1
|
|
|
from typing import Dict, Iterator, List, Tuple, Type |
2
|
|
|
|
3
|
|
|
from abc import ABC, ABCMeta, abstractmethod |
4
|
|
|
|
5
|
|
|
from bson import ObjectId |
6
|
|
|
from bson.objectid import InvalidId |
7
|
|
|
|
8
|
|
|
from .database import MongomanticClient |
9
|
|
|
from .errors import ( |
10
|
|
|
DoesNotExistError, |
11
|
|
|
FieldDoesNotExistError, |
12
|
|
|
InvalidQueryError, |
13
|
|
|
MultipleObjectsReturnedError, |
14
|
|
|
WriteError, |
15
|
|
|
) |
16
|
|
|
from .mongo_model import MongoDBModel |
17
|
|
|
|
18
|
|
|
|
19
|
|
|
class ABRepositoryMeta(ABCMeta): |
20
|
|
|
"""Abstract Base Repository Metaclass |
21
|
|
|
|
22
|
|
|
This Metaclass ensures that any concrete implementations of BaseRepository |
23
|
|
|
include all necessary definitions, in order to decrease user errors. |
24
|
|
|
""" |
25
|
|
|
|
26
|
|
|
def __new__(cls, name, bases, dct): |
27
|
|
|
base_repo = super().__new__(cls, name, bases, dct) |
28
|
|
|
meta = base_repo.__dict__.get("Meta", False) |
29
|
|
|
if not meta: |
30
|
|
|
raise NotImplementedError("Internal 'Meta' not implemented") |
31
|
|
|
else: |
32
|
|
|
# Check existence of model and collection |
33
|
|
|
if not (meta.__dict__.get("model", False) and meta.__dict__.get("collection", False)): |
34
|
|
|
raise NotImplementedError("'model' or 'collection' properties are missing from internal Meta class") |
35
|
|
|
|
36
|
|
|
return base_repo |
37
|
|
|
|
38
|
|
|
|
39
|
|
|
class BaseRepository(metaclass=ABRepositoryMeta): |
40
|
|
|
class Meta: |
41
|
|
|
@property |
42
|
|
|
def model(self) -> Type[MongoDBModel]: |
43
|
|
|
"""Model class that subclasses MongoDBModel""" |
44
|
|
|
raise NotImplementedError |
45
|
|
|
|
46
|
|
|
@property |
47
|
|
|
def collection(self) -> str: |
48
|
|
|
"""String representing the MongoDB collection to use when storing this model""" |
49
|
|
|
raise NotImplementedError |
50
|
|
|
|
51
|
|
|
@classmethod |
52
|
|
|
def process_kwargs(cls, kwargs: Dict) -> Tuple: |
53
|
|
|
"""Update keyword arguments from human readable to mongo specific""" |
54
|
|
|
if "id" in kwargs: |
55
|
|
|
try: |
56
|
|
|
oid = str(kwargs.pop("id")) |
57
|
|
|
oid = ObjectId(oid) |
58
|
|
|
kwargs["_id"] = oid |
59
|
|
|
except InvalidId: |
60
|
|
|
raise InvalidQueryError(f"Invalid ObjectId {oid}.") |
61
|
|
|
|
62
|
|
|
projection = kwargs.pop("projection", None) |
63
|
|
|
skip = kwargs.pop("skip", 0) |
64
|
|
|
limit = kwargs.pop("limit", 0) |
65
|
|
|
|
66
|
|
|
for key in kwargs: |
67
|
|
|
if key not in cls.Meta.model.__fields__: |
68
|
|
|
raise FieldDoesNotExistError(f"Field {key} does not exist for model {cls.Meta.model}") |
69
|
|
|
|
70
|
|
|
return projection, skip, limit |
71
|
|
|
|
72
|
|
|
@classmethod |
73
|
|
|
def save(cls, model) -> Type[MongoDBModel]: |
74
|
|
|
"""Saves object in MongoDB""" |
75
|
|
|
try: |
76
|
|
|
document = model.to_mongo() |
77
|
|
|
res = MongomanticClient.db.__getattr__(cls.Meta.collection).insert_one(document) |
78
|
|
|
except Exception as e: |
79
|
|
|
res = None |
80
|
|
|
raise WriteError(f"Error inserting document: \n{e}") |
81
|
|
|
else: |
82
|
|
|
if res is None: |
83
|
|
|
raise WriteError("Error inserting document") |
84
|
|
|
|
85
|
|
|
document["_id"] = res.inserted_id |
86
|
|
|
return cls.Meta.model.from_mongo(document) |
87
|
|
|
|
88
|
|
|
@classmethod |
89
|
|
|
def get(cls, **kwargs) -> Type[MongoDBModel]: |
90
|
|
|
"""Get a unique document based on some filter. |
91
|
|
|
|
92
|
|
|
Args: |
93
|
|
|
kwargs: Filter keyword arguments |
94
|
|
|
|
95
|
|
|
Raises: |
96
|
|
|
DoesNotExistError: If object not found |
97
|
|
|
MultipleObjectsReturnedError: If more than one object matches filter |
98
|
|
|
|
99
|
|
|
Returns: |
100
|
|
|
Type[MongoDBModel]: Matching model |
101
|
|
|
""" |
102
|
|
|
cls.process_kwargs(kwargs) |
103
|
|
|
|
104
|
|
|
try: |
105
|
|
|
res = MongomanticClient.db.__getattr__(cls.Meta.collection).find(filter=kwargs, limit=2) |
106
|
|
|
document = next(res) |
107
|
|
|
except StopIteration: |
108
|
|
|
raise DoesNotExistError("Document not found") |
109
|
|
|
|
110
|
|
|
try: |
111
|
|
|
res = next(res) |
112
|
|
|
raise MultipleObjectsReturnedError("2 or more items returned, instead of 1") |
113
|
|
|
except StopIteration: |
114
|
|
|
return cls.Meta.model.from_mongo(document) |
115
|
|
|
|
116
|
|
|
@classmethod |
117
|
|
|
def find(cls, **kwargs) -> Iterator[Type[MongoDBModel]]: |
118
|
|
|
"""Queries database and filters on kwargs provided. |
119
|
|
|
|
120
|
|
|
Args: |
121
|
|
|
kwargs: Filter keyword arguments |
122
|
|
|
|
123
|
|
|
Reserved *optional* field names: |
124
|
|
|
projection: can either be a list of field names that should be returned in the result set |
125
|
|
|
or a dict specifying the fields to include or exclude. If projection is a list |
126
|
|
|
“_id” will always be returned. Use a dict to exclude fields from the result |
127
|
|
|
(e.g. projection={‘_id’: False}). |
128
|
|
|
skip: the number of documents to omit when returning results |
129
|
|
|
limit: the maximum number of results to return |
130
|
|
|
|
131
|
|
|
Note that invalid query errors may not be detected until the generator is consumed. |
132
|
|
|
This is because the query is not executed until the result is needed. |
133
|
|
|
|
134
|
|
|
Raises: |
135
|
|
|
InvalidQueryError: In case one or more arguments were invalid |
136
|
|
|
|
137
|
|
|
Yields: |
138
|
|
|
Iterator[Type[MongoDBModel]]: Generator that wraps PyMongo cursor and transforms documents to models |
139
|
|
|
""" |
140
|
|
|
projection, skip, limit = cls.process_kwargs(kwargs) |
141
|
|
|
|
142
|
|
|
try: |
143
|
|
|
results = MongomanticClient.db.__getattr__(cls.Meta.collection).find( |
144
|
|
|
filter=kwargs, projection=projection, skip=skip, limit=limit |
145
|
|
|
) |
146
|
|
|
for result in results: |
147
|
|
|
yield cls.Meta.model.from_mongo(result) |
148
|
|
|
except Exception as e: |
149
|
|
|
raise InvalidQueryError(f"Invalid argument types: {e}") |
150
|
|
|
|
151
|
|
|
@classmethod |
152
|
|
|
def aggregate(cls, pipeline: List[Dict]): |
153
|
|
|
try: |
154
|
|
|
results = MongomanticClient.db.__getattr__(cls.Meta.collection).aggregate(pipeline) |
155
|
|
|
for result in results: |
156
|
|
|
yield cls.Meta.model.from_mongo(result) |
157
|
|
|
except Exception as e: |
158
|
|
|
raise InvalidQueryError(f"Error executing pipeline: {e}") |
159
|
|
|
|