|
1
|
|
|
from copy import deepcopy |
|
2
|
|
|
from typing import Any, Dict |
|
3
|
|
|
import json |
|
4
|
|
|
|
|
5
|
|
|
from omi.base import get_metadata_specification |
|
6
|
|
|
from omi.validation import validate_metadata # parse_metadata |
|
7
|
|
|
from sqlalchemy.engine import Engine |
|
8
|
|
|
|
|
9
|
|
|
from egon.data.metadata import settings |
|
10
|
|
|
|
|
11
|
|
|
|
|
12
|
|
|
class OEMetadataPackage: |
|
13
|
|
|
def __init__(self, version: str = settings.OEMETADATA_VERSION) -> None: |
|
14
|
|
|
self.spec = get_metadata_specification(version) |
|
15
|
|
|
self._doc: Dict[str, Any] = { |
|
16
|
|
|
"@context": "https://raw.githubusercontent.com/OpenEnergyPlatform/oemetadata/production/oemetadata/latest/context.json", # noqa: E501 |
|
17
|
|
|
"name": "", |
|
18
|
|
|
"title": "", |
|
19
|
|
|
"description": "", |
|
20
|
|
|
"@id": "", |
|
21
|
|
|
"resources": [], |
|
22
|
|
|
"metaMetadata": { |
|
23
|
|
|
"metadataVersion": version, |
|
24
|
|
|
"metadataLicense": {"name": "CC0-1.0"}, |
|
25
|
|
|
}, |
|
26
|
|
|
} |
|
27
|
|
|
self._validated = False |
|
28
|
|
|
|
|
29
|
|
|
def set_root( |
|
30
|
|
|
self, |
|
31
|
|
|
*, |
|
32
|
|
|
name: str, |
|
33
|
|
|
title: str = "", |
|
34
|
|
|
description: str = "", |
|
35
|
|
|
id_: str = "", |
|
36
|
|
|
) -> "OEMetadataPackage": |
|
37
|
|
|
self._doc["name"] = name |
|
38
|
|
|
self._doc["title"] = title |
|
39
|
|
|
self._doc["description"] = description |
|
40
|
|
|
self._doc["@id"] = id_ |
|
41
|
|
|
self._validated = False |
|
42
|
|
|
return self |
|
43
|
|
|
|
|
44
|
|
|
def add_resource( |
|
45
|
|
|
self, |
|
46
|
|
|
resource: dict, |
|
47
|
|
|
*, |
|
48
|
|
|
dedupe_by: str = "name", |
|
49
|
|
|
overwrite: bool = True, |
|
50
|
|
|
) -> "OEMetadataPackage": |
|
51
|
|
|
if dedupe_by and overwrite: |
|
52
|
|
|
self._doc["resources"] = [ |
|
53
|
|
|
r |
|
54
|
|
|
for r in self._doc["resources"] |
|
55
|
|
|
if r.get(dedupe_by) != resource.get(dedupe_by) |
|
56
|
|
|
] |
|
57
|
|
|
self._doc["resources"].append(deepcopy(resource)) |
|
58
|
|
|
self._validated = False |
|
59
|
|
|
return self |
|
60
|
|
|
|
|
61
|
|
|
def add_from_full_document( |
|
62
|
|
|
self, full_doc: dict, *, take_root_if_empty: bool = False |
|
63
|
|
|
) -> "OEMetadataPackage": |
|
64
|
|
|
# Optionally fill root if still empty |
|
65
|
|
|
if take_root_if_empty and not self._doc["name"]: |
|
66
|
|
|
for k in ("name", "title", "description", "@id"): |
|
67
|
|
|
if k in full_doc: |
|
68
|
|
|
self._doc[k] = full_doc[k] |
|
69
|
|
|
for r in full_doc.get("resources", []): |
|
70
|
|
|
self.add_resource(r) |
|
71
|
|
|
return self |
|
72
|
|
|
|
|
73
|
|
|
def add_from_table_comment( |
|
74
|
|
|
self, engine: Engine, schema: str, table: str |
|
75
|
|
|
) -> "OEMetadataPackage": |
|
76
|
|
|
sql = """ |
|
77
|
|
|
SELECT obj_description((quote_ident(%s)||'.'||quote_ident(%s))::regclass, 'pg_class') AS comment |
|
78
|
|
|
""" # noqa: E501 |
|
79
|
|
|
with engine.begin() as conn: |
|
80
|
|
|
comment = conn.exec_driver_sql(sql, (schema, table)).scalar() |
|
81
|
|
|
if not comment: |
|
82
|
|
|
return self |
|
83
|
|
|
try: |
|
84
|
|
|
full_doc = json.loads(comment) |
|
85
|
|
|
except Exception: |
|
86
|
|
|
return self |
|
87
|
|
|
# Optionally validate the doc before merging |
|
88
|
|
|
try: |
|
89
|
|
|
validate_metadata(full_doc, check_license=False) |
|
90
|
|
|
except Exception: |
|
91
|
|
|
pass |
|
92
|
|
|
return self.add_from_full_document(full_doc) |
|
93
|
|
|
|
|
94
|
|
|
def finalize(self, *, license_check: bool = True) -> "OEMetadataPackage": |
|
95
|
|
|
validate_metadata(self._doc, check_license=license_check) |
|
96
|
|
|
self._validated = True |
|
97
|
|
|
return self |
|
98
|
|
|
|
|
99
|
|
|
def as_dict(self) -> dict: |
|
100
|
|
|
if not self._validated: |
|
101
|
|
|
self.finalize() |
|
102
|
|
|
return deepcopy(self._doc) |
|
103
|
|
|
|
|
104
|
|
|
def as_json(self) -> str: |
|
105
|
|
|
return json.dumps(self.as_dict(), ensure_ascii=False) |
|
106
|
|
|
|
openego /
eGon-data
