Spaces:
Sleeping
Sleeping
File size: 13,031 Bytes
b401086 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 | """Tests Sprint 41 โ section NER dans le rapport HTML.
Couvre :
1. ``build_ner_summary_html`` rend le tableau rรฉsumรฉ (F1/P/R + totaux).
2. ``build_ner_per_category_html`` rend la heatmap moteur ร catรฉgorie.
3. **Masquage adaptatif** : les deux retournent ``""`` si aucun moteur
n'a de ``aggregated_ner``, ou si le sous-champ ``per_category`` est
absent (pour le second).
4. **Anti-injection** : un nom de moteur ou une catรฉgorie contenant des
balises HTML est รฉchappรฉ.
5. **Intรฉgration template** : le rapport HTML inclut la section quand
au moins un moteur a un ``aggregated_ner``, et l'omet sinon.
6. **i18n** : les clรฉs FR/EN existent et sont utilisรฉes.
"""
from __future__ import annotations
import json
from pathlib import Path
import pytest
from picarones.fixtures import generate_sample_benchmark
from picarones.report.generator import ReportGenerator
from picarones.report.ner_render import (
build_ner_per_category_html,
build_ner_summary_html,
)
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# Fixtures
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def _engine_with_ner(name: str = "tess") -> dict:
return {
"name": name,
"aggregated_ner": {
"global": {
"precision": 0.85, "recall": 0.78, "f1": 0.81, "support": 50,
},
"per_category": {
"PER": {"precision": 0.9, "recall": 0.8, "f1": 0.85, "support": 30},
"LOC": {"precision": 0.7, "recall": 0.6, "f1": 0.65, "support": 15},
"DATE": {"precision": 1.0, "recall": 0.9, "f1": 0.95, "support": 5},
},
"doc_count": 50,
"hallucinated_total": 8,
"missed_total": 11,
},
}
def _engine_without_ner(name: str = "no_ner") -> dict:
return {"name": name, "aggregated_ner": None}
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# 1. build_ner_summary_html
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class TestNerSummaryHtml:
def test_renders_table_with_engine_row(self) -> None:
html = build_ner_summary_html([_engine_with_ner("tess")])
assert "ner-summary" in html
assert "tess" in html
assert "81.0 %" in html # F1
assert "85.0 %" in html # Precision
assert "78.0 %" in html # Recall
assert "50" in html # doc_count
assert "8" in html # hallucinations
assert "11" in html # missed
def test_multiple_engines(self) -> None:
engines = [_engine_with_ner("a"), _engine_with_ner("b")]
html = build_ner_summary_html(engines)
assert "<tr>" in html
# Deux moteurs โ trois <tr> minimum (header + 2 lignes)
assert html.count("<tr>") >= 3
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# 2. build_ner_per_category_html
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class TestNerPerCategoryHtml:
def test_renders_heatmap_with_categories(self) -> None:
html = build_ner_per_category_html([_engine_with_ner("tess")])
assert "ner-per-category" in html
# Trois catรฉgories en en-tรชtes
for cat in ("PER", "LOC", "DATE"):
assert cat in html
# Le F1 est rendu en pourcentage
assert "85.0 %" in html # PER
assert "65.0 %" in html # LOC
assert "95.0 %" in html # DATE
def test_categories_union_across_engines(self) -> None:
e1 = _engine_with_ner("a")
e2 = {
"name": "b",
"aggregated_ner": {
"global": {"precision": 1.0, "recall": 1.0, "f1": 1.0, "support": 5},
"per_category": {
"ORG": {"precision": 1.0, "recall": 1.0, "f1": 1.0, "support": 5},
},
"doc_count": 5, "hallucinated_total": 0, "missed_total": 0,
},
}
html = build_ner_per_category_html([e1, e2])
for cat in ("PER", "LOC", "DATE", "ORG"):
assert cat in html
def test_no_data_marker_for_missing_category(self) -> None:
e1 = _engine_with_ner("a")
# b n'a que ORG, donc ses cellules PER/LOC/DATE doivent afficher "โ"
e2 = {
"name": "b",
"aggregated_ner": {
"global": {"precision": 1.0, "recall": 1.0, "f1": 1.0, "support": 5},
"per_category": {
"ORG": {"precision": 1.0, "recall": 1.0, "f1": 1.0, "support": 5},
},
"doc_count": 5, "hallucinated_total": 0, "missed_total": 0,
},
}
html = build_ner_per_category_html([e1, e2])
# "โ" apparaรฎt dans les cellules vides
assert "โ" in html
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# 3. Masquage adaptatif
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class TestAdaptiveMasking:
def test_summary_empty_when_no_engine_has_ner(self) -> None:
assert build_ner_summary_html([]) == ""
assert build_ner_summary_html([_engine_without_ner()]) == ""
def test_per_category_empty_when_no_engine_has_ner(self) -> None:
assert build_ner_per_category_html([]) == ""
assert build_ner_per_category_html([_engine_without_ner()]) == ""
def test_per_category_empty_when_no_categories(self) -> None:
engine = {
"name": "x",
"aggregated_ner": {
"global": {"precision": 0.0, "recall": 0.0, "f1": 0.0},
"per_category": {}, # vide
"doc_count": 0,
"hallucinated_total": 0, "missed_total": 0,
},
}
assert build_ner_per_category_html([engine]) == ""
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# 4. Anti-injection
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class TestAntiInjection:
def test_engine_name_with_html_chars_escaped(self) -> None:
engine = _engine_with_ner("<script>alert(1)</script>")
html = build_ner_summary_html([engine])
assert "<script>" not in html
assert "<script>" in html
def test_category_label_with_html_chars_escaped(self) -> None:
engine = {
"name": "x",
"aggregated_ner": {
"global": {"precision": 1.0, "recall": 1.0, "f1": 1.0, "support": 1},
"per_category": {
"<img src=x>": {
"precision": 1.0, "recall": 1.0, "f1": 1.0, "support": 1,
},
},
"doc_count": 1, "hallucinated_total": 0, "missed_total": 0,
},
}
html = build_ner_per_category_html([engine])
assert "<img" not in html or "<img" in html
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# 5. Intรฉgration ReportGenerator
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class TestReportIntegration:
def test_section_absent_when_no_ner(self, tmp_path: Path) -> None:
bench = generate_sample_benchmark()
# Aucun engine_reports[i].aggregated_ner par dรฉfaut
for r in bench.engine_reports:
assert r.aggregated_ner is None
out = tmp_path / "report.html"
ReportGenerator(bench).generate(out)
html = out.read_text(encoding="utf-8")
assert "ner-summary" not in html
assert "ner-per-category" not in html
def test_section_present_when_at_least_one_engine_has_ner(
self, tmp_path: Path,
) -> None:
bench = generate_sample_benchmark()
bench.engine_reports[0].aggregated_ner = {
"global": {"precision": 0.9, "recall": 0.85, "f1": 0.87, "support": 30},
"per_category": {
"PER": {"precision": 0.9, "recall": 0.85, "f1": 0.87, "support": 30},
},
"doc_count": 30, "hallucinated_total": 5, "missed_total": 4,
}
out = tmp_path / "report.html"
ReportGenerator(bench).generate(out)
html = out.read_text(encoding="utf-8")
assert "ner-summary" in html
assert "ner-per-category" in html
assert "87.0 %" in html
def test_french_locale_uses_french_labels(self, tmp_path: Path) -> None:
bench = generate_sample_benchmark()
bench.engine_reports[0].aggregated_ner = {
"global": {"precision": 0.9, "recall": 0.85, "f1": 0.87, "support": 30},
"per_category": {
"PER": {"precision": 0.9, "recall": 0.85, "f1": 0.87, "support": 30},
},
"doc_count": 30, "hallucinated_total": 5, "missed_total": 4,
}
out = tmp_path / "report_fr.html"
ReportGenerator(bench, lang="fr").generate(out)
html = out.read_text(encoding="utf-8")
# Labels franรงais
assert "Entitรฉs manquรฉes" in html or "Hallucinations" in html
assert "F1 global" in html
def test_english_locale_uses_english_labels(self, tmp_path: Path) -> None:
bench = generate_sample_benchmark()
bench.engine_reports[0].aggregated_ner = {
"global": {"precision": 0.9, "recall": 0.85, "f1": 0.87, "support": 30},
"per_category": {
"PER": {"precision": 0.9, "recall": 0.85, "f1": 0.87, "support": 30},
},
"doc_count": 30, "hallucinated_total": 5, "missed_total": 4,
}
out = tmp_path / "report_en.html"
ReportGenerator(bench, lang="en").generate(out)
html = out.read_text(encoding="utf-8")
assert "Missed entities" in html or "Hallucinations" in html
assert "Global F1" in html
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# 6. i18n FR/EN โ clรฉs prรฉsentes et non vides
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
REQUIRED_KEYS = (
"h_ner",
"ner_note",
"ner_summary_caption",
"ner_per_category_caption",
"ner_engine_label",
"ner_f1_label",
"ner_precision_label",
"ner_recall_label",
"ner_doc_count_label",
"ner_hallucinated_label",
"ner_missed_label",
"ner_no_data_label",
)
class TestI18NCompleteness:
@pytest.mark.parametrize("lang", ["fr", "en"])
@pytest.mark.parametrize("key", REQUIRED_KEYS)
def test_key_present(self, lang: str, key: str) -> None:
path = (
Path(__file__).parent.parent
/ "picarones" / "report" / "i18n" / f"{lang}.json"
)
data = json.loads(path.read_text(encoding="utf-8"))
assert key in data, f"Clรฉ {key!r} manquante dans {lang}.json"
assert data[key].strip(), f"Clรฉ {key!r} vide dans {lang}.json"
|