Features

deeppath reads, writes, deletes and searches json-like data (nested dictionaries and lists) using short /-separated paths:

from deeppath import ddelete, dget, dsearch, dset, dwalk, flatten, has

data = {
    "users": [
        {"name": "John", "surname": "Doe", "tags": ["a", "b"]},
        {"name": "Jane", "surname": "Doe", "tags": ["c"]},
    ],
    "location": {"city": "London", "country": "United Kingdom"},
}

dget(data, "location/city")        # "London"
dget(data, "users[-1]/name")       # "Jane"
dget(data, "users[*]/name")        # ["John", "Jane"]
dget(data, "users[*]/tags[*]")     # ["a", "b", "c"]
dget(data, "users[:1]/name")       # ["John"]
dget(data, "location/*")           # ["London", "United Kingdom"]
dget(data, "missing", default=0)   # 0

has(data, "users[1]/name")         # True

dset(data, "location/zip", "E1")   # creates missing keys along the way
ddelete(data, "users[*]/surname")  # True, removed from every user

list(dsearch(data, "^name$"))      # [("users[0]/name", "John"), ("users[1]/name", "Jane")]
list(dwalk(data))                  # every (path, leaf value) pair
flatten([1, [2, [3, [4]]]])        # [1, 2, 3, 4]

Summary of the API:

• dget(data, path, default=None, strict=False): return the value at path, or a list of every match when the path contains a wildcard or a slice.

• has(data, path): whether path matches anything. Stops at the first match, so it’s cheap even on large wildcard fan-outs.

• dset(data, path, value): set a value, creating intermediate dicts and lists as needed.

• ddelete(data, path): remove every match. List elements are really removed, so later indices shift down.

• dsearch(data, pattern): find every key matching a regular expression, at any depth.

• dwalk(data): yield (path, value) for every leaf.

• flatten(nested_list): flatten arbitrarily nested lists.

Path syntax:

Note that * only matches dictionary values and [*] only matches list elements: dget(data, "users/*/name") returns [] because users is a list. See the usage guide for details.

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.