2to3 truncates files at formfeed character

[cgohlke]

BPO	11250
Nosy	@warsaw, @benjaminp
Files	formfeedbug.py

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = 'https://github.com/warsaw'
closed_at = <Date 2011-10-07.19:46:01.553>
created_at = <Date 2011-02-19.23:57:24.087>
labels = ['expert-2to3']
title = '2to3 truncates files at formfeed character'
updated_at = <Date 2011-10-07.19:46:01.552>
user = 'https://bugs.python.org/cgohlke'

bugs.python.org fields:

activity = <Date 2011-10-07.19:46:01.552>
actor = 'barry'
assignee = 'barry'
closed = True
closed_date = <Date 2011-10-07.19:46:01.553>
closer = 'barry'
components = ['2to3 (2.x to 3.x conversion tool)']
creation = <Date 2011-02-19.23:57:24.087>
creator = 'cgohlke'
dependencies = []
files = ['20800']
hgrepos = []
issue_num = 11250
keywords = []
message_count = 6.0
messages = ['128885', '129602', '145046', '145095', '145118', '145129']
nosy_count = 4.0
nosy_names = ['barry', 'benjamin.peterson', 'cgohlke', 'python-dev']
pr_nums = []
priority = 'normal'
resolution = 'fixed'
stage = None
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue11250'
versions = ['Python 2.7', 'Python 3.2']

[cgohlke]

Running Tools/Scripts/2to3.py on Python 3.2rc3 or 2.7.1 for Windows on a file that contains a formfeed character (0x0C, <FF>) results in a truncated file.

E.g. a file (attached) with the content

print 1
<FF>
print 2

is incorrectly refactored:

@@ -1,4 +1,1 @@
-print 1
-
-
-print 2
+print(1)

Python 2.6.6 and 3.1.3 correctly refactor the file:

-print 1
+print(1)
 <FF>
-print 2
+print(2)

[benjaminp]

r88661

[warsaw]

Was this patch ever folded into Python 3.2? Looking at the hg repository, I think the answer is "no". It does appear to have made it into Python 2.7 and trunk though (afaict).

In point of fact, this bug is hitting me now with 3.2.2.

[warsaw]

I also noticed that test_parser.py isn't being run in either 3.2 or 3.3. I'll fix that at the same time I port this patch to 3.2.

[warsaw]

Okay, re-enabling test_parser.py introduces a number of test failures in test_all_project_files(). Because I don't want to continue to shave the yak on this one, I'm going to wrap this test in @expectedfailure and open a separate bug for it.

[python-dev]

New changeset 6e0e9237d8e8 by Barry Warsaw in branch '3.2':

• Issue bpo-11250: Back port fix from 3.3 branch, so that 2to3 can handle files
  http://hg.python.org/cpython/rev/6e0e9237d8e8