[Zope-dev] optimised python

[Sidnei da Silva]

On Mon, Dec 19, 2005 at 07:12:44AM -0500, Jim Fulton wrote:
| Sidnei da Silva wrote:
| >On Mon, Dec 19, 2005 at 06:50:19AM -0500, Jim Fulton wrote:
| >| >There is a drawback: Debugging becomes a lot harder and sometimes  
| >| >impossible because you get no or faulty line number information in  
| >| >tracebacks and when using pdb.
| >| 
| >| That isn't true.  Line numbers are included and valid.
| >| 
| >| All -O does is:
| >| 
| >| - cause __debug__ to be false
| >| 
| >| - cause assert statements to be noop.
| >
| >I think Jens was referring to the fact that since some version (which I
| >don't remember correctly), when run with -O python will show the line
| >number of the function definition on tracebacks instead of the real
| >line number where the exception occurred.
| 
| I don't think this is the case for recent versions of Python.

Indeed, was speaking from memory, and understood the SET_LINENO
changes backwards. Mark help me out understand the issue:

"""
actually, when -O was first introduced, -O did magic with the
SET_LINENO opcode.  but in recent (even 2.3 I believe), even
non-optimised Python manages to use the same perf tricks, but keep a
correct lineno. This is reflected if you ever look in the debugger - a
frame's f_lineno structure element is indeed only ever the function
decl.  Magically f_lineno as exposed to Python does get the right
thing though...
"""

-- 
Sidnei da Silva
Enfold Systems, LLC.
http://enfoldsystems.com