[Livin]

right now i've been using try/except but this gets long after two dictionaries.

i found some code on web pages and such but cannot get them to work. any help is appreciated as i need to search multiple dictionaries for keys. yes i know it is not the best way to do it but due to limitations in xbmc button/list functions i have no choice since i need to use dictionaries to efficiently store a lot of data.

here's the code i found but cannot get to work...

       dict_set = (self.dicthsdevices, self.dicthsevents, self.dicthsbtndevices)    <-- creates set of dictionaries

--- code examples to search the set --->

       val = [ d for d in dict_set if d[value] in dict_set ]
or
       for key, value in dict.iteritems(): pass

[BigBellyBilly]

i know you've asked me about this offline, but is this what your after ?

to find a key in multiple dicts:

dicthsdevices = {'device1':'device1_value', 'device2':'device2_value'}
dicthsevents = {'event1':'event1_value', 'event2':'event2_value'}
dicthsbtndevices = {'btndevices1':'btndevices1_value', 'btndevices2':'btndevices2_value'}
dict_set = (dicthsdevices, dicthsevents, dicthsbtndevices)

key = 'event2' # key that we want to find
value = none
for d in dict_set:
try:
value = d[key]
break
except: pass

if value:
print "found, value", value
else:
print 'not found'

its not as compact as a single line, but it works. is that what you were trying to achieve ?

[Asteron]

through the magic of anonymous functions and short-circuiting conditionals the python one liner:

value = reduce(lambda x,y: (not x and y.has_key(key) and y[key]) or x, dict_set , none)

[BigBellyBilly]

would you mind breaking that down in an explanation, whats happening in that statement ?

ah. got of my lazy arse and just read up on reduce() and lambda. - hard to read initially, but nice.

[Asteron]

haha yeah that stuff is powerful / compact but definitely not readable.

lambda/filter/map/reduce require a tutorial on their own... there is lots of talk of them being removed though since filter and map are the same as list comprehensions ( when you say [x.something for x in list] ) which are more clear. you can't do reduce though with a list comprehension but reduce is the hardest to understand of the three.

using reduce with the lambda function and a shortcircuited conditional should make that line effectively impossible to read.

really your example is much better.

[Livin]

i have been using the method that bbb suggested...

       for d in dict_set:
           try:
               if d.has_key(key):
                   item = d[key]
                   try: vals = d[attribs]

but what is the fastest? it must search 3 dictionaries right now, possibly more later.

i read some newsgroup posts saying this is the fastest...

[ d for d in dict_set if d[key] in dict_set ]

but i have not been able to get the syntax correct.

[Asteron]

i think if you want something like that you would use

[ d[key] for d in dict_set if d.has_key(key)]

this will return a list though of the associated value from all dicts with the key in it. if you just want the one value i still recommend bbb's code. looking up keys in maps is plenty fast and you shouldnt worry about performance here too much.

a general software rule is that you spend 80% of your time in 20% of your code.