Issue #26799: Fix python-gdb.py: don't get C types once when the Python code is
loaded, but get C types on demand. The C types can change if python-gdb.py is
loaded before the Python executable. Patch written by Thomas Ilsche.
* Add py-bt-full command
* py-bt now gives an output similar to a regular Python traceback
* py-bt indicates:
- if the garbage collector is running
- if the thread is waiting for the GIL
- detect PyCFunction_Call to get the name of the builtin function
requires them. Disable executable bits and shebang lines in test and
benchmark files in order to prevent using a random system python, and in
source files of modules which don't provide command line interface.
traceback if Python is optimized.
* delay the lookup of the size_t type, it is not available at startup
* The second argument of the PyFrameObjectPtr constructor is optional, as
done in other constructors
* iter_builtins() and iter_globals() methods of PyFrameObjectPtr returns
an empty tuple instead of None if Python is optimized
* Fix py-bt to handle correctly "optimized" frames
* Frame.get_pyop() tries to get the frame pointer from PyEval_EvalCodeEx()
if the pointer is optimized out in PyEval_EvalFrameEx()
svn+ssh://pythondev@svn.python.org/python/branches/py3k
........
r84635 | antoine.pitrou | 2010-09-08 22:57:48 +0200 (mer., 08 sept. 2010) | 5 lines
Issue #9188: The gdb extension now handles correctly narrow (UCS2) as well
as wide (UCS4) unicode builds for both the host interpreter (embedded
inside gdb) and the interpreter under test.
........
r84636 | antoine.pitrou | 2010-09-08 23:07:40 +0200 (mer., 08 sept. 2010) | 4 lines
Add a safety limit to the number of unicode characters we fetch
(followup to r84635, suggested by Dave Malcolm).
........
Forward port some code from Python3:
* join surrogate pairs if sizeof(Py_UNICODE)==2
* Enable non-BMP test on narrow builds using u"\U0001D121" instead of
unichr(0x1D121)