#!/usr/bin/env python3 """TSLaw: buat artikel kesembilanbelas (draft) - Notaris dan PPAT di Surabaya. Tahapan: siapkan foto (crop 1000x567 webp) -> cek semua tautan internal 200 -> cek kanibalisasi -> unggah media + alt -> buat post draft (REST, JSON) -> meta Yoast (REST) -> verifikasi baca ulang. Jalankan tanpa argumen = dry run; dengan --apply = benar-benar membuat di WordPress.""" import base64, json, os, re, sys, time, urllib.request, urllib.error sys.path.insert(0, os.path.dirname(os.path.abspath(__file__))) from tslaw_listikel_19 import A19 DOM = "tslaw.id" USER, PASS = "bambangbot", "EFKm EYGx VQCP Y9go ncPr uM2Y" AUTH = base64.b64encode(f"{USER}:{PASS}".encode()).decode() UA = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/124 Safari/537.36"} OUT = "/home/mantapman/.hermes/profiles/atun/workspace/tslaw_listikel" KATEGORI = [14] # Hukum Bisnis FOTO_ID = "7651557" # Pexels: dua orang memeriksa dokumen di ruang pelayanan FOTO_YBIAS = 0.35 SALINAN = f"{OUT}/artikel19.html" ALT = ("Dua orang memeriksa berkas akta dan sertifikat di ruang pelayanan, gambaran urusan notaris dan PPAT " "di Surabaya") def rest(path, method="GET", data=None): h = {"Authorization": f"Basic {AUTH}", **UA} body = None if data is not None: body = json.dumps(data).encode() h["Content-Type"] = "application/json" r = urllib.request.Request(f"https://{DOM}{path}", data=body, headers=h, method=method) for i in range(4): try: with urllib.request.urlopen(r, timeout=180) as x: return x.status, json.loads(x.read()) except urllib.error.HTTPError as e: return e.code, e.read().decode()[:300] except Exception as e: print(f" (ulang {i+1}: {str(e)[:60]})") time.sleep(8) return 0, "gagal" def potong_foto(pexels_id, nama, ybias=0.0): from PIL import Image u = f"https://images.pexels.com/photos/{pexels_id}/pexels-photo-{pexels_id}.jpeg?auto=compress&cs=tinysrgb&w=1600" data = urllib.request.urlopen(urllib.request.Request(u, headers=UA), timeout=120).read() tmp = f"{OUT}/_tmp_{nama}.jpg" open(tmp, "wb").write(data) im = Image.open(tmp).convert("RGB") tw, th = 1000, 567 w, h = im.size r = tw / th if w / h > r: cw = int(h * r); x0 = (w - cw) // 2 im = im.crop((x0, 0, x0 + cw, h)) else: ch = int(w / r); y0 = int((h - ch) * ybias) im = im.crop((0, y0, w, y0 + ch)) f = f"{OUT}/{nama}" im.resize((tw, th), Image.LANCZOS).save(f, "WEBP", quality=80, method=6) os.remove(tmp) return f def hitung(html): teks = re.sub(r"<(script|style)[^>]*>.*?", " ", html, flags=re.S | re.I) return len([w for w in re.split(r"\s+", re.sub(r"<[^>]+>", " ", teks)) if w]) def main(): a = A19 kata = hitung(a["isi"]) meta_teks = re.sub(r"\s+", " ", a["meta"]) print(f"=== PERSIAPAN ===\n {a['slug']} | {kata} kata | meta {len(meta_teks)} char | judul SEO {len(a['seo'])} char") assert 1000 <= kata <= 1300, f"jumlah kata di luar 1000-1300: {kata}" assert len(meta_teks) <= 158, "meta lebih dari 158" assert a["fokus"] in meta_teks.lower(), "keyphrase tidak ada di meta" assert meta_teks.lower().startswith(a["fokus"]), "keyphrase tidak di awal meta" assert not re.findall(r"\{[A-Za-z_\[\]0-9]+\}", a["isi"]), "masih ada placeholder" assert not re.findall(r"[\u3000-\u9fff\uac00-\ud7af]", a["isi"]), "ada karakter CJK" assert chr(8212) not in a["isi"], "ada em dash" assert "E0T" not in a["isi"] and "E6T" not in a["isi"], "entri mentah masih tertinggal" assert "https://www.tslaw.id/practice-areas/" not in a["isi"], "tautan /practice-areas/ (404) dipakai" assert ".htm" not in a["isi"], "ada tautan .htm" assert '\"draft\"' not in a["isi"], "menyebut status draft di isi" wa = re.findall(r'href="(https://wa\.me/[^"]+)"', a["isi"]) assert wa and all(re.fullmatch(r"https://wa\.me/\d+(\?[^\s\"]*)?", u) for u in wa), f"tautan WA cacat: {wa}" assert os.path.exists(SALINAN), "salinan lokal belum dibuat" print(f" meta final : {meta_teks}") print(f" tombol WA : {wa}") print(f" keyphrase di isi : {a['isi'].lower().count(a['fokus'])} kali") print(f" salinan isi lokal : {SALINAN} ({os.path.getsize(SALINAN)} byte)") print("\n=== CEK TAUTAN INTERNAL ===") for u in sorted(set(re.findall(r'href="(https://www\.tslaw\.id/[^"]*)"', a["isi"]))): try: with urllib.request.urlopen(urllib.request.Request(u, headers=UA), timeout=60) as x: ok = x.status except urllib.error.HTTPError as e: ok = e.code except Exception as e: ok = f"ERR {str(e)[:40]}" print(f" {ok} {u}") assert ok == 200, f"tautan bukan 200: {u} -> {ok}" print("\n=== PENJAGA KANIBALISASI ===") semua, off = [], 0 while off < 300: st, p = rest(f"/wp-json/wp/v2/posts?status=publish,future,draft&per_page=100&offset={off}" "&context=edit&_fields=id,slug,title,status,meta") if not isinstance(p, list) or not p: break semua += p off += 100 print(f" diperiksa: {len(semua)} artikel") kata_ku = set(re.sub(r"[^a-z0-9 ]", "", a["fokus"].lower()).split()) for p in semua: m = p.get("meta") or {} fk = (m.get("_yoast_wpseo_focuskw") or "").lower().strip() t = p["title"]["raw"] if isinstance(p["title"], dict) else str(p["title"]) if fk and fk == a["fokus"]: raise SystemExit(f" STOP: fokus kata kunci sama dengan #{p['id']} ({fk})") kata_p = set(re.sub(r"[^a-z0-9 ]", "", (fk or t).lower()).split()) if kata_ku and kata_p: skor = len(kata_ku & kata_p) / len(kata_ku | kata_p) if skor >= 0.34: print(f" tetangga dekat: #{p['id']} skor {skor:.2f} | fokus='{fk}' | {t[:70]}") print(" tidak ada fokus kembar") foto = potong_foto(FOTO_ID, a["slug"] + ".webp", FOTO_YBIAS) from PIL import Image print(f" foto siap : {foto} ({os.path.getsize(foto)//1024} KB) {Image.open(foto).size}") if "--apply" not in sys.argv: print("\n(dry run - pakai --apply untuk membuat di WordPress)") return st, ada = rest(f"/wp-json/wp/v2/posts?slug={a['slug']}&status=any&_fields=id,slug,status,link") if isinstance(ada, list) and ada: print(f"\n PERINGATAN: slug sudah ada: {ada} -> hentikan supaya tidak menimpa") return print(f"\n=== SLUG ===\n {a['slug']} bebas dipakai") st, kat = rest(f"/wp-json/wp/v2/categories?include={','.join(map(str, KATEGORI))}&_fields=id,name,slug") if isinstance(kat, list): for k in kat: print(f" kategori {k['id']} = {k['name']} ({k['slug']})") st, semua_tag = rest("/wp-json/wp/v2/tags?per_page=100&orderby=count&order=desc&_fields=id,name,slug,count") peta = {t["name"].lower(): t for t in semua_tag} if isinstance(semua_tag, list) else {} print("\n=== TAG ===") tag_ids = [] for nama in a["tag_nama"]: t = peta.get(nama.lower()) if t: tag_ids.append(t["id"]) print(f" pakai {nama:<24} #{t['id']} ({t['count']} artikel)") else: time.sleep(1) st, baru = rest("/wp-json/wp/v2/tags", "POST", {"name": nama}) if isinstance(baru, dict) and "id" in baru: tag_ids.append(baru["id"]) print(f" buat {nama:<24} #{baru['id']}") elif isinstance(baru, str) and "term_exists" in baru: slug = re.sub(r"[^a-z0-9]+", "-", nama.lower()).strip("-") st2, cari = rest(f"/wp-json/wp/v2/tags?slug={slug}&_fields=id,name,slug") if isinstance(cari, list) and cari: tag_ids.append(cari[0]["id"]) print(f" ada {nama:<24} #{cari[0]['id']} (lewat slug)") else: print(f" GAGAL {nama}: term_exists tapi tidak ketemu slug {slug}") else: print(f" GAGAL {nama}: {str(baru)[:120]}") print(f" tag dipakai: {tag_ids}") assert len(tag_ids) == 4, f"tag kurang dari 4: {tag_ids}" print("\n=== FOTO ===") h = {"Authorization": f"Basic {AUTH}", **UA, "Content-Type": "image/webp", "Content-Disposition": f'attachment; filename="{a["slug"]}.webp"'} req = urllib.request.Request(f"https://{DOM}/wp-json/wp/v2/media", data=open(foto, "rb").read(), headers=h, method="POST") try: with urllib.request.urlopen(req, timeout=240) as x: med = json.loads(x.read()) except urllib.error.HTTPError as e: print(f" GAGAL unggah foto: {e.code} {e.read().decode()[:200]}") return mid = med["id"] rest(f"/wp-json/wp/v2/media/{mid}", "POST", {"alt_text": ALT, "title": a["judul"]}) st, cek = rest(f"/wp-json/wp/v2/media/{mid}?context=edit&_fields=id,alt_text,source_url") print(f" media #{mid} -> {med.get('source_url')}") print(f" alt tersimpan: {'ya' if (isinstance(cek, dict) and cek.get('alt_text') == ALT) else 'TIDAK - ' + str(cek)[:80]}") print("\n=== ARTIKEL (draft) ===") payload = {"title": a["judul"], "slug": a["slug"], "content": a["isi"], "status": "draft", "categories": KATEGORI, "tags": tag_ids, "featured_media": mid, "excerpt": meta_teks} st, p = rest("/wp-json/wp/v2/posts", "POST", payload) if st not in (200, 201): print(f" GAGAL buat artikel: {st} {str(p)[:300]}") return pid = p["id"] print(f" artikel #{pid} dibuat -> {p.get('link')} (status: {p.get('status')})") print("\n=== META YOAST ===") st, r = rest(f"/wp-json/wp/v2/posts/{pid}", "POST", {"meta": { "_yoast_wpseo_focuskw": a["fokus"], "_yoast_wpseo_title": a["seo"], "_yoast_wpseo_metadesc": meta_teks}}) st2, c = rest(f"/wp-json/wp/v2/posts/{pid}?context=edit&_fields=meta") m = c.get("meta", {}) if isinstance(c, dict) else {} if m.get("_yoast_wpseo_focuskw") == a["fokus"]: print(f" meta Yoast via REST : tersimpan (HTTP {st}) -> {json.dumps(m, ensure_ascii=False)[:220]}") else: print(f" meta Yoast : REST GAGAL -> {str(m)[:160]}") print("\n=== VERIFIKASI (baca ulang via REST) ===") st, c = rest(f"/wp-json/wp/v2/posts/{pid}?context=edit&_fields=id,slug,link,status,content,title," f"categories,tags,featured_media,meta,date,date_gmt") raw = c.get("content", {}).get("raw", "") if isinstance(c, dict) else "" kata2 = hitung(raw) sisa = re.findall(r"\{[A-Za-z_\[\]0-9]+\}", raw) cjk = re.findall(r"[\u3000-\u9fff\uac00-\ud7af]", raw) taut = len(re.findall(r'href="https://www\.tslaw\.id/', raw)) htm = re.findall(r'href="(https://www\.tslaw\.id/[^"]*\.htm)"', raw) print(f" #{pid} {c.get('status')} | {len(raw)} char | {kata2} kata") print(f" slug : {c.get('slug')} (link: {c.get('link')})") print(f" kategori : {c.get('categories')} (Hukum Bisnis = 14)") print(f" tag : {len(c.get('tags', []))} buah -> {c.get('tags')}") print(f" foto utama : media #{c.get('featured_media')} (harus {mid})") print(f" tombol WA : {'ada' if 'wa.me/6281909222026' in raw else 'TIDAK'}") print(f" FAQPage : {'ada' if 'FAQPage' in raw else 'TIDAK'}") beranda = 'href="https://www.tslaw.id/"' in raw print(f" beranda TS Law : {'ada' if beranda else 'TIDAK'}") print(f" tautan internal: {taut} tautan ke tslaw.id | .htm tersisa: {len(htm)}") print(f" keyphrase di isi : {raw.lower().count(a['fokus'])} kali") print(f" placeholder sisa : {len(sisa)} {'(BERSIH)' if not sisa else sisa[:5]}") print(f" karakter CJK : {len(cjk)} {'(BERSIH)' if not cjk else cjk[:5]}") print(f" em dash : {raw.count(chr(8212))}") patah = 0 for mm in re.finditer(r"<(p|li|h2|h3)(?: [^>]*)?>([^<]*)", raw): for b in mm.group(2).split("\n")[:-1]: if b.strip() and not b.rstrip().endswith((".", "!", "?", ":", ";", ",")): patah += 1 print(f" patahan baris : {patah} {'(BERSIH)' if not patah else 'PERIKSA'}") print(f" meta tersimpan: {json.dumps(c.get('meta', {}), ensure_ascii=False)[:280]}") print(f"\n LAPORAN: judul={a['judul']}") print(f" url draft={c.get('link')}") print(f" kata={kata2} | keyphrase={a['fokus']} | meta={len(meta_teks)} char") json.dump({"id": pid, "link": c.get("link"), "raw": raw, "meta": c.get("meta")}, open(f"{OUT}/_hasil19.json", "w"), ensure_ascii=False) if __name__ == "__main__": main()