#!/usr/bin/env python3
"""hc49: ambil isi 3 artikel terpilih dari WP REST studiosoft.id."""
import json, subprocess, sys

SLUGS = [
    "strategi-seo-terbaik-untuk-klinik-kecantikan-di-surabaya-tingkatkan-pasien-lewat-google",
    "7-strategi-digital-marketing-rumah-sakit-untuk-menambah-pasien-setiap-bulan",
    "jasa-digital-marketing-klinik",
]
for s in SLUGS:
    u = ("https://studiosoft.id/wp-json/wp/v2/posts?slug=%s&_fields=id,date,link,title,content,excerpt" % s)
    r = subprocess.run(["curl", "-4", "-s", "--http1.1", "--max-time", "150", "-A", "Mozilla/5.0", u],
                       capture_output=True, text=True)
    try:
        d = json.loads(r.stdout)
    except Exception as e:
        print("GAGAL", s, e, r.stdout[:200])
        continue
    if not d:
        print("KOSONG", s)
        continue
    p = d[0]
    print("=" * 100)
    print("SLUG:", s)
    print("ID:", p["id"], "| DATE:", p["date"])
    print("LINK:", p["link"])
    print("TITLE:", p["title"]["rendered"])
    print("-" * 60)
    txt = p["content"]["rendered"]
    import re, html
    txt = re.sub(r"<script.*?</script>", " ", txt, flags=re.S)
    txt = re.sub(r"<style.*?</style>", " ", txt, flags=re.S)
    txt = re.sub(r"<h([1-6])[^>]*>", "\n\n## ", txt)
    txt = re.sub(r"</h[1-6]>", "\n", txt)
    txt = re.sub(r"<li[^>]*>", "\n- ", txt)
    txt = re.sub(r"<p[^>]*>", "\n", txt)
    txt = re.sub(r"<[^>]+>", " ", txt)
    txt = html.unescape(txt)
    txt = re.sub(r"[ \t]+", " ", txt)
    txt = re.sub(r"\n\s*\n+", "\n", txt)
    print(txt.strip()[:5000])