Remove remaining legacy Python 2 code (#599)

* Replace OrderedDict with plain dict()

<https://docs.python.org/3/library/collections.html#ordereddict-objects>:
> Ordered dictionaries are just like regular dictionaries but have some
> extra capabilities relating to ordering operations. They have become
> less important now that the built-in dict class gained the ability to
> remember insertion order (this new behavior became guaranteed in
> Python 3.7).

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

* construct/lib/container: Remove unneeded methods

`collections.abc.MutableMapping` already implements `keys()`,
`update()`, `__eq__()`, `__ne__()` and `__contains__()` – most
importantly correctly, e.g. `update()` has many variants which the
current implementation does not handle.

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

* construct: Remove Python 2 compatibility layer

Remove the compatibility layer `py3compat`.

PS: There is #548 to migrate
from the interned version of `construct` back to "Construct 2.10+" now
that it is maintained again. When that happens, this change will become
moot. Until then restore some sanity back into our copy as handling
Python type annotations for both Python 2 and 3 is a major pain.

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

* Drop remaining Python 2 compatibility code

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

* Drop from __future__ import print_function

The project is Python 3 only, so remove the legacy Python 2
compatibility.

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

* Replace `bytelist2string` with `bytes`

The function was used in only one place. Just use the much more
efficient `bytes(list[int])` to convert from `list[int]` to `bytes`.

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

---------

Signed-off-by: Philipp Hahn <phahn-oss@avm.de>

elftools/common/construct_utils.py

@@ -49,9 +49,6 @@ def _sizeof(self, context):
 
 class ULEB128(Construct):
     """A construct based parser for ULEB128 encoding.
-
-       Incompatible with Python 2 - assumes that the return of read()
-       is an indexed collection of numbers.
     """
     def _parse(self, stream, context):
         value = 0
@@ -68,9 +65,6 @@ def _parse(self, stream, context):
 
 class SLEB128(Construct):
     """A construct based parser for SLEB128 encoding.
-
-       Incompatible with Python 2 - assumes that the return of read()
-       is an indexed collection of numbers.
     """
     def _parse(self, stream, context):
         value = 0

elftools/common/utils.py

@@ -23,13 +23,6 @@ def bytes2str(b):
     """Decode a bytes object into a string."""
     return b.decode('latin-1')
 
-def bytelist2string(bytelist):
-    """ Convert a list of byte values (e.g. [0x10 0x20 0x00]) to a bytes object
-        (e.g. b'\x10\x20\x00').
-    """
-    return b''.join(bytes((b,)) for b in bytelist)
-
-
 def struct_parse(struct, stream, stream_pos=None):
     """ Convenience function for using the given struct to parse a stream.
         If stream_pos is provided, the stream is seeked to this position before

elftools/construct/__init__.py

@@ -22,7 +22,7 @@
     ... )
     >>> s.parse("\\x01\\x02\\x03")
     Container(a = 1, b = 515)
-    >>> print s.parse("\\x01\\x02\\x03")
+    >>> print(s.parse("\\x01\\x02\\x03"))
     Container:
         a = 1
         b = 515

elftools/construct/adapters.py

@@ -1,7 +1,8 @@
+from io import BytesIO
+
 from .core import Adapter, AdaptationError, Pass
 from .lib import int_to_bin, bin_to_int, swap_bytes
 from .lib import FlagsContainer, HexString
-from .lib.py3compat import BytesIO, decodebytes
 
 
 #===============================================================================

elftools/construct/core.py

@@ -1,6 +1,6 @@
+from io import BytesIO
 from struct import Struct as Packer
 
-from .lib.py3compat import BytesIO, advance_iterator, bchr
 from .lib import Container, ListContainer, LazyContainer
 
 
@@ -520,13 +520,13 @@ def _build(self, obj, stream, context):
             if self.subcon.conflags & self.FLAG_COPY_CONTEXT:
                 for subobj in obj:
                     if isinstance(obj, bytes):
-                        subobj = bchr(subobj)
+                        subobj = bytes((subobj,))
                     self.subcon._build(subobj, stream, context.__copy__())
                     cnt += 1
             else:
                 for subobj in obj:
                     if isinstance(obj, bytes):
-                        subobj = bchr(subobj)
+                        subobj = bytes((subobj,))
                     self.subcon._build(subobj, stream, context)
                     cnt += 1
         except ConstructError as ex:
@@ -587,7 +587,7 @@ def _build(self, obj, stream, context):
                     break
         else:
             for subobj in obj:
-                subobj = bchr(subobj)
+                subobj = bytes((subobj,))
                 self.subcon._build(subobj, stream, context.__copy__())
                 if self.predicate(subobj, context):
                     terminated = True
@@ -722,7 +722,7 @@ def _build(self, obj, stream, context):
             elif sc.name is None:
                 subobj = None
             else:
-                subobj = advance_iterator(objiter)
+                subobj = next(objiter)
                 context[sc.name] = subobj
             sc._build(subobj, stream, context)
 
@@ -881,7 +881,7 @@ def _parse(self, stream, context):
             except ConstructError:
                 stream.seek(pos)
             else:
-                context.__update__(context2)
+                context.update(context2)
                 if self.include_name:
                     return sc.name, obj
                 else:
@@ -903,7 +903,7 @@ def _build(self, obj, stream, context):
                 except Exception:
                     pass
                 else:
-                    context.__update__(context2)
+                    context.update(context2)
                     stream.write(stream2.getvalue())
                     return
         raise SelectError("no subconstruct matched", obj)

elftools/construct/debug.py

@@ -1,7 +1,6 @@
 """
 Debugging utilities for constructs
 """
-from __future__ import print_function
 import sys
 import traceback
 import pdb
@@ -81,7 +80,7 @@ def printout(self, stream, context):
             frames.reverse()
             for f in frames:
                 a = Container()
-                a.__update__(f.f_locals)
+                a.update(f.f_locals)
                 obj.stack.append(a)
 
         print("=" * 80)

elftools/construct/lib/binary.py

@@ -1,6 +1,3 @@
-from .py3compat import int2byte
-
-
 def int_to_bin(number, width=32):
     r"""
     Convert an integer into its binary representation in a bytes object.
@@ -32,13 +29,6 @@ def int_to_bin(number, width=32):
     1: 1,
     48: 0, # '0'
     49: 1, # '1'
-
-    # The following are for Python 2, in which iteration over a bytes object
-    # yields single-character bytes and not integers.
-    '\x00': 0,
-    '\x01': 1,
-    '0': 0,
-    '1': 1,
     }
 
 def bin_to_int(bits, signed=False):
@@ -81,10 +71,8 @@ def swap_bytes(bits, bytesize=8):
 _char_to_bin = {}
 _bin_to_char = {}
 for i in range(256):
-    ch = int2byte(i)
+    ch = bytes((i,))
     bin = int_to_bin(i, 8)
-    # Populate with for both keys i and ch, to support Python 2 & 3
-    _char_to_bin[ch] = bin
     _char_to_bin[i] = bin
     _bin_to_char[bin] = ch
 
@@ -94,15 +82,15 @@ def encode_bin(data):
     Create a binary representation of the given b'' object. Assume 8-bit
     ASCII. Example:
 
-        >>> encode_bin('ab')
+        >>> encode_bin(b'ab')
         b"\x00\x01\x01\x00\x00\x00\x00\x01\x00\x01\x01\x00\x00\x00\x01\x00"
     """
     return b"".join(_char_to_bin[ch] for ch in data)
 
 
 def decode_bin(data):
     """
-    Locical opposite of decode_bin.
+    Logical opposite of decode_bin.
     """
     if len(data) & 7:
         raise ValueError("Data length must be a multiple of 8")

elftools/construct/lib/container.py

@@ -2,8 +2,9 @@
 Various containers.
 """
 
+from collections.abc import MutableMapping
 from pprint import pformat
-from .py3compat import MutableMapping
+
 
 def recursion_lock(retval, lock_name = "__recursion_lock__"):
     def decorator(func):
@@ -40,45 +41,19 @@ def __delitem__(self, name):
     def __setitem__(self, name, value):
         self.__dict__[name] = value
 
-    def keys(self):
-        return self.__dict__.keys()
+    def __iter__(self):
+        return iter(self.__dict__)
 
     def __len__(self):
         return len(self.__dict__.keys())
 
-    # Extended dictionary interface.
-
-    def update(self, other):
-        self.__dict__.update(other)
-
-    __update__ = update
-
-    def __contains__(self, value):
-        return value in self.__dict__
-
-    # Rich comparisons.
-
-    def __eq__(self, other):
-        try:
-            return self.__dict__ == other.__dict__
-        except AttributeError:
-            return False
-
-    def __ne__(self, other):
-        return not self == other
-
     # Copy interface.
 
     def copy(self):
         return self.__class__(**self.__dict__)
 
     __copy__ = copy
 
-    # Iterator interface.
-
-    def __iter__(self):
-        return iter(self.__dict__)
-
     def __repr__(self):
         return "%s(%s)" % (self.__class__.__name__, repr(self.__dict__))
 

elftools/construct/lib/hex.py

@@ -1,9 +1,6 @@
-from .py3compat import byte2int, int2byte, bytes2str
-
-
 # Map an integer in the inclusive range 0-255 to its string byte representation
 _printable = dict((i, ".") for i in range(256))
-_printable.update((i, bytes2str(int2byte(i))) for i in range(32, 128))
+_printable.update((i, chr(i)) for i in range(32, 128))
 
 
 def hexdump(data, linesize):
@@ -18,8 +15,8 @@ def hexdump(data, linesize):
     fmt = fmt % (3 * linesize - 1,)
     for i in range(0, len(data), linesize):
         line = data[i : i + linesize]
-        hextext = " ".join('%02x' % byte2int(b) for b in line)
-        rawtext = "".join(_printable[byte2int(b)] for b in line)
+        hextext = " ".join('%02x' % b for b in line)
+        rawtext = "".join(_printable[b] for b in line)
         prettylines.append(fmt % (i, str(hextext), str(rawtext)))
     return prettylines
 

elftools/construct/macros.py

@@ -1,4 +1,3 @@
-from .lib.py3compat import int2byte
 from .lib import (BitStreamReader, BitStreamWriter, encode_bin,
     decode_bin)
 from .core import (Struct, MetaField, StaticField, FormatField,
@@ -103,7 +102,7 @@ def Flag(name, truth = 1, falsehood = 0, default = False):
     """
 
     return SymmetricMapping(Field(name, 1),
-        {True : int2byte(truth), False : int2byte(falsehood)},
+        {True : bytes((truth,)), False : bytes((falsehood,))},
         default = default,
     )
 

elftools/dwarf/descriptions.py

@@ -490,7 +490,6 @@ def _import_extra(attr, die, section_offset):
             # stream.
             with preserve_stream_pos(die.stream):
                 ref_die = DIE(cu, die.stream, ref_die_offset)
-            #print '&&& ref_die', ref_die
             return '[Abbrev Number: %s (%s)]' % (
                 ref_die.abbrev_code, ref_die.tag)
 

elftools/dwarf/die.py

@@ -6,7 +6,7 @@
 # Eli Bendersky (eliben@gmail.com)
 # This code is in the public domain
 #-------------------------------------------------------------------------------
-from collections import namedtuple, OrderedDict
+from collections import namedtuple
 import os
 
 from ..common.exceptions import DWARFError, ELFParseError
@@ -86,7 +86,7 @@ def __init__(self, cu, stream, offset):
         self.stream = stream
         self.offset = offset
 
-        self.attributes = OrderedDict()
+        self.attributes = {}
         self.tag = None
         self.has_children = None
         self.abbrev_code = None

elftools/dwarf/dwarf_expr.py

@@ -9,7 +9,7 @@
 from collections import namedtuple
 from io import BytesIO
 
-from ..common.utils import struct_parse, bytelist2string, read_blob
+from ..common.utils import struct_parse, read_blob
 from ..common.exceptions import DWARFError
 
 
@@ -135,7 +135,7 @@ def parse_expr(self, expr):
 
         The list can potentially be nested.
         """
-        stream = BytesIO(bytelist2string(expr))
+        stream = BytesIO(bytes(expr))
         parsed = []
 
         while True: