philiprehberger-dotpath
v0.3.0878 downloadsMIT
Access and mutate deeply nested dicts using dot-notation paths
$
pip install philiprehberger-dotpathREADME
philiprehberger-dotpath
Access and mutate deeply nested dicts using dot-notation paths.
Installation
pip install philiprehberger-dotpath
Usage
from philiprehberger_dotpath import (
get, set, has, delete, flatten, unflatten, pop, merge, paths, search,
)
data = {"users": [{"name": "Alice", "email": "alice@example.com"}]}
# Get a value
get(data, "users[0].name") # "Alice"
# Get with default
get(data, "users[0].phone", default=None) # None
# Set a value (creates intermediate dicts)
set(data, "users[0].address.city", "Berlin")
# Check existence
has(data, "users[0].address.city") # True
# Delete a key
delete(data, "users[0].address.city")
Wildcards
data = {"users": [{"email": "a@b.com"}, {"email": "c@d.com"}]}
get(data, "users[*].email") # ["a@b.com", "c@d.com"]
# Set a field on all items
set(data, "users[*].active", True)
# data["users"] == [{"email": "a@b.com", "active": True}, {"email": "c@d.com", "active": True}]
# Delete a field from all items
delete(data, "users[*].active")
Negative Indexing
data = {"items": [{"name": "a"}, {"name": "b"}, {"name": "c"}]}
get(data, "items[-1].name") # "c"
set(data, "items[-2].name", "B") # updates second-to-last
Pop
data = {"a": {"b": 1, "c": 2}}
pop(data, "a.b") # 1 (key removed)
pop(data, "a.missing", default=0) # 0 (no error)
Merge
data = {"config": {"db": {"host": "localhost", "port": 3306}}}
merge(data, "config.db", {"port": 5432, "name": "mydb"})
# data["config"]["db"] == {"host": "localhost", "port": 5432, "name": "mydb"}
Search
data = {"a": 1, "b": {"c": 2, "d": [10, 20]}}
search(data, lambda v: isinstance(v, int) and v > 5)
# ["b.d[0]", "b.d[1]"]
List All Paths
paths() is a lazy iterator over every leaf path — the keys of flatten()
without materialising the dict.
data = {"a": {"b": 1}, "c": [10, 20]}
list(paths(data))
# ["a.b", "c[0]", "c[1]"]
# Useful for streaming over very large structures
for p in paths(huge_config):
print(p)
Flatten and Unflatten
nested = {"a": {"b": {"c": 1}}, "d": [10, 20]}
flatten(nested)
# {"a.b.c": 1, "d[0]": 10, "d[1]": 20}
unflatten({"a.b.c": 1, "d[0]": 10, "d[1]": 20})
# {"a": {"b": {"c": 1}}, "d": [10, 20]}
API
| Function | Description |
|---|---|
get(data, path, *, default=_MISSING) | Get value at dot path; raises KeyError if missing and no default |
set(data, path, value) | Set value at dot path, creating intermediate dicts; supports wildcards |
delete(data, path) | Delete the key at dot path; supports wildcards |
has(data, path) -> bool | Check whether a path exists |
pop(data, path, *, default=_MISSING) | Remove and return value at path; like dict.pop() |
merge(data, path, value) | Deep-merge a dict into the dict at path |
search(data, predicate) -> list[str] | Find all dot-paths where predicate(value) is True |
paths(data, *, separator=".") -> Iterator[str] | Lazy iterator over every leaf path |
flatten(data, *, separator=".") -> dict | Flatten nested dict to single level |
unflatten(data, *, separator=".") -> dict | Restore flattened dict to nested structure |
Development
pip install -e .
python -m pytest tests/ -v
Support
If you find this project useful: