Complicated list comprehension

Question

Im trying to build a list through list comprehension in python. What I have so far, and it works:

modified_list = [
    {id: metadata}
    for id, metadata in new_resource_map.items()
    if id not in old_resource_map or metadata["lastModified"] != old_resource_map[id]["lastModified"] 
]

My list called: modified_list

Every item in it is dictionary {id: metadata}

I want to add one more thing and it will look like that:

modified_list = [
    {id: metadata}
    for id, metadata in new_resource_map.items()
    if id not in old_resource_map or metadata["lastModified"] != old_resource_map[id]["lastModified"] **or
       metadata["infer_tags"] != old_resource_map[id]["infer_tags"]**
]

The problem is what the last part:

or   metadata["infer_tags"] != old_resource_map[id]["infer_tags"]

The problem is not all of the files have that field ("infer_tags").

I wanna do this last thing only after I check if this field is existing.

Is anyone know to do that?

"I wanna do this last thing only after I check if this field is existing. Is anyone know to do that?" This has nothing to do with list comprehensions. How would you handle the check normally? — Karl Knechtel, Sep 04 '22 at 12:56
not all `metadata` have it or not all `old_resource_map[id]` ? or both ? — azro, Sep 04 '22 at 12:57
From what I understand you need to add a check if this field exist before doing the logic — Yafaa Ben Tili, Sep 04 '22 at 12:58
You may need [`dict.get`](https://docs.python.org/3/library/stdtypes.html#dict.get). — Mechanic Pig, Sep 04 '22 at 12:59
If the key is missing than the logic is like before: modified_list = [ {id: metadata} for id, metadata in new_resource_map.items() if id not in old_resource_map or metadata["lastModified"] != old_resource_map[id]["lastModified"] ] — or.cohenadiv, Sep 04 '22 at 13:04

DraftyHat · Answer 1 · 2022-09-04T14:53:02.873

1

as Mechanic Pig suggests:

    if id not in old_resource_map or metadata["lastModified"] != old_resource_map[id]["lastModified"] or
       metadata.get("infer_tags", np.nan) != old_resource_map[id].get("infer_tags", np.nan)

Note that the default values used in the get() calls must not be valid values for infer_tags fields for this to be reliable.

edited Sep 04 '22 at 14:53

answered Sep 04 '22 at 13:10

DraftyHat

428
1
2
6

1

np.nan seems better for default i.e. `metadatda metadata.get("infer_tags", np.nan) != old_resource_map[id].get("infer_tags", np.nan)`, since np.nan also has the property that `np.nan != np.nan` evaluates to True. – DarrylG Sep 04 '22 at 14:03
Thanks DarrylG, that is better! Answer updated to include np.nan. – DraftyHat Sep 04 '22 at 14:53

Complicated list comprehension

1 Answers1