1#!/usr/bin/env python3
2"""Tests for extract_claims.py CLI."""
3 
4import json
5import os
6import shutil
7import subprocess
8import sys
9import tempfile
10import unittest
11 
12SCRIPT = os.path.join(os.path.dirname(__file__), '..', 'scripts', 'extract_claims.py')
13FIXTURES = os.path.join(os.path.dirname(__file__), 'fixtures')
14 
15 
16def run_ec(*args: str) -> dict | list:
17    """Run extract_claims.py with args."""
18    result = subprocess.run(
19        [sys.executable, SCRIPT, *args],
20        capture_output=True, text=True,
21    )
22    if result.returncode != 0:
23        raise RuntimeError(f'Exit {result.returncode}: {result.stderr}')
24    return json.loads(result.stdout)
25 
26 
27SAMPLE_REPORT = """\
28---
29title: Test Research Report
30---
31 
32## Executive Summary
33 
34This report examines the impact of quantum computing on cryptography [1, 2]. The field has advanced significantly since 2020, with major breakthroughs in error correction.
35 
36## Introduction
37 
38Quantum computing represents a paradigm shift in computational capability. Researchers at Google demonstrated quantum supremacy in 2019 using a 53-qubit processor [3]. This milestone confirmed theoretical predictions made decades earlier.
39 
40## Finding 1
41 
42The Shor algorithm can factor large numbers exponentially faster than classical methods [4]. Current RSA-2048 encryption could be broken by a sufficiently large quantum computer. However, such machines are estimated to require millions of physical qubits [5, 6].
43 
44## Finding 2
45 
46Post-quantum cryptography standards should be adopted within the next 5 years. Organizations should consider hybrid classical-quantum approaches during the transition period. NIST has already standardized several lattice-based algorithms [7].
47 
48## Synthesis
49 
50Taken together, the evidence suggests that quantum computing poses a real but manageable threat to current cryptographic systems. The timeline for practical quantum attacks remains uncertain, but proactive migration reduces risk substantially.
51 
52## Recommendations
53 
54Organizations should begin evaluating post-quantum cryptography solutions immediately. Security teams should conduct a cryptographic inventory to identify vulnerable systems. Companies should consider implementing crypto-agility frameworks to enable rapid algorithm switching.
55 
56## Bibliography
57 
58[1] Smith et al. (2023). Quantum Computing Advances.
59[2] Johnson (2024). Cryptographic Implications.
60"""
61 
62 
63class TestExtract(unittest.TestCase):
64    def setUp(self):
65        self.tmpdir = tempfile.mkdtemp()
66        # Create empty claims.jsonl
67        open(os.path.join(self.tmpdir, 'claims.jsonl'), 'w').close()
68        # Write sample report
69        self.report_path = os.path.join(self.tmpdir, 'report.md')
70        with open(self.report_path, 'w') as f:
71            f.write(SAMPLE_REPORT)
72 
73    def tearDown(self):
74        shutil.rmtree(self.tmpdir, ignore_errors=True)
75 
76    def test_extract_finds_claims(self):
77        out = run_ec('extract', '--report', self.report_path, '--dir', self.tmpdir)
78        self.assertEqual(out['status'], 'ok')
79        self.assertGreater(out['claims_added'], 5)
80 
81    def test_extract_idempotent(self):
82        out1 = run_ec('extract', '--report', self.report_path, '--dir', self.tmpdir)
83        out2 = run_ec('extract', '--report', self.report_path, '--dir', self.tmpdir)
84        self.assertEqual(out2['claims_added'], 0)
85        self.assertEqual(out2['claims_skipped'], out1['claims_added'])
86 
87    def test_claim_types_assigned(self):
88        run_ec('extract', '--report', self.report_path, '--dir', self.tmpdir)
89        out = run_ec('stats', '--dir', self.tmpdir)
90        # Should have at least factual and recommendation types
91        self.assertIn('factual', out['by_type'])
92        self.assertIn('recommendation', out['by_type'])
93 
94    def test_sections_detected(self):
95        run_ec('extract', '--report', self.report_path, '--dir', self.tmpdir)
96        out = run_ec('stats', '--dir', self.tmpdir)
97        self.assertIn('finding_1', out['by_section'])
98        self.assertIn('finding_2', out['by_section'])
99        self.assertIn('recommendations', out['by_section'])
100 
101 
102class TestAdd(unittest.TestCase):
103    def setUp(self):
104        self.tmpdir = tempfile.mkdtemp()
105        open(os.path.join(self.tmpdir, 'claims.jsonl'), 'w').close()
106 
107    def tearDown(self):
108        shutil.rmtree(self.tmpdir, ignore_errors=True)
109 
110    def test_add_and_dedup(self):
111        claim = json.dumps({
112            'section_id': 'finding_1',
113            'text': 'Quantum computers can break RSA encryption.',
114            'claim_type': 'factual',
115        })
116        out1 = run_ec('add', '--json', claim, '--dir', self.tmpdir)
117        self.assertEqual(out1['status'], 'added')
118        self.assertEqual(len(out1['claim_id']), 16)
119 
120        out2 = run_ec('add', '--json', claim, '--dir', self.tmpdir)
121        self.assertEqual(out2['status'], 'duplicate')
122 
123    def test_add_with_sources(self):
124        claim = json.dumps({
125            'section_id': 'finding_1',
126            'text': 'NIST standardized CRYSTALS-Kyber in 2024.',
127            'claim_type': 'factual',
128            'cited_source_ids': ['abcdef0123456789'],
129            'evidence_ids': ['1234567890abcdef'],
130        })
131        out = run_ec('add', '--json', claim, '--dir', self.tmpdir)
132        self.assertEqual(out['status'], 'added')
133 
134 
135class TestListAndStats(unittest.TestCase):
136    def setUp(self):
137        self.tmpdir = tempfile.mkdtemp()
138        open(os.path.join(self.tmpdir, 'claims.jsonl'), 'w').close()
139        # Add mixed claims
140        for sec, text, ctype in [
141            ('finding_1', 'The sky appears blue due to Rayleigh scattering.', 'factual'),
142            ('finding_1', 'Light wavelengths scatter differently in the atmosphere.', 'factual'),
143            ('synthesis', 'Overall, atmospheric optics explains most visual phenomena.', 'synthesis'),
144            ('recommendations', 'Researchers should investigate polarization effects further.', 'recommendation'),
145        ]:
146            run_ec('add', '--json', json.dumps({
147                'section_id': sec, 'text': text, 'claim_type': ctype,
148            }), '--dir', self.tmpdir)
149 
150    def tearDown(self):
151        shutil.rmtree(self.tmpdir, ignore_errors=True)
152 
153    def test_list_all(self):
154        out = run_ec('list', '--dir', self.tmpdir)
155        self.assertEqual(out['count'], 4)
156 
157    def test_list_by_section(self):
158        out = run_ec('list', '--dir', self.tmpdir, '--section', 'finding_1')
159        self.assertEqual(out['count'], 2)
160 
161    def test_list_by_type(self):
162        out = run_ec('list', '--dir', self.tmpdir, '--type', 'recommendation')
163        self.assertEqual(out['count'], 1)
164 
165    def test_stats(self):
166        out = run_ec('stats', '--dir', self.tmpdir)
167        self.assertEqual(out['total'], 4)
168        self.assertEqual(out['by_type']['factual'], 2)
169        self.assertEqual(out['by_type']['synthesis'], 1)
170        self.assertEqual(out['by_type']['recommendation'], 1)
171 
172 
173class TestClaimID(unittest.TestCase):
174    """Unit tests for compute_claim_id."""
175 
176    @classmethod
177    def setUpClass(cls):
178        sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..', 'scripts'))
179        from extract_claims import compute_claim_id, classify_claim
180        cls.compute_id = staticmethod(compute_claim_id)
181        cls.classify = staticmethod(classify_claim)
182 
183    def test_deterministic(self):
184        id1 = self.compute_id('finding_1', 'Test claim.')
185        id2 = self.compute_id('finding_1', 'Test claim.')
186        self.assertEqual(id1, id2)
187 
188    def test_section_matters(self):
189        id1 = self.compute_id('finding_1', 'Same text.')
190        id2 = self.compute_id('finding_2', 'Same text.')
191        self.assertNotEqual(id1, id2)
192 
193    def test_classify_recommendation(self):
194        self.assertEqual(
195            self.classify('Organizations should adopt PQC immediately.', 'recommendations'),
196            'recommendation',
197        )
198 
199    def test_classify_factual(self):
200        self.assertEqual(
201            self.classify('RSA-2048 uses 2048-bit keys.', 'finding_1'),
202            'factual',
203        )
204 
205    def test_classify_synthesis(self):
206        self.assertEqual(
207            self.classify('Taken together, the results indicate a clear trend.', 'synthesis'),
208            'synthesis',
209        )
210 
211 
212if __name__ == '__main__':
213    unittest.main()
214