feat: linha do tempo processual dos autos vinculados Pet16704

Substitui o gitGraph do Mermaid (ilegível com 3.437 commits) por uma timeline interativa com eixo temporal real, filtros por instituição, tipo de documento e pessoas mencionadas (extraidas via NER local com pierreguillou/ner-bert-base-cased-pt-lenerbr).
2026-09-13 10:37:10 +02:00
commit 891d58ec55
10 changed files with 1559 additions and 0 deletions
@@ -0,0 +1,144 @@
+import re, json, sys
+
+root = "/Users/polux/tmp/Pet16704-pt2-md"
+
+# 1. Parse gitgraph.md to get id -> branch (lane), preserving sequence order.
+branch_of = {}
+order_of = {}
+current_branch = "STF · Relatoria"
+idx = 0
+with open(f"{root}/gitgraph.md", encoding="utf-8") as f:
+    for line in f:
+        line = line.strip()
+        m = re.match(r'checkout "([^"]*)"', line)
+        if m:
+            current_branch = m.group(1)
+            continue
+        m = re.match(r'commit id: "([^"]*)"', line)
+        if m:
+            cid = m.group(1)
+            branch_of[cid] = current_branch
+            order_of[cid] = idx
+            idx += 1
+
+print(f"gitgraph commits: {len(branch_of)}", file=sys.stderr)
+
+# 2. Parse TIMELINE.md: sections "## <Case>" then lines "- **NNNNN** (YYYY-MM-DD) — text — [`file`](<path>)"
+records = []
+current_case = None
+line_re = re.compile(
+    r'^- \*\*(\d+)\*\*(?: \((\d{4}-\d{2}-\d{2})\))? — (.*?) — \[`[^`]*`\]\(<([^>]*)>\)\s*$'
+)
+with open(f"{root}/TIMELINE.md", encoding="utf-8") as f:
+    for line in f:
+        line = line.rstrip("\n")
+        m = re.match(r'^## (.+)$', line)
+        if m:
+            current_case = m.group(1).strip()
+            continue
+        m = line_re.match(line)
+        if m:
+            num, date, text, path = m.groups()
+            cid = f"{current_case}-{num}"
+            records.append({
+                "id": cid,
+                "date": date,
+                "text": text,
+                "file": path,
+            })
+
+# disambiguate duplicate ids (multiple documents filed under the same number)
+# to match gitgraph.md's "-a"/"-b"/"-c" suffix convention
+from collections import Counter
+id_counts = Counter(r["id"] for r in records)
+seen = Counter()
+suffixes = "abcdefgh"
+for r in records:
+    base = r["id"]
+    if id_counts[base] > 1:
+        r["id"] = f"{base}-{suffixes[seen[base]]}"
+        seen[base] += 1
+
+print(f"timeline records: {len(records)}", file=sys.stderr)
+
+# 3a. Extract & normalize document type from the source filename.
+TYPE_MAP = {
+    "Peticao": "Petição",
+    "Peticao inicial": "Petição inicial",
+    "Recibo de peticao eletronica": "Recibo de petição eletrônica",
+    "Documentos comprobatorios": "Documento comprobatório",
+    "Documento comprobatorio": "Documento comprobatório",
+    "Comunicacao assinada": "Comunicação assinada",
+    "Procuracao": "Procuração",
+    "Despacho": "Despacho",
+    "Certidao": "Certidão",
+    "Vista a PGR": "Vista à PGR",
+    "Intimacao": "Intimação",
+    "Decisao monocratica": "Decisão monocrática",
+    "Mandado de intimacao": "Mandado de intimação",
+    "Termo de disponibilizacao de autos": "Termo de disponibilização de autos",
+    "Documentos de identificacao": "Documentos de identificação",
+    "Aviso de recebimento": "Aviso de recebimento",
+    "Outras pecas": "Outras peças",
+    "Manifestacao": "Manifestação",
+    "Certidao de transito em julgado": "Certidão de trânsito em julgado",
+    "Certidao de julgamento": "Certidão de julgamento",
+    "Oficio": "Ofício",
+    "Documento Sigiloso": "Documento sigiloso",
+    "Manifestacao da PGR": "Manifestação da PGR",
+    "Pedido de ingresso como interessado": "Pedido de ingresso como interessado",
+    "Volume": "Volume",
+    "Inteiro teor do acordao": "Inteiro teor do acórdão",
+    "Inteiro teor do acordao (completo)": "Inteiro teor do acórdão",
+    "Mandado de prisao": "Mandado de prisão",
+    "Pedido de reconsideracao": "Pedido de reconsideração",
+    "Informacao": "Informação",
+    "Peticao de Interposicao de Agravo Regimental": "Petição de interposição de agravo regimental",
+    "Peticao de renuncia ao mandato": "Petição de renúncia ao mandato",
+}
+
+def extract_type(file_path):
+    base = re.sub(r'\.md$', '', file_path.split('/')[-1])
+    m = re.match(r'^\d+\s+(.*)$', base)
+    if not m:
+        return "Outro"
+    rest = re.sub(r'_[0-9a-fA-F]{6,}$', '', m.group(1))
+    if ' - ' in rest:
+        rest = rest.split(' - ')[0]
+    rest = rest.strip()
+    return TYPE_MAP.get(rest, rest)
+
+# 3b. Merge
+merged = []
+missing_branch = 0
+for r in records:
+    branch = branch_of.get(r["id"])
+    if branch is None:
+        missing_branch += 1
+        branch = "STF · Relatoria"
+    merged.append({
+        "id": r["id"],
+        "branch": branch,
+        "date": r["date"],
+        "text": r["text"],
+        "file": r["file"],
+        "type": extract_type(r["file"]),
+        "order": order_of.get(r["id"], -1),
+    })
+
+print(f"missing branch matches: {missing_branch}", file=sys.stderr)
+
+# also report ids in gitgraph not in timeline (should just be "raiz")
+timeline_ids = {r["id"] for r in records}
+only_in_gitgraph = [cid for cid in branch_of if cid not in timeline_ids]
+print(f"only in gitgraph (not in timeline): {only_in_gitgraph[:10]} (total {len(only_in_gitgraph)})", file=sys.stderr)
+
+merged.sort(key=lambda r: r["order"])
+
+with open("/private/tmp/claude-501/-Users-polux-tmp-Pet16704-pt2-md/a294e67f-4337-4617-9903-54ea4711fb63/scratchpad/docs.json", "w", encoding="utf-8") as f:
+    json.dump(merged, f, ensure_ascii=False)
+
+branches = sorted(set(branch_of.values()))
+print("branches:", branches, file=sys.stderr)
+dated = sum(1 for r in merged if r["date"])
+print(f"dated: {dated}, undated: {len(merged)-dated}", file=sys.stderr)
@@ -0,0 +1,54 @@
+import json, re, time
+from collections import Counter, defaultdict
+from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
+
+docs = json.load(open("docs.json"))
+texts = [d["text"] for d in docs]
+print(f"{len(texts)} documents to process", flush=True)
+
+name = "pierreguillou/ner-bert-base-cased-pt-lenerbr"
+tok = AutoTokenizer.from_pretrained(name)
+model = AutoModelForTokenClassification.from_pretrained(name)
+nlp = pipeline("ner", model=model, tokenizer=tok, aggregation_strategy="simple", device=-1)
+
+def merge_entities(ents, group="PESSOA"):
+    ents = [e for e in ents if e["entity_group"] == group]
+    ents.sort(key=lambda e: e["start"])
+    merged = []
+    for e in ents:
+        if merged and e["start"] - merged[-1]["end"] <= 1:
+            merged[-1]["end"] = e["end"]
+        else:
+            merged.append({"start": e["start"], "end": e["end"]})
+    return merged
+
+t0 = time.time()
+BATCH = 32
+all_people_per_doc = []  # list aligned with docs: list of names
+for i in range(0, len(texts), BATCH):
+    batch = texts[i:i+BATCH]
+    results = nlp(batch)
+    for text, ents in zip(batch, results):
+        merged = merge_entities(ents, "PESSOA")
+        names = [text[m["start"]:m["end"]].strip() for m in merged]
+        names = [n for n in names if len(n) >= 3]
+        all_people_per_doc.append(names)
+    if (i//BATCH) % 10 == 0:
+        elapsed = time.time()-t0
+        print(f"{i+len(batch)}/{len(texts)} docs, {elapsed:.1f}s elapsed", flush=True)
+
+print("total time", time.time()-t0, "s")
+
+# save raw per-doc results
+out = []
+for d, names in zip(docs, all_people_per_doc):
+    out.append({"id": d["id"], "names": names})
+json.dump(out, open("ner_people_per_doc.json","w",encoding="utf-8"), ensure_ascii=False)
+
+counter = Counter()
+for names in all_people_per_doc:
+    for n in names:
+        counter[n] += 1
+print("distinct raw person strings:", len(counter))
+for n,c in counter.most_common(50):
+    print(c, n)