mirror of https://github.com/python/cpython
cd16bf6404
svn+ssh://pythondev@svn.python.org/python/branches/p3yk ................ r55837 | guido.van.rossum | 2007-06-08 16:04:42 -0700 (Fri, 08 Jun 2007) | 2 lines PEP 3119 -- the abc module. ................ r55838 | guido.van.rossum | 2007-06-08 17:38:55 -0700 (Fri, 08 Jun 2007) | 2 lines Implement part of PEP 3119 -- One Trick Ponies. ................ r55847 | guido.van.rossum | 2007-06-09 08:28:06 -0700 (Sat, 09 Jun 2007) | 2 lines Different way to do one trick ponies, allowing registration (per PEP strawman). ................ r55849 | guido.van.rossum | 2007-06-09 18:06:38 -0700 (Sat, 09 Jun 2007) | 3 lines Make sure that the magic looking for __hash__ (etc.) doesn't apply to real subclasses of Hashable. ................ r55852 | guido.van.rossum | 2007-06-10 08:29:51 -0700 (Sun, 10 Jun 2007) | 2 lines Add some more examples, e.g. generators and dict views. ................ r55853 | guido.van.rossum | 2007-06-10 08:31:59 -0700 (Sun, 10 Jun 2007) | 2 lines keys() and items() *are* containers -- just values() isn't. ................ r55864 | georg.brandl | 2007-06-10 15:29:40 -0700 (Sun, 10 Jun 2007) | 2 lines PEP 3127: new octal literals, binary literals. ................ r55865 | georg.brandl | 2007-06-10 15:31:37 -0700 (Sun, 10 Jun 2007) | 2 lines Some octal literal fixes in Tools. ................ r55866 | georg.brandl | 2007-06-10 15:37:43 -0700 (Sun, 10 Jun 2007) | 2 lines Tokenizer changes for PEP 3127. ................ r55867 | georg.brandl | 2007-06-10 15:37:55 -0700 (Sun, 10 Jun 2007) | 2 lines Some docs for PEP 3127. ................ r55868 | georg.brandl | 2007-06-10 15:44:39 -0700 (Sun, 10 Jun 2007) | 2 lines Missed a place in intobject.c. Is that used anymore anyway? ................ r55871 | neal.norwitz | 2007-06-10 18:31:49 -0700 (Sun, 10 Jun 2007) | 182 lines Merged revisions 55729-55868 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r55731 | neal.norwitz | 2007-06-01 00:29:12 -0700 (Fri, 01 Jun 2007) | 7 lines SF 1668596/1720897: distutils now copies data files even if package_dir is empty. This needs to be backported. I'm too tired tonight. It would be great if someone backports this if the buildbots are ok with it. Otherwise, I will try to get to it tomorrow. ........ r55732 | georg.brandl | 2007-06-01 04:33:33 -0700 (Fri, 01 Jun 2007) | 2 lines Bug #1722484: remove docstrings again when running with -OO. ........ r55735 | georg.brandl | 2007-06-01 12:20:27 -0700 (Fri, 01 Jun 2007) | 2 lines Fix wrong issue number. ........ r55739 | brett.cannon | 2007-06-01 20:02:29 -0700 (Fri, 01 Jun 2007) | 3 lines Have configure raise an error when building on AtheOS. Code specific to AtheOS will be removed in Python 2.7. ........ r55746 | neal.norwitz | 2007-06-02 11:33:53 -0700 (Sat, 02 Jun 2007) | 1 line Update expected birthday of 2.6 ........ r55751 | neal.norwitz | 2007-06-03 13:32:50 -0700 (Sun, 03 Jun 2007) | 10 lines Backout the original 'fix' to 1721309 which had no effect. Different versions of Berkeley DB handle this differently. The comments and bug report should have the details. Memory is allocated in 4.4 (and presumably earlier), but not in 4.5. Thus 4.5 has the free error, but not earlier versions. Mostly update comments, plus make the free conditional. This fix was already applied to the 2.5 branch. ........ r55752 | brett.cannon | 2007-06-03 16:13:41 -0700 (Sun, 03 Jun 2007) | 6 lines Make _strptime.TimeRE().pattern() use ``\s+`` for matching whitespace instead of ``\s*``. This prevents patterns from "stealing" bits from other patterns in order to make a match work. Closes bug #1730389. Will be backported. ........ r55766 | hyeshik.chang | 2007-06-05 11:16:52 -0700 (Tue, 05 Jun 2007) | 4 lines Fix build on FreeBSD. Bluetooth HCI API in FreeBSD is quite different from Linux's. Just fix the build for now but the code doesn't support the complete capability of HCI on FreeBSD yet. ........ r55770 | hyeshik.chang | 2007-06-05 11:58:51 -0700 (Tue, 05 Jun 2007) | 4 lines Bug #1728403: Fix a bug that CJKCodecs StreamReader hangs when it reads a file that ends with incomplete sequence and sizehint argument for .read() is specified. ........ r55775 | hyeshik.chang | 2007-06-05 12:28:15 -0700 (Tue, 05 Jun 2007) | 2 lines Fix for Windows: close a temporary file before trying to delete it. ........ r55783 | guido.van.rossum | 2007-06-05 14:24:47 -0700 (Tue, 05 Jun 2007) | 2 lines Patch by Tim Delany (missing DECREF). SF #1731330. ........ r55785 | collin.winter | 2007-06-05 17:17:35 -0700 (Tue, 05 Jun 2007) | 3 lines Patch #1731049: make threading.py use a proper "raise" when checking internal state, rather than assert statements (which get stripped out by -O). ........ r55786 | facundo.batista | 2007-06-06 08:13:37 -0700 (Wed, 06 Jun 2007) | 4 lines FTP.ntransfercmd method now uses create_connection when passive, using the timeout received in connection time. ........ r55792 | facundo.batista | 2007-06-06 10:15:23 -0700 (Wed, 06 Jun 2007) | 7 lines Added an optional timeout parameter to function urllib2.urlopen, with tests in test_urllib2net.py (must have network resource enabled to execute them). Also modified test_urllib2.py because testing mock classes must take it into acount. Docs are also updated. ........ r55793 | thomas.heller | 2007-06-06 13:19:19 -0700 (Wed, 06 Jun 2007) | 1 line Build _ctypes and _ctypes_test in the ReleaseAMD64 configuration. ........ r55802 | georg.brandl | 2007-06-07 06:23:24 -0700 (Thu, 07 Jun 2007) | 3 lines Disallow function calls like foo(None=1). Backport from py3k rev. 55708 by Guido. ........ r55804 | georg.brandl | 2007-06-07 06:30:24 -0700 (Thu, 07 Jun 2007) | 2 lines Make reindent.py executable. ........ r55805 | georg.brandl | 2007-06-07 06:34:10 -0700 (Thu, 07 Jun 2007) | 2 lines Patch #1667860: Fix UnboundLocalError in urllib2. ........ r55821 | kristjan.jonsson | 2007-06-07 16:53:49 -0700 (Thu, 07 Jun 2007) | 1 line Fixing changes to getbuildinfo.c that broke linux builds ........ r55828 | thomas.heller | 2007-06-08 09:10:27 -0700 (Fri, 08 Jun 2007) | 1 line Make this test work with older Python releases where struct has no 't' format character. ........ r55829 | martin.v.loewis | 2007-06-08 10:29:20 -0700 (Fri, 08 Jun 2007) | 3 lines Bug #1733488: Fix compilation of bufferobject.c on AIX. Will backport to 2.5. ........ r55831 | thomas.heller | 2007-06-08 11:20:09 -0700 (Fri, 08 Jun 2007) | 2 lines [ 1715718 ] x64 clean compile patch for _ctypes, by Kristj?n Valur with small modifications. ........ r55832 | thomas.heller | 2007-06-08 12:01:06 -0700 (Fri, 08 Jun 2007) | 1 line Fix gcc warnings intruduced by passing Py_ssize_t to PyErr_Format calls. ........ r55833 | thomas.heller | 2007-06-08 12:08:31 -0700 (Fri, 08 Jun 2007) | 2 lines Fix wrong documentation, and correct the punktuation. Closes [1700455]. ........ r55834 | thomas.heller | 2007-06-08 12:14:23 -0700 (Fri, 08 Jun 2007) | 1 line Fix warnings by using proper function prototype. ........ r55839 | neal.norwitz | 2007-06-08 20:36:34 -0700 (Fri, 08 Jun 2007) | 7 lines Prevent expandtabs() on string and unicode objects from causing a segfault when a large width is passed on 32-bit platforms. Found by Google. It would be good for people to review this especially carefully and verify I don't have an off by one error and there is no other way to cause overflow. ........ r55841 | neal.norwitz | 2007-06-08 21:48:22 -0700 (Fri, 08 Jun 2007) | 1 line Use macro version of GET_SIZE to avoid Coverity warning (#150) about a possible error. ........ r55842 | martin.v.loewis | 2007-06-09 00:42:52 -0700 (Sat, 09 Jun 2007) | 3 lines Patch #1733960: Allow T_LONGLONG to accept ints. Will backport to 2.5. ........ r55843 | martin.v.loewis | 2007-06-09 00:58:05 -0700 (Sat, 09 Jun 2007) | 2 lines Fix Windows build. ........ r55845 | martin.v.loewis | 2007-06-09 03:10:26 -0700 (Sat, 09 Jun 2007) | 2 lines Provide LLONG_MAX for S390. ........ r55854 | thomas.heller | 2007-06-10 08:59:17 -0700 (Sun, 10 Jun 2007) | 4 lines First version of build scripts for Windows/AMD64 (no external components are built yet, and 'kill_python' is disabled). ........ r55855 | thomas.heller | 2007-06-10 10:55:51 -0700 (Sun, 10 Jun 2007) | 3 lines For now, disable the _bsddb, _sqlite3, _ssl, _testcapi, _tkinter modules in the ReleaseAMD64 configuration because they do not compile. ........ r55856 | thomas.heller | 2007-06-10 11:27:54 -0700 (Sun, 10 Jun 2007) | 1 line Need to set the environment variables, otherwise devenv.com is not found. ........ r55860 | thomas.heller | 2007-06-10 14:01:17 -0700 (Sun, 10 Jun 2007) | 1 line Revert commit 55855. ........ ................ r55880 | neal.norwitz | 2007-06-10 22:07:36 -0700 (Sun, 10 Jun 2007) | 5 lines Fix the refleak counter on test_collections. The ABC metaclass creates a registry which must be cleared on each run. Otherwise, there *seem* to be refleaks when there really aren't any. (The class is held within the registry even though it's no longer needed.) ................ r55884 | neal.norwitz | 2007-06-10 22:46:33 -0700 (Sun, 10 Jun 2007) | 1 line These tests have been removed, so they are no longer needed here ................ r55886 | georg.brandl | 2007-06-11 00:26:37 -0700 (Mon, 11 Jun 2007) | 3 lines Optimize access to True and False in the compiler (if True) and the peepholer (LOAD_NAME True). ................ r55905 | georg.brandl | 2007-06-11 10:02:26 -0700 (Mon, 11 Jun 2007) | 5 lines Remove __oct__ and __hex__ and use __index__ for converting non-ints before formatting in a base. Add a bin() builtin. ................ r55906 | georg.brandl | 2007-06-11 10:04:44 -0700 (Mon, 11 Jun 2007) | 2 lines int(x, 0) does not "guess". ................ r55907 | georg.brandl | 2007-06-11 10:05:47 -0700 (Mon, 11 Jun 2007) | 2 lines Add a comment to explain that nb_oct and nb_hex are nonfunctional. ................ r55908 | guido.van.rossum | 2007-06-11 10:49:18 -0700 (Mon, 11 Jun 2007) | 2 lines Get rid of unused imports and comment. ................ r55910 | guido.van.rossum | 2007-06-11 13:05:17 -0700 (Mon, 11 Jun 2007) | 2 lines _Abstract.__new__ now requires either no arguments or __init__ overridden. ................ r55911 | guido.van.rossum | 2007-06-11 13:07:49 -0700 (Mon, 11 Jun 2007) | 7 lines Move the collections ABCs to a separate file, _abcoll.py, in order to avoid needing to import _collections.so during the bootstrap (this will become apparent in the next submit of os.py). Add (plain and mutable) ABCs for Set, Mapping, Sequence. ................ r55912 | guido.van.rossum | 2007-06-11 13:09:31 -0700 (Mon, 11 Jun 2007) | 2 lines Rewrite the _Environ class to use the new collections ABCs. ................ r55913 | guido.van.rossum | 2007-06-11 13:59:45 -0700 (Mon, 11 Jun 2007) | 72 lines Merged revisions 55869-55912 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r55869 | neal.norwitz | 2007-06-10 17:42:11 -0700 (Sun, 10 Jun 2007) | 1 line Add Atul Varma for patch # 1667860 ........ r55870 | neal.norwitz | 2007-06-10 18:22:03 -0700 (Sun, 10 Jun 2007) | 1 line Ignore valgrind problems on Ubuntu from ld ........ r55872 | neal.norwitz | 2007-06-10 18:48:46 -0700 (Sun, 10 Jun 2007) | 2 lines Ignore config.status.lineno which seems new (new autoconf?) ........ r55873 | neal.norwitz | 2007-06-10 19:14:39 -0700 (Sun, 10 Jun 2007) | 1 line Prevent these tests from running on Win64 since they don\'t apply there either ........ r55874 | neal.norwitz | 2007-06-10 19:16:10 -0700 (Sun, 10 Jun 2007) | 5 lines Fix a bug when there was a newline in the string expandtabs was called on. This also catches another condition that can overflow. Will backport. ........ r55879 | neal.norwitz | 2007-06-10 21:52:37 -0700 (Sun, 10 Jun 2007) | 1 line Prevent hang if the port cannot be opened. ........ r55881 | neal.norwitz | 2007-06-10 22:28:45 -0700 (Sun, 10 Jun 2007) | 4 lines Add all of the distuils modules that don't seem to have explicit tests. :-( Move an import in mworkscompiler so that this module can be imported on any platform. Hopefully this works on all platforms. ........ r55882 | neal.norwitz | 2007-06-10 22:35:10 -0700 (Sun, 10 Jun 2007) | 4 lines SF #1734732, lower case the module names per PEP 8. Will backport. ........ r55885 | neal.norwitz | 2007-06-10 23:16:48 -0700 (Sun, 10 Jun 2007) | 4 lines Not sure why this only fails sometimes on Unix machines. Better to disable it and only import msvccompiler on Windows since that's the only place it can work anyways. ........ r55887 | neal.norwitz | 2007-06-11 00:29:43 -0700 (Mon, 11 Jun 2007) | 4 lines Bug #1734723: Fix repr.Repr() so it doesn't ignore the maxtuple attribute. Will backport ........ r55889 | neal.norwitz | 2007-06-11 00:36:24 -0700 (Mon, 11 Jun 2007) | 1 line Reflow long line ........ r55896 | thomas.heller | 2007-06-11 08:58:33 -0700 (Mon, 11 Jun 2007) | 3 lines Use "O&" in calls to PyArg_Parse when we need a 'void*' instead of "k" or "K" codes. ........ r55901 | facundo.batista | 2007-06-11 09:27:08 -0700 (Mon, 11 Jun 2007) | 5 lines Added versionchanged flag to all the methods which received a new optional timeout parameter, and a versionadded flag to the socket.create_connection function. ........ ................ r55914 | guido.van.rossum | 2007-06-11 14:19:50 -0700 (Mon, 11 Jun 2007) | 3 lines New super() implementation, for PEP 3135 (though the PEP is not yet updated to this design, and small tweaks may still be made later). ................ r55923 | guido.van.rossum | 2007-06-11 21:15:24 -0700 (Mon, 11 Jun 2007) | 4 lines I'm guessing this module broke when Neal ripped out the types module -- it used 'list' both as a local variable and as the built-in list type. Renamed the local variable since that was easier. ................ r55924 | guido.van.rossum | 2007-06-11 21:20:05 -0700 (Mon, 11 Jun 2007) | 5 lines Change all occurrences of super(<thisclass>, <firstarg>) to super(). Seems to have worked, all the tests still pass. Exception: test_descr and test_descrtut, which have tons of these and are there to test the various usages. ................ r55939 | collin.winter | 2007-06-12 13:57:33 -0700 (Tue, 12 Jun 2007) | 1 line Patch #1735485: remove StandardError from the exception hierarchy. ................ r55954 | neal.norwitz | 2007-06-12 21:56:32 -0700 (Tue, 12 Jun 2007) | 51 lines Merged revisions 55913-55950 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r55926 | marc-andre.lemburg | 2007-06-12 02:09:58 -0700 (Tue, 12 Jun 2007) | 3 lines Apply patch #1734945 to support TurboLinux as distribution. ........ r55927 | marc-andre.lemburg | 2007-06-12 02:26:49 -0700 (Tue, 12 Jun 2007) | 3 lines Add patch #1726668: Windows Vista support. ........ r55929 | thomas.heller | 2007-06-12 08:36:22 -0700 (Tue, 12 Jun 2007) | 1 line Checkout, but do not yet try to build, exernal sources. ........ r55930 | thomas.heller | 2007-06-12 09:08:27 -0700 (Tue, 12 Jun 2007) | 6 lines Add bufferoverflowU.lib to the libraries needed by _ssl (is this the right thing to do?). Set the /XP64 /RETAIL build enviroment in the makefile when building ReleaseAMD64. ........ r55931 | thomas.heller | 2007-06-12 09:23:19 -0700 (Tue, 12 Jun 2007) | 5 lines Revert this change, since it breaks the win32 build: Add bufferoverflowU.lib to the libraries needed by _ssl (is this the right thing to do?). ........ r55934 | thomas.heller | 2007-06-12 10:28:31 -0700 (Tue, 12 Jun 2007) | 3 lines Specify the bufferoverflowU.lib to the makefile on the command line (for ReleaseAMD64 builds). ........ r55937 | thomas.heller | 2007-06-12 12:02:59 -0700 (Tue, 12 Jun 2007) | 3 lines Add bufferoverflowU.lib to PCBuild\_bsddb.vcproj. Build sqlite3.dll and bsddb. ........ r55938 | thomas.heller | 2007-06-12 12:56:12 -0700 (Tue, 12 Jun 2007) | 2 lines Don't rebuild Berkeley DB if not needed (this was committed by accident). ........ r55948 | martin.v.loewis | 2007-06-12 20:42:19 -0700 (Tue, 12 Jun 2007) | 3 lines Provide PY_LLONG_MAX on all systems having long long. Will backport to 2.5. ........ ................ r55959 | guido.van.rossum | 2007-06-13 09:22:41 -0700 (Wed, 13 Jun 2007) | 2 lines Fix a compilation warning. ................ |
||
---|---|---|
.. | ||
package | ||
Arithmetic.py | ||
Calls.py | ||
CommandLine.py | ||
Constructs.py | ||
Dict.py | ||
Exceptions.py | ||
Imports.py | ||
Instances.py | ||
LICENSE | ||
Lists.py | ||
Lookups.py | ||
NewInstances.py | ||
Numbers.py | ||
README | ||
Setup.py | ||
Strings.py | ||
Tuples.py | ||
Unicode.py | ||
clockres.py | ||
pybench.py | ||
systimes.py |
README
________________________________________________________________________ PYBENCH - A Python Benchmark Suite ________________________________________________________________________ Extendable suite of of low-level benchmarks for measuring the performance of the Python implementation (interpreter, compiler or VM). pybench is a collection of tests that provides a standardized way to measure the performance of Python implementations. It takes a very close look at different aspects of Python programs and let's you decide which factors are more important to you than others, rather than wrapping everything up in one number, like the other performance tests do (e.g. pystone which is included in the Python Standard Library). pybench has been used in the past by several Python developers to track down performance bottlenecks or to demonstrate the impact of optimizations and new features in Python. The command line interface for pybench is the file pybench.py. Run this script with option '--help' to get a listing of the possible options. Without options, pybench will simply execute the benchmark and then print out a report to stdout. Micro-Manual ------------ Run 'pybench.py -h' to see the help screen. Run 'pybench.py' to run the benchmark suite using default settings and 'pybench.py -f <file>' to have it store the results in a file too. It is usually a good idea to run pybench.py multiple times to see whether the environment, timers and benchmark run-times are suitable for doing benchmark tests. You can use the comparison feature of pybench.py ('pybench.py -c <file>') to check how well the system behaves in comparison to a reference run. If the differences are well below 10% for each test, then you have a system that is good for doing benchmark testings. Of you get random differences of more than 10% or significant differences between the values for minimum and average time, then you likely have some background processes running which cause the readings to become inconsistent. Examples include: web-browsers, email clients, RSS readers, music players, backup programs, etc. If you are only interested in a few tests of the whole suite, you can use the filtering option, e.g. 'pybench.py -t string' will only run/show the tests that have 'string' in their name. This is the current output of pybench.py --help: """ ------------------------------------------------------------------------ PYBENCH - a benchmark test suite for Python interpreters/compilers. ------------------------------------------------------------------------ Synopsis: pybench.py [option] files... Options and default settings: -n arg number of rounds (10) -f arg save benchmark to file arg () -c arg compare benchmark with the one in file arg () -s arg show benchmark in file arg, then exit () -w arg set warp factor to arg (10) -t arg run only tests with names matching arg () -C arg set the number of calibration runs to arg (20) -d hide noise in comparisons (0) -v verbose output (not recommended) (0) --with-gc enable garbage collection (0) --with-syscheck use default sys check interval (0) --timer arg use given timer (time.time) -h show this help text --help show this help text --debug enable debugging --copyright show copyright --examples show examples of usage Version: 2.0 The normal operation is to run the suite and display the results. Use -f to save them for later reuse or comparisons. Available timers: time.time time.clock systimes.processtime Examples: python2.1 pybench.py -f p21.pybench python2.5 pybench.py -f p25.pybench python pybench.py -s p25.pybench -c p21.pybench """ License ------- See LICENSE file. Sample output ------------- """ ------------------------------------------------------------------------------- PYBENCH 2.0 ------------------------------------------------------------------------------- * using Python 2.4.2 * disabled garbage collection * system check interval set to maximum: 2147483647 * using timer: time.time Calibrating tests. Please wait... Running 10 round(s) of the suite at warp factor 10: * Round 1 done in 6.388 seconds. * Round 2 done in 6.485 seconds. * Round 3 done in 6.786 seconds. ... * Round 10 done in 6.546 seconds. ------------------------------------------------------------------------------- Benchmark: 2006-06-12 12:09:25 ------------------------------------------------------------------------------- Rounds: 10 Warp: 10 Timer: time.time Machine Details: Platform ID: Linux-2.6.8-24.19-default-x86_64-with-SuSE-9.2-x86-64 Processor: x86_64 Python: Executable: /usr/local/bin/python Version: 2.4.2 Compiler: GCC 3.3.4 (pre 3.3.5 20040809) Bits: 64bit Build: Oct 1 2005 15:24:35 (#1) Unicode: UCS2 Test minimum average operation overhead ------------------------------------------------------------------------------- BuiltinFunctionCalls: 126ms 145ms 0.28us 0.274ms BuiltinMethodLookup: 124ms 130ms 0.12us 0.316ms CompareFloats: 109ms 110ms 0.09us 0.361ms CompareFloatsIntegers: 100ms 104ms 0.12us 0.271ms CompareIntegers: 137ms 138ms 0.08us 0.542ms CompareInternedStrings: 124ms 127ms 0.08us 1.367ms CompareLongs: 100ms 104ms 0.10us 0.316ms CompareStrings: 111ms 115ms 0.12us 0.929ms CompareUnicode: 108ms 128ms 0.17us 0.693ms ConcatStrings: 142ms 155ms 0.31us 0.562ms ConcatUnicode: 119ms 127ms 0.42us 0.384ms CreateInstances: 123ms 128ms 1.14us 0.367ms CreateNewInstances: 121ms 126ms 1.49us 0.335ms CreateStringsWithConcat: 130ms 135ms 0.14us 0.916ms CreateUnicodeWithConcat: 130ms 135ms 0.34us 0.361ms DictCreation: 108ms 109ms 0.27us 0.361ms DictWithFloatKeys: 149ms 153ms 0.17us 0.678ms DictWithIntegerKeys: 124ms 126ms 0.11us 0.915ms DictWithStringKeys: 114ms 117ms 0.10us 0.905ms ForLoops: 110ms 111ms 4.46us 0.063ms IfThenElse: 118ms 119ms 0.09us 0.685ms ListSlicing: 116ms 120ms 8.59us 0.103ms NestedForLoops: 125ms 137ms 0.09us 0.019ms NormalClassAttribute: 124ms 136ms 0.11us 0.457ms NormalInstanceAttribute: 110ms 117ms 0.10us 0.454ms PythonFunctionCalls: 107ms 113ms 0.34us 0.271ms PythonMethodCalls: 140ms 149ms 0.66us 0.141ms Recursion: 156ms 166ms 3.32us 0.452ms SecondImport: 112ms 118ms 1.18us 0.180ms SecondPackageImport: 118ms 127ms 1.27us 0.180ms SecondSubmoduleImport: 140ms 151ms 1.51us 0.180ms SimpleComplexArithmetic: 128ms 139ms 0.16us 0.361ms SimpleDictManipulation: 134ms 136ms 0.11us 0.452ms SimpleFloatArithmetic: 110ms 113ms 0.09us 0.571ms SimpleIntFloatArithmetic: 106ms 111ms 0.08us 0.548ms SimpleIntegerArithmetic: 106ms 109ms 0.08us 0.544ms SimpleListManipulation: 103ms 113ms 0.10us 0.587ms SimpleLongArithmetic: 112ms 118ms 0.18us 0.271ms SmallLists: 105ms 116ms 0.17us 0.366ms SmallTuples: 108ms 128ms 0.24us 0.406ms SpecialClassAttribute: 119ms 136ms 0.11us 0.453ms SpecialInstanceAttribute: 143ms 155ms 0.13us 0.454ms StringMappings: 115ms 121ms 0.48us 0.405ms StringPredicates: 120ms 129ms 0.18us 2.064ms StringSlicing: 111ms 127ms 0.23us 0.781ms TryExcept: 125ms 126ms 0.06us 0.681ms TryRaiseExcept: 133ms 137ms 2.14us 0.361ms TupleSlicing: 117ms 120ms 0.46us 0.066ms UnicodeMappings: 156ms 160ms 4.44us 0.429ms UnicodePredicates: 117ms 121ms 0.22us 2.487ms UnicodeProperties: 115ms 153ms 0.38us 2.070ms UnicodeSlicing: 126ms 129ms 0.26us 0.689ms ------------------------------------------------------------------------------- Totals: 6283ms 6673ms """ ________________________________________________________________________ Writing New Tests ________________________________________________________________________ pybench tests are simple modules defining one or more pybench.Test subclasses. Writing a test essentially boils down to providing two methods: .test() which runs .rounds number of .operations test operations each and .calibrate() which does the same except that it doesn't actually execute the operations. Here's an example: ------------------ from pybench import Test class IntegerCounting(Test): # Version number of the test as float (x.yy); this is important # for comparisons of benchmark runs - tests with unequal version # number will not get compared. version = 1.0 # The number of abstract operations done in each round of the # test. An operation is the basic unit of what you want to # measure. The benchmark will output the amount of run-time per # operation. Note that in order to raise the measured timings # significantly above noise level, it is often required to repeat # sets of operations more than once per test round. The measured # overhead per test round should be less than 1 second. operations = 20 # Number of rounds to execute per test run. This should be # adjusted to a figure that results in a test run-time of between # 1-2 seconds (at warp 1). rounds = 100000 def test(self): """ Run the test. The test needs to run self.rounds executing self.operations number of operations each. """ # Init the test a = 1 # Run test rounds # # NOTE: Use xrange() for all test loops unless you want to face # a 20MB process ! # for i in xrange(self.rounds): # Repeat the operations per round to raise the run-time # per operation significantly above the noise level of the # for-loop overhead. # Execute 20 operations (a += 1): a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 a += 1 def calibrate(self): """ Calibrate the test. This method should execute everything that is needed to setup and run the test - except for the actual operations that you intend to measure. pybench uses this method to measure the test implementation overhead. """ # Init the test a = 1 # Run test rounds (without actually doing any operation) for i in xrange(self.rounds): # Skip the actual execution of the operations, since we # only want to measure the test's administration overhead. pass Registering a new test module ----------------------------- To register a test module with pybench, the classes need to be imported into the pybench.Setup module. pybench will then scan all the symbols defined in that module for subclasses of pybench.Test and automatically add them to the benchmark suite. Breaking Comparability ---------------------- If a change is made to any individual test that means it is no longer strictly comparable with previous runs, the '.version' class variable should be updated. Therefafter, comparisons with previous versions of the test will list as "n/a" to reflect the change. Version History --------------- 2.0: rewrote parts of pybench which resulted in more repeatable timings: - made timer a parameter - changed the platform default timer to use high-resolution timers rather than process timers (which have a much lower resolution) - added option to select timer - added process time timer (using systimes.py) - changed to use min() as timing estimator (average is still taken as well to provide an idea of the difference) - garbage collection is turned off per default - sys check interval is set to the highest possible value - calibration is now a separate step and done using a different strategy that allows measuring the test overhead more accurately - modified the tests to each give a run-time of between 100-200ms using warp 10 - changed default warp factor to 10 (from 20) - compared results with timeit.py and confirmed measurements - bumped all test versions to 2.0 - updated platform.py to the latest version - changed the output format a bit to make it look nicer - refactored the APIs somewhat 1.3+: Steve Holden added the NewInstances test and the filtering option during the NeedForSpeed sprint; this also triggered a long discussion on how to improve benchmark timing and finally resulted in the release of 2.0 1.3: initial checkin into the Python SVN repository Have fun, -- Marc-Andre Lemburg mal@lemburg.com