For loops and conditionals in Python-CodePudding

I am new to Python and I was wondering if there was a way I could shorten/optimise the below loops:

for breakdown in data_breakdown:
    for data_source in data_source_ids:
       for camera in camera_ids:
           if (camera.get("id") == data_source.get("parent_id")) and (data_source.get("id") == breakdown.get('parent_id')):
               for res in result:
                   if res.get("camera_id") == camera.get("id"):
                       res.get('data').update({breakdown.get('name'): breakdown.get('total')})

I tried this oneliner, but it doesn't seem to work:

res.get('data').update({breakdown.get('name'): breakdown.get('total')}) for camera in camera_ids if (camera.get("id") == data_source.get("parent_id")) and (data_source.get("id") == breakdown.get('parent_id'))

CodePudding user response：

If anything, to optimize the performance of those loops, you should make them longer and more nested, with the data_source.get("id") == breakdown.get('parent_id') happening outside of the camera loop.

But there is perhaps an alternative, where you could change the structure of your data so that you don't need to loop nearly as much to find matching ID values. Convert each of your current lists (of dicts) into a single dict with its keys equal to the 'id' value you'll be trying to match in that loop, and the value being whole dict.

sources_dict = {source.get("id"): source for source in data_source_ids}
cameras_dict = {camera.get("id"): camera for camera in camera_ids}
results_dict = {res.get("camera_id"): res for res in result}

Now the whole loop only needs one level:

for breakdown in data_breakdown:
    source = sources_dict[breakdown["parent_id"]]
    camera = cameras_dict[source["parent_id"]]
    res = results_dict[camera["id"]]
    res.data[breakdown["name"]] = breakdown["total"]

This code assumes that all the lookups with get in your current code were going to succeed in getting a value. You weren't actually checking if any of the values you were getting from a get call was None, so there probably wasn't much benefit to it.

I'd further note that it's not clear if the camera loop in your original code was at all necessary. You might have been able to skip it and just directly compare data_source['parent_id'] against res['camera_id'] without comparing them both to a camera['id'] in between. In my updated version, that would translate to leaving out the creation of the cameras_dict and just directly indexing results_dict with source["parent_id"] rather than indexing to find camera first.

CodePudding user response：

You can use itertools.product to handle the nested loops for you, and I think (although I'm not sure because I can't see your data) you can skip all the .get and .update and just use the [] operator:

from itertools import product

for b, d, c in product(data_breakdown, data_source_ids, camera_ids):
    if c["id"] != d["parent_id"] or d["id"] != b["parent_id"]:
        continue
    for res in result:
        if res["camera_id"] == c["id"]:
            res['data'][b['name']] = b['total']