如果使用python列表理解嵌套嵌套循环

Question

Can anyone help me figure the list comprehension way of producing following output -

Let given list be -

results = [
    {"id":  1, "name":  "input"},
    {"name": "status", "desc": "Status"},
    {"name": "entity", "fields": [
        {"id": 101, "name": "value"},
        {"id": 102, "name": "address"}]
    }
]

And I am looking for output in the form of list. The code to get the output is:

output = []
for eachDict in results:
    if "fields" in eachDict:
        for field in eachDict["fields"]:
            output.append(eachDict["name"]+"."+field["name"])
    else:
        output.append(eachDict["name"])

Thus the output using above code is -

['input', 'status', 'entity.value', 'entity.address']

Is it possible to get similar output using if else nested for loops in list Comprehension?

I am having trouble trying to get access to that inner for loop in if condition of list Comprehension

My attempt -

output = [eachDict["name"]+"."+field["name"] for field in eachDict["fields"] if "fields" in eachDict else eachDict["name"] for eachDict in results]

Answer 1

One way to transform your code into workable code would be to make the inner loop produce lists, and then flatten the result afterward.

sum(([d['name'] + '.' + f['name'] for f in d['fields']] 
    if d.get('fields') else [d['name']] for d in results), [])

Answer 2

A list comprehension has a fixed number of (nested) loops. So must have two loops here, one over the top-level dictionaries, and one over the fields . The trick is to produce one iteration in the nested loop if there are no fields:

[d['name'] + fieldname 
 for d in results
 for fieldname in (
    ('.' + sub['name'] for sub in d['fields']) if 'fields' in d else
    ('',))
]

The for fieldname loop loops either over the names of the fields sub-directories (with '.' prefixed), or over a tuple with just a single empty string in it.

Note that this isn't actually all that readable. I'd delegate producing the fieldname loop to a helper generator function to improve that:

def fieldnames(d):
    if 'fields' in d:
        for sub in d['fields']:
            yield '.' + sub['name']
    else:
        yield ''

[d['name'] + fieldname for d in results for fieldname in fieldnames(d)]

Demo:

>>> def fieldnames(d):
...     if 'fields' in d:
...         for sub in d['fields']:
...             yield '.' + sub['name']
...     else:
...         yield ''
...
>>> [d['name'] + fieldname for d in results for fieldname in fieldnames(d)]
['input', 'status', 'entity.value', 'entity.address']