Op 04-10-13 23:14, Terry Reedy schreef:
On 10/4/2013 6:46 AM, Ian Kelly wrote:
On the other hand, if you start optimizing every tail call and not
just the recursive functions, then I can see where that could start to
get problematic for debugging -- as arbitrary functions get removed
from the stack traces just because they happened to end in tail calls.
The idea of CPython space-optimizing tail calls when the call is made
has been suggested on python-ideas. Guido verified that it is
technically possible with the current bytecode interpreter but rejected
it because it would arbitrarily mess up stack traces.
What does this mean?
Does it mean that a naive implementation would arbitrarily mess up
stack traces and he wasn't interested in investigating more
sophisticated implementations?
Does it mean he just didn't like the idea a stack trace wouldn't be a
100% represenatation of the active call history?
Does it mean he investigated more sphisticated implementations but found
them to have serious short comings that couldn't be remedied easily?
--
Antoon Pardon
--
https://mail.python.org/mailman/listinfo/python-list
