Merge pull request #533 from python-rope/unicode-handling

Remove usage of `unicode` type

rope/base/ast.py

@@ -3,15 +3,10 @@
 
 from rope.base import fscommands
 
-try:
-    unicode
-except NameError:
-    unicode = str
-
 
 def parse(source, filename="<string>"):
     # NOTE: the raw string should be given to `compile` function
-    if isinstance(source, unicode):
+    if isinstance(source, str):
         source = fscommands.unicode_to_file_data(source)
     if b"\r" in source:
         source = source.replace(b"\r\n", b"\n").replace(b"\r", b"\n")

rope/base/change.py

@@ -2,9 +2,11 @@
 import difflib
 import os
 import time
+from typing import Union
 
 import rope.base.fscommands
-from rope.base import taskhandle, exceptions, utils
+from rope.base import exceptions, taskhandle, utils
+from rope.base.fscommands import FileContent
 
 
 class Change:
@@ -329,11 +331,15 @@ def _get_fscommands(self, resource):
             return self.direct_commands
         return self.fscommands
 
-    def write_file(self, resource, contents):
-        data = rope.base.fscommands.unicode_to_file_data(
-            contents,
-            newlines=resource.newlines,
-        )
+    def write_file(self, resource, contents: Union[str, FileContent]):
+        data: FileContent
+        if not isinstance(contents, bytes):
+            data = rope.base.fscommands.unicode_to_file_data(
+                contents,
+                newlines=resource.newlines,
+            )
+        else:
+            data = contents
         fscommands = self._get_fscommands(resource)
         fscommands.write(resource.real_path, data)
         for observer in list(self.project.observers):

rope/base/fscommands.py

@@ -12,11 +12,10 @@
 import subprocess
 
 import rope.base.utils.pycompat as pycompat
+import typing
 
-try:
-    unicode
-except NameError:
-    unicode = str
+
+FileContent = typing.NewType("FileContent", bytes)
 
 
 def create_fscommands(root):
@@ -214,19 +213,18 @@ def _execute(args, cwd=None):
     return process.returncode
 
 
-def unicode_to_file_data(contents, encoding=None, newlines=None):
-    if not isinstance(contents, unicode):
-        return contents
+def unicode_to_file_data(contents: str, encoding=None, newlines=None) -> FileContent:
+    assert isinstance(contents, str)
     if newlines and newlines != "\n":
         contents = contents.replace("\n", newlines)
     if encoding is None:
         encoding = read_str_coding(contents)
     if encoding is not None:
-        return contents.encode(encoding)
+        return FileContent(contents.encode(encoding))
     try:
-        return contents.encode()
+        return FileContent(contents.encode())
     except UnicodeEncodeError:
-        return contents.encode("utf-8")
+        return FileContent(contents.encode("utf-8"))
 
 
 def file_data_to_unicode(data, encoding=None):
@@ -242,7 +240,7 @@ def file_data_to_unicode(data, encoding=None):
 
 
 def _decode_data(data, encoding):
-    if isinstance(data, unicode):
+    if isinstance(data, str):
         return data
     if encoding is None:
         encoding = read_str_coding(data)

rope/base/pyobjects.py

@@ -1,4 +1,5 @@
-from rope.base.fscommands import _decode_data
+from typing import Optional
+
 from rope.base import ast, exceptions, utils
 
 
@@ -224,13 +225,14 @@ def get_module(self):
             current_object = current_object.parent
         return current_object
 
-    def get_doc(self):
+    def get_doc(self) -> Optional[str]:
         if len(self.get_ast().body) > 0:
             expr = self.get_ast().body[0]
             if isinstance(expr, ast.Expr) and isinstance(expr.value, ast.Str):
                 docstring = expr.value.s
-                coding = self.get_module().coding
-                return _decode_data(docstring, coding)
+                assert isinstance(docstring, str)
+                return docstring
+        return None
 
     def _get_defined_objects(self):
         if self.defineds is None:

rope/base/pyobjectsdef.py

@@ -17,11 +17,6 @@
 )
 from rope.base.utils import pycompat
 
-try:
-    unicode
-except NameError:
-    unicode = str
-
 
 class PyFunction(pyobjects.PyFunction):
     def __init__(self, pycore, ast_node, parent):
@@ -204,7 +199,7 @@ def _init_source(self, pycore, source_code, resource):
                 source_bytes = resource.read_bytes()
                 source_code, _ = fscommands.file_data_to_unicode(source_bytes)
             else:
-                if isinstance(source_code, unicode):
+                if isinstance(source_code, str):
                     source_bytes = fscommands.unicode_to_file_data(source_code)
                 else:
                     source_bytes = source_code

ropetest/contrib/codeassisttest.py

@@ -16,11 +16,6 @@
 )
 from ropetest import testutils
 
-try:
-    unicode
-except NameError:
-    unicode = str
-
 
 class CodeAssistTest(unittest.TestCase):
     def setUp(self):
@@ -731,7 +726,7 @@ def test_get_pydoc_unicode(self):
             def foo():
               u"юникод-объект"''')
         doc = get_doc(self.project, src, src.index("foo") + 1)
-        self.assertTrue(isinstance(doc, unicode))
+        self.assertTrue(isinstance(doc, str))
         self.assertTrue("юникод-объект" in doc)
 
     def test_get_pydoc_utf8_bytestring(self):
@@ -740,7 +735,7 @@ def test_get_pydoc_utf8_bytestring(self):
             def foo():
               "байтстринг"''')
         doc = get_doc(self.project, src, src.index("foo") + 1)
-        self.assertTrue(isinstance(doc, unicode))
+        self.assertTrue(isinstance(doc, str))
         self.assertTrue("байтстринг" in doc)
 
     def test_get_pydoc_for_functions(self):