| 4b8c0f6 | Guido van Rossum | 30 August 2001, 18:31:30 UTC | More stuff discovered while writing the simplest of testcases: tupledealloc(): only feed the free list when the type is really a tuple, not a subtype. Otherwise, use PyObject_GC_Del(). _PyTuple_Resize(): disallow using this for tuple subtypes. | 30 August 2001, 18:31:30 UTC |
| 46add98 | Guido van Rossum | 30 August 2001, 16:06:23 UTC | Do the int inlining only if the type is really an int, not whenever PyInt_Check() succeeds. That returns true for subtypes of int, which may override __add__ or __sub__. | 30 August 2001, 16:06:23 UTC |
| 13228a6 | Guido van Rossum | 30 August 2001, 15:54:44 UTC | Ah, the joy of writing test cases... long_subtype_new(): fix a typo (type->ob_size instead of tmp->ob_size). | 30 August 2001, 15:54:44 UTC |
| f71b5fe | Jeremy Hylton | 30 August 2001, 15:50:34 UTC | spurious pop | 30 August 2001, 15:50:34 UTC |
| 4042c69 | Neil Schemenauer | 30 August 2001, 15:38:01 UTC | Add news about GC API change. Explain how to upgrade extension modules. | 30 August 2001, 15:38:01 UTC |
| 55cdc88 | Neil Schemenauer | 30 August 2001, 15:24:17 UTC | Update documentation for GC API. Closes SF patch #421893. | 30 August 2001, 15:24:17 UTC |
| 60250e2 | Guido van Rossum | 30 August 2001, 15:07:44 UTC | win_getpass(): if sys.stdin is not sys.__stdin__, use default_getpass(). This should prevent hanging when it is called in IDLE. Fixes SF bug #455648. | 30 August 2001, 15:07:44 UTC |
| 8019913 | Michael W. Hudson | 30 August 2001, 14:50:20 UTC | fix for part of bug #453523: disable unmarshalling of code objects in restricted execution mode. | 30 August 2001, 14:50:20 UTC |
| 3c0fc84 | Fred Drake | 30 August 2001, 14:42:40 UTC | Update the links to the FIPS document that defines the Secure Hash Algorithm. This closes SF bug #454917. | 30 August 2001, 14:42:40 UTC |
| 89dfe9e | Sjoerd Mullender | 30 August 2001, 14:37:07 UTC | Removed unreachable return to silence SGI compiler. | 30 August 2001, 14:37:07 UTC |
| 6f848c1 | Sjoerd Mullender | 30 August 2001, 14:15:38 UTC | Removed an unreachable break statement to silence SGI compiler. | 30 August 2001, 14:15:38 UTC |
| a2c2ae6 | Sjoerd Mullender | 30 August 2001, 14:06:45 UTC | Removed unreachable goto statement to silence SGI compiler. | 30 August 2001, 14:06:45 UTC |
| 2f38f81 | Sjoerd Mullender | 30 August 2001, 14:05:20 UTC | Removed some unreachable break statements to silence SGI compiler. | 30 August 2001, 14:05:20 UTC |
| 38b88c2 | Sjoerd Mullender | 30 August 2001, 13:58:58 UTC | Removed some unreachable break statements to silence SGI compiler. | 30 August 2001, 13:58:58 UTC |
| 34eaf86 | Jack Jansen | 30 August 2001, 13:26:53 UTC | Started on the 2.2a2 installer | 30 August 2001, 13:26:53 UTC |
| 52e0717 | Tim Peters | 30 August 2001, 06:15:32 UTC | Give the internal immutable list type .extend and .pop methods (they "should have" been added here when they were added to lists). | 30 August 2001, 06:15:32 UTC |
| 6923234 | Tim Peters | 30 August 2001, 05:16:13 UTC | Add a new function imp.lock_held(), and use it to skip test_threaded_import when that test is doomed to deadlock. | 30 August 2001, 05:16:13 UTC |
| c414187 | Guido van Rossum | 30 August 2001, 04:43:35 UTC | Safety measures now that str and tuple are subclassable: If tp_itemsize of the basetype is nonzero, only allow empty __slots__ (declaring that no __dict__ should be added), and don't add a weakref offset. | 30 August 2001, 04:43:35 UTC |
| 31bcff8 | Guido van Rossum | 30 August 2001, 04:37:15 UTC | Make 'super' subclassable. (Not sure how useful this is yet. :-) | 30 August 2001, 04:37:15 UTC |
| e023fe0 | Guido van Rossum | 30 August 2001, 03:12:59 UTC | Make unicode subclassable. | 30 August 2001, 03:12:59 UTC |
| ae960af | Guido van Rossum | 30 August 2001, 03:11:59 UTC | Make str and tuple types subclassable. | 30 August 2001, 03:11:59 UTC |
| 147b13c | Guido van Rossum | 30 August 2001, 03:10:36 UTC | Make getset subclassable. | 30 August 2001, 03:10:36 UTC |
| d93dce1 | Guido van Rossum | 30 August 2001, 03:09:31 UTC | Fix typo: double semicolons. | 30 August 2001, 03:09:31 UTC |
| 5eef77a | Guido van Rossum | 30 August 2001, 03:08:07 UTC | Make the Py<type>_Check() macro use PyObject_TypeCheck(). | 30 August 2001, 03:08:07 UTC |
| deb77e8 | Tim Peters | 30 August 2001, 00:32:51 UTC | Squash new compiler wng in debug build. | 30 August 2001, 00:32:51 UTC |
| 45c9ae5 | Neil Schemenauer | 30 August 2001, 00:13:38 UTC | gcmodule is now always compiled | 30 August 2001, 00:13:38 UTC |
| 7d0e19e | Neil Schemenauer | 30 August 2001, 00:12:32 UTC | gcmodule is now enabled here | 30 August 2001, 00:12:32 UTC |
| 43411b5 | Neil Schemenauer | 30 August 2001, 00:05:51 UTC | Make more things internal to this file. Remove visit_finalizer_reachable since it's the same as visit_reachable. Rename visit_reachable to visit_move. Objects can now have the GC type flag set, reachable by tp_traverse and not be in a GC linked list. This should make the collector more robust and easier to use by extension module writers. Add memory management functions for container objects (new, del, resize). | 30 August 2001, 00:05:51 UTC |
| cf22946 | Neil Schemenauer | 29 August 2001, 23:58:47 UTC | Always build gcmodule. | 29 August 2001, 23:58:47 UTC |
| 69374e4 | Neil Schemenauer | 29 August 2001, 23:57:22 UTC | Flush output more aggressively. This makes things look better if the setup script is running from inside Vim. | 29 August 2001, 23:57:22 UTC |
| c806c88 | Neil Schemenauer | 29 August 2001, 23:54:54 UTC | Use new GC API. Remove usage of BASICSIZE macros. | 29 August 2001, 23:54:54 UTC |
| e83c00e | Neil Schemenauer | 29 August 2001, 23:54:21 UTC | Use new GC API. | 29 August 2001, 23:54:21 UTC |
| fd34369 | Neil Schemenauer | 29 August 2001, 23:54:03 UTC | Remove GC related code. It lives in gcmodule now. | 29 August 2001, 23:54:03 UTC |
| 4f4817f | Neil Schemenauer | 29 August 2001, 23:52:17 UTC | Make frames a PyVarObject. Use new GC API. | 29 August 2001, 23:52:17 UTC |
| 01b66a8 | Neil Schemenauer | 29 August 2001, 23:50:42 UTC | Remove bogus PyGC_HEAD_SIZE. | 29 August 2001, 23:50:42 UTC |
| 74b5ade | Neil Schemenauer | 29 August 2001, 23:49:28 UTC | Change the GC type flag since the API has changed. Allow types using the old flag to still compile. Remove the PyType_BASICSIZE and PyType_SET_BASICSIZE macros. Add PyObject_GC_New, PyObject_GC_NewVar, PyObject_GC_Resize, PyObject_GC_Del, PyObject_GC_Track, PyObject_GC_UnTrack. Part of SF patch #421893. | 29 August 2001, 23:49:28 UTC |
| 31ec142 | Neil Schemenauer | 29 August 2001, 23:46:35 UTC | Change the GC type flag since the API has changed. Allow types using the old flag to still compile. | 29 August 2001, 23:46:35 UTC |
| 251ead8 | Neil Schemenauer | 29 August 2001, 23:45:25 UTC | Make frames a PyVarObject instead of a PyObject. | 29 August 2001, 23:45:25 UTC |
| 9c63e6d | Neil Schemenauer | 29 August 2001, 23:44:38 UTC | Always compile gcmodule. | 29 August 2001, 23:44:38 UTC |
| e4685ec | Jeremy Hylton | 29 August 2001, 22:30:09 UTC | Track the block stack more reasonably in order to handle continue in try/except or try/finally. Previous versions had only track SETUP_LOOP blocks and ignored the exception part. This meant that it allowed continue inside a try/except but generated buggy code. Now it does the right thing. | 29 August 2001, 22:30:09 UTC |
| 9263848 | Jeremy Hylton | 29 August 2001, 22:27:14 UTC | Improve stack depth computation for try/except and try/finally Add CONTINUE_LOOP to the list of unconditional transfers | 29 August 2001, 22:27:14 UTC |
| 4bd4ddd | Jeremy Hylton | 29 August 2001, 22:26:35 UTC | Add __getitem__() handler for use by visitContinue() | 29 August 2001, 22:26:35 UTC |
| 7c10008 | Jack Jansen | 29 August 2001, 22:08:06 UTC | Added xx and xxsubtype modules, for completeness and because xxsubtype is used by the test suite. | 29 August 2001, 22:08:06 UTC |
| 028f2d5 | Jack Jansen | 29 August 2001, 22:04:08 UTC | Started on release notes and readme for 2.2a2. | 29 August 2001, 22:04:08 UTC |
| 79248aa | Tim Peters | 29 August 2001, 21:37:10 UTC | SF bug [#456252] Python should never stomp on [u]intptr_t. pyport.h: typedef a new Py_intptr_t type. DELICATE ASSUMPTION: That HAVE_UINTPTR_T implies intptr_t is available as well as uintptr_t. If that turns out not to be true, things must get uglier (C99 wants both, so I think it's an assumption we're *likely* to get away with). thread_nt.h, PyThread_start_new_thread: MS _beginthread is documented as returning unsigned long; no idea why uintptr_t was being used. Others: Always use Py_[u]intptr_t, never [u]intptr_t directly. | 29 August 2001, 21:37:10 UTC |
| 1936745 | Jeremy Hylton | 29 August 2001, 20:57:43 UTC | Generate SET_LINENO for list and tuple literals when the open paren starts a new line. Also fix undetected typo in visitDict() -- uncovered by recent change to add lineno attrs to atoms. | 29 August 2001, 20:57:43 UTC |
| 7845cf8 | Jeremy Hylton | 29 August 2001, 20:56:30 UTC | Make sure that atoms (Tuple, List, etc.) have lineno attributes | 29 August 2001, 20:56:30 UTC |
| 4ba9001 | Jeremy Hylton | 29 August 2001, 20:55:17 UTC | Fix off-by-one errors in code to find depth of stack. XXX The code is still widely inaccurate, but most (all?) of the time it's an overestimate. | 29 August 2001, 20:55:17 UTC |
| 8779787 | Jack Jansen | 29 August 2001, 20:26:24 UTC | Workaround by Tim Peters to skip this test if run from test.autotest, in which case it will hang because the import lock is already held by the main thread. | 29 August 2001, 20:26:24 UTC |
| bf77c46 | Jeremy Hylton | 29 August 2001, 19:45:33 UTC | Undo change from list to dict for handling varnames, consts, etc. As the doc string for _lookupName() explains: This routine uses a list instead of a dictionary, because a dictionary can't store two different keys if the keys have the same value but different types, e.g. 2 and 2L. The compiler must treat these two separately, so it does an explicit type comparison before comparing the values. | 29 August 2001, 19:45:33 UTC |
| 5a9ac97 | Jeremy Hylton | 29 August 2001, 18:17:22 UTC | Change default() to use getChildNodes() instead of getChildren() | 29 August 2001, 18:17:22 UTC |
| 94afe32 | Jeremy Hylton | 29 August 2001, 18:14:39 UTC | Support // and //= Generate SET_LINENO for del statements. Define klass=1 for PyFlowGraph constructor for a class statement. A class has no varnames. | 29 August 2001, 18:14:39 UTC |
| 7abf520 | Jeremy Hylton | 29 August 2001, 18:12:30 UTC | Add support for // and //=. Avoid if/elif/elif/else tests where the final else is supposed to handle exactly one case instead of all other cases. When the list of operators is extended, the catchall else treats all new operators as the last operator in the set of tests. Instead, raise an exception if an unexpected operator occurs. | 29 August 2001, 18:12:30 UTC |
| d4be10d | Jeremy Hylton | 29 August 2001, 18:10:51 UTC | Add generator detection to symbol table. Fix bug in handling of statements like "l[x:y] = 2". The visitor was treating this as assignments to l, x, and y! | 29 August 2001, 18:10:51 UTC |
| e4e9cd4 | Jeremy Hylton | 29 August 2001, 18:09:50 UTC | Modify name conversion to be (hopefully) a bit more efficient. Use a dictionary instead of a list to map objects to their offsets in a const/name tuple of a code object. XXX The conversion is perhaps incomplete, in that we shouldn't have to do the list2dict to start. | 29 August 2001, 18:09:50 UTC |
| 5477f52 | Jeremy Hylton | 29 August 2001, 18:08:02 UTC | Revise implementations of getChildren() and getChildNodes(). Add support for floor division (// and //=) The implementation of getChildren() and getChildNodes() is intended to be faster, because it avoids calling flatten() on every return value. But it's not clear that it is a lot faster, because constructing a tuple with just the right values ends up being slow. (Too many attribute lookups probably.) The ast.txt file is much more complicated, with funny characters at the ends of names (*, &, !) to indicate the types of each child node. The astgen script is also much more complex, making me wonder if it's still useful. | 29 August 2001, 18:08:02 UTC |
| 96d68d5 | Jeremy Hylton | 29 August 2001, 18:02:21 UTC | Add opcodes for floor division and true division (PEP 238) | 29 August 2001, 18:02:21 UTC |
| 4de8df9 | Jeremy Hylton | 29 August 2001, 17:50:27 UTC | Add tests for augmented floor division | 29 August 2001, 17:50:27 UTC |
| da8db8c | Jeremy Hylton | 29 August 2001, 17:19:02 UTC | Don't include doc string of class in its code child | 29 August 2001, 17:19:02 UTC |
| 8aea0cc | Guido van Rossum | 29 August 2001, 15:48:43 UTC | Now that int is subclassable, have to change a test that tests for non-subclassability. (More tests for number subclassing should follow.) | 29 August 2001, 15:48:43 UTC |
| bef1417 | Guido van Rossum | 29 August 2001, 15:47:46 UTC | Make int, long and float subclassable. This uses a slightly wimpy and wasteful approach, but it works. :-) | 29 August 2001, 15:47:46 UTC |
| e705ef1 | Guido van Rossum | 29 August 2001, 15:47:06 UTC | Fix super() so that it is usable for static methods (like __new__) as well. In particular, the second argument can now be a subclass of the first as well (normally it must be an instance though). | 29 August 2001, 15:47:06 UTC |
| c16fcdf | Guido van Rossum | 29 August 2001, 15:45:32 UTC | Make the PyXXX_Check() macros for the numeric types inheritance-aware. | 29 August 2001, 15:45:32 UTC |
| c51395d | Jack Jansen | 29 August 2001, 15:24:53 UTC | GUSI on the Mac creates threads with a default stack size of 20KB, which is not enough for Python. Increased the stacksize to a (somewhat arbitrary) 64KB. | 29 August 2001, 15:24:53 UTC |
| f9dc043 | Fred Drake | 29 August 2001, 02:34:10 UTC | Fix a minor markup typo. | 29 August 2001, 02:34:10 UTC |
| 8211237 | Tim Peters | 29 August 2001, 02:28:42 UTC | marshal.c r_long64: When reading a TYPE_INT64 value on a box with 32-bit ints, convert to PyLong (rather than throwing away the high-order 32 bits). | 29 August 2001, 02:28:42 UTC |
| da21ce3 | Barry Warsaw | 29 August 2001, 01:41:58 UTC | On Fred's suggestion, convert sprintf() examples to use PyString_FromFormat(). Also fixed one grammar problem, and a few other mark-up issues. Sample code not checked. | 29 August 2001, 01:41:58 UTC |
| 0ab31b8 | Andrew M. Kuchling | 29 August 2001, 01:16:54 UTC | Add METH_O and METH_NOARGS change Rewrite snprintf() item a bit Bump version number | 29 August 2001, 01:16:54 UTC |
| 19ef62d | Tim Peters | 28 August 2001, 22:21:18 UTC | pickle.py, load_int(): Match cPickle's just-repaired ability to unpickle 64-bit INTs on 32-bit boxes (where they become longs). Also exploit that int(str) and long(str) will ignore a trailing newline (saves creating a new string at the Python level). pickletester.py: Simulate reading a pickle produced by a 64-bit box. | 28 August 2001, 22:21:18 UTC |
| 12778e3 | Tim Peters | 28 August 2001, 22:08:34 UTC | load_int: The fallback to long ints was coded in such a way that it couldn't succeed. Fixed. | 28 August 2001, 22:08:34 UTC |
| 08f9956 | Barry Warsaw | 28 August 2001, 21:26:33 UTC | Update an email address. | 28 August 2001, 21:26:33 UTC |
| c6d9581 | Tim Peters | 28 August 2001, 20:56:27 UTC | Note change in fp literal syntax (e.g. "3e-" worked by accident before). | 28 August 2001, 20:56:27 UTC |
| 5592e4d | Guido van Rossum | 28 August 2001, 18:28:21 UTC | Fix a typo in SLOT0 macro for the declaration of cache_str. Dunno why I didn't catch this before. | 28 August 2001, 18:28:21 UTC |
| ce129a5 | Guido van Rossum | 28 August 2001, 18:23:24 UTC | Fix the test again due to fewer calls to __getattr__. | 28 August 2001, 18:23:24 UTC |
| 2730b13 | Guido van Rossum | 28 August 2001, 18:22:14 UTC | Finish the previous checkin: also avoid getattr when calling the method directly. | 28 August 2001, 18:22:14 UTC |
| b5f17f2 | Fred Drake | 28 August 2001, 18:09:11 UTC | Revise comments about TeX & LaTeX for accuracy. This closes SF bug #456222. | 28 August 2001, 18:09:11 UTC |
| a5be8ed | Guido van Rossum | 28 August 2001, 17:58:55 UTC | Fix one test to reflect the change in method lookup policy. | 28 August 2001, 17:58:55 UTC |
| 6071873 | Guido van Rossum | 28 August 2001, 17:47:51 UTC | Change in policy: when a slot_tp_xxx function looks for the __xxx__ method, don't use getattr, but only look in the dict of the type and base types. This prevents picking up all sorts of weird stuff, including things defined by the metaclass when the object is a class (type). For this purpose, a helper function lookup_method() was added. One or two other places also use this. | 28 August 2001, 17:47:51 UTC |
| 2ac9c3e | Jeremy Hylton | 28 August 2001, 17:28:33 UTC | Make sure the JUMP_ABSOLUTE and POP_BLOCK at the end of a for loop are contiguous. | 28 August 2001, 17:28:33 UTC |
| cbfc855 | Guido van Rossum | 28 August 2001, 16:37:51 UTC | The "O!" format code should implement an isinstance() test rather than a type equality test. | 28 August 2001, 16:37:51 UTC |
| 63db7b9 | Jeremy Hylton | 28 August 2001, 16:36:12 UTC | XXX_NAME ops should affect varnames varnames should list all the local variables (with arguments first). The XXX_NAME ops typically occur at the module level and assignment ops should create locals. | 28 August 2001, 16:36:12 UTC |
| f354575 | Jeremy Hylton | 28 August 2001, 16:35:18 UTC | Generate FOR_ITER-based loops instead of old FOR_LOOP-based loops | 28 August 2001, 16:35:18 UTC |
| 318e167 | Jeremy Hylton | 28 August 2001, 15:32:48 UTC | FOR_ITER is a jrel_op() not a plain old def_op() | 28 August 2001, 15:32:48 UTC |
| d800cff | Fred Drake | 28 August 2001, 14:56:05 UTC | Added explanation that [...] * n generates shallow copies of [...], so the contents will be shared by multiple references. This closes SF bug #455694. | 28 August 2001, 14:56:05 UTC |
| 49a806e | Jack Jansen | 28 August 2001, 14:49:00 UTC | Added list of tests expected to be skipped on the mac. | 28 August 2001, 14:49:00 UTC |
| 5cb29a4 | Fred Drake | 28 August 2001, 14:25:03 UTC | Fix a number of minor markup errors, and improve the consistency a bit. | 28 August 2001, 14:25:03 UTC |
| 46ccd1d | Eric S. Raymond | 28 August 2001, 12:50:03 UTC | Plant a mention in the description of backreferences of the fact that while \0 doesn't do what one might expect, \g<0> does. | 28 August 2001, 12:50:03 UTC |
| 8c64a54 | Barry Warsaw | 28 August 2001, 02:32:04 UTC | Refcount info for PyString_FromFormat() and PyString_FromFormatV(). Closes SF patch #455666. | 28 August 2001, 02:32:04 UTC |
| c86aa57 | Barry Warsaw | 28 August 2001, 02:31:28 UTC | Documentation for PyString_FromFormat() and PyString_FromFormatV(). Closes SF patch #455666. | 28 August 2001, 02:31:28 UTC |
| 3070ee6 | Jack Jansen | 27 August 2001, 23:16:34 UTC | Experimental feature: allow \n as well as \r as newline for text files, by breaking in to the lowlevel I/O system. Can be disabled by defining WITHOUT_UNIX_NEWLINES. | 27 August 2001, 23:16:34 UTC |
| c59e220 | Jeremy Hylton | 27 August 2001, 22:56:16 UTC | Handle private names (Hard to believe these were never handled before) Add misc.mangle() that mangles based on the rules in compile.c. XXX Need to test the corner cases Update CodeGenerator with a class_name attribute bound to None. If a particular instance is created within a class scope, the instance's class_name is bound to that class's name. Add mangle() method to CodeGenerator that mangles if the class_name has a class_name in it. Modify the FunctionCodeGenerator family to handle an extra argument-- the class_name. Wrap all name ops and attrnames in calls to self.mangle() | 27 August 2001, 22:56:16 UTC |
| e7d8322 | Jack Jansen | 27 August 2001, 22:50:41 UTC | exceptions was missing from the module list. Added. | 27 August 2001, 22:50:41 UTC |
| 535c524 | Jack Jansen | 27 August 2001, 22:31:58 UTC | A quick hack to make the test pass on the Mac (similar to the quick hack to make it pass on Windows:-). | 27 August 2001, 22:31:58 UTC |
| 80ea40d | Jeremy Hylton | 27 August 2001, 21:58:09 UTC | emit SET_LINENO for augmented assignments | 27 August 2001, 21:58:09 UTC |
| 2afff32 | Jeremy Hylton | 27 August 2001, 21:51:52 UTC | Many changes -- bug fixes and sundry improvements Make nested scopes enabled by default Add is_constant_false() helper so that compiled code and symbols are consistent with builtin compiler's handling of "if 0:" Fix doc string handling to be consistent with recent change that eliminates the doc string from the Module's node attribute. Add fix to print handling from Evan & Shane. Track change to visitor api by making "verbose" explicit. Comment out setting CO_NESTED flag (it's unnecessary in 2.2). | 27 August 2001, 21:51:52 UTC |
| 9f44815 | Tim Peters | 27 August 2001, 21:50:42 UTC | Fix another test still expecting overflow on big int literals. | 27 August 2001, 21:50:42 UTC |
| c15a828 | Tim Peters | 27 August 2001, 21:45:32 UTC | Change test_overflow to test_no_overflow; looks like big int literals are auto-coerced to longs now, but this test still expected OverflowError. I can't imagine this test failure was unique to Windows. | 27 August 2001, 21:45:32 UTC |
| 3c06b9a | Jack Jansen | 27 August 2001, 21:41:23 UTC | Use the new macresource module to open the accompanying resource file (if needed). | 27 August 2001, 21:41:23 UTC |
| a5d7da5 | Jack Jansen | 27 August 2001, 21:37:45 UTC | need() now returns the refno of the resource file opened, or None if the specified resource was already available and no file was opened. | 27 August 2001, 21:37:45 UTC |
| de3226f | Jack Jansen | 27 August 2001, 21:21:07 UTC | Module to help other modules locate their resources. It currently knows about resources in applets and separate OS9 style resource files, but it will eventually also be thought the hoops to jump through on OSX/MachO. | 27 August 2001, 21:21:07 UTC |
