Journal_Utilities/scripts/enrich_metadata.py at main · ActiveInferenceInstitute/Journal_Utilities · GitHub

History

686 lines (603 loc) · 28 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

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

#!/usr/bin/env python3

"""

Enrich ActiveInferenceJournal metadata.json files (v2.1 schema) from:

1. the Coda (Superhuman Docs) livestream table — full API fetch with

fallback to the checked-in snapshot,

2. data/input/*_split.txt session split-files (multi-talk videos),

3. data/output/channel_videos.json (fills empty parts[] on split items).

Dry-run by default; pass --apply to write. Never creates journal items —

unmatched Coda rows are reported only. Idempotent: re-running --apply on an

enriched tree changes nothing.

Usage:

python scripts/enrich_metadata.py # dry run, API fetch

python scripts/enrich_metadata.py --snapshot-only

python scripts/enrich_metadata.py --apply --report report.json

"""

import argparse

import json

import logging

import sys

from collections import defaultdict

from pathlib import Path

from typing import Any

REPO = Path(__file__).resolve().parent.parent

sys.path.insert(0, str(REPO / "src"))

from dotenv import load_dotenv # noqa: E402

from journal_utilities.data.coda_client import load_rows # noqa: E402

from journal_utilities.data.enrichment import ( # noqa: E402

PER_PART_KEYS,

map_coda_row,

merge_enrichment,

parse_split_file,

prefer_url,

)

from journal_utilities.youtube.categorizer import categorize_name # noqa: E402

from journal_utilities.youtube.chapter_generator import validate_chapters # noqa: E402

from journal_utilities.youtube.youtube import extract_youtube_id # noqa: E402

logger = logging.getLogger("enrich_metadata")

SRC_PREFIX = "data/video/activeinferenceinstitute" # matches refactor_journal.py

GITHUB_BASE = "https://github.com/ActiveInferenceInstitute/ActiveInferenceJournal/tree/main"

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

"""session_name -> record for whole-video sessions (per-talk _sessNN skipped)."""

if not path.exists():

return {}

sessions = json.loads(path.read_text(encoding="utf-8")).get("sessions", [])

return {s["session_name"]: s for s in sessions if "_sess" not in s["session_name"]}

def github_link(rel: str) -> str:

from urllib.parse import quote

return f"{GITHUB_BASE}/{quote(SRC_PREFIX + '/' + rel, safe='/')}"

def load_journal_index(journal: Path) -> dict[str, dict]:

"""item_rel_path -> {path, meta}; item_rel_path is relative to SRC_PREFIX."""

root = journal / SRC_PREFIX

items = {}

for meta_path in sorted(root.rglob("metadata.json")):

rel = str(meta_path.parent.relative_to(root))

items[rel] = {"path": meta_path, "meta": json.loads(meta_path.read_text(encoding="utf-8"))}

return items

def video_index(items: dict[str, dict]) -> dict[str, str]:

"""video_id -> item_rel_path (first claim wins; collisions logged)."""

index: dict[str, str] = {}

for rel, entry in items.items():

for part in entry["meta"].get("parts", []):

vid = part.get("video_id", "")

if vid and vid not in index:

index[vid] = rel

elif vid:

logger.debug("video id %s in both %s and %s", vid, index[vid], rel)

return index

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

videos = json.loads(path.read_text(encoding="utf-8")).get("videos", [])

return {v["id"]: v for v in videos if v.get("id")}

def manifest_part(video: dict) -> dict:

part = {

"video_id": video["id"],

"url": video.get("url", f"https://www.youtube.com/watch?v={video['id']}"),

"title": video.get("title", ""),

}

if video.get("duration"):

part["duration"] = video["duration"]

if video.get("upload_date"):

part["upload_date"] = video["upload_date"]

return part

def find_secondary_item(vid: str, canonical: str, items: dict) -> str | None:

"""Locate the non-canonical item carrying this video id (slug-named Other/ dup)."""

for rel, entry in items.items():

if rel == canonical:

continue

if any(p.get("video_id") == vid for p in entry["meta"].get("parts", [])):

return rel

return None

def process_split_files(

split_dir: Path, items: dict, manifest: dict, report: dict

) -> dict[str, dict]:

"""Parse *_split.txt files -> {item_rel: {enrichment, fill_parts}}; mark Other/ duplicates."""

pending: dict[str, dict] = {}

for split_path in sorted(split_dir.glob("*_split.txt")):

result = parse_split_file(split_path.read_text(encoding="utf-8"))

target = f"{result.category}/{result.series}"

if target not in items:

report["errors"].append(

f"{split_path.name}: target item {target!r} not found in journal"

)

continue

enrichment = {

"sessions": result.sessions,

"enriched_from": ["split_file"],

}

if result.description:

enrichment["description"] = result.description

fill = [manifest_part(manifest[result.video_id])] if result.video_id in manifest else None

pending[target] = {"enrichment": enrichment, "fill_parts": fill}

report["split_files"].append(

{

"file": split_path.name,

"item": target,

"sessions": len(result.sessions),

"parts_filled": bool(fill),

}

)

duplicate = find_secondary_item(result.video_id, target, items)

if duplicate:

pending[duplicate] = {

"enrichment": {"duplicate_of": target, "enriched_from": ["split_file"]},

"fill_parts": None,

}

report["duplicates_marked"].append(duplicate)

return pending

def match_coda_rows(

rows: list[dict], vid_index: dict, items: dict, report: dict

) -> dict[str, list]:

"""Return item_rel -> [(video_id_or_None, mapped_fields), ...]."""

matched: dict[str, list] = defaultdict(list)

claimed_fallback: set[str] = set()

for row in rows:

values = row.get("values", {})

name = row.get("name", "") or values.get("Unique event name", "")

vid = extract_youtube_id(values.get("YouTube", "") or "")

fields = map_coda_row(values)

if not fields:

continue

if vid and vid in vid_index:

matched[vid_index[vid]].append((vid, fields))

report["matched_by_id"] += 1

continue

# Fallback: reconstruct the item path from the Unique event name.

category, series, _ = categorize_name(values.get("Unique event name", "") or "", True)

rel = f"{category}/{series}" if category and series else None

if rel and rel in items and rel not in claimed_fallback and rel not in matched:

claimed_fallback.add(rel)

matched[rel].append((vid, fields))

report["matched_by_fallback"].append({"row": name, "item": rel})

else:

report["unmatched"].append({"row": name, "video_id": vid or None})

return matched

# List-valued fields whose cross-row differences merge by order-preserving union.

UNION_KEYS = ("guests", "other_participants", "keywords")

# Human-verified corrections (2026-07-16): the refactor's 11-char id regex

# matched asset filename fragments for these items; the real uploads sat

# uncategorized in Other/. Maps item -> its actual YouTube video id.

CURATED_PARTS = {

"Applied Active Inference Symposium/2023 Ecosystem Symposium/First_Interval": "rIemcswLfGg",

"Applied Active Inference Symposium/2023 Ecosystem Symposium/Second_Interval": "PVeyvHSAwmk",

# Social Sciences course lectures: curated items had empty parts while the

# real uploads sat uncategorized (verified 2026-07-17).

"Courses/ActiveInferenceForTheSocialSciences/ActInf_Basics_Lecture": "BNLnbOFdgc0",

"Courses/ActiveInferenceForTheSocialSciences/Introduction_Lecture": "JkORFoyk8o8",

}

# (CURATED_GUESTS retired 2026-07-16: the swapped guest lists for the two

# 2023 Symposium videos were fixed upstream in the Coda/Superhuman table.)

def process_curated_parts(items: dict, manifest: dict, pending: dict, report: dict) -> None:

"""Replace bogus refactor-derived parts with the real video; mark Other/ duplicates."""

for rel, vid in CURATED_PARTS.items():

if rel not in items or vid not in manifest:

report["errors"].append(f"curated parts: {rel!r} or video {vid} unavailable")

continue

pending[rel] = {

"enrichment": {"enriched_from": ["curated"]},

"fill_parts": [manifest_part(manifest[vid])],

"replace_parts": True,

"overrides": {"sessions": CURATED_SESSIONS[rel]} if rel in CURATED_SESSIONS else {},

}

duplicate = find_secondary_item(vid, rel, items)

if duplicate:

pending[duplicate] = {

"enrichment": {"duplicate_of": rel, "enriched_from": ["curated"]},

"fill_parts": None,

}

report["duplicates_marked"].append(duplicate)

report["curated_parts"].append({"item": rel, "video_id": vid})

def _fmt_start(seconds: float) -> str:

total = int(seconds)

return f"{total // 3600}:{total % 3600 // 60:02d}:{total % 60:02d}"

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

"""Load a chapters cache: video_id -> payload (list of chapters, or a

provenance dict ``{source, model, generated_at, chapters: [...]}``).

Trust decisions are made per-part in chapter_sessions, where the

part's video duration is available for the coverage rule.

"""

if not path.exists():

return {}

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

def _usable_part_chapters(payload: Any, duration: float | None) -> list | None: # noqa: ANN401 - cache payload is list-or-dict by design

"""Return seedable chapters for one video part, or None to skip.

Trust model:

- ``{"source": "youtube", "chapters": [...]}`` or a plain list (the

video_chapters.json shape — only fetch_chapters.py writes it, from

YouTube's own chapter data) → trusted; needs >=2 chapters.

- Anything else with provenance (``source`` != youtube, i.e. the LLM

cache) must pass validate_chapters with the part's duration — lists

without a duration can never prove coverage, so they are rejected.

"""

source = ""

entries: list

if isinstance(payload, dict) and "chapters" in payload:

source = str(payload.get("source") or "")

entries = payload["chapters"]

elif isinstance(payload, list):

entries = payload

sources = {c.get("source", "") for c in entries if isinstance(c, dict)}

if sources == {"youtube"}:

source = "youtube"

else:

return None

if source == "youtube":

return entries if len(entries) >= 2 else None

if duration is None:

return None

if not validate_chapters(entries, duration_seconds=duration).passed:

return None

return entries

def chapter_sessions(meta: dict, chapters: dict[str, Any]) -> list[dict]:

"""Seed sessions[] from trusted chapter lists, sequential across parts.

Per part, chapters are seedable when their source is ``youtube`` or they

pass ``validate_chapters`` (journal part duration required for the

coverage rule). Unprovenanced lists — e.g. the legacy LLM cache — are

never seeded. Only fires when a part has >=2 chapters (a single chapter

is not a session structure). Speaker attribution is left to humans — the

journal owns sessions after seeding.

"""

sessions: list[dict] = []

for part in meta.get("parts", []):

vid = part.get("video_id", "")

if not vid:

continue

usable = _usable_part_chapters(chapters.get(vid), part.get("duration"))

if not usable:

continue

for chapter in usable:

index = len(sessions) + 1

sessions.append(

{

"index": index,

"session_name": f"{vid}_sess{index:02d}",

"start": _fmt_start(chapter["start"]),

"title": chapter["title"],

}

)

return sessions

def _sess(vid: str, index: int, start: str, guests: list[str], title: str = "") -> dict:

session = {"index": index, "session_name": f"{vid}_sess{index:02d}", "start": start}

if title:

session["title"] = title

if guests:

session["guests"] = guests

return session

# Per-talk segments from the video descriptions (verified 2026-07-16).

CURATED_SESSIONS = {

"Applied Active Inference Symposium/2023 Ecosystem Symposium/First_Interval": [

_sess("rIemcswLfGg", 1, "0:00:00", ["André Bastos"]),

_sess("rIemcswLfGg", 2, "0:44:19", ["Keith Duggar"]),

_sess("rIemcswLfGg", 3, "1:03:54", ["Sanjeev Namjoshi"]),

_sess("rIemcswLfGg", 4, "1:23:35", ["Inês Hipólito"]),

_sess("rIemcswLfGg", 5, "1:58:05", ["Aswin Paul"]),

_sess("rIemcswLfGg", 6, "3:19:30", ["Takuya Isomura"]),

_sess("rIemcswLfGg", 7, "3:56:55", ["Shanna Dobson"]),

_sess("rIemcswLfGg", 8, "4:28:25", ["Nynke Boiten"]),

"Applied Active Inference Symposium/2023 Ecosystem Symposium/Second_Interval": [

_sess("PVeyvHSAwmk", 1, "0:00:00", ["Jean-François Cloutier"]),

_sess("PVeyvHSAwmk", 2, "1:01:25", ["Conor Heins"]),

_sess("PVeyvHSAwmk", 3, "1:31:24", ["Bert de Vries", "Dmitry Bagaev", "Bart van Erp"]),

_sess("PVeyvHSAwmk", 4, "2:19:49", [], title="Active Inference Institute"),

_sess("PVeyvHSAwmk", 5, "3:01:20", ["Rafael Kaufmann"]),

_sess("PVeyvHSAwmk", 6, "3:31:10", ["Avel Guénin-Carlut"]),

_sess("PVeyvHSAwmk", 7, "4:00:40", ["Pablo Fernandez-Maquieira"]),

_sess("PVeyvHSAwmk", 8, "4:30:00", ["Mahault Albarracin"]),

_sess(

"PVeyvHSAwmk",

"5:02:30",

[

"Bert de Vries",

"Rafael Kaufmann",

"Anna Lembke",

"Curt Jaimungal",

"Karl J Friston",

"Guillaume Dumas",

title="Roundtable",

}

# Human-reviewed disagreements where the Coda value was confirmed correct;

# suppressed from the review report. (item, field, db names as parsed)

RESOLVED_NAME_DIFFS = {

# Verified against the video and website 2026-07-16: guest is Austin Cook.

("GuestStream/GuestStream_072", "guests", ("John Cook",)),

}

def _collect_name_diffs(rec: dict, enrichment: dict, rel: str, report: dict) -> None:

"""Record (not apply) people-name disagreements between legacy DB and Coda."""

from journal_utilities.data.enrichment import split_names

for field in ("guests", "other_participants"):

db_names = split_names(rec.get(field) or "")

coda_names = enrichment.get(field) or []

if (rel, field, tuple(db_names)) in RESOLVED_NAME_DIFFS:

continue

if db_names and coda_names and not set(db_names) <= set(coda_names):

report["name_diffs"].append(

{"item": rel, "field": field, "db": db_names, "coda": coda_names}

)

def write_private_registry(

journal: Path, db_records: dict, vid_index: dict, apply: bool, report: dict

) -> None:

"""Document private/unlisted channel videos absent from the journal."""

private = [

{"video_id": vid, "title": rec.get("title", "")}

for vid, rec in sorted(db_records.items())

if rec.get("is_private") and vid not in vid_index

]

if not private:

return

path = journal / SRC_PREFIX / "private_videos.json"

payload = (

json.dumps(

{

"description": "Private/unlisted videos known to the Institute but absent from "

"this public corpus (from the legacy session database).",

"videos": private,

indent=2,

ensure_ascii=False,

)

+ "\n"

)

changed = not path.exists() or path.read_text(encoding="utf-8") != payload

if changed and apply:

path.write_text(payload, encoding="utf-8")

report["private_registry"] = {"videos": len(private), "written": changed and apply}

def combine_rows(entries: list, meta: dict, report: dict, rel: str) -> tuple[dict, dict]:

"""Merge multiple matched rows: shared values -> item level, list fields

union, per-part fields attach to their part, rest first-row-wins."""

if len(entries) == 1:

return entries[0][1], {}

item_level: dict = {}

part_updates: dict = defaultdict(dict)

keys = {key for _, fields in entries for key in fields}

for key in keys:

values = [fields.get(key) for _, fields in entries if fields.get(key)]

unique = {json.dumps(v, sort_keys=True) for v in values}

if len(unique) == 1:

item_level[key] = values[0]

elif key in UNION_KEYS:

item_level[key] = list(dict.fromkeys(name for value in values for name in value))

elif key in PER_PART_KEYS:

for vid, fields in entries:

if vid and fields.get(key):

part_updates[vid][key] = fields[key]

report["per_part_attachments"].append({"item": rel, "field": key})

else:

item_level[key] = values[0]

report["conflicts"].append({"item": rel, "field": key, "kept": "first row"})

return item_level, dict(part_updates)

def main() -> int:

parser = argparse.ArgumentParser(

description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter

)

parser.add_argument("--journal", type=Path, default=REPO.parent / "ActiveInferenceJournal")

parser.add_argument("--snapshot-only", action="store_true", help="skip the Coda API fetch")

parser.add_argument(

"--coda-cache", type=Path, default=REPO / "data/input/livestream_fulldata_full.json"

)

parser.add_argument("--split-dir", type=Path, default=REPO / "data/input")

parser.add_argument("--manifest", type=Path, default=REPO / "data/output/channel_videos.json")

parser.add_argument(

"--db-export", type=Path, default=REPO / "data/input/session_db_export.json"

)

parser.add_argument("--chapters", type=Path, default=REPO / "data/input/video_chapters.json")

parser.add_argument("--apply", action="store_true", help="write changes (default: dry run)")

parser.add_argument("--report", type=Path, help="write full JSON report to this path")

args = parser.parse_args()

logging.basicConfig(level=logging.INFO, format="%(levelname)s %(message)s")

load_dotenv(REPO / ".env")

report: dict = {

"source": "",

"rows_total": 0,

"matched_by_id": 0,

"matched_by_fallback": [],

"unmatched": [],

"split_files": [],

"duplicates_marked": [],

"per_part_attachments": [],

"conflicts": [],

"items_enriched": 0,

"items_unchanged": 0,

"errors": [],

"db_export": "absent",

"db_slides": [],

"name_diffs": [],

"private_registry": None,

"curated_parts": [],

"chapter_seeded": [],

"chapters_cache": "absent",

"per_series": defaultdict(lambda: {"enriched": 0, "total": 0}),

}

items = load_journal_index(args.journal)

if not items:

logger.error("no journal items under %s/%s", args.journal, SRC_PREFIX)

return 1

manifest = load_manifest(args.manifest)

# Split files and curated corrections first, so symposium parts are

# filled/replaced before id-matching.

pending = process_split_files(args.split_dir, items, manifest, report)

process_curated_parts(items, manifest, pending, report)

vid_index = video_index(items)

for rel, work in pending.items():

for part in work.get("fill_parts") or []:

vid_index[part["video_id"]] = rel # symposium item wins over Other/ duplicate

rows, source = load_rows(

snapshot_path=REPO / "data/input/livestream_fulldata_table.json",

cache_path=args.coda_cache,

fetch=not args.snapshot_only,

)

report["source"] = source

report["rows_total"] = len(rows)

matched = match_coda_rows(rows, vid_index, items, report)

db_records = load_db_export(args.db_export)

db_by_item: dict[str, list] = defaultdict(list)

for vid, rec in db_records.items():

if vid in vid_index:

db_by_item[vid_index[vid]].append(rec)

report["db_export"] = (

f"{args.db_export.name} ({len(db_records)} sessions)" if db_records else "absent"

)

chapters = load_chapters(args.chapters)

report["chapters_cache"] = (

f"{sum(1 for c in chapters.values() if c)} videos with chapters" if chapters else "absent"

)

for rel, entry in items.items():

series = rel.split("/")[0]

report["per_series"][series]["total"] += 1

# Provenance is sticky: a source that contributed in a past run stays

# recorded even when this run has nothing new from it.

enrichment: dict = {"enriched_from": list(entry["meta"].get("enriched_from", []))}

part_updates: dict = {}

fill_parts = None

replace_parts = False

if rel in matched:

row_fields, part_updates = combine_rows(matched[rel], entry["meta"], report, rel)

row_fields.pop("enriched_from", None)

enrichment.update(row_fields)

enrichment["enriched_from"] = list(

dict.fromkeys(enrichment["enriched_from"] + ["coda"])

)

if rel in pending:

work = pending[rel]

for key, value in work["enrichment"].items():

if key == "enriched_from":

enrichment["enriched_from"] = list(

dict.fromkeys(enrichment.get("enriched_from", []) + value)

)

else:

enrichment.setdefault(key, value)

for key, value in work.get("overrides", {}).items():

enrichment[key] = value

fill_parts = work["fill_parts"]

replace_parts = work.get("replace_parts", False)

if fill_parts:

enrichment["enriched_from"] = list(

dict.fromkeys(enrichment.get("enriched_from", []) + ["youtube"])

)

# YouTube chapters seed sessions[] when nothing else has (journal-owned

# after seeding — hand-edits there are never overwritten).

if not enrichment.get("sessions") and not entry["meta"].get("sessions"):

seeded = chapter_sessions(entry["meta"], chapters)

if seeded:

enrichment["sessions"] = seeded

enrichment["enriched_from"] = list(

dict.fromkeys(enrichment.get("enriched_from", []) + ["youtube"])

)

report["chapter_seeded"].append(rel)

# Legacy-DB overlay: slides links the current Coda table lost or holds junk for.

# Single-part items take them at item level; multi-part items per-part.

db_recs = db_by_item.get(rel, [])

multi_part = len(entry["meta"].get("parts", [])) > 1

for rec in db_recs:

db_slides = (rec.get("slides_url") or "").strip()

_collect_name_diffs(rec, enrichment, rel, report)

if not db_slides.startswith(("http://", "https://")):

continue

applied = False

if multi_part:

vid = rec["session_name"]

existing = next(

(

p.get("slides_url", "")

for p in entry["meta"]["parts"]

if p.get("video_id") == vid

"",

)

if prefer_url(existing, db_slides) != existing:

part_updates.setdefault(vid, {})["slides_url"] = db_slides

report["db_slides"].append(

{"item": rel, "part": vid, "replaced": existing or None}

)

applied = True

else:

current = enrichment.get("slides_url") or entry["meta"].get("slides_url") or ""

if prefer_url(current, db_slides) != current:

enrichment["slides_url"] = db_slides

report["db_slides"].append(

{"item": rel, "part": None, "replaced": current or None}

)

applied = True

already_db = db_slides in (

[enrichment.get("slides_url"), entry["meta"].get("slides_url")]

+ [p.get("slides_url") for p in entry["meta"].get("parts", [])]

)

if applied or already_db:

enrichment["enriched_from"] = list(

dict.fromkeys(enrichment.get("enriched_from", []) + ["db"])

)

# Canonical self-link into this repo's current layout (Coda's are pre-v2, stale).

enrichment["github"] = github_link(rel)

enrichment["enriched_from"] = list(

dict.fromkeys(enrichment.get("enriched_from", []) + ["generated"])

)

new_meta, changed = merge_enrichment(

entry["meta"], enrichment, part_updates, fill_parts, replace_parts

)

if changed:

report["items_enriched"] += 1

report["per_series"][series]["enriched"] += 1

if args.apply:

entry["path"].write_text(

json.dumps(new_meta, indent=2, ensure_ascii=False) + "\n", encoding="utf-8"

)

else:

report["items_unchanged"] += 1

write_private_registry(args.journal, db_records, vid_index, args.apply, report)

report["per_series"] = dict(sorted(report["per_series"].items()))

_print_summary(report, dry_run=not args.apply)

if args.report:

args.report.write_text(

json.dumps(report, indent=2, ensure_ascii=False) + "\n", encoding="utf-8"

)

logger.info("full report -> %s", args.report)

return 0 if not report["errors"] else 1

def _print_summary(report: dict, dry_run: bool) -> None:

mode = "DRY RUN" if dry_run else "APPLIED"

print(f"\n=== enrich_metadata {mode} ===")

print(f"coda source: {report['source']}")

print(f"coda rows: {report['rows_total']}")

print(f" matched by id: {report['matched_by_id']}")

print(f" matched by fallback:{len(report['matched_by_fallback']):>4}")

print(f" unmatched: {len(report['unmatched'])}")

print(

f"split files: {len(report['split_files'])} "

f"({sum(s['sessions'] for s in report['split_files'])} sessions)"

)

print(f"duplicates marked: {report['duplicates_marked']}")

print(f"curated part fixes: {[c['item'].split('/')[-1] for c in report['curated_parts']]}")

Journal_Utilities/scripts/enrich_metadata.py at main · ActiveInferenceInstitute/Journal_Utilities · GitHub

Expand file tree

Breadcrumbs

enrich_metadata.py

Latest commit

History

Breadcrumbs

enrich_metadata.py

File metadata and controls