Mailman 3 June 2008 - Python-Dev

Python 2.6 SSL
by Roger wenham 29 Jun '08

29 Jun '08

HI, I have used the SSL lib to wrap sockets in existing client server code, but I find that the max send size is 16K. In other words I send 20K, the write call returns 20K but the receiving end only gets 16K. I remove the wraper and everything worksagain. I even modified the unit test to send >16K, but got the same result. Is therea patch for this ? I hope this the correct place to report this... Regards Roger -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer

3 2

Disable tests in unittest (issue3202)
by Justin Mazzola Paluska 28 Jun '08

28 Jun '08

Hi, I just reported issue3202 in the bugracker (http://bugs.python.org/issue3202) with a patch that implements a way to disable unittest.TestCases using a decorator. Benjamin Peterson suggested that I bring up the issue on python-dev. The idea behind the patch is that it’s sometimes useful to disable tests without removing them from the TestCase. For example, a co-worker and I have a module with a few tests that will fail for the forseeable future because we haven’t had a chance to implement the features the tests are testing. The tracebacks for these “not implemented yet” tests obscure real failures in other tests. Normally, we’d just rename the failing methods from something starting with “test” to something like “_test”, so unittest ignores them. However, doing so removes all traces of the tests when you re-run the test suite, so the disabled tests are easy to forget. Instead, issue3202 allows you to write something like: from unittest import main, TestCase, disabled class SampleTest(TestCase): def test_addition(self): self.assertEqual(2, 1+1) def test_broken(self): # this is a real failure in our code self.assertEqual(5, 2+2) @disabled def test_not_implemented(self): # Test of a feature not implemented yet. doit() if __name__ == '__main__': main() which has one test disabled. Running the test suite outputs %python sample_tests2.py .FD ====================================================================== FAIL: test_broken (__main__.SampleTest) ---------------------------------------------------------------------- Traceback (most recent call last): File "sample_tests2.py", line 12, in test_broken self.assertEqual(5, 2+2) AssertionError: 5 != 4 ---------------------------------------------------------------------- Ran 3 tests in 0.001s FAILED (failures=1) showing only the real failure and a simple “D” for the disabled test. The disabled test is still there, but doesn’t get in your way. JUnit4 has a similar decorator for its tests, @Ignore. The patch in issue3202 implements the decorator @disabled and a few modifications to the classes in the unittest module. What does the Python hivemind think? I’m also open to enhancing it if the list has any ideas. —Justin

4 5

Summary of Python tracker Issues
by Python tracker 28 Jun '08

28 Jun '08

ACTIVITY SUMMARY (06/20/08 - 06/27/08) Python tracker at http://bugs.python.org/ To view or respond to any of the issues listed below, click on the issue number. Do NOT respond to this message. 1933 open (+40) / 13106 closed (+25) / 15039 total (+65) Open issues with patches: 600 Average duration of open issues: 707 days. Median duration of open issues: 1532 days. Open Issues Breakdown open 1908 (+39) pending 25 ( +1) Issues Created Or Reopened (69) _______________________________ Move Demo/classes/Rat.py to Lib/fractions.py and fix it up. 06/25/08 http://bugs.python.org/issue1682 reopened marketdickinson patch pickling of large recursive structures crashes cPickle 06/25/08 http://bugs.python.org/issue2702 reopened facundobatista patch Let bin/oct/hex show floats 06/25/08 http://bugs.python.org/issue3008 reopened rhettinger patch ast module docs document private attribute 06/20/08 CLOSED http://bugs.python.org/issue3152 created brett.cannon easy, 26backport sqlite leaks on error 06/20/08 http://bugs.python.org/issue3153 created Rhamphoryncus "Quick search" box renders too long on FireFox 3 06/20/08 http://bugs.python.org/issue3154 created tebeka Python should expose a pthread_cond_timedwait API for threading 06/20/08 http://bugs.python.org/issue3155 created gregory.p.smith bytes type has inconsistent methods (append, insert) 06/20/08 http://bugs.python.org/issue3156 created akuchling patch sqlite3 minor documentation issues 06/21/08 CLOSED http://bugs.python.org/issue3157 created slewis Doctest fails to find doctests in extension modules 06/21/08 http://bugs.python.org/issue3158 created fer_perez glob.py improvements 06/21/08 http://bugs.python.org/issue3159 created Indy patch Building a Win32 binary installer crashes 06/21/08 http://bugs.python.org/issue3160 created complex Missing import for sys in _abcoll 06/21/08 CLOSED http://bugs.python.org/issue3161 created hodgestar patch ssl.SSLSocket implements methods that are overriden by socket.so 06/21/08 http://bugs.python.org/issue3162 created hodgestar module struct support for ssize_t and size_t 06/21/08 http://bugs.python.org/issue3163 created MrJean1 patch cPickle calls to save_string and save_unicode with unicode objec 06/21/08 CLOSED http://bugs.python.org/issue3164 created cuerty patch cPickle recursion problem 06/21/08 CLOSED http://bugs.python.org/issue3165 created cuerty patch Make conversions from long to float correctly rounded. 06/21/08 http://bugs.python.org/issue3166 created marketdickinson patch math test fails on Solaris 10 06/21/08 http://bugs.python.org/issue3167 created MrJean1 cmath test fails on Solaris 10 06/21/08 http://bugs.python.org/issue3168 created MrJean1 email/header.py doesn't handle Base64 headers that have been ins 06/21/08 http://bugs.python.org/issue3169 created jasonjwwilliams test_pydoc has no way to regenerate pristine data 06/22/08 http://bugs.python.org/issue3170 created brett.cannon operator.*slice() should be deprecated in 2.6 06/22/08 http://bugs.python.org/issue3171 created scoder test macos skipped "unexpected" on linux2 06/22/08 CLOSED http://bugs.python.org/issue3172 created MrJean1 external strftime for Python? 06/23/08 http://bugs.python.org/issue3173 created skip.montanaro patch, patch 3.0b1 doesn't seem to install on macs 06/23/08 http://bugs.python.org/issue3174 created erickt multiprocessing build fails on Fedora 8 and Xubuntu 8 + solution 06/23/08 CLOSED http://bugs.python.org/issue3175 created mark result error for 2/5 and 4/5 06/23/08 CLOSED http://bugs.python.org/issue3176 created hiweed implement os.startfile on posix and MacOSX 06/23/08 http://bugs.python.org/issue3177 created ganadist patch __radd__(self,other) isn't called if self and other are of the s 06/23/08 CLOSED http://bugs.python.org/issue3178 created Frosburn cPickle is seriously broken 06/23/08 CLOSED http://bugs.python.org/issue3179 created schmir patch Interrupts are lost during readline PyOS_InputHook processing 06/23/08 http://bugs.python.org/issue3180 created Araneidae patch ConfigParsers are classic classes 06/23/08 http://bugs.python.org/issue3181 created hauva patch 2to3 Slight Patch 06/23/08 http://bugs.python.org/issue3182 created nedds sha modules & Modules/Setup.dist 06/23/08 http://bugs.python.org/issue3183 created rwgk The elementtree.ElementTree Module doc is missing 06/23/08 CLOSED http://bugs.python.org/issue3184 created gladstein Documentation of time.time() differs from source documentation 06/24/08 CLOSED http://bugs.python.org/issue3185 created cgrohmann bin(long) doesn't have a trailing 'L' 06/24/08 CLOSED http://bugs.python.org/issue3186 created marketdickinson os.listdir can return byte strings 06/24/08 http://bugs.python.org/issue3187 reopened amaury.forgeotdarc patch float('infinity') should be valid 06/24/08 http://bugs.python.org/issue3188 created marketdickinson patch Py3k DeprecationWarning in difflib 06/24/08 CLOSED http://bugs.python.org/issue3189 created vdupras patch Pydoc should ignore __package__ attributes 06/24/08 http://bugs.python.org/issue3190 created ncoghlan round docstring is inaccurate 06/24/08 http://bugs.python.org/issue3191 created marketdickinson exec(open(filename)) doesn't work 06/24/08 CLOSED http://bugs.python.org/issue3192 created ms Python 2.5.2 - Python 06/24/08 CLOSED http://bugs.python.org/issue3193 created peterdilley Demo/loop.c passing "char *" instead of "wchar_t *" 06/25/08 http://bugs.python.org/issue3194 created zanella patch invalid conversion xml.etree.ElementTree.Element object to boole 06/25/08 CLOSED http://bugs.python.org/issue3195 created xaka Option in pydoc to show docs from private methods 06/25/08 http://bugs.python.org/issue3196 created mortenlj patch Documentation for fractions module needs work 06/25/08 http://bugs.python.org/issue3197 created marketdickinson patch strings don't seem to roundtrip with repr() 06/25/08 CLOSED http://bugs.python.org/issue3198 created mark 2.6b1 Build Fails On OSX 10.5 06/25/08 CLOSED http://bugs.python.org/issue3199 created johna Python's crash in heavy multithreading IO operations 06/25/08 CLOSED http://bugs.python.org/issue3200 created sadit struct.Struct size attribute undocumented 06/25/08 CLOSED http://bugs.python.org/issue3201 created ericvw Wish: disable tests in unittest 06/25/08 http://bugs.python.org/issue3202 created jmp patch sphinx - table of contents doesn't render correctly (html) 06/25/08 http://bugs.python.org/issue3203 created grflanagan operator module docs are not updated to 3.0 06/26/08 CLOSED http://bugs.python.org/issue3204 created mishok13 patch bz2 iterator fails silently on MemoryError 06/26/08 http://bugs.python.org/issue3205 created mdehoon patch Multiprocessing Array and sharedctypes.Array error in docs/imple 06/26/08 http://bugs.python.org/issue3206 created mishok13 patch file.write() after file.readline() in mode "r+" 06/26/08 CLOSED http://bugs.python.org/issue3207 created peterdemin function annotation for builtin and C function 06/26/08 http://bugs.python.org/issue3208 created bhy Grammar error in UserDict module docs 06/26/08 CLOSED http://bugs.python.org/issue3209 created technel subprocess.Popen does not release process handles if process can 06/26/08 http://bugs.python.org/issue3210 created gjb1002 warnings.warn_explicit raises SystemError 06/26/08 CLOSED http://bugs.python.org/issue3211 created exarkun ssl module - should test for a wrong cert 06/26/08 http://bugs.python.org/issue3212 created jonas.wagner patch "pydoc -p" should listen to [::] if IPv6 is supported 06/27/08 http://bugs.python.org/issue3213 created ndim patch Suggest change to glossary explanation: "Duck Typing" 06/27/08 http://bugs.python.org/issue3214 created paddy3118 Can't import sqlite3 in Python 2.6b1 06/27/08 http://bugs.python.org/issue3215 created craigneuro errors in msilib documentation 06/27/08 http://bugs.python.org/issue3216 created sjoerd IndexError: Add bad index to msg 06/23/08 http://bugs.python.org/issue1534607 reopened benjamin.peterson Issues Now Closed (52) ______________________ simple patch, improving unreachable bytecode removing 232 days http://bugs.python.org/issue1394 gvanrossum patch os.system() fails for commands with multiple quoted file names 208 days http://bugs.python.org/issue1524 d4rk1 Failure when calling __str__ for MIMEBase(message, rfc822) objec 199 days http://bugs.python.org/issue1556 facundobatista module-cgi: handling GET and POST together 161 days http://bugs.python.org/issue1817 facundobatista patch, easy shutil: NameError (WindowsError) when moving from ext3 to fat32 80 days http://bugs.python.org/issue2549 draghuram easy default_scheme in urlparse.urlparse() useless 75 days http://bugs.python.org/issue2569 facundobatista csv.reader accepts string instead of file object (duck typing go 56 days http://bugs.python.org/issue2701 facundobatista os.getcwd fails for long path names on linux 54 days http://bugs.python.org/issue2722 facundobatista patch, easy curses.textpad loses characters at the end of lines 52 days http://bugs.python.org/issue2732 akuchling pprint produces different output in 2.6 and 3.0 36 days http://bugs.python.org/issue2888 facundobatista patch Verify doc updates for the decimal module 21 days http://bugs.python.org/issue3017 facundobatista Recursion bug in deepcopy 16 days http://bugs.python.org/issue3043 facundobatista heapq change breaking compatibility 13 days http://bugs.python.org/issue3051 rhettinger patch Let set.union and set.intersection accept multiple arguments 13 days http://bugs.python.org/issue3069 rhettinger patch chapter 17.1.3.5 'Replacing os.popen*' in the Python library ref 11 days http://bugs.python.org/issue3085 benjamin.peterson patch str.format("{0:n}") poss. bug with setlocale() 5 days http://bugs.python.org/issue3140 eric.smith urllib docs don't match the new modules 4 days http://bugs.python.org/issue3142 georg.brandl Sphinx/LaTeX fails on Python 3.0b1 documentation 3 days http://bugs.python.org/issue3146 georg.brandl ast module docs document private attribute 0 days http://bugs.python.org/issue3152 loewis easy, 26backport sqlite3 minor documentation issues 2 days http://bugs.python.org/issue3157 georg.brandl Missing import for sys in _abcoll 1 days http://bugs.python.org/issue3161 rhettinger patch cPickle calls to save_string and save_unicode with unicode objec 1 days http://bugs.python.org/issue3164 facundobatista patch cPickle recursion problem 4 days http://bugs.python.org/issue3165 facundobatista patch test macos skipped "unexpected" on linux2 0 days http://bugs.python.org/issue3172 benjamin.peterson multiprocessing build fails on Fedora 8 and Xubuntu 8 + solution 0 days http://bugs.python.org/issue3175 benjamin.peterson result error for 2/5 and 4/5 0 days http://bugs.python.org/issue3176 amaury.forgeotdarc __radd__(self,other) isn't called if self and other are of the s 0 days http://bugs.python.org/issue3178 amaury.forgeotdarc cPickle is seriously broken 3 days http://bugs.python.org/issue3179 jcea patch The elementtree.ElementTree Module doc is missing 0 days http://bugs.python.org/issue3184 gladstein Documentation of time.time() differs from source documentation 0 days http://bugs.python.org/issue3185 loewis bin(long) doesn't have a trailing 'L' 0 days http://bugs.python.org/issue3186 marketdickinson Py3k DeprecationWarning in difflib 0 days http://bugs.python.org/issue3189 rhettinger patch exec(open(filename)) doesn't work 2 days http://bugs.python.org/issue3192 georg.brandl Python 2.5.2 - Python 0 days http://bugs.python.org/issue3193 peterdilley invalid conversion xml.etree.ElementTree.Element object to boole 0 days http://bugs.python.org/issue3195 benjamin.peterson strings don't seem to roundtrip with repr() 0 days http://bugs.python.org/issue3198 amaury.forgeotdarc 2.6b1 Build Fails On OSX 10.5 0 days http://bugs.python.org/issue3199 marketdickinson Python's crash in heavy multithreading IO operations 0 days http://bugs.python.org/issue3200 amaury.forgeotdarc struct.Struct size attribute undocumented 0 days http://bugs.python.org/issue3201 benjamin.peterson operator module docs are not updated to 3.0 0 days http://bugs.python.org/issue3204 georg.brandl patch file.write() after file.readline() in mode "r+" 0 days http://bugs.python.org/issue3207 amaury.forgeotdarc Grammar error in UserDict module docs 0 days http://bugs.python.org/issue3209 georg.brandl warnings.warn_explicit raises SystemError 0 days http://bugs.python.org/issue3211 brett.cannon POP3 lib timeout 1602 days http://bugs.python.org/issue888830 facundobatista xrange overflows 1417 days http://bugs.python.org/issue1003935 rhettinger docs for Py_UNICODE are wrong 1349 days http://bugs.python.org/issue1044479 facundobatista easy functions replaced by subprocess should point to its docs 1287 days http://bugs.python.org/issue1083895 facundobatista BaseHTTPServer uses deprecated mimetools.Message 1169 days http://bugs.python.org/issue1180470 benjamin.peterson installation problem with python 2.4.1 on Win2k system 1137 days http://bugs.python.org/issue1199808 loewis poplib.py list interface 633 days http://bugs.python.org/issue1567948 facundobatista File Read/Write Flushing Patch 527 days http://bugs.python.org/issue1636874 amaury.forgeotdarc patch Add Tkinter.Checkbutton get() and set(value) 317 days http://bugs.python.org/issue1774370 gpolo patch Top Issues Most Discussed (10) ______________________________ 12 math test fails on Solaris 10 6 days open http://bugs.python.org/issue3167 12 implement PEP 3134 exception reporting 13 days open http://bugs.python.org/issue3112 11 sys.getsizeof() gives an AttributeError for _sre objects. 11 days open http://bugs.python.org/issue3122 9 pickling of large recursive structures crashes cPickle 2 days open http://bugs.python.org/issue2702 8 Let bin/oct/hex show floats 3 days open http://bugs.python.org/issue3008 8 No float formatting in PyString_FromFormat 48 days open http://bugs.python.org/issue2813 8 email.MIMEText.MIMEText.as_string incorrectly folding long subj 149 days open http://bugs.python.org/issue1974 7 exec(open(filename)) doesn't work 2 days closed http://bugs.python.org/issue3192 7 cPickle is seriously broken 3 days closed http://bugs.python.org/issue3179 6 rlcompleter add "(" to callables feature 2515 days open http://bugs.python.org/issue449227

1 0

Vacation
by Thomas Heller 27 Jun '08

27 Jun '08

FYI: I'm going to a vacation for the next two or three weeks. I will shutdown my buildbots, and restart them when I'm back: x86 osx.5, x86 XP-3, amd64 XP. Sorry for the inconvenience, Thomas

1 0

deepcopy
by Walter Bender 27 Jun '08

27 Jun '08

<mailto:python-dev@python.org>Hello togehter, I get an error while doing a deep copy of an "event". An event is an object, with has a lot of __dict__["attr"] = Event("SAMPLE") OTHER Events added to the object (no self refence) and store additional __dict__["attr2"] = 2 informations. deep copy is failing with no proper error message. nevent = copy.deepcopy(event); File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 174, in deepcopy y = copier(x, memo) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 305, in _deepcopy_inst state = deepcopy(state, memo) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 174, in deepcopy y = copier(x, memo) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 268, in _deepcopy_dict y[deepcopy(key, memo)] = deepcopy(value, memo) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 174, in deepcopy y = copier(x, memo) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 241, in _deepcopy_list y.append(deepcopy(a, memo)) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 174, in deepcopy y = copier(x, memo) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 241, in _deepcopy_list y.append(deepcopy(a, memo)) File "/net/software_cms/slc4_ia32_gcc345/external/python/2.4.2-CMS3q/lib/python2.4/copy.py", line 185, in deepcopy y = copier(x, memo) TypeError: __deepcopy__() takes no arguments (1 given cheers, Walter -- Walter Bender bender(a)physik.rwth-aachen.de RWTH Aachen III. Physikalisches Institut A Tel. +49-241-8027286

2 1

Undocumenting test.support in 3.x (was Py3k DeprecationWarning in stdlib)
by Benjamin Peterson 27 Jun '08

27 Jun '08

On Wed, Jun 25, 2008 at 12:18 PM, Brett Cannon <brett(a)python.org> wrote: > If you want, but Benjamin plans to undocument this for users along > with all other test.support stuff (which I agree with). Most of the > APIs in test.support were just quickly written and have not > necessarily been thought through in order to make sure that the APIs > makes sense (like in this case). Thanks for bringing that up, Brett. Is making test_support non-public a API in 3.0 fine with everybody? -- Cheers, Benjamin Peterson "There's no place like 127.0.0.1."

4 9

C API for gc.enable() and gc.disable()
by Alexandre Vassalotti 27 Jun '08

27 Jun '08

Would anyone mind if I did add a public C API for gc.disable() and gc.enable()? I would like to use it as an optimization for the pickle module (I found out that I get a good 2x speedup just by disabling the GC while loading large pickles). Of course, I could simply import the gc module and call the functions there, but that seems overkill to me. I included the patch below for review. -- Alexandre Index: Include/objimpl.h =================================================================== --- Include/objimpl.h (revision 63766) +++ Include/objimpl.h (working copy) @@ -221,8 +221,10 @@ * ========================== */ -/* C equivalent of gc.collect(). */ +/* C equivalent of gc.collect(), gc.enable() and gc.disable(). */ PyAPI_FUNC(Py_ssize_t) PyGC_Collect(void); +PyAPI_FUNC(void) PyGC_Enable(void); +PyAPI_FUNC(void) PyGC_Disable(void); /* Test if a type has a GC head */ #define PyType_IS_GC(t) PyType_HasFeature((t), Py_TPFLAGS_HAVE_GC) Index: Modules/gcmodule.c =================================================================== --- Modules/gcmodule.c (revision 63766) +++ Modules/gcmodule.c (working copy) @@ -1252,6 +1252,18 @@ return n; } +void +PyGC_Disable(void) +{ + enabled = 0; +} + +void +PyGC_Enable(void) +{ + enabled = 1; +} + /* for debugging */ void _PyGC_Dump(PyGC_Head *g)

14 40

Py3k DeprecationWarning in stdlib
by Virgil Dupras 27 Jun '08

27 Jun '08

Hi Devs, Today, I saw the Py3k beta 1 announcement, so I thought "Well, I gotta do my duty and try to migrate my own code base to see if I can uncover any problems". I'm just at the step where I run my test suite with the "-3" flag, and I noticed something peculiar: I get DeprecationWarning from stdlib code (namely, difflib, logging and unittest). Is it something to be expected or should I create tickets for those? I'm sure that the stdlib on the 3k side is fixed, but shouldn't the 2.6 side be fixed as well? Sorry if this question is already answered somewhere else, I searched for an answer but I couldn't find it. Additionally, one thing that strikes me as strange is how hard it is to find a "Migration guide" on the website. I had to dig in the PEP 3000 to find the brief bullet list telling me to run py26 with the -3 flag. Virgil Dupras

9 25

Re: [Python-Dev] [Python-checkins] r64424 - in python/trunk:Include/object.h Lib/test/test_sys.py Misc/NEWSObjects/intobject.c Objects/longobject.c Objects/typeobject.cPython/bltinmodule.c
by Nick Coghlan 26 Jun '08

26 Jun '08

Raymond Hettinger wrote: > FYI, I had already updated the patch to incorporate your suggestion. > It checks __index__ first and then __float__ if index doesn't exist. > The result makes it work with Decimal inputs. Only sort of - the domain of the output is then constrained by what the builtin float type can represent, which strikes me as fairly inappropriate for what should be a lossless display operation (both fractions.Fraction and decimal.Decimal can represent a lot of numbers that float can't deal with properly). To use the old faithful example of this, given: >>> float(Decimal("1.1")) 1.1000000000000001 what happens for bin(Decimal("1.1")) with the patch version that uses __float__? As I've said elsewhere in this thread, I'm certainly not averse to the basic idea of letting bin/oct/hex handle rational numbers as well as integers, I just don't see it as something we have to have in the current release cycle, and nor do I see it really working properly without adding a __rational__ special method that returns a numerator/denominator 2-tuple (which would obviously require a PEP). Cheers, Nick. -- Nick Coghlan | ncoghlan(a)gmail.com | Brisbane, Australia --------------------------------------------------------------- http://www.boredomandlaziness.org

1 0

Re: [Python-Dev] Undocumenting test.support in 3.x (was Py3k DeprecationWarning in stdlib)
by Simon Cross 26 Jun '08

26 Jun '08

On Thu, Jun 26, 2008 at 12:27 AM, Brett Cannon <brett(a)python.org> wrote: > Right, but I would think all core developers know about test.support > and are capable of reading the code and docstring. Docstrings don't help for things other than Modules, Classes and Functions. So, for example, are constants like FUZZ intended to be part of the external API provided by the module or not? Also, there are already a fair number of functions in test_support that don't have docstrings. Are these intended to be part of the API? Obviously one can grep around the existing tests to see what they do, but then one ends up perpetuating any errors others have made. While one could possibly attempt to make a case for "all core developers know" [1] it certainly isn't true that "all future core developers know" [2] or that "all people who could put patches into the issue tracker know" [3]. [1] Although personally I would bet against it until proven -- especially given a sensible definition of "core developer" to mean, in this case, anyone with the right to commit tests to the repository. [2] Since many of these future developers probably don't even use Python yet. [3] And we do want patches to issues to include tests, right?

2 1