cms-hte-ial2-reader 0.0.2__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,73 @@
1
+ Metadata-Version: 2.4
2
+ Name: cms-hte-ial2-reader
3
+ Version: 0.0.2
4
+ Summary: Reads IAL2 token, verifies digital signature and returns FHIR Patient resource
5
+ Author-email: Imran Qureshi <imran.qureshi@bwell.com>
6
+ Project-URL: Repository, https://github.com/icanbwell/cms-hte-ial2-reader
7
+ Classifier: Development Status :: 4 - Beta
8
+ Classifier: Programming Language :: Python :: 3
9
+ Classifier: License :: OSI Approved :: Apache Software License
10
+ Classifier: Operating System :: OS Independent
11
+ Requires-Python: >=3.10
12
+ Description-Content-Type: text/markdown
13
+ License-File: LICENSE
14
+ License-File: NOTICE
15
+ Requires-Dist: pyjwt>=2.8.0
16
+ Requires-Dist: cryptography>=43.0.0
17
+ Requires-Dist: fhirschemapy>=0.0.11
18
+ Dynamic: license-file
19
+
20
+ # cms-hte-ial2-reader
21
+
22
+ Reads IAL2 token, verifies digital signature and returns FHIR Patient resource
23
+
24
+ Accepts a signed JWT (ID Token) from a Credential Service Provider (CSP),
25
+ verifies the signature via JWKS, extracts demographics per the CSP payload
26
+ specification V7.1, and returns a FHIR R4 Patient resource.
27
+
28
+ ## Usage
29
+
30
+ ```python
31
+ from cmshteial2reader import IAL2Extractor, TokenVerifier
32
+
33
+ verifier = TokenVerifier(
34
+ jwks_uri="https://idp.example.com/.well-known/jwks.json",
35
+ audience="my-client-id",
36
+ )
37
+ extractor = IAL2Extractor(verifier=verifier)
38
+ patient = await extractor.extract(token_string)
39
+ # patient is a FHIR R4 Patient resource dict
40
+ ```
41
+
42
+ To accept tokens from any issuer on an explicit allow list (rather than a
43
+ single pinned issuer), use `MultiIssuerTokenVerifier` in place of
44
+ `TokenVerifier`:
45
+
46
+ ```python
47
+ from cmshteial2reader import IAL2Extractor, MultiIssuerTokenVerifier
48
+
49
+ verifier = MultiIssuerTokenVerifier.from_env(audience="my-client-id")
50
+ extractor = IAL2Extractor(verifier=verifier)
51
+ patient = await extractor.extract(token_string)
52
+ ```
53
+
54
+ `MultiIssuerTokenVerifier.from_env` reads its whitelist of trusted JWKS URLs
55
+ from the `IAL2_ALLOWED_JWKS_URLS` environment variable (comma-separated).
56
+
57
+ ## Quickstart
58
+ - `make init` to set up the local dev environment.
59
+ - `make up` to start the Docker dev stack.
60
+ - `make tests` to run tests in the dev container.
61
+
62
+ ## Common Tasks
63
+ - `make shell` to open a shell in the dev container.
64
+ - `make run-pre-commit` to run linting and formatting hooks.
65
+ - `make build` to build the package artifacts.
66
+
67
+ ## Publishing
68
+ - Set a token: `export TWINE_PASSWORD=pypi-xxx`
69
+ - `make package` to publish to PyPI or `make testpackage` for TestPyPI.
70
+
71
+ ## Project Metadata
72
+ - Repo: https://github.com/icanbwell/cms-hte-ial2-reader
73
+ - Author: Imran Qureshi <imran.qureshi@bwell.com>
@@ -0,0 +1,13 @@
1
+ cms_hte_ial2_reader-0.0.2.dist-info/licenses/LICENSE,sha256=8bqmh5rx8pyajtpI25opYOAsjey5tYgml_i-v1IzlHI,11928
2
+ cms_hte_ial2_reader-0.0.2.dist-info/licenses/NOTICE,sha256=xB4gR4vwAHrDAixnYEexN3a2j3jVR0KvCB2gRKovp7k,1289
3
+ cmshteial2reader/__init__.py,sha256=0hdBjWKIt12dc29P98_vC9IkZnWmvDtJvtH1YPRq6K4,787
4
+ cmshteial2reader/claims_model.py,sha256=PsrmcSECbgvh3ykVaA1gXE44Atzec4maQFW3DNoPNO0,7109
5
+ cmshteial2reader/fhir_converter.py,sha256=gPTLngLUt0_V1de0Ux3DhmGH_qblRUJf9eigAo2SRwY,8444
6
+ cmshteial2reader/ial2_extractor.py,sha256=1XprEVgK6Kb-J9F1HNS2qCTmf6bRD1bomMXxxeeVkX8,2962
7
+ cmshteial2reader/multi_issuer_verifier.py,sha256=ivFP7X9YoURVQswCx9JYGilXZXAjDjx10SGgZt7EPIs,7374
8
+ cmshteial2reader/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
9
+ cmshteial2reader/token_verifier.py,sha256=GqXdSmE6JDmpogb9Z7lZauWHIHnfjwV2WciEG8jKC_w,7299
10
+ cms_hte_ial2_reader-0.0.2.dist-info/METADATA,sha256=JMEsTtL03IPOg1w3nrFsMxquMoLBid6y0PG0MbGBhxg,2521
11
+ cms_hte_ial2_reader-0.0.2.dist-info/WHEEL,sha256=YVMoNqKzERt-wjUZwJ33xBGAwnFl-4cqbYkTtWa4itE,91
12
+ cms_hte_ial2_reader-0.0.2.dist-info/top_level.txt,sha256=kZHQ9HpkkY1Ei3nv7TiSP_XPeMD-lz0FnXqIqFware0,17
13
+ cms_hte_ial2_reader-0.0.2.dist-info/RECORD,,
@@ -0,0 +1,5 @@
1
+ Wheel-Version: 1.0
2
+ Generator: setuptools (84.0.0)
3
+ Root-Is-Purelib: true
4
+ Tag: py3-none-any
5
+
@@ -0,0 +1,215 @@
1
+ Copyright 2026 b.well Connected Health, Inc.
2
+
3
+ Licensed under the Apache License, Version 2.0 (the "License");
4
+ you may not use this file except in compliance with the License.
5
+ You may obtain a copy of the License at
6
+
7
+ http://www.apache.org/licenses/LICENSE-2.0
8
+
9
+ Unless required by applicable law or agreed to in writing, software
10
+ distributed under the License is distributed on an "AS IS" BASIS,
11
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12
+ See the License for the specific language governing permissions and
13
+ limitations under the License.
14
+
15
+ Apache License
16
+ Version 2.0, January 2004
17
+ http://www.apache.org/licenses/
18
+
19
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
20
+
21
+ 1. Definitions.
22
+
23
+ "License" shall mean the terms and conditions for use, reproduction,
24
+ and distribution as defined by Sections 1 through 9 of this document.
25
+
26
+ "Licensor" shall mean the copyright owner or entity authorized by
27
+ the copyright owner that is granting the License.
28
+
29
+ "Legal Entity" shall mean the union of the acting entity and all
30
+ other entities that control, are controlled by, or are under common
31
+ control with that entity. For the purposes of this definition,
32
+ "control" means (i) the power, direct or indirect, to cause the
33
+ direction or management of such entity, whether by contract or
34
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
35
+ outstanding shares, or (iii) beneficial ownership of such entity.
36
+
37
+ "You" (or "Your") shall mean an individual or Legal Entity
38
+ exercising permissions granted by this License.
39
+
40
+ "Source" form shall mean the preferred form for making modifications,
41
+ including but not limited to software source code, documentation
42
+ source, and configuration files.
43
+
44
+ "Object" form shall mean any form resulting from mechanical
45
+ transformation or translation of a Source form, including but
46
+ not limited to compiled object code, generated documentation,
47
+ and conversions to other media types.
48
+
49
+ "Work" shall mean the work of authorship, whether in Source or
50
+ Object form, made available under the License, as indicated by a
51
+ copyright notice that is included in or attached to the work
52
+ (an example is provided in the Appendix below).
53
+
54
+ "Derivative Works" shall mean any work, whether in Source or Object
55
+ form, that is based on (or derived from) the Work and for which the
56
+ editorial revisions, annotations, elaborations, or other modifications
57
+ represent, as a whole, an original work of authorship. For the purposes
58
+ of this License, Derivative Works shall not include works that remain
59
+ separable from, or merely link (or bind by name) to the interfaces of,
60
+ the Work and Derivative Works thereof.
61
+
62
+ "Contribution" shall mean any work of authorship, including
63
+ the original version of the Work and any modifications or additions
64
+ to that Work or Derivative Works thereof, that is intentionally
65
+ submitted to Licensor for inclusion in the Work by the copyright owner
66
+ or by an individual or Legal Entity authorized to submit on behalf of
67
+ the copyright owner. For the purposes of this definition, "submitted"
68
+ means any form of electronic, verbal, or written communication sent
69
+ to the Licensor or its representatives, including but not limited to
70
+ communication on electronic mailing lists, source code control systems,
71
+ and issue tracking systems that are managed by, or on behalf of, the
72
+ Licensor for the purpose of discussing and improving the Work, but
73
+ excluding communication that is conspicuously marked or otherwise
74
+ designated in writing by the copyright owner as "Not a Contribution."
75
+
76
+ "Contributor" shall mean Licensor and any individual or Legal Entity
77
+ on behalf of whom a Contribution has been received by Licensor and
78
+ subsequently incorporated within the Work.
79
+
80
+ 2. Grant of Copyright License. Subject to the terms and conditions of
81
+ this License, each Contributor hereby grants to You a perpetual,
82
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
83
+ copyright license to reproduce, prepare Derivative Works of,
84
+ publicly display, publicly perform, sublicense, and distribute the
85
+ Work and such Derivative Works in Source or Object form.
86
+
87
+ 3. Grant of Patent License. Subject to the terms and conditions of
88
+ this License, each Contributor hereby grants to You a perpetual,
89
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
90
+ (except as stated in this section) patent license to make, have made,
91
+ use, offer to sell, sell, import, and otherwise transfer the Work,
92
+ where such license applies only to those patent claims licensable
93
+ by such Contributor that are necessarily infringed by their
94
+ Contribution(s) alone or by combination of their Contribution(s)
95
+ with the Work to which such Contribution(s) was submitted. If You
96
+ institute patent litigation against any entity (including a
97
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
98
+ or a Contribution incorporated within the Work constitutes direct
99
+ or contributory patent infringement, then any patent licenses
100
+ granted to You under this License for that Work shall terminate
101
+ as of the date such litigation is filed.
102
+
103
+ 4. Redistribution. You may reproduce and distribute copies of the
104
+ Work or Derivative Works thereof in any medium, with or without
105
+ modifications, and in Source or Object form, provided that You
106
+ meet the following conditions:
107
+
108
+ (a) You must give any other recipients of the Work or
109
+ Derivative Works a copy of this License; and
110
+
111
+ (b) You must cause any modified files to carry prominent notices
112
+ stating that You changed the files; and
113
+
114
+ (c) You must retain, in the Source form of any Derivative Works
115
+ that You distribute, all copyright, patent, trademark, and
116
+ attribution notices from the Source form of the Work,
117
+ excluding those notices that do not pertain to any part of
118
+ the Derivative Works; and
119
+
120
+ (d) If the Work includes a "NOTICE" text file as part of its
121
+ distribution, then any Derivative Works that You distribute must
122
+ include a readable copy of the attribution notices contained
123
+ within such NOTICE file, excluding those notices that do not
124
+ pertain to any part of the Derivative Works, in at least one
125
+ of the following places: within a NOTICE text file distributed
126
+ as part of the Derivative Works; within the Source form or
127
+ documentation, if provided along with the Derivative Works; or,
128
+ within a display generated by the Derivative Works, if and
129
+ wherever such third-party notices normally appear. The contents
130
+ of the NOTICE file are for informational purposes only and
131
+ do not modify the License. You may add Your own attribution
132
+ notices within Derivative Works that You distribute, alongside
133
+ or as an addendum to the NOTICE text from the Work, provided
134
+ that such additional attribution notices cannot be construed
135
+ as modifying the License.
136
+
137
+ You may add Your own copyright statement to Your modifications and
138
+ may provide additional or different license terms and conditions
139
+ for use, reproduction, or distribution of Your modifications, or
140
+ for any such Derivative Works as a whole, provided Your use,
141
+ reproduction, and distribution of the Work otherwise complies with
142
+ the conditions stated in this License.
143
+
144
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
145
+ any Contribution intentionally submitted for inclusion in the Work
146
+ by You to the Licensor shall be under the terms and conditions of
147
+ this License, without any additional terms or conditions.
148
+ Notwithstanding the above, nothing herein shall supersede or modify
149
+ the terms of any separate license agreement you may have executed
150
+ with Licensor regarding such Contributions.
151
+
152
+ 6. Trademarks. This License does not grant permission to use the trade
153
+ names, trademarks, service marks, or product names of the Licensor,
154
+ except as required for reasonable and customary use in describing the
155
+ origin of the Work and reproducing the content of the NOTICE file.
156
+
157
+ 7. Disclaimer of Warranty. Unless required by applicable law or
158
+ agreed to in writing, Licensor provides the Work (and each
159
+ Contributor provides its Contributions) on an "AS IS" BASIS,
160
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
161
+ implied, including, without limitation, any warranties or conditions
162
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
163
+ PARTICULAR PURPOSE. You are solely responsible for determining the
164
+ appropriateness of using or redistributing the Work and assume any
165
+ risks associated with Your exercise of permissions under this License.
166
+
167
+ 8. Limitation of Liability. In no event and under no legal theory,
168
+ whether in tort (including negligence), contract, or otherwise,
169
+ unless required by applicable law (such as deliberate and grossly
170
+ negligent acts) or agreed to in writing, shall any Contributor be
171
+ liable to You for damages, including any direct, indirect, special,
172
+ incidental, or consequential damages of any character arising as a
173
+ result of this License or out of the use or inability to use the
174
+ Work (including but not limited to damages for loss of goodwill,
175
+ work stoppage, computer failure or malfunction, or any and all
176
+ other commercial damages or losses), even if such Contributor
177
+ has been advised of the possibility of such damages.
178
+
179
+ 9. Accepting Warranty or Additional Liability. While redistributing
180
+ the Work or Derivative Works thereof, You may choose to offer,
181
+ and charge a fee for, acceptance of support, warranty, indemnity,
182
+ or other liability obligations and/or rights consistent with this
183
+ License. However, in accepting such obligations, You may act only
184
+ on Your own behalf and on Your sole responsibility, not on behalf
185
+ of any other Contributor, and only if You agree to indemnify,
186
+ defend, and hold each Contributor harmless for any liability
187
+ incurred by, or claims asserted against, such Contributor by reason
188
+ of your accepting any such warranty or additional liability.
189
+
190
+ END OF TERMS AND CONDITIONS
191
+
192
+ APPENDIX: How to apply the Apache License to your work.
193
+
194
+ To apply the Apache License to your work, attach the following
195
+ boilerplate notice, with the fields enclosed by brackets "[]"
196
+ replaced with your own identifying information. (Don't include
197
+ the brackets!) The text should be enclosed in the appropriate
198
+ comment syntax for the file format. We also recommend that a
199
+ file or class name and description of purpose be included on the
200
+ same "printed page" as the copyright notice for easier
201
+ identification within third-party archives.
202
+
203
+ Copyright [yyyy] [name of copyright owner]
204
+
205
+ Licensed under the Apache License, Version 2.0 (the "License");
206
+ you may not use this file except in compliance with the License.
207
+ You may obtain a copy of the License at
208
+
209
+ http://www.apache.org/licenses/LICENSE-2.0
210
+
211
+ Unless required by applicable law or agreed to in writing, software
212
+ distributed under the License is distributed on an "AS IS" BASIS,
213
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
214
+ See the License for the specific language governing permissions and
215
+ limitations under the License.
@@ -0,0 +1,10 @@
1
+ cms-hte-ial2-reader
2
+ Copyright 2026 b.well Connected Health, Inc.
3
+
4
+ This product includes software developed by b.well Connected Health, Inc. (https://www.bwell.com).
5
+
6
+ Portions of this software may incorporate third-party open source components; see THIRD-PARTY-NOTICES.md for a complete list of those components and their respective licenses.
7
+
8
+ b.well Connected Health, Inc. is not affiliated with, and this software is not endorsed or certified by, the Centers for Medicare & Medicaid Services (CMS) or HL7 International. This software verifies signed JWT ID Tokens issued by third-party Credential Service Providers (CSPs) at NIST 800-63-3 Identity Assurance Level 2 (IAL2) and extracts demographics per those CSPs' payload specifications into a FHIR R4 Patient resource; references to CMS, IAL2, JWT/JWKS, and FHIR are solely for interoperability description and do not imply endorsement by, or affiliation with, any CSP, NIST, or HL7 International.
9
+
10
+ This software may be subject to U.S. export control laws, including the Export Administration Regulations (EAR). You may not use, export, or re-export this software in violation of applicable export laws, or to any country, entity, or individual subject to U.S. sanctions or listed on a U.S. government restricted- or denied-party list.
@@ -0,0 +1 @@
1
+ cmshteial2reader
@@ -0,0 +1,23 @@
1
+ """cmshteial2reader: Extract and verify IAL2 token demographics as FHIR Patient resources.
2
+
3
+ Accepts a signed JWT (ID Token) from a Credential Service Provider (CSP),
4
+ verifies the signature via JWKS, extracts demographics per the CSP payload
5
+ specification V7.1, and returns a FHIR R4 Patient resource.
6
+ """
7
+
8
+ from .claims_model import IAL2Address, IAL2Claims, IAL2LegalId
9
+ from .fhir_converter import IAL2ToFhirConverter
10
+ from .ial2_extractor import IAL2Extractor
11
+ from .multi_issuer_verifier import MultiIssuerTokenVerifier
12
+ from .token_verifier import TokenVerificationError, TokenVerifier
13
+
14
+ __all__ = [
15
+ "IAL2Address",
16
+ "IAL2Claims",
17
+ "IAL2Extractor",
18
+ "IAL2LegalId",
19
+ "IAL2ToFhirConverter",
20
+ "MultiIssuerTokenVerifier",
21
+ "TokenVerificationError",
22
+ "TokenVerifier",
23
+ ]
@@ -0,0 +1,211 @@
1
+ """Data models for IAL2 token claims per CSP Payload Specification V7.1.
2
+
3
+ Handles field-name aliases across CSPs (e.g. name_first vs given_name)
4
+ and normalizes them into a single canonical representation.
5
+ """
6
+
7
+ from __future__ import annotations
8
+
9
+ from dataclasses import dataclass
10
+ from typing import Any
11
+
12
+
13
+ @dataclass
14
+ class IAL2Address:
15
+ """A physical address extracted from an IAL2 token.
16
+
17
+ Attributes:
18
+ address_line1: Primary street address.
19
+ address_line2: Secondary address line (apt, suite, etc.).
20
+ city: City or locality.
21
+ state: State, province, or region.
22
+ postal_code: ZIP or postal code.
23
+ country: Country.
24
+ full_address: Composite formatted address string.
25
+ """
26
+
27
+ address_line1: str | None = None
28
+ address_line2: str | None = None
29
+ city: str | None = None
30
+ state: str | None = None
31
+ postal_code: str | None = None
32
+ country: str | None = None
33
+ full_address: str | None = None
34
+
35
+
36
+ @dataclass
37
+ class IAL2LegalId:
38
+ """A government-issued legal identity document.
39
+
40
+ Attributes:
41
+ issuer: Issuing authority (e.g. state DMV, federal government).
42
+ id_type: Type of document (e.g. drivers_license, passport).
43
+ identifier: The document number/identifier.
44
+ """
45
+
46
+ issuer: str | None = None
47
+ id_type: str | None = None
48
+ identifier: str | None = None
49
+
50
+
51
+ @dataclass
52
+ class IAL2Claims:
53
+ """Normalized demographic claims from an IAL2 token.
54
+
55
+ Fields follow the CSP Payload Specification V7.1 (Feb 2026).
56
+ Alias resolution (e.g. given_name -> name_first) happens in from_token_claims().
57
+ """
58
+
59
+ # OIDC standard claims
60
+ iss: str = ""
61
+ sub: str = ""
62
+ aud: str = ""
63
+ exp: int = 0
64
+ iat: int = 0
65
+ jti: str = ""
66
+ at_hash: str | None = None
67
+ auth_time: int | None = None
68
+ identity_assurance_level: str | None = None
69
+
70
+ # Core person attributes
71
+ name_first: str = ""
72
+ name_middle: str | None = None
73
+ name_last: str = ""
74
+ name_full: str | None = None
75
+ name_historical: list[str] | None = None
76
+ birth_date: str = ""
77
+ sex_legal: str | None = None
78
+
79
+ # Address information
80
+ address: IAL2Address | None = None
81
+ address_historical: list[IAL2Address] | None = None
82
+
83
+ # Contact information
84
+ email: str | None = None
85
+ phone_number: str | None = None
86
+ phone_number_historical: list[str] | None = None
87
+
88
+ # Legal identifiers
89
+ uuid: str | None = None
90
+ ssn_itin: str | None = None
91
+ ssn_itin_short: str | None = None
92
+ legal_id: IAL2LegalId | None = None
93
+ suffix: str | None = None
94
+
95
+ @classmethod
96
+ def from_token_claims(cls, claims: dict[str, Any]) -> IAL2Claims:
97
+ """Create an IAL2Claims instance from raw JWT claims.
98
+
99
+ Resolves field-name aliases used by different CSPs:
100
+ - given_name -> name_first
101
+ - middle_name -> name_middle
102
+ - last_name -> name_last
103
+ - street_address -> address_line1
104
+ - locality -> city
105
+ - region -> state
106
+ - postal_code / code_postal -> postal_code
107
+ - formatted -> full_address
108
+ """
109
+ address = _extract_address(claims)
110
+ historical_addresses = _extract_historical_addresses(
111
+ claims.get("address_historical")
112
+ )
113
+ legal_id = _extract_legal_id(claims)
114
+
115
+ return cls(
116
+ # OIDC standard
117
+ iss=claims.get("iss", ""),
118
+ sub=claims.get("sub", ""),
119
+ aud=claims.get("aud", ""),
120
+ exp=claims.get("exp", 0),
121
+ iat=claims.get("iat", 0),
122
+ jti=claims.get("jti", ""),
123
+ at_hash=claims.get("at_hash"),
124
+ auth_time=claims.get("auth_time"),
125
+ identity_assurance_level=claims.get("identity_assurance_level"),
126
+ # Core person
127
+ name_first=claims.get("name_first") or claims.get("given_name", ""),
128
+ name_middle=claims.get("name_middle") or claims.get("middle_name"),
129
+ name_last=claims.get("name_last") or claims.get("last_name", ""),
130
+ name_full=claims.get("name_full"),
131
+ name_historical=claims.get("name_historical"),
132
+ birth_date=claims.get("birth_date") or claims.get("date_birth", ""),
133
+ sex_legal=claims.get("sex_legal") or claims.get("gender"),
134
+ suffix=claims.get("suffix"),
135
+ # Address
136
+ address=address,
137
+ address_historical=historical_addresses,
138
+ # Contact
139
+ email=claims.get("email"),
140
+ phone_number=claims.get("phone_number"),
141
+ phone_number_historical=claims.get("phone_number_historical"),
142
+ # Legal identifiers
143
+ uuid=claims.get("UUID") or claims.get("uuid"),
144
+ ssn_itin=claims.get("ssn_itin"),
145
+ ssn_itin_short=claims.get("ssn_itin_short"),
146
+ legal_id=legal_id,
147
+ )
148
+
149
+
150
+ def _extract_address(claims: dict[str, Any]) -> IAL2Address | None:
151
+ """Extract the primary address from claims, resolving aliases."""
152
+ line1 = claims.get("address_line1") or claims.get("street_address")
153
+ city = claims.get("city") or claims.get("locality")
154
+ state = claims.get("state") or claims.get("region")
155
+ postal = claims.get("postal_code") or claims.get("code_postal")
156
+
157
+ if not any([line1, city, state, postal]):
158
+ return None
159
+
160
+ return IAL2Address(
161
+ address_line1=line1,
162
+ address_line2=claims.get("address_line2"),
163
+ city=city,
164
+ state=state,
165
+ postal_code=postal,
166
+ country=claims.get("country"),
167
+ full_address=claims.get("full_address") or claims.get("formatted"),
168
+ )
169
+
170
+
171
+ def _extract_historical_addresses(
172
+ raw: Any,
173
+ ) -> list[IAL2Address] | None:
174
+ """Parse historical addresses from the claims payload."""
175
+ if not raw or not isinstance(raw, list):
176
+ return None
177
+
178
+ addresses: list[IAL2Address] = []
179
+ for entry in raw:
180
+ if not isinstance(entry, dict):
181
+ continue
182
+
183
+ addresses.append(
184
+ IAL2Address(
185
+ address_line1=entry.get("address_line1") or entry.get("street_address"),
186
+ address_line2=entry.get("address_line2"),
187
+ city=entry.get("city") or entry.get("locality"),
188
+ state=entry.get("state") or entry.get("region"),
189
+ postal_code=entry.get("postal_code") or entry.get("code_postal"),
190
+ country=entry.get("country"),
191
+ full_address=entry.get("full_address") or entry.get("formatted"),
192
+ )
193
+ )
194
+ return addresses if addresses else None
195
+
196
+
197
+ def _extract_legal_id(claims: dict[str, Any]) -> IAL2LegalId | None:
198
+ """Extract legal identity document info from claims."""
199
+ legal_id_raw = claims.get("legal_id") or claims.get("legal_id_issuer")
200
+ if not legal_id_raw:
201
+ return None
202
+
203
+ if isinstance(legal_id_raw, dict):
204
+ return IAL2LegalId(
205
+ issuer=legal_id_raw.get("issuer"),
206
+ id_type=legal_id_raw.get("type"),
207
+ identifier=legal_id_raw.get("identifier"),
208
+ )
209
+
210
+ # If it's a string, treat it as the issuer identifier
211
+ return IAL2LegalId(issuer=str(legal_id_raw))
@@ -0,0 +1,267 @@
1
+ """Converts IAL2 demographic claims into a FHIR R4 Patient resource.
2
+
3
+ Produces a dict conforming to the FHIR R4 Patient resource structure
4
+ (https://hl7.org/fhir/R4/patient.html).
5
+ """
6
+
7
+ from __future__ import annotations
8
+
9
+ from typing import Any
10
+
11
+ from fhirschemapy.R4B.patient import Patient
12
+
13
+ from .claims_model import IAL2Address, IAL2Claims
14
+
15
+ # Mapping from CSP sex_legal values to FHIR administrative gender codes
16
+ _SEX_TO_FHIR_GENDER: dict[str, str] = {
17
+ "male": "male",
18
+ "m": "male",
19
+ "female": "female",
20
+ "f": "female",
21
+ "other": "other",
22
+ "unknown": "unknown",
23
+ }
24
+
25
+ # https://terminology.hl7.org/CodeSystem-v2-0203.html
26
+ _IDENTIFIER_TYPE_CODING_SYSTEM = "http://terminology.hl7.org/CodeSystem/v2-0203"
27
+ _SSN_IDENTIFIER_SYSTEM = "http://hl7.org/fhir/sid/us-ssn"
28
+
29
+ # FHIR Identifier.type coding codes (from the v2-0203 code system above)
30
+ _IDENTIFIER_CODE_RESOURCE_IDENTIFIER = "RI"
31
+ _IDENTIFIER_CODE_SOCIAL_SECURITY_NUMBER = "SS"
32
+ _IDENTIFIER_CODE_DRIVERS_LICENSE_NUMBER = "DL"
33
+
34
+ # FHIR HumanName.use / Address.use / ContactPoint.use values
35
+ _USE_OFFICIAL = "official"
36
+ _USE_HOME = "home"
37
+ _USE_MOBILE = "mobile"
38
+ _USE_OLD = "old"
39
+
40
+ # FHIR ContactPoint.system values
41
+ _CONTACT_SYSTEM_EMAIL = "email"
42
+ _CONTACT_SYSTEM_PHONE = "phone"
43
+
44
+
45
+ class IAL2ToFhirConverter:
46
+ """Converts IAL2Claims into a FHIR R4 Patient resource dictionary."""
47
+
48
+ def convert(self, claims: IAL2Claims) -> dict[str, Any]:
49
+ """Convert IAL2 claims to a FHIR Patient resource.
50
+
51
+ Args:
52
+ claims: The normalized IAL2 demographic claims.
53
+
54
+ Returns:
55
+ A dictionary representing a FHIR R4 Patient resource, validated
56
+ through fhirschemapy.
57
+
58
+ Raises:
59
+ pydantic.ValidationError: If the built resource doesn't conform
60
+ to the FHIR R4 Patient schema.
61
+ """
62
+ patient: dict[str, Any] = {
63
+ "resourceType": "Patient",
64
+ }
65
+
66
+ # Identifiers
67
+ identifiers = self._build_identifiers(claims)
68
+ if identifiers:
69
+ patient["identifier"] = identifiers
70
+
71
+ # Name
72
+ names = self._build_names(claims)
73
+ if names:
74
+ patient["name"] = names
75
+
76
+ # Birth date
77
+ if claims.birth_date:
78
+ patient["birthDate"] = claims.birth_date
79
+
80
+ # Gender
81
+ if claims.sex_legal:
82
+ gender = _SEX_TO_FHIR_GENDER.get(claims.sex_legal.lower(), "unknown")
83
+ patient["gender"] = gender
84
+
85
+ # Telecom (email, phone)
86
+ telecoms = self._build_telecoms(claims)
87
+ if telecoms:
88
+ patient["telecom"] = telecoms
89
+
90
+ # Address
91
+ addresses = self._build_addresses(claims)
92
+ if addresses:
93
+ patient["address"] = addresses
94
+
95
+ validated = Patient.model_validate(patient)
96
+ result: dict[str, Any] = validated.model_dump(exclude_none=True, by_alias=True)
97
+ return result
98
+
99
+ @staticmethod
100
+ def _build_identifiers(claims: IAL2Claims) -> list[dict[str, Any]]:
101
+ """Build FHIR Identifier entries from claims."""
102
+ identifiers: list[dict[str, Any]] = []
103
+
104
+ if claims.uuid:
105
+ identifiers.append(
106
+ {
107
+ "system": claims.iss,
108
+ "value": claims.uuid,
109
+ "type": {
110
+ "coding": [
111
+ {
112
+ "system": _IDENTIFIER_TYPE_CODING_SYSTEM,
113
+ "code": _IDENTIFIER_CODE_RESOURCE_IDENTIFIER,
114
+ "display": "Resource identifier",
115
+ }
116
+ ]
117
+ },
118
+ }
119
+ )
120
+
121
+ if claims.ssn_itin:
122
+ identifiers.append(
123
+ {
124
+ "system": _SSN_IDENTIFIER_SYSTEM,
125
+ "value": claims.ssn_itin,
126
+ "type": {
127
+ "coding": [
128
+ {
129
+ "system": _IDENTIFIER_TYPE_CODING_SYSTEM,
130
+ "code": _IDENTIFIER_CODE_SOCIAL_SECURITY_NUMBER,
131
+ "display": "Social Security number",
132
+ }
133
+ ]
134
+ },
135
+ }
136
+ )
137
+
138
+ if claims.legal_id:
139
+ lid = claims.legal_id
140
+ identifier_entry: dict[str, Any] = {
141
+ "type": {
142
+ "coding": [
143
+ {
144
+ "system": _IDENTIFIER_TYPE_CODING_SYSTEM,
145
+ "code": _IDENTIFIER_CODE_DRIVERS_LICENSE_NUMBER,
146
+ "display": "Driver's license number",
147
+ }
148
+ ]
149
+ },
150
+ }
151
+ if lid.identifier:
152
+ identifier_entry["value"] = lid.identifier
153
+ if lid.issuer:
154
+ identifier_entry["assigner"] = {"display": lid.issuer}
155
+ identifiers.append(identifier_entry)
156
+
157
+ return identifiers
158
+
159
+ @staticmethod
160
+ def _build_names(claims: IAL2Claims) -> list[dict[str, Any]]:
161
+ """Build FHIR HumanName entries from claims."""
162
+ names: list[dict[str, Any]] = []
163
+
164
+ # Official / current name
165
+ official: dict[str, Any] = {"use": _USE_OFFICIAL}
166
+ if claims.name_last:
167
+ official["family"] = claims.name_last
168
+ given: list[str] = []
169
+ if claims.name_first:
170
+ given.append(claims.name_first)
171
+ if claims.name_middle:
172
+ given.append(claims.name_middle)
173
+ if given:
174
+ official["given"] = given
175
+ if claims.suffix:
176
+ official["suffix"] = [claims.suffix]
177
+ if claims.name_full:
178
+ official["text"] = claims.name_full
179
+ if official.get("family") or official.get("given"):
180
+ names.append(official)
181
+
182
+ # Historical names
183
+ if claims.name_historical:
184
+ for hist_name in claims.name_historical:
185
+ names.append(
186
+ {
187
+ "use": _USE_OLD,
188
+ "text": hist_name,
189
+ }
190
+ )
191
+
192
+ return names
193
+
194
+ @staticmethod
195
+ def _build_telecoms(claims: IAL2Claims) -> list[dict[str, Any]]:
196
+ """Build FHIR ContactPoint entries from claims."""
197
+ telecoms: list[dict[str, Any]] = []
198
+
199
+ if claims.email:
200
+ telecoms.append(
201
+ {
202
+ "system": _CONTACT_SYSTEM_EMAIL,
203
+ "value": claims.email,
204
+ "use": _USE_HOME,
205
+ }
206
+ )
207
+
208
+ if claims.phone_number:
209
+ telecoms.append(
210
+ {
211
+ "system": _CONTACT_SYSTEM_PHONE,
212
+ "value": claims.phone_number,
213
+ "use": _USE_MOBILE,
214
+ }
215
+ )
216
+
217
+ if claims.phone_number_historical:
218
+ for phone in claims.phone_number_historical:
219
+ telecoms.append(
220
+ {
221
+ "system": _CONTACT_SYSTEM_PHONE,
222
+ "value": phone,
223
+ "use": _USE_OLD,
224
+ }
225
+ )
226
+
227
+ return telecoms
228
+
229
+ @staticmethod
230
+ def _build_addresses(claims: IAL2Claims) -> list[dict[str, Any]]:
231
+ """Build FHIR Address entries from claims."""
232
+ addresses: list[dict[str, Any]] = []
233
+
234
+ if claims.address:
235
+ addresses.append(_address_to_fhir(claims.address, use=_USE_HOME))
236
+
237
+ if claims.address_historical:
238
+ for hist_addr in claims.address_historical:
239
+ addresses.append(_address_to_fhir(hist_addr, use=_USE_OLD))
240
+
241
+ return addresses
242
+
243
+
244
+ def _address_to_fhir(addr: IAL2Address, *, use: str) -> dict[str, Any]:
245
+ """Convert an IAL2Address to a FHIR Address dict."""
246
+ fhir_addr: dict[str, Any] = {"use": use}
247
+
248
+ lines: list[str] = []
249
+ if addr.address_line1:
250
+ lines.append(addr.address_line1)
251
+ if addr.address_line2:
252
+ lines.append(addr.address_line2)
253
+ if lines:
254
+ fhir_addr["line"] = lines
255
+
256
+ if addr.city:
257
+ fhir_addr["city"] = addr.city
258
+ if addr.state:
259
+ fhir_addr["state"] = addr.state
260
+ if addr.postal_code:
261
+ fhir_addr["postalCode"] = addr.postal_code
262
+ if addr.country:
263
+ fhir_addr["country"] = addr.country
264
+ if addr.full_address:
265
+ fhir_addr["text"] = addr.full_address
266
+
267
+ return fhir_addr
@@ -0,0 +1,97 @@
1
+ """Main orchestrator for IAL2 token extraction.
2
+
3
+ Combines token verification and FHIR conversion into a single
4
+ entry point.
5
+ """
6
+
7
+ from __future__ import annotations
8
+
9
+ import logging
10
+ from typing import Any, Protocol
11
+
12
+ from .claims_model import IAL2Claims
13
+ from .fhir_converter import IAL2ToFhirConverter
14
+
15
+ logger = logging.getLogger(__name__)
16
+
17
+
18
+ class TokenVerifierProtocol(Protocol):
19
+ """Structural interface for anything IAL2Extractor can verify tokens with.
20
+
21
+ Both TokenVerifier (single issuer) and MultiIssuerTokenVerifier
22
+ (whitelisted issuers) satisfy this without inheritance.
23
+ """
24
+
25
+ async def verify(self, token: str) -> dict[str, Any]: ...
26
+
27
+
28
+ class IAL2Extractor:
29
+ """Accepts an IAL2 JWT, verifies it, and returns a FHIR Patient resource.
30
+
31
+ Args:
32
+ verifier: A TokenVerifier (single issuer) or MultiIssuerTokenVerifier
33
+ (whitelisted issuers) for signature validation.
34
+ converter: An IAL2ToFhirConverter instance. If None, a default
35
+ converter is created.
36
+
37
+ Example::
38
+
39
+ verifier = TokenVerifier(
40
+ jwks_uri="https://idp.example.com/.well-known/jwks.json",
41
+ audience="my-client-id",
42
+ )
43
+ extractor = IAL2Extractor(verifier=verifier)
44
+ patient = await extractor.extract(token_string)
45
+ # patient is a FHIR R4 Patient resource dict
46
+ """
47
+
48
+ def __init__(
49
+ self,
50
+ *,
51
+ verifier: TokenVerifierProtocol,
52
+ converter: IAL2ToFhirConverter | None = None,
53
+ ) -> None:
54
+ self._verifier = verifier
55
+ self._converter = converter or IAL2ToFhirConverter()
56
+
57
+ async def extract(self, token: str) -> dict[str, Any]:
58
+ """Verify an IAL2 token and return a FHIR Patient resource.
59
+
60
+ Args:
61
+ token: The encoded JWT string from the CSP.
62
+
63
+ Returns:
64
+ A FHIR R4 Patient resource as a dictionary.
65
+
66
+ Raises:
67
+ TokenVerificationError: If the token signature or claims
68
+ are invalid.
69
+ """
70
+ logger.debug("Verifying IAL2 token")
71
+ raw_claims = await self._verifier.verify(token)
72
+
73
+ logger.debug("Extracting claims from verified token")
74
+ claims = IAL2Claims.from_token_claims(raw_claims)
75
+
76
+ logger.debug("Converting claims to FHIR Patient resource")
77
+ patient = self._converter.convert(claims)
78
+
79
+ return patient
80
+
81
+ async def extract_claims(self, token: str) -> IAL2Claims:
82
+ """Verify an IAL2 token and return the normalized claims.
83
+
84
+ Useful when you need the structured claims without FHIR conversion.
85
+
86
+ Args:
87
+ token: The encoded JWT string from the CSP.
88
+
89
+ Returns:
90
+ An IAL2Claims dataclass with normalized demographics.
91
+
92
+ Raises:
93
+ TokenVerificationError: If the token signature or claims
94
+ are invalid.
95
+ """
96
+ raw_claims = await self._verifier.verify(token)
97
+ return IAL2Claims.from_token_claims(raw_claims)
@@ -0,0 +1,198 @@
1
+ """Multi-issuer JWT verification restricted to a whitelist of JWKS URLs.
2
+
3
+ Resolves an incoming token's issuer via OIDC discovery, then only proceeds
4
+ with signature verification if the discovered JWKS URL is on an explicit
5
+ allow list -- an unrecognized issuer is rejected before any signing key is
6
+ fetched.
7
+ """
8
+
9
+ from __future__ import annotations
10
+
11
+ import asyncio
12
+ import ipaddress
13
+ import os
14
+ import socket
15
+ from typing import Any
16
+ from urllib.parse import urlparse
17
+
18
+ import jwt
19
+
20
+ from .token_verifier import TokenVerificationError, TokenVerifier
21
+
22
+ _IpAddress = ipaddress.IPv4Address | ipaddress.IPv6Address
23
+
24
+ ALLOWED_JWKS_URLS_ENV_VAR = "IAL2_ALLOWED_JWKS_URLS"
25
+
26
+
27
+ class MultiIssuerTokenVerifier:
28
+ """Verifies IAL2 JWTs from any issuer whose JWKS URL is whitelisted.
29
+
30
+ Unlike TokenVerifier, which is pinned to a single jwks_uri, this class
31
+ accepts tokens from multiple issuers by resolving each token's issuer
32
+ via OIDC discovery and checking the discovered JWKS URL against an
33
+ explicit allow list before trusting it.
34
+
35
+ Args:
36
+ audience: Expected ``aud`` claim value, applied to every issuer.
37
+ allowed_jwks_uris: JWKS URLs this instance will trust.
38
+ algorithms: Allowed signing algorithms. Defaults to RS256.
39
+
40
+ Example::
41
+
42
+ verifier = MultiIssuerTokenVerifier.from_env(audience="my-client-id")
43
+ claims = await verifier.verify(token_string)
44
+ """
45
+
46
+ def __init__(
47
+ self,
48
+ *,
49
+ audience: str,
50
+ allowed_jwks_uris: list[str],
51
+ algorithms: list[str] | None = None,
52
+ ) -> None:
53
+ self._audience = audience
54
+ self._allowed_jwks_uris = set(allowed_jwks_uris)
55
+ self._algorithms = algorithms
56
+ self._verifiers_by_issuer: dict[str, TokenVerifier] = {}
57
+
58
+ @classmethod
59
+ def from_env(
60
+ cls,
61
+ *,
62
+ audience: str,
63
+ env_var: str = ALLOWED_JWKS_URLS_ENV_VAR,
64
+ algorithms: list[str] | None = None,
65
+ ) -> MultiIssuerTokenVerifier:
66
+ """Build the verifier from a comma-separated env var of JWKS URLs.
67
+
68
+ Args:
69
+ audience: Expected ``aud`` claim value.
70
+ env_var: Name of the env var holding the whitelist.
71
+ algorithms: Allowed signing algorithms.
72
+
73
+ Raises:
74
+ ValueError: If the env var is unset or contains no URLs.
75
+ """
76
+ raw = os.environ.get(env_var, "")
77
+ allowed = [url.strip() for url in raw.split(",") if url.strip()]
78
+ if not allowed:
79
+ raise ValueError(f"{env_var} is not set or contains no JWKS URLs")
80
+ return cls(audience=audience, allowed_jwks_uris=allowed, algorithms=algorithms)
81
+
82
+ async def verify(self, token: str) -> dict[str, Any]:
83
+ """Verify a token from any whitelisted issuer.
84
+
85
+ Args:
86
+ token: The encoded JWT string.
87
+
88
+ Returns:
89
+ The decoded claims dictionary.
90
+
91
+ Raises:
92
+ TokenVerificationError: If the issuer's JWKS URL isn't
93
+ whitelisted, or the token fails verification.
94
+ """
95
+ issuer = self._peek_issuer(token)
96
+
97
+ verifier = self._verifiers_by_issuer.get(issuer)
98
+ if verifier is None:
99
+ verifier = await self._build_verifier_for_issuer(issuer)
100
+ self._verifiers_by_issuer[issuer] = verifier
101
+
102
+ return await verifier.verify(token)
103
+
104
+ @staticmethod
105
+ def _peek_issuer(token: str) -> str:
106
+ """Read the ``iss`` claim without verifying the signature.
107
+
108
+ This claim is untrusted at this point -- it is only used to select
109
+ which JWKS URL to check against the whitelist, never to make a
110
+ trust decision on its own. The real verification happens in
111
+ TokenVerifier.verify() once the whitelist check passes.
112
+ """
113
+ try:
114
+ unverified = jwt.decode(token, options={"verify_signature": False})
115
+ except jwt.InvalidTokenError as e:
116
+ raise TokenVerificationError(f"Could not parse token: {e}") from e
117
+
118
+ issuer = unverified.get("iss")
119
+ if not issuer:
120
+ raise TokenVerificationError("Token is missing the 'iss' claim")
121
+ return str(issuer)
122
+
123
+ async def _build_verifier_for_issuer(self, issuer: str) -> TokenVerifier:
124
+ """Discover the issuer's JWKS URL and check it against the whitelist."""
125
+ pinned_ip = await self._validate_issuer_url(issuer)
126
+
127
+ verifier = await TokenVerifier.from_oidc_discovery(
128
+ discovery_url=f"{issuer.rstrip('/')}/.well-known/openid-configuration",
129
+ audience=self._audience,
130
+ issuer=issuer,
131
+ algorithms=self._algorithms,
132
+ pinned_ip=pinned_ip,
133
+ )
134
+ if verifier.jwks_uri not in self._allowed_jwks_uris:
135
+ raise TokenVerificationError(
136
+ f"Issuer '{issuer}' resolved to JWKS URL "
137
+ f"'{verifier.jwks_uri}', which is not whitelisted"
138
+ )
139
+ return verifier
140
+
141
+ @staticmethod
142
+ async def _validate_issuer_url(issuer: str) -> str:
143
+ """Reject issuer URLs that could be used for SSRF before any
144
+ discovery or JWKS request is made, and return the validated IP to
145
+ pin the actual discovery connection to.
146
+
147
+ The issuer claim is unverified at this point -- read via
148
+ _peek_issuer() before signature verification, purely to select a
149
+ JWKS URL. Without this check, an attacker could set 'iss' to an
150
+ internal service or cloud metadata endpoint and have this process
151
+ make a request to it during OIDC discovery. Returning the validated
152
+ IP (rather than just validating and discarding it) lets the caller
153
+ pin the actual connection to it, closing the DNS-rebinding window
154
+ between this check and the request.
155
+ """
156
+ parsed = urlparse(issuer)
157
+ if parsed.scheme != "https":
158
+ raise TokenVerificationError(f"Issuer URL must use https, got: {issuer!r}")
159
+
160
+ hostname = parsed.hostname
161
+ if not hostname:
162
+ raise TokenVerificationError(f"Issuer URL is missing a host: {issuer!r}")
163
+
164
+ addresses = await MultiIssuerTokenVerifier._resolve_addresses(hostname)
165
+ for address in addresses:
166
+ if (
167
+ address.is_private
168
+ or address.is_loopback
169
+ or address.is_link_local
170
+ or address.is_multicast
171
+ or address.is_reserved
172
+ or address.is_unspecified
173
+ ):
174
+ raise TokenVerificationError(
175
+ f"Issuer '{issuer}' resolves to a non-public address "
176
+ f"({address}); rejected"
177
+ )
178
+ return str(addresses[0])
179
+
180
+ @staticmethod
181
+ async def _resolve_addresses(hostname: str) -> list[_IpAddress]:
182
+ """Resolve a hostname to the IP address(es) it points to.
183
+
184
+ Guards against DNS rebinding: a hostname that looks external in
185
+ text form can still resolve to an internal address.
186
+ """
187
+ try:
188
+ return [ipaddress.ip_address(hostname)]
189
+ except ValueError:
190
+ pass
191
+
192
+ try:
193
+ infos = await asyncio.to_thread(socket.getaddrinfo, hostname, None)
194
+ except socket.gaierror as e:
195
+ raise TokenVerificationError(
196
+ f"Could not resolve issuer host '{hostname}': {e}"
197
+ ) from e
198
+ return [ipaddress.ip_address(info[4][0]) for info in infos]
File without changes
@@ -0,0 +1,205 @@
1
+ """JWT signature verification for IAL2 tokens using JWKS.
2
+
3
+ Fetches the CSP's public keys from their OIDC discovery endpoint
4
+ and verifies the token signature, expiration, and audience.
5
+ """
6
+
7
+ from __future__ import annotations
8
+
9
+ import asyncio
10
+ import http.client
11
+ import json
12
+ import logging
13
+ import socket
14
+ import ssl
15
+ from typing import Any
16
+ from urllib.parse import urlparse
17
+ from urllib.request import urlopen
18
+
19
+ import jwt
20
+ from jwt import PyJWKClient
21
+
22
+ logger = logging.getLogger(__name__)
23
+
24
+
25
+ class TokenVerificationError(Exception):
26
+ """Raised when an IAL2 token fails signature or claims verification."""
27
+
28
+
29
+ class _PinnedHTTPSConnection(http.client.HTTPSConnection):
30
+ """An HTTPSConnection that dials a pre-resolved IP instead of
31
+ re-resolving its host via DNS, while still using that original host for
32
+ TLS SNI and certificate verification.
33
+
34
+ Used to close the DNS-rebinding TOCTOU window between validating a
35
+ hostname's resolved address and actually connecting to it: a plain
36
+ HTTPSConnection re-resolves the hostname at connect() time, so an
37
+ attacker controlling that hostname's DNS could serve a different
38
+ (internal) address than the one that was validated moments earlier.
39
+ """
40
+
41
+ # Declared here for mypy: real attributes set by HTTPSConnection.__init__
42
+ # (typeshed doesn't expose them, since they're conventionally private).
43
+ _context: ssl.SSLContext
44
+ source_address: tuple[str, int] | None
45
+
46
+ def __init__(self, host: str, pinned_ip: str, **kwargs: Any) -> None:
47
+ super().__init__(host, **kwargs)
48
+ self._pinned_ip = pinned_ip
49
+
50
+ def connect(self) -> None:
51
+ sock = socket.create_connection(
52
+ (self._pinned_ip, self.port), self.timeout, self.source_address
53
+ )
54
+ self.sock = self._context.wrap_socket(sock, server_hostname=self.host)
55
+
56
+
57
+ def _fetch_json_pinned(
58
+ url: str, pinned_ip: str, *, timeout: float = 10.0
59
+ ) -> dict[str, Any]:
60
+ """Fetch and parse JSON from `url`, connecting to `pinned_ip` rather
61
+ than letting the request re-resolve the URL's hostname via DNS."""
62
+ parsed = urlparse(url)
63
+ conn = _PinnedHTTPSConnection(
64
+ parsed.hostname or "", pinned_ip, port=parsed.port, timeout=timeout
65
+ )
66
+ try:
67
+ conn.request("GET", parsed.path or "/", headers={"Accept": "application/json"})
68
+ response = conn.getresponse()
69
+ body = response.read()
70
+ finally:
71
+ conn.close()
72
+ result: dict[str, Any] = json.loads(body)
73
+ return result
74
+
75
+
76
+ class TokenVerifier:
77
+ """Verifies IAL2 JWT tokens against a CSP's JWKS endpoint.
78
+
79
+ Args:
80
+ jwks_uri: URL of the JWKS endpoint (e.g.
81
+ ``https://idp.example.com/.well-known/jwks.json``).
82
+ audience: Expected ``aud`` claim value.
83
+ issuer: Expected ``iss`` claim value. If provided, the token's
84
+ issuer must match exactly.
85
+ algorithms: Allowed signing algorithms. Defaults to RS256.
86
+ """
87
+
88
+ def __init__(
89
+ self,
90
+ *,
91
+ jwks_uri: str,
92
+ audience: str,
93
+ issuer: str | None = None,
94
+ algorithms: list[str] | None = None,
95
+ ) -> None:
96
+ self._jwks_uri = jwks_uri
97
+ self._audience = audience
98
+ self._issuer = issuer
99
+ self._algorithms = algorithms or ["RS256"]
100
+ self._jwks_client = PyJWKClient(jwks_uri)
101
+
102
+ @property
103
+ def jwks_uri(self) -> str:
104
+ """The JWKS endpoint this verifier fetches signing keys from."""
105
+ return self._jwks_uri
106
+
107
+ async def verify(self, token: str) -> dict[str, Any]:
108
+ """Verify the token signature and standard claims.
109
+
110
+ Args:
111
+ token: The encoded JWT string.
112
+
113
+ Returns:
114
+ The decoded claims dictionary.
115
+
116
+ Raises:
117
+ TokenVerificationError: If the token is invalid, expired,
118
+ or fails audience/issuer checks.
119
+ """
120
+ return await asyncio.to_thread(self._verify_sync, token)
121
+
122
+ def _verify_sync(self, token: str) -> dict[str, Any]:
123
+ """Blocking body of verify() -- PyJWT/PyJWKClient have no async API,
124
+ so verify() offloads this to a thread instead of blocking the event
125
+ loop for the JWKS fetch (network I/O) this does internally."""
126
+ try:
127
+ signing_key = self._jwks_client.get_signing_key_from_jwt(token)
128
+
129
+ decoded: dict[str, Any] = jwt.decode(
130
+ token,
131
+ signing_key.key,
132
+ algorithms=self._algorithms,
133
+ audience=self._audience,
134
+ issuer=self._issuer,
135
+ options={
136
+ "require": ["exp", "iat", "iss", "sub", "aud", "jti"],
137
+ },
138
+ )
139
+ return decoded
140
+
141
+ except jwt.ExpiredSignatureError as e:
142
+ raise TokenVerificationError(f"Token has expired: {e}") from e
143
+ except jwt.InvalidAudienceError as e:
144
+ raise TokenVerificationError(f"Invalid audience: {e}") from e
145
+ except jwt.InvalidIssuerError as e:
146
+ raise TokenVerificationError(f"Invalid issuer: {e}") from e
147
+ except jwt.PyJWKClientError as e:
148
+ raise TokenVerificationError(
149
+ f"Failed to fetch signing key from JWKS: {e}"
150
+ ) from e
151
+ except jwt.InvalidTokenError as e:
152
+ raise TokenVerificationError(f"Token verification failed: {e}") from e
153
+
154
+ @classmethod
155
+ async def from_oidc_discovery(
156
+ cls,
157
+ *,
158
+ discovery_url: str,
159
+ audience: str,
160
+ issuer: str | None = None,
161
+ algorithms: list[str] | None = None,
162
+ pinned_ip: str | None = None,
163
+ ) -> TokenVerifier:
164
+ """Create a verifier by fetching the JWKS URI from OIDC discovery.
165
+
166
+ Args:
167
+ discovery_url: The ``/.well-known/openid-configuration`` URL.
168
+ audience: Expected ``aud`` claim value.
169
+ issuer: Expected ``iss`` claim value.
170
+ algorithms: Allowed signing algorithms.
171
+ pinned_ip: If given, connect to this IP instead of letting the
172
+ discovery fetch resolve discovery_url's host via DNS. Used
173
+ by callers (e.g. MultiIssuerTokenVerifier) that have already
174
+ validated an untrusted host's resolved address and need the
175
+ actual connection pinned to it, closing the window between
176
+ that validation and the request.
177
+
178
+ Returns:
179
+ A configured TokenVerifier instance.
180
+
181
+ Raises:
182
+ TokenVerificationError: If discovery metadata cannot be fetched.
183
+ """
184
+
185
+ def _fetch_metadata() -> dict[str, Any]:
186
+ if pinned_ip:
187
+ return _fetch_json_pinned(discovery_url, pinned_ip)
188
+ with urlopen(discovery_url) as response: # nosec B310
189
+ result: dict[str, Any] = json.loads(response.read())
190
+ return result
191
+
192
+ try:
193
+ metadata = await asyncio.to_thread(_fetch_metadata)
194
+ jwks_uri = metadata["jwks_uri"]
195
+ except Exception as e:
196
+ raise TokenVerificationError(
197
+ f"Failed to fetch OIDC discovery metadata from {discovery_url}: {e}"
198
+ ) from e
199
+
200
+ return cls(
201
+ jwks_uri=jwks_uri,
202
+ audience=audience,
203
+ issuer=issuer or metadata.get("issuer"),
204
+ algorithms=algorithms,
205
+ )