Appending random items from one list to another- while excluding items that were already appended to a previous list.

Hey Steven, thanks so much for your response! It was very helpful.

I tried the list comprehension method. I wrote this code:

def balance_teams():
    panthers_players.append(random.sample(cleaned_player_list, k=num_players_team))
    potential_bandits_players = [player for player in cleaned_player_list if player not in panthers_players]
    bandits_players.append(random.sample(potential_bandits_players, k= num_players_team))

As you can see, my plan was to:

• add 6 random players from cleaned_player_list to panthers_players.

• create the list potential_bandits_players using list comprehension. This list would contain all players NOT currently in panthers_players.

• add 6 random players from potential_bandits_players to bandits_players. This way, I can be sure any players added to bandits_players won't already be on panthers_players.

However, this is not working as intended. I think the problem is with the list comprehension for potential_bandits_players.

When I run my script and call potential_bandits_players, I see an empty list in the console (as opposed to a list that contains all players in cleaned_player_list, except for players in panthers_players.)

Here is my script:

import constants
import random

cleaned_player_list = []

cleaned_player_set = set()

panthers_players = []

bandits_players = []

potential_bandits_players = []

warriors_players = []

num_players_team = int(len(constants.PLAYERS) / len(constants.TEAMS))


def clean_data():
    for player in constants.PLAYERS:
        fixed = {}
        fixed["name"] = player["name"]
        fixed["guardians"] = player["guardians"]
        fixed["experience"] = player["experience"]
        fixed["height"] = player["height"]
        fixed["height"] = int(player["height"][0:2])
        if player["experience"] == "YES":
            fixed["experience"] = True
        else:
            fixed["experience"] = False
        cleaned_player_list.append(fixed)
    return cleaned_player_list

def balance_teams():
    panthers_players.append(random.sample(cleaned_player_list, k=num_players_team))
    potential_bandits_players = [player for player in cleaned_player_list if player not in panthers_players]
    bandits_players.append(random.sample(potential_bandits_players, k= num_players_team))

Any idea where I'm going wrong here? I think my balance_teams function would work if the list comprehension method was working for potential_bandits_players, but I'm stuck on how to fix that.

I also have a second question, if you would be kind enough to answer!

When I saw your comment "But if these were sets instead of lists...", I tried to take the contents of cleaned_player_list and turn it into a set.

Cleaned_player_list is a single list that contains multiple dictionaries. I tried this function:

#def convert_to_set():
    cleaned_player_set = set()
    for player in cleaned_player_list:
        cleaned_player_set.add(player)
    return(cleaned_player_set)

Then I got the TypeError: unhashable type: 'dict'.

If I understand correctly, converting cleaned_player_list to a set requires converting its contents (each dictionary) into a hashable data type- like a tuple.

So, I updated the function to this:

def convert_to_set():
    for player in cleaned_player_list:
        cleaned_player_tuple = player.items()
    cleaned_player_set = set(cleaned_player_tuple)
    print(cleaned_player_set)

When I print cleaned_player_set, though, I only get 4 tuples:

>>> convert_to_set()
{('height', 41), ('experience', False), ('name', 'Kimmy Stein'), ('guardians', 'Bill Stein and Hillary Stein')}

Here's my line of thinking as to why this is happening:

• Sets can't have duplicate values
• Cleaned_player_tuple contains many tuples representing key:value pairs
• In those tuples, the "keys" are all the same (name, guardian, experience, and height.)
• The set constructor function is considering those tuples as duplicates (I'm a bit confused here, as I'm not sure why this is happening. No tuple within cleaned_player_tuple is exactly the same, as the values are different.)

That being said, would you advise that I not use sets in this scenario? Or am I creating my set in a problematic way? It seems that my balance_teams function would be much easier if I could use set operators.

Again, thank you so much! I appreciate you greatly.

Since the lists contain complex objects, simple comparisons (like what "not in" uses) aren't going to work unless you redefine the comparison method (__eq__) for the object class to make it compare data (like just the name) instead of references.

This would probably also be necessary for the set operators to work AND you'd need to define a __hash__ operation.

I think getting remove to work will also require custom comparisons. Until you're ready to redefine the internal comparison mechanisms in a class, how about managing a list of just index values and populating the object lists from it?

def balance_teams():
    available_ids = range(len(constants.PLAYERS))
    # pick first team
    panthers_ids = random.sample(available_ids, num_players_team)
    # remove choices from available list
    available_ids = [id for id in available_ids if id not in panthers_ids]
    # repeat for next team
    bandits_ids = random.sample(available_ids, num_players_team)
    available_ids = [id for id in available_ids if id not in bandits_ids]
    # anyone not picked goes in last team
    warriors_ids = available_ids
    # finally, populate the object lists
    global panthers_players, warriors_players, bandits_players
    panthers_players = [constants.PLAYERS[id] for id in panthers_ids]
    bandits_players  = [constants.PLAYERS[id] for id in bandits_ids]
    warriors_players = [constants.PLAYERS[id] for id in warriors_ids]

Wow, thank you Steven! That is a brilliant solution. It didn't even occur to me to conceptualize solving the problem in that way. It performs the balances_team function perfectly.

I learned so much about Python from this thread, all thanks to you. I appreciate you! Thank you again for your time!

To add a comment onto another comment, use the "Add Comment" on the original answer.
See the new comment I added to my answer that way.