Merged revisions 79306,79311,79325 via svnmerge from

svn+ssh://pythondev@svn.python.org/python/trunk

................
  r79306 | benjamin.peterson | 2010-03-22 17:40:06 -0500 (Mon, 22 Mar 2010) | 21 lines

  Merged revisions 79077,79137,79304-79305 via svnmerge from
  svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3

  ........
    r79077 | benjamin.peterson | 2010-03-18 18:05:29 -0500 (Thu, 18 Mar 2010) | 1 line

    port detect_encoding improvements from py3k
  ........
    r79137 | benjamin.peterson | 2010-03-20 11:12:53 -0500 (Sat, 20 Mar 2010) | 1 line

    add a fixer for setting sys.exitfunc #2356
  ........
    r79304 | benjamin.peterson | 2010-03-22 17:20:22 -0500 (Mon, 22 Mar 2010) | 1 line

    fix test_parser when it's run in a path with spaces #7666
  ........
    r79305 | benjamin.peterson | 2010-03-22 17:27:07 -0500 (Mon, 22 Mar 2010) | 1 line

    normalize whitespace
  ........
................
  r79311 | benjamin.peterson | 2010-03-22 17:54:42 -0500 (Mon, 22 Mar 2010) | 9 lines

  Merged revisions 79309 via svnmerge from
  svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3

  ........
    r79309 | benjamin.peterson | 2010-03-22 17:50:47 -0500 (Mon, 22 Mar 2010) | 1 line

    pass correct symbol in
  ........
................
  r79325 | benjamin.peterson | 2010-03-22 22:03:55 -0500 (Mon, 22 Mar 2010) | 13 lines

  Merged revisions 79313,79324 via svnmerge from
  svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3

  ........
    r79313 | benjamin.peterson | 2010-03-22 17:59:57 -0500 (Mon, 22 Mar 2010) | 1 line

    another case where a symbol is needed
  ........
    r79324 | benjamin.peterson | 2010-03-22 21:59:47 -0500 (Mon, 22 Mar 2010) | 1 line

    use unicode literals
  ........
................
This commit is contained in:
Benjamin Peterson 2010-03-23 03:22:05 +00:00
parent d96b2f249d
commit 0af9398d27
4 changed files with 170 additions and 10 deletions

View File

@ -0,0 +1,70 @@
"""
Convert use of sys.exitfunc to use the atexit module.
"""
# Author: Benjamin Peterson
from lib2to3 import pytree, fixer_base
from lib2to3.fixer_util import Name, Attr, Call, Comma, Newline, syms
class FixExitfunc(fixer_base.BaseFix):
PATTERN = """
(
sys_import=import_name<'import'
('sys'
|
dotted_as_names< (any ',')* 'sys' (',' any)* >
)
>
|
expr_stmt<
power< 'sys' trailer< '.' 'exitfunc' > >
'=' func=any >
)
"""
def __init__(self, *args):
super(FixExitfunc, self).__init__(*args)
def start_tree(self, tree, filename):
super(FixExitfunc, self).start_tree(tree, filename)
self.sys_import = None
def transform(self, node, results):
# First, find a the sys import. We'll just hope it's global scope.
if "sys_import" in results:
if self.sys_import is None:
self.sys_import = results["sys_import"]
return
func = results["func"].clone()
func.prefix = ""
register = pytree.Node(syms.power,
Attr(Name("atexit"), Name("register"))
)
call = Call(register, [func], node.prefix)
node.replace(call)
if self.sys_import is None:
# That's interesting.
self.warning(node, "Can't find sys import; Please add an atexit "
"import at the top of your file.")
return
# Now add an atexit import after the sys import.
names = self.sys_import.children[1]
if names.type == syms.dotted_as_names:
names.append_child(Comma())
names.append_child(Name("atexit", " "))
else:
containing_stmt = self.sys_import.parent
position = containing_stmt.children.index(self.sys_import)
stmt_container = containing_stmt.parent
new_import = pytree.Node(syms.import_name,
[Name("import"), Name("atexit", " ")]
)
new = pytree.Node(syms.simple_stmt, [new_import])
containing_stmt.insert_child(position + 1, Newline())
containing_stmt.insert_child(position + 2, new)

View File

@ -253,14 +253,16 @@ def detect_encoding(readline):
in. in.
It detects the encoding from the presence of a utf-8 bom or an encoding It detects the encoding from the presence of a utf-8 bom or an encoding
cookie as specified in pep-0263. If both a bom and a cookie are present, cookie as specified in pep-0263. If both a bom and a cookie are present, but
but disagree, a SyntaxError will be raised. If the encoding cookie is an disagree, a SyntaxError will be raised. If the encoding cookie is an invalid
invalid charset, raise a SyntaxError. charset, raise a SyntaxError. Note that if a utf-8 bom is found,
'utf-8-sig' is returned.
If no encoding is specified, then the default of 'utf-8' will be returned. If no encoding is specified, then the default of 'utf-8' will be returned.
""" """
bom_found = False bom_found = False
encoding = None encoding = None
default = 'utf-8'
def read_or_stop(): def read_or_stop():
try: try:
return readline() return readline()
@ -287,17 +289,16 @@ def detect_encoding(readline):
if codec.name != 'utf-8': if codec.name != 'utf-8':
# This behaviour mimics the Python interpreter # This behaviour mimics the Python interpreter
raise SyntaxError('encoding problem: utf-8') raise SyntaxError('encoding problem: utf-8')
else: encoding += '-sig'
# Allow it to be properly encoded and decoded.
encoding = 'utf-8-sig'
return encoding return encoding
first = read_or_stop() first = read_or_stop()
if first.startswith(BOM_UTF8): if first.startswith(BOM_UTF8):
bom_found = True bom_found = True
first = first[3:] first = first[3:]
default = 'utf-8-sig'
if not first: if not first:
return 'utf-8', [] return default, []
encoding = find_cookie(first) encoding = find_cookie(first)
if encoding: if encoding:
@ -305,13 +306,13 @@ def detect_encoding(readline):
second = read_or_stop() second = read_or_stop()
if not second: if not second:
return 'utf-8', [first] return default, [first]
encoding = find_cookie(second) encoding = find_cookie(second)
if encoding: if encoding:
return encoding, [first, second] return encoding, [first, second]
return 'utf-8', [first, second] return default, [first, second]
def untokenize(iterable): def untokenize(iterable):
"""Transform tokens back into Python source code. """Transform tokens back into Python source code.

View File

@ -4285,3 +4285,91 @@ class Test_operator(FixerTestCase):
def test_bare_sequenceIncludes(self): def test_bare_sequenceIncludes(self):
s = "sequenceIncludes(x, y)" s = "sequenceIncludes(x, y)"
self.warns_unchanged(s, "You should use operator.contains here.") self.warns_unchanged(s, "You should use operator.contains here.")
class Test_exitfunc(FixerTestCase):
fixer = "exitfunc"
def test_simple(self):
b = """
import sys
sys.exitfunc = my_atexit
"""
a = """
import sys
import atexit
atexit.register(my_atexit)
"""
self.check(b, a)
def test_names_import(self):
b = """
import sys, crumbs
sys.exitfunc = my_func
"""
a = """
import sys, crumbs, atexit
atexit.register(my_func)
"""
self.check(b, a)
def test_complex_expression(self):
b = """
import sys
sys.exitfunc = do(d)/a()+complex(f=23, g=23)*expression
"""
a = """
import sys
import atexit
atexit.register(do(d)/a()+complex(f=23, g=23)*expression)
"""
self.check(b, a)
def test_comments(self):
b = """
import sys # Foo
sys.exitfunc = f # Blah
"""
a = """
import sys
import atexit # Foo
atexit.register(f) # Blah
"""
self.check(b, a)
b = """
import apples, sys, crumbs, larry # Pleasant comments
sys.exitfunc = func
"""
a = """
import apples, sys, crumbs, larry, atexit # Pleasant comments
atexit.register(func)
"""
self.check(b, a)
def test_in_a_function(self):
b = """
import sys
def f():
sys.exitfunc = func
"""
a = """
import sys
import atexit
def f():
atexit.register(func)
"""
self.check(b, a)
def test_no_sys_import(self):
b = """sys.exitfunc = f"""
a = """atexit.register(f)"""
msg = ("Can't find sys import; Please add an atexit import at the "
"top of your file.")
self.warns(b, a, msg)
def test_unchanged(self):
s = """f(sys.exitfunc)"""
self.unchanged(s)

View File

@ -206,6 +206,7 @@ def diff(fn, result):
finally: finally:
f.close() f.close()
try: try:
return os.system("diff -u %r @" % fn) fn = fn.replace('"', '\\"')
return os.system('diff -u "%s" @' % fn)
finally: finally:
os.remove("@") os.remove("@")