Spaces:
Running on Zero
Running on Zero
File size: 9,733 Bytes
f7b61d8 5ab38b4 f7b61d8 5ab38b4 f7b61d8 5ab38b4 f7b61d8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 | from __future__ import annotations
import csv
from pathlib import Path
import pytest
from scripts import run_small_classification
from gcmd_classifier.articles import validate_article_records
from gcmd_classifier.config import ModelSettings
from gcmd_classifier.llm import FakeModelClient
from gcmd_classifier.models import (
ArticleClassificationOutcome,
ArticleProcessingStatus,
ArticleRecord,
ArticleResult,
ReviewStatus,
)
from gcmd_classifier.vocabulary import load_vocabulary
FIXTURE_PATH = Path("tests/fixtures/gcmd_hierarchy_small.json")
def test_small_run_script_imports_without_live_api_calls() -> None:
assert callable(run_small_classification.main)
assert callable(run_small_classification.fake_scripted_responses)
def test_limit_article_load_uses_first_valid_articles() -> None:
load_result = validate_article_records(
[
{"DOI": "10.example/1", "Title": "One", "Year": 2025, "Abstract": ""},
{"DOI": "10.example/2", "Title": "Two", "Year": 2025, "Abstract": ""},
{"DOI": "10.example/3", "Title": "Three", "Year": 2025, "Abstract": ""},
]
)
limited = run_small_classification.limit_article_load(load_result, 2)
assert [article.DOI for article in limited.articles] == ["10.example/1", "10.example/2"]
assert limited.source_count == 3
def test_fake_scripted_responses_support_requested_article_count() -> None:
vocabulary = load_vocabulary(FIXTURE_PATH)
actions = run_small_classification.fake_scripted_responses(vocabulary, article_count=3)
assert actions[0]["selected"][0]["candidate_id"].startswith("topic_")
assert actions[1]["selected"][0]["candidate_id"].startswith("term_")
no_topic_count = sum(1 for action in actions if action.get("no_selection_reason"))
assert no_topic_count == 2
def test_build_model_client_fake_returns_fake_model() -> None:
client = run_small_classification.build_model_client(
settings=ModelSettings(provider="fake"),
fake=True,
vocabulary=load_vocabulary(FIXTURE_PATH),
article_count=1,
)
assert isinstance(client, FakeModelClient)
def test_live_mode_requires_openai_provider() -> None:
with pytest.raises(SystemExit, match="requires --provider openai"):
run_small_classification.build_model_client(
settings=ModelSettings(provider="fake"),
fake=False,
vocabulary=load_vocabulary(FIXTURE_PATH),
article_count=1,
)
def test_live_mode_requires_credentials(monkeypatch) -> None:
monkeypatch.delenv("OPENAI_API_KEY", raising=False)
with pytest.raises(SystemExit, match="requires OPENAI_API_KEY"):
run_small_classification.build_model_client(
settings=ModelSettings(provider="openai", api_key_env_var="OPENAI_API_KEY"),
fake=False,
vocabulary=load_vocabulary(FIXTURE_PATH),
article_count=1,
)
def test_console_summary_includes_classification_and_no_classification(capsys) -> None:
classified = ArticleResult(
DOI="10.example/classified",
Title="Classified title",
Year=2025,
Abstract="Text.",
processing_status=ArticleProcessingStatus.COMPLETED,
classification_outcome=ArticleClassificationOutcome.CLASSIFIED,
classifications=(
{
"UUID": "topic-atmosphere",
"name": "ATMOSPHERE",
"level": "Topic",
"canonical_path": "ATMOSPHERE",
"path_components": ("ATMOSPHERE",),
"topic": "ATMOSPHERE",
"deterministic_validation": {"valid": True},
"final_status": "accepted",
"review_required": True,
"warnings": (
{
"code": "REVIEW_RECOMMENDED_WEAK_SUPPORT",
"message": "Manual scientific review is recommended.",
},
),
},
),
review_status=ReviewStatus.NOT_REQUIRED,
)
not_classified = ArticleResult(
DOI="10.example/no",
Title="No title",
Year=2025,
Abstract="",
processing_status=ArticleProcessingStatus.COMPLETED,
classification_outcome=ArticleClassificationOutcome.NOT_CLASSIFIED,
classifications=(),
no_classification_reason="No Topic selected.",
review_status=ReviewStatus.NOT_REQUIRED,
)
run_small_classification.print_console_summary((classified, not_classified))
output = capsys.readouterr().out
assert "topic-atmosphere" in output
assert "ATMOSPHERE" in output
assert "No Topic selected." in output
def test_review_csv_path_follows_output_dir(tmp_path: Path) -> None:
output_dir = tmp_path / "small_run_gpt56"
result = ArticleResult(
DOI="10.example/no",
Title="No title",
Year=2025,
Abstract="",
processing_status=ArticleProcessingStatus.COMPLETED,
classification_outcome=ArticleClassificationOutcome.NOT_CLASSIFIED,
classifications=(),
no_classification_reason="No Topic selected.",
review_status=ReviewStatus.NOT_REQUIRED,
)
csv_path = run_small_classification.write_review_csv((result,), output_dir)
assert csv_path == output_dir / "review_table.csv"
assert csv_path.exists()
def test_review_csv_contains_classification_and_article_level_rows(tmp_path: Path) -> None:
classified = ArticleResult(
DOI="10.example/classified",
Title="Classified title",
Year=2025,
Abstract="Text.",
processing_status=ArticleProcessingStatus.COMPLETED,
classification_outcome=ArticleClassificationOutcome.CLASSIFIED,
classifications=(
{
"UUID": "topic-atmosphere",
"name": "ATMOSPHERE",
"level": "Topic",
"canonical_path": "ATMOSPHERE",
"path_components": ("ATMOSPHERE",),
"topic": "ATMOSPHERE",
"classifier_evidence": "Evidence text.",
"support_type": "explicit",
"reason_for_stopping": "Stopped here.",
"confidence": {"final": 0.7},
"deterministic_validation": {"valid": True},
"final_status": "accepted",
"review_required": True,
"warnings": (
{
"code": "REVIEW_RECOMMENDED_WEAK_SUPPORT",
"message": "Manual scientific review is recommended.",
},
),
},
),
review_status=ReviewStatus.NOT_REQUIRED,
)
failed = ArticleResult(
DOI="10.example/failed",
Title="Failed title",
Year=2025,
Abstract="Text.",
processing_status=ArticleProcessingStatus.FAILED,
classification_outcome=None,
classifications=(),
errors=({"code": "MODEL_ERROR", "message": "Model failed."},),
review_status=ReviewStatus.NOT_REQUIRED,
)
csv_path = run_small_classification.write_review_csv((classified, failed), tmp_path)
rows = list(csv.DictReader(csv_path.open()))
assert len(rows) == 2
assert rows[0]["DOI"] == "10.example/classified"
assert rows[0]["level"] == "Topic"
assert rows[0]["UUID"] == "topic-atmosphere"
assert rows[0]["canonical_path"] == "ATMOSPHERE"
assert rows[0]["support_type"] == "explicit"
assert rows[0]["confidence_final"] == "0.7"
assert rows[0]["classifier_evidence"] == "Evidence text."
assert rows[0]["reason_for_stopping"] == "Stopped here."
assert rows[0]["deterministic_valid"] == "True"
assert rows[0]["review_required"] == "True"
assert "REVIEW_RECOMMENDED_WEAK_SUPPORT" in rows[0]["warnings"]
assert rows[1]["DOI"] == "10.example/failed"
assert rows[1]["UUID"] == ""
assert rows[1]["errors"] == "MODEL_ERROR: Model failed."
def test_progress_helpers_print_safe_truncated_output(capsys) -> None:
article = ArticleRecord(
DOI="10.example/progress",
Title="A very long title " * 12,
Year=2025,
Abstract="Do not print this abstract.",
)
result = ArticleResult(
DOI=article.DOI,
Title=article.Title,
Year=article.Year,
Abstract=article.Abstract,
processing_status=ArticleProcessingStatus.COMPLETED,
classification_outcome=ArticleClassificationOutcome.NOT_CLASSIFIED,
classifications=(),
no_classification_reason="No Topic selected.",
review_status=ReviewStatus.NOT_REQUIRED,
)
run_small_classification.print_article_start(0, 3, article)
run_small_classification.print_article_finish(0, 3, result, 1.234)
output = capsys.readouterr().out
assert "Starting article 1/3" in output
assert "DOI=10.example/progress" in output
assert "..." in output
assert "Finished article 1/3" in output
assert "processing_status=completed" in output
assert "classification_outcome=not_classified" in output
assert "accepted_classifications=0" in output
assert "elapsed_seconds=1.23" in output
assert "Do not print this abstract" not in output
def test_truncate_title_uses_one_line_and_limit() -> None:
title = "Title with\nnewlines and " + "x" * 100
truncated = run_small_classification.truncate_title(title, max_length=30)
assert "\n" not in truncated
assert len(truncated) == 30
assert truncated.endswith("...")
def test_one_line_compacts_tabs_and_newlines() -> None:
assert run_small_classification.one_line("A\tlong\n title") == "A long title"
|