homepage

After a fresh clone, configure && make fails due to a fairly obvious bug in the code. Note that if you `hg revert --all`, configure && make will succeed, probably because the timestamps get updated enough to fool make.
@resist[~/projects/python:1043]% hg clone ssh://hg@hg.python.org/releasing/3.3.0
[...]
@resist[~/projects/python:1044]% cd 3.3.0/
@resist[~/projects/python/3.3.0:1045]% ./configure && make
[...]
gcc -pthread -c -Wno-unused-result -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -I. -I./Include -DPy_BUILD_CORE -o Python/_warnings.o Python/_warnings.c
./Parser/asdl_c.py -h ./Include ./Parser/Python.asdl
./Parser/asdl_c.py -c ./Python ./Parser/Python.asdl
Error visitingSum([Constructor(Load, []), Constructor(Store, []), Constructor(Del, []), Constructor(AugLoad, []), Constructor(AugStore, []), Constructor(Param, [])], [])
not all arguments converted during string formatting
Traceback (most recent call last):
 File "/home/barry/projects/python/3.Parser/asdl.py", line 309, in visit
 meth(object, *args)
 File "./Parser/asdl_c.py", line 1043, in visitSum
 self.simpleSum(sum, name)
 File "./Parser/asdl_c.py", line 1067, in simpleSum
 self.emit("default:" % name, 2)
TypeError: not all arguments converted during string formatting
make: *** [Python/Python-ast.c] Error 1

It's really puzzling that this has ever worked; the buggy code has been there since 2008, yet apparently executed successfully many times since.

It's most probably related to issue15801: if 'name' is an instance of Python-defined class, then it was considered as a mapping...

See also #15964.

New changeset 8f07ab82de92 by Ezio Melotti in branch '2.7':
#15923: fix a mistake in asdl_c.py that resulted in a TypeError after 2801bf875a24 (see #15801).
http://hg.python.org/cpython/rev/8f07ab82de92
New changeset cb988d601803 by Ezio Melotti in branch '3.2':
#15923: fix a mistake in asdl_c.py that resulted in a TypeError after 2801bf875a24 (see #15801).
http://hg.python.org/cpython/rev/cb988d601803
New changeset e7a55a393a80 by Ezio Melotti in branch '3.3':
#15923: merge with 3.2.
http://hg.python.org/cpython/rev/e7a55a393a80
New changeset b392a39970bb by Ezio Melotti in branch 'default':
#15923: merge with 3.3.
http://hg.python.org/cpython/rev/b392a39970bb 

This is now fixed.
self.emit("default:" % name, 2) worked until #15801 got fixed,
Since asdl_c.py is executed using the system Python, this was failing only when the system Python included the fix.