Make tools/ compatible with both, Python 2 and 3 without 2to3-conversion.

git-svn-id: http://svn.code.sf.net/p/docutils/code/trunk/docutils@7442 929543f6-e4f2-0310-98a6-ba3bd3dd1d04

3 files changed, 42 insertions, 32 deletions

diff --git a/tools/dev/create_unimap.py b/tools/dev/create_unimap.py
index 0f733c69f..9fb75bcc7 100755
--- a/tools/dev/create_unimap.py
+++ b/tools/dev/create_unimap.py
@@ -13,8 +13,15 @@ from xml.dom import minidom
 import sys
 import pprint
 
+if sys.version_info >= (3,0):
+    unicode = str
+else:
+    bytes = str
+    chr = unichr
+
+
 def w(s):
-    if isinstance(s, unicode):
+    if sys.version_info >= (3,0) and isinstance(s, unicode):
         s = s.encode('utf8')
     sys.stdout.write(s)
 
@@ -39,9 +46,9 @@ class Visitor:
                     continue
                 latex_code = n.childNodes[0].nodeValue.encode('ascii').strip()
                 if node.attributes['mode'].value == 'math':
-                    math_map[unichr(int(code))] = '$%s$' % latex_code
+                    math_map[chr(int(code))] = '$%s$' % latex_code
                 else:
-                    text_map[unichr(int(code))] = '{%s}' % latex_code
+                    text_map[chr(int(code))] = '{%s}' % latex_code
 
 def call_visitor(node, visitor=Visitor()):
     if isinstance(node, minidom.Text):
@@ -63,16 +70,16 @@ unicode_map.update(text_map)
 # Now unicode_map contains the text entries plus dollar-enclosed math
 # entries for those chars for which no text entry exists.
 
-print '# $%s$' % 'Id'
-print '# Author: Lea Wiemann <LeWiemann@gmail.com>'
-print '# Copyright: This file has been placed in the public domain.'
-print
-print '# This is a mapping of Unicode characters to LaTeX equivalents.'
-print '# The information has been extracted from'
-print '# <http://www.w3.org/2003/entities/xml/unicode.xml>, written by'
-print '# David Carlisle and Sebastian Rahtz.'
-print '#'
-print '# The extraction has been done by the "create_unimap.py" script'
-print '# located at <http://docutils.sf.net/tools/dev/create_unimap.py>.'
-print
-print 'unicode_map = %s' % pprint.pformat(unicode_map, indent=0)
+print('# $%s$' % 'Id')
+print('# Author: Lea Wiemann <LeWiemann@gmail.com>')
+print('# Copyright: This file has been placed in the public domain.')
+print('')
+print('# This is a mapping of Unicode characters to LaTeX equivalents.')
+print('# The information has been extracted from')
+print('# <http://www.w3.org/2003/entities/xml/unicode.xml>, written by')
+print('# David Carlisle and Sebastian Rahtz.')
+print('#')
+print('# The extraction has been done by the "create_unimap.py" script')
+print('# located at <http://docutils.sf.net/tools/dev/create_unimap.py>.')
+print('')
+print('unicode_map = %s' % pprint.pformat(unicode_map, indent=0))
diff --git a/tools/dev/profile_docutils.py b/tools/dev/profile_docutils.py
index 90b083787..d31ba6fd0 100755
--- a/tools/dev/profile_docutils.py
+++ b/tools/dev/profile_docutils.py
@@ -8,25 +8,25 @@ import os.path
 import docutils.core
 import hotshot.stats
 
-print 'Profiler started.'
+print('Profiler started.')
 
 os.chdir(os.path.join(os.path.dirname(docutils.__file__), '..'))
 
-print 'Profiling...'
+print('Profiling...')
 
 prof = hotshot.Profile('docutils.prof')
 prof.runcall(docutils.core.publish_file, source_path='HISTORY.txt',
              destination_path='prof.HISTORY.html', writer_name='html')
 prof.close()
 
-print 'Loading statistics...'
+print('Loading statistics...')
 
-print """
+print("""
 stats = hotshot.stats.load('docutils.prof')
 stats.strip_dirs()
 stats.sort_stats('time')  # 'cumulative'; 'calls'
 stats.print_stats(40)
-"""
+""")
 
 stats = hotshot.stats.load('docutils.prof')
 stats.strip_dirs()
diff --git a/tools/dev/unicode2rstsubs.py b/tools/dev/unicode2rstsubs.py
index 22ac07ecc..b4e65e109 100755
--- a/tools/dev/unicode2rstsubs.py
+++ b/tools/dev/unicode2rstsubs.py
@@ -27,12 +27,12 @@ import re
 from xml.parsers.expat import ParserCreate
 
 
-usage_msg = """Usage: %s [unicode.xml]"""
+usage_msg = """Usage: %s [unicode.xml]\n"""
 
 def usage(prog, status=0, msg=None):
-    print >>sys.stderr, usage_msg % prog
+    sys.stderr.write(usage_msg % prog)
     if msg:
-        print >>sys.stderr, msg
+        sys.stderr.write(msg + '\n')
     sys.exit(status)
 
 def main(argv=None):
@@ -47,7 +47,10 @@ def main(argv=None):
         inpath = 'unicode.xml'
     if not os.path.isfile(inpath):
         usage(argv[0], 1, 'No such file: "%s".' % inpath)
-    infile = open(inpath)
+    if sys.version_info >= (3,0):
+        infile = open(inpath, mode='rb')
+    else:
+        infile = open(inpath)
     process(infile)
 
 def process(infile):
@@ -130,7 +133,7 @@ class CharacterEntitySetExtractor:
         if not set:
             return
         if set not in self.sets:
-            print 'bad set: %r' % set
+            print('bad set: %r' % set)
             return
         entity = attributes['id']
         assert (entity not in self.sets[set]
@@ -159,7 +162,7 @@ class CharacterEntitySetExtractor:
         return name
 
     def write_sets(self):
-        sets = self.sets.keys()
+        sets = list(self.sets.keys())
         sets.sort()
         for set_name in sets:
             self.write_set(set_name)
@@ -170,8 +173,8 @@ class CharacterEntitySetExtractor:
         else:
             outname = set_name + '.txt'
         outfile = open(outname, 'w')
-        print 'writing file "%s"' % outname
-        print >>outfile, self.header
+        print('writing file "%s"' % outname)
+        outfile.write(self.header + '\n')
         set = self.sets[set_name]
         entities = [(e.lower(), e) for e in set.keys()]
         entities.sort()
@@ -193,9 +196,9 @@ class CharacterEntitySetExtractor:
                 if int(code, 16) > 0xFFFF:
                     return 1            # wide-Unicode character
         codes = ' '.join(['U+%s' % code for code in charid[1:].split('-')])
-        print >>outfile, ('.. %-*s unicode:: %s .. %s'
-                          % (longest + 2, '|' + entity_name + '|',
-                             codes, self.descriptions[charid]))
+        outfile.write('.. %-*s unicode:: %s .. %s\n'
+                      % (longest + 2, '|' + entity_name + '|',
+                         codes, self.descriptions[charid]))
 
 
 if __name__ == '__main__':