Python: append dictionary into dictionary list-CodePudding

What I'm trying to achieve is a dictionary containing sensors as keys and a list of dictionaries as values. The list of dictionary must be formatted {"value": xx,"timestamp": EpochTimeInMs}. The code is called cyclically to append new values to each sensor (key). The final result should be like this:

{
  "temperature": [
    {"value": 10,"timestamp": 1634336087000},
    {"value": 11,"timestamp": 1634336088765}
  ],
  "humidity": [
    {"value": 90,"timestamp": 1634336087000},
    {"value": 95,"timestamp": 1634336088765}
  ]
}'

To do so, I've tried this code:

import time

####################
my_sensors = ["temperature", "humidity"]
my_dict = {}.fromkeys(my_sensors, [])
print(my_dict)
val_template = ["value", "timestamp"]
my_val = {}.fromkeys(val_template)
my_val["timestamp"] = int(time.time()*1000)
print(my_val)

#temperature
my_val["value"] = 1234
print(my_val)
my_dict['temperature'].append(my_val.copy())

#humidity
my_val["value"] = 4321
print(my_val)
my_dict['humidity'].append(my_val.copy())

print(my_dict)

But each append seems to proceed to all keys. Here is the result from terminal:

{'temperature': [], 'humidity': []}
{'value': None, 'timestamp': 1651676483130}
{'value': 1234, 'timestamp': 1651676483130}
{'value': 4321, 'timestamp': 1651676483130}
{'temperature': [{'value': 1234, 'timestamp': 1651676483130}, {'value': 4321, 'timestamp': 1651676483130}], 'humidity': [{'value': 1234, 'timestamp': 1651676483130}, {'value': 4321, 'timestamp': 1651676483130}]}

Some help would be much appreciated.

CodePudding user response：

Appending to one list seems to appends to all lists because they are all the same list! You can check this by id(my_dict['humidity']) == id(my_dict['temperature']), which gives True.

You need to create a new list for every key, so do that:

my_dict = {s: [] for s in my_sensors}

CodePudding user response：

import time

my_sensors = ["temperature", "humidity"]
my_dict = {"temperature":[],"humidity":[]}

val_template = ["value", "timestamp"]
my_val = {"value": None, "timestamp": None}

my_val["timestamp"] = int(time.time()*1000)


#temperature
new_dict = my_val.copy()
new_dict["value"] = 1234
my_dict['temperature'].append(new_dict)


#humidity
new_dict = my_val.copy()
new_dict["value"] = 4321
my_dict['humidity'].append(new_dict)

print(my_dict)

Python internally uses pointers. Hence, the empty list ([]) in the fromkeys() function is a single list that is provided by default to each key.

CodePudding user response：

From https://docs.python.org/3/library/stdtypes.html:

classmethod fromkeys(iterable[, value])
Create a new dictionary with keys from iterable and values set to value.

fromkeys() is a class method that returns a new dictionary. value defaults to None. All of the values refer to just a single instance,
so it generally doesn’t make sense for value to be a mutable object such as an empty list. To get distinct values, use a dict comprehension instead.

Important to note the statement : "All of the values refer to just a single instance"

You can verify it by getting the address of the values:

print(id(my_dict['temperature']))
print(id(my_dict['humidity']))

2606880003968
2606880003968

You should use dict comprehension to create your dictionary like:

my_dict = dict([(sensor, []) for sensor in my_sensors])