.

tests/string.py tests/check_string.py

@@ -1,4 +1,5 @@
-
+from __future__ import (print_function, unicode_literals, absolute_import,
+                        division)
 
 import glob
 import io
@@ -215,8 +216,12 @@ def main():
             failed.append(test_name)
 
     def print_result(name, passed, total):
-        print('{:<13} {:>3} / {:>3} ({:>4.0%})'
-              .format(name, passed, total, passed / total), file=destination)
+        if total == 0:
+            print('<no tests found>: check README.md file for instructions')
+        else:
+            print('{:<13} {:>3} / {:>3} ({:>4.0%})'.format(
+                name, passed, total, passed / total),
+                file=destination)
 
     print('Failed tests:')
     for test_name in failed: