HydroModPy/evaluation/generate_report.py at dev · HydroModPy/HydroModPy · GitHub

History

585 lines (512 loc) · 22.2 KB

Raw

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

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

from __future__ import annotations

import argparse

import json

import sys

from collections import defaultdict

from pathlib import Path

from typing import Any

REPO_ROOT = Path(__file__).resolve().parents[1]

if str(REPO_ROOT) not in sys.path:

sys.path.insert(0, str(REPO_ROOT))

try:

import pandas as pd

except ImportError as exc: # pragma: no cover - optional dependency

pd = None

PANDAS_ERROR = exc

else:

PANDAS_ERROR = None

try:

import matplotlib.pyplot as plt

except ImportError as exc: # pragma: no cover - optional dependency

plt = None

MPL_ERROR = exc

else:

MPL_ERROR = None

from evaluation._utils import resolve_repository_root

from evaluation.cohesion import compute_cohesion

from evaluation.coupling import compute_cbo

from evaluation.extract_architecture import collect_architecture

from evaluation.radon_metrics import compute_radon_metrics

from evaluation.repository_summary import summarize_repository

TOP_N_PACKAGES = 15

TOP_N_MODULES = 20

TOP_N_BOXPLOT_PACKAGES = 10

PACKAGE_GROUP_DEPTH = 1

def load_json(path: Path) -> dict[str, Any]:

if not path.exists():

return {}

return json.loads(path.read_text(encoding="utf-8"))

def package_summary_frame(summary: dict[str, Any]) -> pd.DataFrame:

packages = summary.get("packages") or []

if packages:

return pd.DataFrame(packages)

files = pd.DataFrame(summary.get("files", []))

if files.empty:

return pd.DataFrame()

if "package" not in files.columns:

files["package"] = files["module"].apply(

lambda name: name.rsplit(".", 1)[0] if "." in name else name

)

grouped = files.groupby("package", dropna=False).agg(

file_count=("path", "count"),

module_count=("module", "nunique"),

public_module_count=("public", "sum") if "public" in files.columns else ("module", "count"),

class_count=("classes", "sum") if "classes" in files.columns else ("module", "count"),

method_count=("methods", "sum") if "methods" in files.columns else ("module", "count"),

function_count=("functions", "sum")

if "functions" in files.columns

else ("module", "count"),

line_count=("lines", "sum") if "lines" in files.columns else ("module", "count"),

)

return grouped.reset_index()

def top_modules_frame(summary: dict[str, Any]) -> pd.DataFrame:

top_modules = summary.get("top_modules") or []

if top_modules:

return pd.DataFrame(top_modules)

files = pd.DataFrame(summary.get("files", []))

if files.empty:

return pd.DataFrame()

if "public" in files.columns:

files = files[files["public"]]

if files.empty:

files = pd.DataFrame(summary.get("files", []))

return files.sort_values(["lines", "module"], ascending=[False, True]).head(TOP_N_MODULES)

def flatten_coupling(data: dict[str, Any]) -> list[dict[str, Any]]:

rows: list[dict[str, Any]] = []

for file_info in data.get("files", []):

module = file_info.get("module")

package = file_info.get("package")

public = file_info.get("public")

for cls in file_info.get("classes", []):

rows.append(

{

"module": module,

"package": package,

"public": public,

"path": file_info.get("path"),

**cls,

}

)

return rows

def flatten_cohesion(data: dict[str, Any]) -> list[dict[str, Any]]:

rows: list[dict[str, Any]] = []

for file_info in data.get("files", []):

module = file_info.get("module")

package = file_info.get("package")

public = file_info.get("public")

for cls in file_info.get("classes", []):

rows.append(

{

"module": module,

"package": package,

"public": public,

"path": file_info.get("path"),

**cls,

}

)

return rows

def _module_to_package(module: str, depth: int | None = None) -> str:

package = module.rsplit(".", 1)[0] if "." in module else module

if depth is not None:

package = ".".join(package.split(".")[:depth]) or package

return package

def build_package_dependency_frame(

architecture: pd.DataFrame | dict[str, Any], depth: int | None = None

) -> pd.DataFrame:

if isinstance(architecture, dict):

architecture = pd.DataFrame(architecture.get("modules", []))

if architecture.empty or "module" not in architecture.columns:

return pd.DataFrame()

edge_weights: dict[tuple[str, str], int] = defaultdict(int)

for _, module_info in architecture.iterrows():

source_module = module_info.get("module")

if not source_module:

continue

source_package = _module_to_package(source_module, depth)

for dependency in module_info.get("internal_dependencies") or []:

target_package = _module_to_package(dependency, depth)

if source_package and target_package and source_package != target_package:

edge_weights[(source_package, target_package)] += 1

rows = [

{"source": source, "target": target, "weight": weight}

for (source, target), weight in sorted(edge_weights.items())

]

return pd.DataFrame(rows)

def public_only(frame: pd.DataFrame) -> pd.DataFrame:

if frame.empty:

return frame

if "public" in frame.columns:

public_frame = frame[frame["public"].fillna(False)]

if not public_frame.empty:

return public_frame

return frame

def package_group_column(frame: pd.DataFrame, depth: int, source_col: str = "package") -> pd.Series:

if frame.empty or source_col not in frame.columns:

return pd.Series(dtype=object)

return (

frame[source_col]

.fillna("")

.astype(str)

.apply(lambda value: ".".join(value.split(".")[:depth]) or value)

)

def select_package_order(counts: pd.Series, top_n: int, pinned: list[str]) -> list[str]:

ordered = counts.sort_values(ascending=False)

selected = list(ordered.head(top_n).index)

for package in pinned:

if package in ordered.index and package not in selected:

selected.append(package)

return selected

def dependency_cycles_text(graph: Any, n: int = 5) -> str:

# Strongly connected components with more than one node are exactly the

# packages caught in an import cycle. This is O(V+E) via Tarjan's

# algorithm, unlike enumerating every simple cycle which can blow up

# combinatorially on a real dependency graph.

import networkx as nx

cycle_groups = [

sorted(group) for group in nx.strongly_connected_components(graph) if len(group) > 1

]

if not cycle_groups:

return ""

lines = [" <-> ".join(group) for group in cycle_groups[:n]]

return "Cycles de dependances:\n" + "\n".join(lines)

def top_outliers_text(

frame: pd.DataFrame, value_col: str, label_cols: list[str], n: int = 5

) -> str:

if frame.empty or value_col not in frame.columns:

return ""

subset = frame.dropna(subset=[value_col])

if subset.empty:

return ""

subset = subset.nlargest(n, value_col)

lines: list[str] = []

for _, row in subset.iterrows():

label = "::".join(str(row[col]) for col in label_cols if col in row and pd.notna(row[col]))

lines.append(f"{label} = {row[value_col]:.1f}")

return "Valeurs extremes:\n" + "\n".join(lines)

def explode_complexity_frame(radon: pd.DataFrame) -> pd.DataFrame:

if radon.empty or "complexity" not in radon.columns:

return pd.DataFrame()

rows: list[dict[str, Any]] = []

for _, row in radon.iterrows():

for block in row.get("complexity", []) or []:

complexity = block.get("complexity")

if isinstance(complexity, (int, float)):

rows.append(

{

"module": row.get("module"),

"package": row.get("package"),

"public": row.get("public", False),

"path": row.get("path"),

"name": block.get("name"),

"complexity": float(complexity),

}

)

return pd.DataFrame(rows)

def write_excel(output_path: Path, frames: tuple[Any, ...]) -> None:

if pd is None:

raise SystemExit(f"pandas is required: {PANDAS_ERROR}")

output_path.parent.mkdir(parents=True, exist_ok=True)

with pd.ExcelWriter(output_path, engine="openpyxl") as writer:

sheet_names = ["summary", "radon", "coupling", "cohesion", "architecture"]

for sheet_name, frame in zip(sheet_names, frames, strict=False):

frame.to_excel(writer, sheet_name=sheet_name, index=False)

def write_charts(

output_dir: Path,

frames: tuple[Any, ...],

package_depth: int = PACKAGE_GROUP_DEPTH,

pinned_packages: list[str] | None = None,

) -> None:

if plt is None or pd is None:

return

pinned_packages = pinned_packages or []

output_dir.mkdir(parents=True, exist_ok=True)

summary, radon, coupling, cohesion, architecture = frames

if not summary.empty and "package" in summary.columns:

summary_grouped = summary.copy()

summary_grouped["package_group"] = package_group_column(summary_grouped, package_depth)

columns = [

col

for col in ["line_count", "class_count", "function_count"]

if col in summary_grouped.columns

]

if columns:

grouped = summary_grouped.groupby("package_group", as_index=True)[columns].sum()

rank_col = "line_count" if "line_count" in columns else columns[0]

order = select_package_order(grouped[rank_col], TOP_N_PACKAGES, pinned_packages)

top_packages = grouped.reindex(order)

ax = top_packages[columns].plot(kind="bar", figsize=(14, 7))

ax.set_title("Top packages by size")

ax.set_xlabel("Package")

ax.set_ylabel("Count")

plt.tight_layout()

plt.savefig(output_dir / "summary.png", dpi=200)

plt.close()

top_modules = summary.attrs.get("top_modules", pd.DataFrame())

if (

not top_modules.empty

and "lines" in top_modules.columns

and "module" in top_modules.columns

ax = top_modules.set_index("module")["lines"].plot(kind="bar", figsize=(14, 7))

ax.set_title("Top public modules by lines")

ax.set_xlabel("Module")

ax.set_ylabel("LOC")

plt.tight_layout()

plt.savefig(output_dir / "top_modules.png", dpi=200)

plt.close()

if not radon.empty:

complexity = explode_complexity_frame(radon)

if not complexity.empty:

ax = complexity["complexity"].plot(

kind="hist", bins=20, figsize=(10, 6), title="Cyclomatic complexity distribution"

)

ax.set_xlabel("Complexity")

plt.tight_layout()

plt.savefig(output_dir / "complexity_distribution.png", dpi=200)

plt.close()

complexity["package_group"] = package_group_column(complexity, package_depth)

package_counts = complexity.groupby("package_group").size()

package_order = select_package_order(

package_counts, TOP_N_BOXPLOT_PACKAGES, pinned_packages

)

if package_order:

data = [

complexity.loc[complexity["package_group"] == package, "complexity"]

.dropna()

.tolist()

for package in package_order

]

fig, ax = plt.subplots(figsize=(14, 7))

ax.boxplot(data, tick_labels=package_order, showmeans=True)

ax.set_title("Cyclomatic complexity by package")

ax.set_ylabel("Complexity")

ax.tick_params(axis="x", rotation=30)

complexity_subset = complexity[complexity["package_group"].isin(package_order)]

outliers_text = top_outliers_text(

complexity_subset, "complexity", ["module", "name"]

)

if outliers_text:

ax.text(

1.02,

0.98,

outliers_text,

transform=ax.transAxes,

fontsize=7,

va="top",

ha="left",

bbox=dict(boxstyle="round", facecolor="white", edgecolor="gray", alpha=0.9),

)

plt.tight_layout()

plt.savefig(output_dir / "complexity_boxplot.png", dpi=200, bbox_inches="tight")

plt.close(fig)

if "maintainability_index" in radon.columns:

series = radon["maintainability_index"].dropna()

if not series.empty:

ax = series.plot(

kind="hist",

bins=20,

figsize=(10, 6),

title="Maintainability Index distribution",

)

ax.set_xlabel("Maintainability Index")

plt.tight_layout()

plt.savefig(output_dir / "maintainability.png", dpi=200)

plt.close()

if not coupling.empty and "cbo" in coupling.columns:

coupling = public_only(coupling).copy()

cbo_values = coupling["cbo"].dropna()

if not cbo_values.empty:

ax = cbo_values.plot(kind="hist", bins=20, figsize=(10, 6), title="CBO distribution")

ax.set_xlabel("CBO")

plt.tight_layout()

plt.savefig(output_dir / "coupling_distribution.png", dpi=200)

plt.close()

coupling["package_group"] = package_group_column(coupling, package_depth)

packages = coupling.groupby("package_group").size()

order = select_package_order(packages, TOP_N_BOXPLOT_PACKAGES, pinned_packages)

if order:

data = [

coupling.loc[coupling["package_group"] == package, "cbo"].dropna().tolist()

for package in order

]

fig, ax = plt.subplots(figsize=(14, 7))

ax.boxplot(data, tick_labels=order, showmeans=True)

ax.set_title("CBO by package")

ax.set_ylabel("CBO")

ax.tick_params(axis="x", rotation=30)

coupling_subset = coupling[coupling["package_group"].isin(order)]

outliers_text = top_outliers_text(coupling_subset, "cbo", ["module", "class"])

if outliers_text:

ax.text(

1.02,

0.98,

outliers_text,

transform=ax.transAxes,

fontsize=7,

va="top",

ha="left",

bbox=dict(boxstyle="round", facecolor="white", edgecolor="gray", alpha=0.9),

)

plt.tight_layout()

plt.savefig(output_dir / "coupling_boxplot.png", dpi=200, bbox_inches="tight")

plt.close(fig)

if not cohesion.empty and "lcom" in cohesion.columns:

cohesion = public_only(cohesion).copy()

lcom_values = cohesion["lcom"].dropna()

if not lcom_values.empty:

ax = lcom_values.plot(kind="hist", bins=20, figsize=(10, 6), title="LCOM distribution")

ax.set_xlabel("LCOM")

plt.tight_layout()

plt.savefig(output_dir / "cohesion_distribution.png", dpi=200)

plt.close()

cohesion["package_group"] = package_group_column(cohesion, package_depth)

packages = cohesion.groupby("package_group").size()

order = select_package_order(packages, TOP_N_BOXPLOT_PACKAGES, pinned_packages)

if order:

data = [

cohesion.loc[cohesion["package_group"] == package, "lcom"].dropna().tolist()

for package in order

]

fig, ax = plt.subplots(figsize=(14, 7))

ax.boxplot(data, tick_labels=order, showmeans=True)

ax.set_title("LCOM by package")

ax.set_ylabel("LCOM")

ax.tick_params(axis="x", rotation=30)

cohesion_subset = cohesion[cohesion["package_group"].isin(order)]

outliers_text = top_outliers_text(cohesion_subset, "lcom", ["module", "class"])

if outliers_text:

ax.text(

1.02,

0.98,

outliers_text,

transform=ax.transAxes,

fontsize=7,

va="top",

ha="left",

bbox=dict(boxstyle="round", facecolor="white", edgecolor="gray", alpha=0.9),

)

plt.tight_layout()

plt.savefig(output_dir / "cohesion_boxplot.png", dpi=200, bbox_inches="tight")

plt.close(fig)

architecture_graph = build_package_dependency_frame(architecture, package_depth)

if not architecture_graph.empty:

packages = sorted(

set(architecture_graph["source"]).union(set(architecture_graph["target"]))

)

index = {package: i for i, package in enumerate(packages)}

matrix = [[0 for _ in packages] for _ in packages]

for _, edge in architecture_graph.iterrows():

matrix[index[edge["source"]]][index[edge["target"]]] += int(edge["weight"])

fig, ax = plt.subplots(figsize=(12, 10))

image = ax.imshow(matrix, cmap="Blues", aspect="auto")

ax.set_title("Package dependency heatmap")

ax.set_xticks(range(len(packages)))

ax.set_yticks(range(len(packages)))

ax.set_xticklabels(packages, rotation=90, fontsize=7)

ax.set_yticklabels(packages, fontsize=7)

fig.colorbar(image, ax=ax, fraction=0.046, pad=0.04)

plt.tight_layout()

plt.savefig(output_dir / "architecture_package_heatmap.png", dpi=200)

plt.close(fig)

try:

import networkx as nx

except ImportError:

return

graph = nx.DiGraph()

for _, edge in architecture_graph.iterrows():

graph.add_edge(edge["source"], edge["target"], weight=edge["weight"])

if graph.number_of_nodes() > 0:

fig, ax = plt.subplots(figsize=(14, 10))

positions = nx.spring_layout(graph, seed=42)

nx.draw_networkx(

graph,

positions,

ax=ax,

node_size=400,

font_size=7,

arrows=True,

edge_color="#888888",

)

ax.set_title("Package architecture graph")

ax.axis("off")

cycles_text = dependency_cycles_text(graph)

if cycles_text:

ax.text(

1.02,

0.98,

cycles_text,

transform=ax.transAxes,

fontsize=7,

va="top",

ha="left",

bbox=dict(

boxstyle="round", facecolor="#fff3cd", edgecolor="#856404", alpha=0.9

)

plt.tight_layout()

plt.savefig(output_dir / "architecture_package_graph.png", dpi=200, bbox_inches="tight")

plt.close(fig)

def _build_frames(

summary: dict[str, Any],

radon: dict[str, Any],

coupling: dict[str, Any],

cohesion: dict[str, Any],

architecture: dict[str, Any],

) -> tuple[Any, Any, Any, Any, Any]:

summary_frame = package_summary_frame(summary)

summary_frame.attrs["top_modules"] = top_modules_frame(summary)

return (

summary_frame,

pd.DataFrame(radon.get("files", [])),

pd.DataFrame(flatten_coupling(coupling)),

pd.DataFrame(flatten_cohesion(cohesion)),

pd.DataFrame(architecture.get("modules", [])),

)

def build_frames_from_repo(root: Path):

summary = summarize_repository(root)

radon = compute_radon_metrics(root)

coupling = compute_cbo(root)

cohesion = compute_cohesion(root)

architecture = collect_architecture(root)

architecture = architecture.get("data", architecture)

return _build_frames(summary, radon, coupling, cohesion, architecture)

def build_frames(input_dir: Path):

if pd is None:

raise SystemExit(f"pandas is required: {PANDAS_ERROR}")

summary = load_json(input_dir / "summary.json")

radon = load_json(input_dir / "radon.json")

coupling = load_json(input_dir / "coupling.json")

cohesion = load_json(input_dir / "cohesion.json")

architecture = load_json(input_dir / "architecture.json")

return _build_frames(summary, radon, coupling, cohesion, architecture)

def main() -> None:

parser = argparse.ArgumentParser(description="Generate quality report")

parser.add_argument("--repo", type=str)

parser.add_argument("--output-dir", type=Path, required=True)

parser.add_argument("--excel", type=Path, default=None)

parser.add_argument(

"--package-depth",

type=int,

default=PACKAGE_GROUP_DEPTH,

help="Number of dotted segments used to group packages in charts (1 = top-level package)",

)

parser.add_argument(

"--pin-package",

action="append",

default=None,

help="Package (at --package-depth granularity) to always include in the top-N charts, can be repeated",

)

args = parser.parse_args()

if pd is None:

raise SystemExit(f"pandas is required: {PANDAS_ERROR}")

with resolve_repository_root(None, args.repo, None) as root:

frames = build_frames_from_repo(root)

args.output_dir.mkdir(parents=True, exist_ok=True)

names = ["summary", "radon", "coupling", "cohesion", "architecture"]

for name, frame in zip(names, frames, strict=False):

frame.to_csv(args.output_dir / f"{name}.csv", index=False)

excel_path = args.excel if args.excel is not None else args.output_dir / "evaluation.xlsx"

write_excel(excel_path, frames)

write_charts(args.output_dir / "charts", frames, args.package_depth, args.pin_package)

if __name__ == "__main__":

main()

HydroModPy/evaluation/generate_report.py at dev · HydroModPy/HydroModPy · GitHub

Uh oh!

Expand file tree

Breadcrumbs

generate_report.py

Latest commit

History

Breadcrumbs

generate_report.py

File metadata and controls