New submission from Humberto Diogenes:
This patch addresses 3 simple errors (1 syntax, 1 runtime, 1 import) when
doing a "make altinstall" on Mac OS X, including the error listed in issue
1078.
--
components: Installation, Macintosh
files: py3k-mac-install-1.patch
messages:
Humberto Diogenes added the comment:
One more patch, for:
1. Writing str to binary file;
2. Using string exceptions.
Traceback for #2:
/usr/bin/install -c -s ../python.exe
"/Library/Frameworks/Python.framework/Versions/3.0/Resources/Python.app/
Contents/MacOS/Python"
DYLD_FRAM
Humberto Diogenes added the comment:
Patch 3: some more trivial fixes: new syntax for print() and octal
numbers.
__
Tracker <[EMAIL PROTECTED]>
<http://bugs.python.org/issue1091>
__Index: Mac/PythonLauncher
Humberto Diogenes added the comment:
And thanks for accepting them! (without complaining about my errors :-)
__
Tracker <[EMAIL PROTECTED]>
<http://bugs.python.org/
New submission from Humberto Diogenes:
Running this on Mac OS X:
$ ./configure --with-pydebug --enable-framework
$ make
I get this:
(...)
ar cr libpython3.0.a Modules/config.o Modules/getpath.o Modules/main.o
Modules/gcmodule.o
ar cr libpython3.0.a Modules/threadmodule.o Modules
Humberto Diogenes added the comment:
I'm using these versions:
intel: macosx 10.4.10, xcode 2.4.1 (gcc 4.0.1 build 5367)
Tried again with current revision (58153) and with --enable-universalsdk
it built just fine. Without it, I still get the same
Humberto Diogenes added the comment:
Luís, please analyse the response from the web server to see if it really is a
valid HTTP response - AFAIK CPanel web services do not always give valid HTTP
responses.
--
components: +Library (Lib)
type: resource usage -> behav
New submission from Humberto Diogenes :
On some systems, datetime.strftime() accepts a %s format string that returns
epoch / UNIX timestamp, but this behavior is not documented at
http://docs.python.org/library/datetime.html
Python 2.7 (r27:82508, Jul 3 2010, 21:12:11)
[GCC 4.0.1 (Apple Inc
Humberto Diogenes added the comment:
David, as discussed on the IRC channel: maybe we could just add pointers to the
OS-specific docs. Something like:
"""
If you want to use platform-specific format strings, search for strftime in
your OS documentation (`man strftime` o
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
A quick search for rfc822 imports in py3k reveals this:
$ find . -name '*.py' | xargs egrep -nr '(from|import) rfc822'
./Demo/scripts/mboxconvert.py:7:import rfc822
./Lib/cgi.py:39:import rfc822
./Lib/mimetool
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Correction: mimetools is also being removed from py3k (issue 2848).
__
Tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Added file: http://bugs.python.org/file10375/remove_rfc822-test_urllib2.py
__
Tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Tried to just replace rfc822.Message with email.message_from_file in
cgi.py but it didn't work. I still have to figure out how to fix
FieldStorage.read_multi.
Added file: http://bugs.python.org/file10376/remove_rfc822-cgi-in
New submission from Humberto Diogenes <[EMAIL PROTECTED]>:
StringIO was merged into the `io` module, but the old stringio.rst docs
are still there.
--
assignee: georg.brandl
components: Documentation
files: io.stringio-docs.patch
keywords: patch
messages: 67515
nosy: georg.
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Added file: http://bugs.python.org/file10473/remove-stringio.rst.patch
___
Python tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
--
nosy: +hdiogenes
___
Python tracker <[EMAIL PROTECTED]>
<http://bugs.python.org/issue2918>
___
__
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Seems that removing rfc822 from `cgi` is not an easy task -- please see issue
1112856.
___
Python tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Added a patch that successfully removes rfc822 dependency from the cgi
module, using ideas from issue 1112856. I had to change one test, as the
email FeedParser ignores trailing spaces in subparts.
--
keywords: +patch
Adde
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
> Er - while reverting it makes the code work again, I'm
> _really_ unhappy with this as a long-term solution.
I've addressed almost everything that's discussed here in issue 2849,
implementing Josh's sug
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
With the attached patches, rfc822 won't be used anywhere inside Lib/
anymore (with the exception of mimetools, which is going away too).
Is there any reason to convert the files below? Shouldn't they be removed
from 3
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
The only use of mimetools inside cgi is in parse_multipart, which is
untested and has huge warnings like these ones:
"""
XXX This does not parse nested multipart parts -- use FieldStorage for
that.
XXX This should
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
I'm sending a patch that removes mimetools from urllib and test_urllib. It
now returns an email.message.Message instead of mimetools.Message.
I also moved some imports to the top of the file. If that's a problem, I
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Added file:
http://bugs.python.org/file10516/remove_mimetools_from_urllib.patch-2
___
Python tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Removed file:
http://bugs.python.org/file10515/remove_mimetools_from_urllib.patch
___
Python tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
It's relatively easy to replace mimetools in most places: it's just a
matter of changing mimetools.Message to email.message_from_file, but it
gets a lot more complicated when there's inheritance involved.
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Added file: http://bugs.python.org/file10537/email.parser-small_fix.patch
___
Python tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Attached a patch to rename all calls to the mimetools-specific
.getheader() to .get(), which works on both libraries (mimetools and
email.Message). That eases transition without breaking any tests.
Added file: http://bugs.pyth
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
As we're moving away from using .getheader() everywhere (file 10538), I
think it's a good time to make HTTPResponse conform to PEP8 and define
.get_header (with underscore) instead.
Added file:
http://bugs.pyth
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Raymond, thanks for the parser solution! It worked -- just broke some
tests because of the nature of FeedParser. While rfc822.Message can read
the file and stop at some point in the middle (meaning that self.fp can
still b
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
mimetools removal is almost complete, with the exception of only two
broken tests: test_urllib2_localnet and test_xmlrpc. I'm not sure if it
can make it to this beta, but at least it's really close.
Added file: http:/
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
[msg68060]
>> Why does it need to be in 2.6? mimetools is still there.
>
> I guess you're right, it doesn't.
So, does it make sense to backport this too?
___
Python track
New submission from Humberto Diogenes <[EMAIL PROTECTED]>:
The new multiprocessing module is not being installed:
$ python3.0
Python 3.0b1+ (py3k:64417, Jun 19 2008, 21:25:46)
[GCC 4.0.1 (Apple Inc. build 5483)] on darwin
Type "help", "copyright", "credits&quo
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Jesse, not install, but altinstall:
sudo rm -fr /usr/local/lib/python3.0
make clean
make && sudo make altinstall
Forgot to mention that it was originally reported by Rodrigo Fenrrir.
___
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
This was solved in Python 3.0: http.server now uses email.parser (issue2848),
which is not
vulnerable to the DoS attack mentioned by Nick.
Should we backport these changes to 2.6? As HTTPMessage's base class was
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
The patch is not applying cleanly.
BTW, urlparse has been renamed to urllib.parse, but the old docs are still
there (urlparse.rst).
--
nosy: +hdiogenes
___
Python tracker <[EMAIL
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Correction: the patch does apply cleanly to 2.6, without breaking tests.
It just needs to be ported to 3.0 (new urllib package).
___
Python tracker <[EMAIL PROTECTED]>
<http://
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
It looks like there are no automated tests for pydoc; it's even listed in
test_sundry.py.
There's only one file Lib/test/pydocfodder.py which defines "Something just to
look at
via pydoc", but isn't use
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Amaury, seems that there's duplicated from render_doc inside pydoc.doc. I
removed it and everything kept working, including the new basic tests.
Added file: http://bugs.python.org/file10103/py3k-pydoc.doc-cl
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
According to Georg, the "adaptation" referred by this issue is just about
changing the URLs in the documentation to point appropriately to the URLs
generated by the new doc system.
Anyway, the doc-cleanup.patch
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
synx's patch wasn't applying cleanly and broke 2 maildir tests. I'm
posting one with updated tests and documentation. Still need to get rid of
rfc822 in test_mailbox, though.
Question: mailbox.py has one section ma
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Skip, I had the same problem as you, with the same solution. On a MacBook
Pro running Mac OS X 10.5.2, running test_ctypes and test_sqlite together
would kill my machine. Found another installation of sqlite3 on
/usr/local, remo
New submission from Humberto Diogenes <[EMAIL PROTECTED]>:
mailbox.py has one section marked as "classes from the original
module (for backward compatibility)", which are all listed as deprecated
in the module documentation.
In issue 756982, A. M. Kuchling agreed that &qu
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Added file: http://bugs.python.org/file10277/mailbox-remove_deprecated_doc.patch
__
Tracker <[EMAIL PROTECTED]>
<http://bugs.pytho
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
I created issue 2814 with a patch to remove those old classes.
Tracker <[EMAIL PROTECTED]>
<http://bugs.python.
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Andrews, this discussion would be more appropriate to the mailing list,
but anyway: I believe you're mixing things up. When PEP 3108 says "remove
symtable/_symtable", it must be talking only about symtable.py and
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
--
keywords: +patch
Added file: http://bugs.python.org/file10279/mailbox-replace_rfc822.patch
Tracker <[EMAIL PROTECTED]>
<http://bugs.python.
Changes by Humberto Diogenes <[EMAIL PROTECTED]>:
Removed file: http://bugs.python.org/file10272/mailbox.py.patch2
Tracker <[EMAIL PROTECTED]>
<http://bugs.python.
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
> rfc822 is replaced by None in the patch here; is that safe to do?
Yes. That's what mailbox documentation says:
Parameter factory is a callable object that accepts a file-like message
representation [...] and re
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
Georg, any special reason for not removing rfc822 references from
test_mailbox? That section of the patch was not merged.
Tracker <[EMAIL PROTECTED]>
<http://bugs.python.
Humberto Diogenes <[EMAIL PROTECTED]> added the comment:
I think we should add a deprecation warning for those classes in 2.6,
right?
As we're not removing the whole module, what's the right place to put
that? On the __init__ method of each class?
__
Humberto Diogenes added the comment:
How is this going?
--
nosy: +hdiogenes
___
Python tracker
<http://bugs.python.org/issue2919>
___
___
Python-bugs-list mailin
51 matches
Mail list logo
