feat: Remove dependency on six (#1134)

* feat: Remove six

* chore: Use context managers

* chore: Remove unnecessary ValueError suppression

requirements.txt

@@ -9,5 +9,4 @@ pycountry>=22.3.5
 pyyaml>=5.3.1
 requests>=2.5.0
 setuptools>=24.2.0
-six>=1.4.1
 wcwidth>=0.2.5

setup.cfg

@@ -44,7 +44,6 @@ install_requires = platformdirs>=3.6.0
     pyyaml>=5.3.1
     requests>=2.5.0
     setuptools>=24.2.0
-    six>=1.4.1
     wcwidth>=0.2.5
 
 [options.package_data]

xml2rfc/parser.py

@@ -12,7 +12,6 @@
 import re
 import requests
 import shutil
-import six
 import time
 import xml2rfc.log
 import xml2rfc.utils
@@ -72,7 +71,7 @@ def __init__(self, cache_path=None, library_dirs=None, source=None,
 
         # Get directory of source
         if self.source:
-            if isinstance(self.source, six.string_types):
+            if isinstance(self.source, str):
                 self.source_dir = os.path.abspath(os.path.dirname(self.source))
             else:
                 self.source_dir = os.path.abspath(os.path.dirname(self.source.name))                
@@ -458,7 +457,7 @@ def get(self, key, default=None):
         if value == default:
             return value
         else:
-            return six.text_type(value)
+            return str(value)
 
 class XmlRfcParser:
 
@@ -543,48 +542,43 @@ def parse(self, remove_comments=True, remove_pis=False, quiet=False, strip_cdata
                 text = text.replace(b'<rfc ', b'<rfc xmlns:%s="%s" ' % (b'xi', self.nsmap[b'xi']), 1)
 
         # Get an iterating parser object
-        file = six.BytesIO(text)
-        file.name = self.source
-        context = lxml.etree.iterparse(file,
-                                      dtd_validation=False,
-                                      load_dtd=True,
-                                      attribute_defaults=True,
-                                      no_network=self.no_network,
-                                      remove_comments=remove_comments,
-                                      remove_pis=remove_pis,
-                                      remove_blank_text=True,
-                                      resolve_entities=False,
-                                      strip_cdata=strip_cdata,
-                                      events=("start",),
-                                      tag="rfc",
-                                  )
-        # resolver without knowledge of rfc_number:
-        caching_resolver = CachingResolver(cache_path=self.cache_path,
-                                        library_dirs=self.library_dirs,
-                                        templates_path=self.templates_path,
-                                        source=self.source,
-                                        no_network=self.no_network,
-                                        network_locs=self.network_locs,
-                                        verbose=self.verbose,
-                                        quiet=self.quiet,
-                                        options=self.options,
-                                     )
-        context.resolvers.add(caching_resolver)
-
-        # Get hold of the rfc number (if any) in the rfc element, so we can
-        # later resolve the "&rfc.number;" entity.
-        self.rfc_number = None
-        self.format_version = None
-        try:
+        with io.BytesIO(text) as file:
+            file.name = self.source
+            context = lxml.etree.iterparse(file,
+                                          dtd_validation=False,
+                                          load_dtd=True,
+                                          attribute_defaults=True,
+                                          no_network=self.no_network,
+                                          remove_comments=remove_comments,
+                                          remove_pis=remove_pis,
+                                          remove_blank_text=True,
+                                          resolve_entities=False,
+                                          strip_cdata=strip_cdata,
+                                          events=("start",),
+                                          tag="rfc",
+                                      )
+            # resolver without knowledge of rfc_number:
+            caching_resolver = CachingResolver(cache_path=self.cache_path,
+                                            library_dirs=self.library_dirs,
+                                            templates_path=self.templates_path,
+                                            source=self.source,
+                                            no_network=self.no_network,
+                                            network_locs=self.network_locs,
+                                            verbose=self.verbose,
+                                            quiet=self.quiet,
+                                            options=self.options,
+                                         )
+            context.resolvers.add(caching_resolver)
+
+            # Get hold of the rfc number (if any) in the rfc element, so we can
+            # later resolve the "&rfc.number;" entity.
+            self.rfc_number = None
+            self.format_version = None
             for action, element in context:
                 if element.tag == "rfc":
                     self.rfc_number = element.attrib.get("number", None)
                     self.format_version = element.attrib.get("version", None)
                     break
-        except ValueError as e:
-            if e.message=="I/O operation on closed file":
-                pass
-
         if self.format_version == "3":
             self.default_dtd_path = None
 
@@ -621,9 +615,9 @@ def parse(self, remove_comments=True, remove_pis=False, quiet=False, strip_cdata
         parser.set_element_class_lookup(element_lookup)
 
         # Parse the XML file into a tree and create an rfc instance
-        file = six.BytesIO(text)
-        file.name = self.source
-        tree = lxml.etree.parse(file, parser)
+        with io.BytesIO(text) as file:
+            file.name = self.source
+            tree = lxml.etree.parse(file, parser)
         xmlrfc = XmlRfc(tree, self.default_dtd_path, nsmap=self.nsmap)
         xmlrfc.source = self.source
 

xml2rfc/run.py

@@ -10,7 +10,6 @@
 import lxml.etree
 import os
 import pycountry
-import six
 import sys
 
 # If this script is renamed to 'xml2rfc.py' on a Windows system, the import
@@ -68,10 +67,7 @@ def print_country_help(options, parser):
     ids = list(country_ids.values())
     ids.sort()
     print('Known country codes and country names for use with <country>:\n')
-    if six.PY3:
-        print(('\n'.join([ '  '+'  -  '.join(v) for v in ids])))
-    else:
-        print(('\n'.join([ '  '+'  -  '.join(v) for v in ids])).encode('utf-8'))
+    print(('\n'.join([ '  '+'  -  '.join(v) for v in ids])))
     sys.exit()
 
 
@@ -767,12 +763,8 @@ def main():
             xmlrfc.tree = prep.prep()
             if xmlrfc.tree:
                 info = extract_anchor_info(xmlrfc.tree)
-                if six.PY2:
-                    with open(filename, 'w') as fp:
-                        json.dump(info, fp, indent=2, ensure_ascii=False, encoding='utf-8')
-                else:
-                    with io.open(filename, 'w', encoding='utf-8') as fp:
-                        json.dump(info, fp, indent=2, ensure_ascii=False)
+                with io.open(filename, 'w', encoding='utf-8') as fp:
+                    json.dump(info, fp, indent=2, ensure_ascii=False)
                 if not options.quiet:
                     xml2rfc.log.write('Created file', filename)
 

xml2rfc/util/unicode.py

@@ -6,7 +6,6 @@
 # Unicode operations
 
 import re
-import six
 import unicodedata
 
 try:
@@ -117,8 +116,8 @@ def name(e):
         try:
             names = []
             for c in e.text:
-                if isinstance(c, six.binary_type):
-                    c = six.text_type(c, encoding='latin1')
+                if isinstance(c, bytes):
+                    c = c.decode('latin1')
                 names.append(unicodedata.name(c, 'U+%04x'%ord(c)))
             return ', '.join(names)
         except ValueError as exc:
@@ -180,7 +179,7 @@ def ascii(e):
 def isascii(u):
     if u is None:
         return True
-    if isinstance(u, six.text_type):
+    if isinstance(u, str):
         t = u+''
         for ch in [ '\u00a0', '\u200B', '\u2011', '\u2028', '\u2060', ]:
             if ch in t:
@@ -438,5 +437,5 @@ def downcode(str, replacements=None, use_charrefs=True):
     u'\u2010': '-',
 }
 
-controlchars = dict( (six.text_type(chr(i)), ' ') for i in range(0, 32) if not i in [ 9, 10, 13 ] )
+controlchars = dict( (str(chr(i)), ' ') for i in range(0, 32) if not i in [ 9, 10, 13 ] )
 

xml2rfc/utils.py

@@ -6,17 +6,13 @@
 
 import base64
 import re
-import six
 import sys
 import textwrap
 
 from collections import OrderedDict
 from lxml.etree import _Comment, _ProcessingInstruction
 
-if six.PY2:
-    from urllib import quote
-else:
-    from urllib.request import quote
+from urllib.request import quote
 
 try:
     from xml2rfc import debug

xml2rfc/writers/html.py

@@ -5,8 +5,6 @@
 import lxml
 import os
 import re
-#import sys
-import six
 import unicodedata
 import xml2rfc
 
@@ -15,12 +13,8 @@
 from lxml.html import html_parser
 from lxml.html.builder import ElementMaker
 
-if six.PY2:
-    from urllib import urlopen
-    from urlparse import urlparse, urljoin
-elif six.PY3:
-    from urllib.request import urlopen
-    from urllib.parse import urlparse, urljoin
+from urllib.request import urlopen
+from urllib.parse import urlparse, urljoin
 
 try:
     from xml2rfc import debug
@@ -221,7 +215,7 @@ def get_bidi_alignment(address):
         line = address[field]
         if line:
             for ch in line:
-                if isinstance(ch, six.text_type):
+                if isinstance(ch, str):
                     dir = unicodedata.bidirectional(ch)
                     if dir in ['R', 'AL']:
                         return 'right'

xml2rfc/writers/paginated_txt.py

@@ -4,7 +4,6 @@
 
 # Python libs
 import calendar
-import six
 
 try:
     import debug
@@ -100,7 +99,7 @@ def write_with_break_hint(self, writer, type, *args, **kwargs):
 
     def needLines(self, count):
         """Deal with the PI directive needLines"""
-        if isinstance(count, six.string_types):
+        if isinstance(count, str):
             if count.isdigit():
                 count = int(count)
             else:

xml2rfc/writers/preptool.py

@@ -6,7 +6,6 @@
 import datetime
 import os
 import re
-import six
 import sys
 import traceback as tb
 import unicodedata
@@ -21,12 +20,8 @@
 except ImportError:
     pass
 
-if six.PY2:
-    from urlparse import urlsplit, urlunsplit, urljoin, urlparse
-    from urllib import urlopen
-elif six.PY3:
-    from urllib.parse import urlsplit, urlunsplit, urljoin, urlparse
-    from urllib.request import urlopen
+from urllib.parse import urlsplit, urlunsplit, urljoin, urlparse
+from urllib.request import urlopen
 
 from lxml import etree
 
@@ -1851,10 +1846,7 @@ def element_artwork(self, e, p):
                     if scheme in ['http', 'https']:
                         with closing(urlopen(src)) as f:
                             data = f.read()
-                            if six.PY2:
-                                mediatype = f.info().gettype()
-                            else:
-                                mediatype = f.info().get_content_type()
+                            mediatype = f.info().get_content_type()
                         src = build_dataurl(mediatype, data)
                         e.set('src', src)
                     elif scheme == 'file':

xml2rfc/writers/text.py

@@ -6,7 +6,6 @@
 import inspect
 import re
 import sys
-import six
 import textwrap
 
 from codecs import open
@@ -54,7 +53,7 @@
 # We don't use namedtuple for Line, because the resulting objects would be immutable:
 class Line(object):
     def __init__(self, text, elem):
-        assert isinstance(text, six.text_type)
+        assert isinstance(text, str)
         self.text = text
         self.elem = elem
         self.page = None
@@ -177,22 +176,22 @@ def align(lines, how, width):
     return lines
 
 def mklines(arg, e):
-    if isinstance(arg, six.text_type):
+    if isinstance(arg, str):
         # \u2028 and \u2029 are eliminated here, through splitlines()
         lines = [ Line(t, e) for t in arg.splitlines() ]
     else:
         lines = arg
     return lines
 
 def mktextblock(arg):
-    if isinstance(arg, six.text_type):
+    if isinstance(arg, str):
         text = arg
     else:
         text = '\u2028'.join([ l.text for l in arg ])
     return text
 
 def mktext(arg):
-    if isinstance(arg, six.text_type):
+    if isinstance(arg, str):
         text = arg
     else:
         text = '\n'.join([ l.text for l in arg ])
@@ -313,15 +312,9 @@ def process(self):
                     tag  = l.elem.tag  if l.elem!=None else '-'
                     page = l.elem.page if l.elem!=None else '-'
                     if l.block:
-                        if six.PY2:
-                            sys.stderr.write(("%3d %10s %3d-%3d [%4s] %s\n" % (i, tag, l.block.beg, l.block.end, page, l.text)).encode('utf8'))
-                        else:
-                            sys.stderr.write(("%3d %10s %3d-%3d [%4s] %s\n" % (i, tag, l.block.beg, l.block.end, page, l.text)))
+                        sys.stderr.write(("%3d %10s %3d-%3d [%4s] %s\n" % (i, tag, l.block.beg, l.block.end, page, l.text)))
                     else:
-                        if six.PY2:
-                            sys.stderr.write(("%3d %10s         [%4s] %s\n" % (i, tag,                           page, l.text)).encode('utf8'))
-                        else:
-                            sys.stderr.write(("%3d %10s         [%4s] %s\n" % (i, tag,                           page, l.text)))
+                        sys.stderr.write(("%3d %10s         [%4s] %s\n" % (i, tag,                           page, l.text)))
             for i, l in enumerate(lines):
                 length = len(l.text)
                 if length > MAX_WIDTH:
@@ -525,7 +518,7 @@ def tjoin(self, text, e, width, **kwargs):
         Render element e, then format and join it to text using the
         appropriate settings in joiners.
         '''
-        assert isinstance(text, six.text_type)
+        assert isinstance(text, str)
         joiners = kwargs['joiners']
         j = joiners[e.tag] if e.tag in joiners else joiners[None]
         width -= j.indent + j.hang
@@ -3792,7 +3785,7 @@ def rreplace(s, old, new, max):
             text = fill(text, width=width, **kwargs)
             lines = mklines(text, e)
         else:
-            if isinstance(text, six.binary_type):
+            if isinstance(text, bytes):
                 text = text.decode('utf-8')
             lines = [ Line(text, e) ]
         return lines