FAQ
We've released a release candidate for the next bugfix release of Python,
2.2.1.

Get the scoop (and the files) here:

http://www.python.org/2.2.1/

In order to make 2.2.1 a solid release, please help by

+ Building the release, and running the test suite on your platform.
+ Building your extension modules and applications against this
release, and running *their* test suites.
+ Reporting any problems to the bug tracker at sf:

http://sourceforge.net/bugs/?group_idT70

This being a bugfix release, there are no exciting new features -- we just
fixed a lot of bugs. For a moderately complete list, please see:

http://http://sourceforge.net/project/shownotes.php?release_id�208

Depending on how many problems are found in this release, 2.2.1 final or a
second release candidate will follow next week.

Extra note for keenies:

One of the bugs addressed in this release relates to the behaviour of
floating point numbers with respect to under- and over-flow. This has
traditionally been an obscure platform-dependent mess. To make sure that
this is still the case, I would ask those of you with access to the more
unusual platforms (i.e. not Linux/x86, Win32/x86) to run test_math in
verbose mode and *email me* the results, including details of platform and
OS tested. Thanks.

Cheers,
The Python development team.

Search Discussions

  • Skip Montanaro at Mar 19, 2002 at 5:47 am
    mh> I would ask those of you with access to the more unusual platforms
    mh> (i.e. not Linux/x86, Win32/x86) to run test_math in verbose mode

    Seemed like a good chance for me to get some exposure to the SF compile
    farm, so I logged in, selected MacOS X, downloaded and built 2.2.1c1, and
    got this when I tried the math regression test:

    ./python Lib/test/test_math.py
    math module, testing with eps 1e-05
    constants
    acos
    asin
    atan
    atan2
    ceil
    cos
    cosh
    exp
    fabs
    floor
    fmod
    frexp
    hypot
    ldexp
    log
    log10
    modf
    pow
    sin
    sinh
    sqrt
    tan
    tanh
    exceptions
    Traceback (most recent call last):
    File "Lib/test/test_math.py", line 195, in ?
    test_exceptions()
    File "Lib/test/test_math.py", line 192, in test_exceptions
    raise TestFailed("sqrt(-1) didn't raise ValueError")
    test_support.TestFailed: sqrt(-1) didn't raise ValueError

    --
    Skip Montanaro (skip at pobox.com - http://www.mojam.com/)
  • Steven Majewski at Mar 19, 2002 at 4:30 pm
    On Mon, 18 Mar 2002, Skip Montanaro wrote:

    ...
    sinh
    sqrt
    tan
    tanh
    exceptions
    Traceback (most recent call last):
    File "Lib/test/test_math.py", line 195, in ?
    test_exceptions()
    File "Lib/test/test_math.py", line 192, in test_exceptions
    raise TestFailed("sqrt(-1) didn't raise ValueError")
    test_support.TestFailed: sqrt(-1) didn't raise ValueError
    I got the same on my osx build.
    Specifically: math.sqrt( -1 ) returns NaN.


    When I do:
    import fpectl
    fpectl.turnon_sigfpe()
    I get:
    'Operation not implemented'


    Are those two connected?


    -- Steve Majewski
  • Tim Peters at Mar 19, 2002 at 4:45 pm
    [Steven Majewski]
    I got the same on my osx build.
    Specifically: math.sqrt( -1 ) returns NaN.
    Please email your results to Michael Hudson.
    When I do:
    import fpectl
    fpectl.turnon_sigfpe()
    I get:
    'Operation not implemented'


    Are those two connected?
    The fpectl module tries to fiddle with HW fp exception masks, in necessarily
    platform-specific ways. Whether your platform libm pays attention to the
    setting of HW fp exception masks depends entirely on your platform, but
    fpectl doesn't know how to set them on your platform anyway. IOW, if you
    taught fpectl how to enable HW fp exceptions, it's still a crapshoot whether
    your libm sqrt would raise a HW fp exception in this case. In any case,
    Python never relies on fpectl by default, so even if fpectl did know how to
    fiddle your platform's flags, and your platform libm played along, it still
    wouldn't make any difference unless you explicitly turned on fpectl
    yourself.
  • Michael Hudson at Mar 19, 2002 at 9:14 am

    Michael Hudson <mwh at python.org> writes:

    This being a bugfix release, there are no exciting new features -- we just
    fixed a lot of bugs. For a moderately complete list, please see:

    http://http://sourceforge.net/project/shownotes.php?release_id�208
    That's a bit embarrassing. Oh well.

    here's-hoping-that's-the-worst-problem-of-this-release-ly
    M.

    --
    C is not clean -- the language has _many_ gotchas and traps, and
    although its semantics are _simple_ in some sense, it is not any
    cleaner than the assembly-language design it is based on.
    -- Erik Naggum, comp.lang.lisp
  • Gerhard Häring at Mar 19, 2002 at 11:31 am

    Michael Hudson wrote:
    I would ask those of you with access to the more
    unusual platforms (i.e. not Linux/x86, Win32/x86) to run test_math in
    verbose mode
    How would I do that?

    Gerhard
  • Tim Peters at Mar 19, 2002 at 3:42 am
    [Michael Hudson]
    I would ask those of you with access to the more
    unusual platforms (i.e. not Linux/x86, Win32/x86) to run test_math in
    verbose mode
    [Gerhard H?ring]
    How would I do that?
    python Lib/test/test_math.py

    For example, on Windows:

    C:\Python22>python Lib/test/test_math.py
    math module, testing with eps 1e-05
    constants
    acos
    asin
    atan
    atan2
    ceil
    cos
    cosh
    exp
    fabs
    floor
    fmod
    frexp
    hypot
    ldexp
    log
    log10
    modf
    pow
    sin
    sinh
    sqrt
    tan
    tanh
    exceptions

    C:\Python22>

    The last part ("exceptions") isn't normally run. If it's unhappy, it will
    blow up (raise some sort of exception).
  • Gerhard Häring at Mar 19, 2002 at 11:57 am

    Tim Peters wrote:
    [Michael Hudson]
    I would ask those of you with access to the more
    unusual platforms (i.e. not Linux/x86, Win32/x86) to run test_math in
    verbose mode

    [Gerhard H?ring]
    How would I do that?

    python Lib/test/test_math.py
    [...]
    The last part ("exceptions") isn't normally run. If it's unhappy, it will
    blow up (raise some sort of exception).
    That's why I asked. I get this:

    bash-2.05$ ../../python test_math.py
    math module, testing with eps 1e-05
    constants
    acos
    asin
    atan
    atan2
    ceil
    cos
    cosh
    exp
    fabs
    floor
    fmod
    frexp
    hypot
    ldexp
    log
    log10
    modf
    pow
    sin
    sinh
    sqrt
    tan
    tanh
    exceptions
    Traceback (most recent call last):
    File "test_math.py", line 195, in ?
    test_exceptions()
    File "test_math.py", line 192, in test_exceptions
    raise TestFailed("sqrt(-1) didn't raise ValueError")
    test_support.TestFailed: sqrt(-1) didn't raise ValueError
    bash-2.05$ ../../python
    Python 2.2.1c1 (#2, Mar 19 2002, 03:54:46)
    [GCC 2.95.3 20010315 (release) [FreeBSD]] on freebsd4
    Type "help", "copyright", "credits" or "license" for more information.
    import math
    math.sqrt(-1)
    NaN
    >>>

    Guess that's normal? I get the same with Python 2.2 on FreeBSD. Python
    2.1 raises an OverflowError instead.

    Gerhard
  • Tim Peters at Mar 19, 2002 at 4:50 am
    [Gerhard H?ring]
    That's why I asked. I get this:
    As Michael asked, please email results to him.
  • Michael Hudson at Mar 19, 2002 at 6:55 pm

    Steven Majewski <sdm7g at Virginia.EDU> writes:

    I got the same on my osx build.
    Specifically: math.sqrt( -1 ) returns NaN.
    Yes, it seems BSD libm returns NaN and does not set errno when
    sqrt(-1) is called.

    That's good -- this is the sort of thing I was hoping to learn.

    People can stop reporting this particular case of it now though,
    thanks.

    Cheers,
    M.

    --
    You sound surprised. We're talking about a government department
    here - they have procedures, not intelligence.
    -- Ben Hutchings, cam.misc
  • Michael Hudson at Mar 25, 2002 at 8:25 pm

    Michael Hudson <mwh at python.org> writes:

    Depending on how many problems are found in this release, 2.2.1
    final or a second release candidate will follow next week.
    It will be 2.2.1c2 and it will be released in about 22 hours time.

    Cheers,
    M.

    --
    I'm not sure that the ability to create routing diagrams
    similar to pretzels with mad cow disease is actually a
    marketable skill. -- Steve Levin
    -- http://home.xnet.com/~raven/Sysadmin/ASR.Quotes.html
  • Mo at Mar 25, 2002 at 9:55 pm
    Mo
    Michael Hudson <mwh at python.org> writes:
    Depending on how many problems are found in this release, 2.2.1
    final or a second release candidate will follow next week.
    It will be 2.2.1c2 and it will be released in about 22 hours time.
    This is like pacing outside a delivery room. Thank you for this
    update. More importantly, thanks to all of you "Doctors" laboring on
    our behalf.

    Mo
  • Peter Hansen at Mar 26, 2002 at 12:46 am

    mo wrote:
    Michael Hudson <mwh at python.org> writes:
    Depending on how many problems are found in this release, 2.2.1
    final or a second release candidate will follow next week.
    It will be 2.2.1c2 and it will be released in about 22 hours time.
    This is like pacing outside a delivery room. Thank you for this
    update. More importantly, thanks to all of you "Doctors" laboring on
    our behalf.
    Since when do _doctors_ do the "labor"? ;-)

    -yo-momma-ly yr's,
    Peter

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouppython-list @
categoriespython
postedMar 18, '02 at 9:35p
activeMar 26, '02 at 12:46a
posts13
users8
websitepython.org

People

Translate

site design / logo © 2022 Grokbase