[issue10335] tokenize.open_python(): open a Python file with the right encoding

[STINNER Victor]

STINNER Victor <victor.stinner at haypocalc.com> added the comment:

On Saturday 06 November 2010 17:00:15 you wrote:
> Note that it is useful for opening any text file with an encoding cookie,
> not only python source code, so "tokenize.open()" sounds attractive.

Ok, the new patch (tokenize_open-2.patch) uses tokenize.open() name and adds a 
test for BOM without coding cookie (test utf-8-sig encoding).

----------
Added file: http://bugs.python.org/file19529/tokenize_open-2.patch

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue10335>
_______________________________________
-------------- next part --------------
Index: Doc/library/tokenize.rst
===================================================================
--- Doc/library/tokenize.rst	(révision 86167)
+++ Doc/library/tokenize.rst	(copie de travail)
@@ -101,16 +101,18 @@
     If no encoding is specified, then the default of ``'utf-8'`` will be
     returned.
 
-    :func:`detect_encoding` is useful for robustly reading Python source files.
-    A common pattern for this follows::
+    Use :func:`open` to open Python script: it uses :func:`detect_encoding` to
+    detect the file encoding.
 
-        def read_python_source(file_name):
-            with open(file_name, "rb") as fp:
-                encoding = tokenize.detect_encoding(fp.readline)[0]
-            with open(file_name, "r", encoding=encoding) as fp:
-                return fp.read()
 
+.. function:: open(filename)
 
+   Open a Python script in read mode using the encoding detected by
+   :func:`detect_encoding`.
+
+   .. versionadded:: 3.2
+
+
 Example of a script rewriter that transforms float literals into Decimal
 objects::
 
@@ -153,4 +155,3 @@
                 result.append((toknum, tokval))
         return untokenize(result).decode('utf-8')
 
-
Index: Lib/py_compile.py
===================================================================
--- Lib/py_compile.py	(révision 86167)
+++ Lib/py_compile.py	(copie de travail)
@@ -104,9 +104,7 @@
     byte-compile all installed files (or all files in selected
     directories).
     """
-    with open(file, "rb") as f:
-        encoding = tokenize.detect_encoding(f.readline)[0]
-    with open(file, encoding=encoding) as f:
+    with tokenize.open(file) as f:
         try:
             timestamp = int(os.fstat(f.fileno()).st_mtime)
         except AttributeError:
Index: Lib/tabnanny.py
===================================================================
--- Lib/tabnanny.py	(révision 86167)
+++ Lib/tabnanny.py	(copie de travail)
@@ -93,11 +93,8 @@
                 check(fullname)
         return
 
-    with open(file, 'rb') as f:
-        encoding, lines = tokenize.detect_encoding(f.readline)
-
     try:
-        f = open(file, encoding=encoding)
+        f = tokenize.open(file)
     except IOError as msg:
         errprint("%r: I/O Error: %s" % (file, msg))
         return
Index: Lib/tokenize.py
===================================================================
--- Lib/tokenize.py	(révision 86167)
+++ Lib/tokenize.py	(copie de travail)
@@ -29,6 +29,7 @@
 from token import *
 from codecs import lookup, BOM_UTF8
 import collections
+from io import TextIOWrapper
 cookie_re = re.compile("coding[:=]\s*([-\w.]+)")
 
 import token
@@ -335,6 +336,20 @@
     return default, [first, second]
 
 
+_builtin_open = open
+
+def open(filename):
+    """
+    Open a Python script in read mode with the right encoding.
+    """
+    buffer = _builtin_open(filename, 'rb')
+    encoding, line = detect_encoding(buffer.readline)
+    buffer.seek(0)
+    text = TextIOWrapper(buffer, encoding, line_buffering=True)
+    text.mode = 'r'
+    return text
+
+
 def tokenize(readline):
     """
     The tokenize() generator requires one argment, readline, which
Index: Lib/trace.py
===================================================================
--- Lib/trace.py	(révision 86167)
+++ Lib/trace.py	(copie de travail)
@@ -419,10 +419,9 @@
 def find_executable_linenos(filename):
     """Return dict where keys are line numbers in the line number table."""
     try:
-        with io.FileIO(filename, 'r') as file:
-            encoding, lines = tokenize.detect_encoding(file.readline)
-        with open(filename, "r", encoding=encoding) as f:
+        with tokenize.open_python(filename) as f:
             prog = f.read()
+            encoding = f.encoding
     except IOError as err:
         print(("Not printing coverage data for %r: %s"
                               % (filename, err)), file=sys.stderr)
Index: Lib/linecache.py
===================================================================
--- Lib/linecache.py	(révision 86167)
+++ Lib/linecache.py	(copie de travail)
@@ -123,9 +123,7 @@
         else:
             return []
     try:
-        with open(fullname, 'rb') as fp:
-            coding, line = tokenize.detect_encoding(fp.readline)
-        with open(fullname, 'r', encoding=coding) as fp:
+        with tokenize.open(fullname) as fp:
             lines = fp.readlines()
     except IOError:
         return []
Index: Lib/test/test_tokenize.py
===================================================================
--- Lib/test/test_tokenize.py	(révision 86167)
+++ Lib/test/test_tokenize.py	(copie de travail)
@@ -564,7 +564,8 @@
 
 from test import support
 from tokenize import (tokenize, _tokenize, untokenize, NUMBER, NAME, OP,
-                     STRING, ENDMARKER, tok_name, detect_encoding)
+                     STRING, ENDMARKER, tok_name, detect_encoding,
+                     open as tokenize_open)
 from io import BytesIO
 from unittest import TestCase
 import os, sys, glob
@@ -857,6 +858,27 @@
         readline = self.get_readline((b'# coding: bad\n',))
         self.assertRaises(SyntaxError, detect_encoding, readline)
 
+    def test_open(self):
+        filename = support.TESTFN + '.py'
+        self.addCleanup(support.unlink, filename)
+
+        #Â test coding cookie
+        for encoding in ('iso-8859-15', 'utf-8'):
+            with open(filename, 'w', encoding=encoding) as fp:
+                print("# coding: %s" % encoding, file=fp)
+                print("print('euro:\u20ac')", file=fp)
+            with tokenize_open(filename) as fp:
+                assert fp.encoding == encoding
+                assert fp.mode == 'r'
+
+        #Â test BOM (no coding cookie)
+        encoding = 'utf-8-sig'
+        with open(filename, 'w', encoding=encoding) as fp:
+            print("print('euro:\u20ac')", file=fp)
+        with tokenize_open(filename) as fp:
+            assert fp.encoding == encoding
+            assert fp.mode == 'r'
+
 class TestTokenize(TestCase):
 
     def test_tokenize(self):