Modernize and add python3 support

This had a dependency on PyXML, which does not support python3 (actually
does not even support python2.7 byt we have been using a patched version).
To drop this dependency, we use lxml sax interface directly.

Modernize several things:
 - enable the doctest
 - use lxml xml comparison to compare XML
 - use python3 compatible syntax

CHANGES.txt → CHANGES.rst

 History
 =======
 
-0.5 (unreleased)
+0.5 (2022-09-14)
 ----------------
-
+ - Support python3 and python2.7 by dropping dependency on PyXML
 
 0.4 (2010-01-21)
 ----------------

MANIFEST.in

+include *.py
+include *.rst
+include MANIFEST.in

README.txt → README.rst

@@ -4,16 +4,18 @@ Introduction
 Apply xupdate diff on XML documents.
 
 
-Installation
-============
+Testing
+=======
 
-python setup.py install
+To run tests::
 
+    python -m unittest discover src
 
-Testing
-=======
+or, using ``zc.buildout`` and ``zope.testrunner``::
+
+    buildout
+    ./bin/test
 
-python setup.py test
 
 Usage
 =====

buildout.cfg

@@ -5,4 +5,3 @@ parts = test
 [test]
 recipe = zc.recipe.testrunner
 eggs = xupdate_processor
-initialization = from shutil import copy2; from glob import glob; [copy2(f, '../../parts/test/%s' % f.split('/')[-1]) for f in glob('../../*.xml')]
\ No newline at end of file

setup.py

@@ -3,15 +3,16 @@
 
 from setuptools import setup, find_packages
 
-version = '0.4'
+version = '0.5'
 
 def read(name):
-    return open(name).read()
+    with open(name) as f:
+        return f.read()
 
 long_description=(
-        read('README.txt')
+        read('README.rst')
         + '\n' +
-        read('CHANGES.txt')
+        read('CHANGES.rst')
     )
 
 setup(name="xupdate-processor",
@@ -20,17 +21,16 @@ setup(name="xupdate-processor",
       long_description=long_description,
       author="Nicolas DELABY",
       author_email="nicolas@nexedi.com",
-      url="http://nexedi.com",
+      url="https://lab.nexedi.com/nexedi/xupdate_processor",
       license="GPL",
       packages=find_packages('src'),
       package_dir={'': 'src'},
       scripts=["xuproc"],
-      install_requires=['PyXML', 'lxml', 'erp5diff >= 0.7'],
+      install_requires=['six', 'lxml', 'erp5diff >= 0.7'],
       classifiers=['License :: OSI Approved :: GNU General Public License (GPL)',
                   'Operating System :: OS Independent',
                   'Topic :: Text Processing :: Markup :: XML',
                   'Topic :: Utilities'],
       include_package_data=True,
       zip_safe=False,
-      test_suite='xupdate_processor',
      )

src/xupdate_processor/__init__.py

 # -*- coding: utf-8 -*-
-from xuproc import applyXUpdate
+from __future__ import absolute_import
+from __future__ import print_function
+from .xuproc import applyXUpdate
 from lxml import etree
+import sys
 
 def main():
   from optparse import OptionParser
@@ -10,12 +13,13 @@ def main():
   options, args = parser.parse_args()
 
   if len(args) != 2:
-    print parser.print_help()
+    parser.print_help()
     parser.error('incorrect number of arguments')
+    sys.exit(2)
 
   xu_xml_name = args[0]
   doc_xml_name = args[1]
 
-  print etree.tostring(applyXUpdate(xml_xu_filename=xu_xml_name,
+  print(etree.tostring(applyXUpdate(xml_xu_filename=xu_xml_name,
                                     xml_doc_filename=doc_xml_name),
-                       pretty_print=True)
+                       pretty_print=True))

src/xupdate_processor/content_handler.py

 # -*- coding: utf-8 -*-
+from __future__ import absolute_import
 from lxml import etree
 from lxml.etree import Element, _ElementTree, ProcessingInstruction, Comment
 from xml.sax.handler import ContentHandler

src/xupdate_processor/tests/xupdate_processor_test_suite.py → src/xupdate_processor/tests/test_xupdate_processor.py

 # -*- coding: utf-8 -*-
+from __future__ import absolute_import
 import unittest
 import pkg_resources
+from io import BytesIO
+import doctest
 
 from ERP5Diff import ERP5Diff
 import xupdate_processor
 from xupdate_processor import applyXUpdate
+import lxml.doctestcompare
 from lxml import etree
-from cStringIO import StringIO
 
 class TestXUpdateProcessor(unittest.TestCase):
   """
@@ -17,9 +20,14 @@ class TestXUpdateProcessor(unittest.TestCase):
     """
     result_tree = applyXUpdate(xml_xu_string=xml_xu_string,
                               xml_doc_string=xml_doc_string)
-    result_string = etree.tostring(result_tree, pretty_print=True)
-    self.assertEquals(result_string, expected_result_string,
-                      '\n%s\n\n%s' % (result_string, expected_result_string))
+    result_string = etree.tostring(result_tree, pretty_print=True, encoding="unicode")
+    checker = lxml.doctestcompare.LXMLOutputChecker()
+    if not checker.check_output(expected_result_string, result_string, 0):
+      self.fail(
+        checker.output_difference(
+          doctest.Example("", expected_result_string),
+          result_string,
+          0))
 
 
   def test_textNodes(self):
@@ -683,13 +691,15 @@ class TestXUpdateProcessor(unittest.TestCase):
     result_tree = applyXUpdate(xml_xu_filename=xu_filename,
                                xml_doc_filename=orig_filename)
     destination_tree = etree.parse(destination_filename)
-    destination_buffer = StringIO()
-    result_buffer = StringIO()
+    destination_buffer = BytesIO()
+    result_buffer = BytesIO()
     destination_tree.write_c14n(destination_buffer)
+    destination_buffer.seek(0)
     result_tree.write_c14n(result_buffer)
+    result_buffer.seek(0)
     erp5diff = ERP5Diff()
-    erp5diff.compare(result_buffer.getvalue(), destination_buffer.getvalue())
-    self.assertEquals(etree.tostring(erp5diff._result),
+    erp5diff.compare(result_buffer, destination_buffer)
+    self.assertEqual(etree.tostring(erp5diff._result, encoding='unicode'),
     '<xupdate:modifications xmlns:xupdate="http://www.xmldb.org/xupdate" version="1.0"/>')
 
   def test_OOofiles2(self):
@@ -706,16 +716,23 @@ class TestXUpdateProcessor(unittest.TestCase):
                             'tests/dest_ooo_2.xml')
     result_tree = applyXUpdate(xml_xu_filename=xu_filename, xml_doc_filename=orig_filename)
     destination_tree = etree.parse(destination_filename)
-    destination_buffer = StringIO()
-    result_buffer = StringIO()
+    destination_buffer = BytesIO()
+    result_buffer = BytesIO()
     destination_tree.write_c14n(destination_buffer)
+    destination_buffer.seek(0)
     result_tree.write_c14n(result_buffer)
+    result_buffer.seek(0)
 
     erp5diff = ERP5Diff()
-    erp5diff.compare(result_buffer.getvalue(), destination_buffer.getvalue())
-    self.assertEquals(etree.tostring(erp5diff._result),
+    erp5diff.compare(result_buffer, destination_buffer)
+    self.assertEqual(etree.tostring(erp5diff._result, encoding='unicode'),
     '<xupdate:modifications xmlns:xupdate="http://www.xmldb.org/xupdate" version="1.0"/>')
 
 
+
+def load_tests(loader, tests, ignore):
+    tests.addTests(doctest.DocFileSuite("test_doctest.txt"))
+    return tests
+
 if __name__ == '__main__':
   unittest.main()
\ No newline at end of file

src/xupdate_processor/xuproc.py

 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-import xml.sax
+from __future__ import absolute_import
+from __future__ import print_function
+import lxml.sax
 from lxml import etree
-from cStringIO import StringIO
-from content_handler import XUpdateHandler
+from six import StringIO
+from .content_handler import XUpdateHandler
 
 def applyXUpdate(xml_xu_file=None,
                  xml_doc_file=None,
@@ -26,16 +28,12 @@ def applyXUpdate(xml_xu_file=None,
   if xml_doc_filename:
     xml_doc = xml_doc_filename
   original_tree = etree.parse(xml_doc, etree_parser)
-  parser = xml.sax.make_parser()
-  parser.setFeature(xml.sax.handler.feature_namespaces, True)
-  parser.setFeature(xml.sax.handler.feature_namespace_prefixes, True)
-  parser.setContentHandler(XUpdateHandler(original_tree=original_tree))
-  parser.parse(xml_xu)
-  content_handler = parser.getContentHandler()
+  content_handler = XUpdateHandler(original_tree=original_tree)
+  lxml.sax.saxify(etree.parse(xml_xu, etree_parser), content_handler)
   return content_handler.result_tree
 
 if __name__ == '__main__':
   import sys
   doc_xml_name = sys.argv[1]
   xu_xml_name = sys.argv[2]
-  print etree.tostring(applyXUpdate(xml_xu_filename=xu_xml_name, xml_doc_filename=doc_xml_name), pretty_print=True)
+  print(etree.tostring(applyXUpdate(xml_xu_filename=xu_xml_name, xml_doc_filename=doc_xml_name), pretty_print=True))