[Tutor] script too slow

[Paul Tremblay]

Thanks everyone for your help. I re-wrote the script using the
suggestions of several people.

I put a snippet of the code at the bottom. 

The old script took 1 minutes and 30 seconds to run. The new one takes
57 seconds. That is a dramatic improvement.

Remember my fear was that the Python version would run 300 percent
slower than the perl one. However, this is how things are shaping up:

tokenze:

perl => 7 seconds
python => 22 seconds 
(Wow! perl really is fast here.)

processing tokens:

perl => 27 seconds
Python => 36
(Pyton might be faster in other areas later on.)

So Python is 40 percent slower in the second phase. 

I'm not saying this is in any way a benchmark. However, at this point I
am getting the script to run 40 percent slower. I can live with that.
That means a 2 minutes in perl *might* take less than 3 in Python.
Before I was looking at 6 minutes

*********************************************************


# rewritten script uses dictionaries as 'case' statements
# This replaces the multiple if elif statements
# The resulting code is amazing simple and clean

def initiate_dict(self):
        self.dict_token = 
	{

        '*'                     :   ('*', self.default_func),
        ':'                     :   (':', self.default_func),
        '{'                     :   ('{', self.ob_func),
        '}'                     :   ('}', self.cb_func),
	# ect hundreds of lines of dictionary entries
	}


def process_cw(self, token):
	# some minor preliminary stuff for small exceptions
	token, action = self.dict_token.get(token, (None, None))
        if action:   
            to_print = action(token, num) 
            return to_print

def process_tokens(self):
	# open file and read one token at a time...
	...
	    if token[0:1] == "\\":
                line = self.process_cw(token)
                if line != None:
                    write_obj.write(line)
            else:
                line = 'tx<nu<nu<nu<nu<%s\n' % token
                write_obj.write(line)
		# can I do:
		# write_obj.write('tx<nu<nu<nu<nu%s\n' % token) ?
		# probably--will have to try



# That's it! I was able to take out maybe 50 lines of code. 
# The result is code that runs 200 to 300 percent faster

**********************

-- 

************************
*Paul Tremblay         *
*phthenry@earthlink.net*
************************