Django 查询集：从查询集构建字典，具有公共元素

Question

I'm trying to construct a dictionary from my database, that will separate my data into values with common time stamps.

data_point:
    time: <timestamp>
    value: integer

I have 66k data points, out of which around 7k share timestamps (meaning the measurement was taken at the same time.

I need to make a dict that would look like:

{
  "data_array": [
    {
      "time": "2018-05-11T10:34:43.826Z",
      "values": [
        13560465,
        87856595,
        78629348
      ]
    },
    {
      "time": "2018-05-11T10:34:43.882Z",
      "values": [
        13560689,
        78237945,
        92378456
      ]
    }
  ]
}

There are other keys in the dictionary, but I'm just having a bit of a struggle with this particular key.

The idea is, look at my data queryset, and group up objects that share a timestamp , then add a key "time" to my dict, with the value being the timestamp , and an array "values" with the value being a list of those data.value objects

I'm not experienced enough to build this without looping a lot and probably being very innefficient. Some kind of "while timestamp doesn't change: append value to list", though I'm not sure how to go about that either.

Ideally, if I can do this with queries (should be faster, right?) I would prefer that.

Answer 1

Why not use collections.defaultdict ?

from collections import defaultdict

data = defaultdict(list)   

# qs is your queryset
for time, value in qs.values_list('time', 'value'):
    data[time].append(value)

In that case data looks like:

{
    'time_1': [
        value_1_1,
        value_1_2,
        ...
    ],
    'time_2': [
        value_2_1,
        value_2_2,
        ...
    ],
    ....
}

at this point you can build any output format you want