laporan-generator 2.6.1

package/scripts/report-doctor.py

@@ -0,0 +1,211 @@
+#!/usr/bin/env python3
+"""report-doctor.py: Audit komprehensif kesehatan dan integritas proyek laporan."""
+
+import glob
+import os
+import re
+import shutil
+import sys
+
+GREEN = "\033[0;32m"
+BLUE = "\033[0;34m"
+CYAN = "\033[0;36m"
+YELLOW = "\033[1;33m"
+RED = "\033[0;31m"
+BOLD = "\033[1m"
+NC = "\033[0m"
+
+
+def check_dependencies():
+    tools = [
+        ("pandoc", "Pandoc Document Engine", True),
+        ("typst", "Typst Typesetting Engine", True),
+        ("python3", "Python 3 Runtime", True),
+        ("magick", "ImageMagick 7", False),
+        ("convert", "ImageMagick 6", False),
+        ("pdftotext", "Poppler PDF Utils", False),
+        ("soffice", "LibreOffice", False),
+    ]
+
+    results = []
+    has_im = False
+    for cmd, name, req in tools:
+        found = shutil.which(cmd) is not None
+        if "ImageMagick" in name and found:
+            has_im = True
+        results.append((cmd, name, req, found))
+
+    return results, has_im
+
+
+def check_citations(bib_path, markdown_files):
+    bib_keys = set()
+    if os.path.isfile(bib_path):
+        with open(bib_path, "r", encoding="utf-8", errors="ignore") as f:
+            bib_keys = set(re.findall(r"@\w+\s*\{\s*([a-zA-Z0-9_:-]+)\s*,", f.read()))
+
+    cited_keys = {}
+    for mf in markdown_files:
+        with open(mf, "r", encoding="utf-8", errors="ignore") as f:
+            for line_no, line in enumerate(f, 1):
+                # Abaikan contoh dalam inline code
+                clean_line = re.sub(r"`.*?`", "", line)
+                matches = re.findall(r"@([a-zA-Z0-9_:-]+)", clean_line)
+                for m in matches:
+                    if m not in cited_keys:
+                        cited_keys[m] = []
+                    cited_keys[m].append((os.path.basename(mf), line_no))
+
+    missing_keys = {}
+    for key, locs in cited_keys.items():
+        if key not in bib_keys:
+            missing_keys[key] = locs
+
+    orphan_keys = bib_keys - set(cited_keys.keys())
+
+    return bib_keys, cited_keys, missing_keys, orphan_keys
+
+
+def check_images(markdown_files):
+    broken_images = []
+    total_images = 0
+    for mf in markdown_files:
+        with open(mf, "r", encoding="utf-8", errors="ignore") as f:
+            for line_no, line in enumerate(f, 1):
+                clean_line = re.sub(r"`.*?`", "", line)
+                matches = re.findall(r"!\[.*?\]\((.*?)\)", clean_line)
+                for img_path in matches:
+                    total_images += 1
+                    clean_path = img_path.split()[0].strip("<>\"'")
+                    if not os.path.isfile(clean_path):
+                        broken_images.append((clean_path, os.path.basename(mf), line_no))
+    return total_images, broken_images
+
+
+def check_style_anomalies(markdown_files):
+    anomalies = []
+    box_draw_regex = re.compile(r"[┌─┐│└┘├┤┬┴┼═║╔╗╚╝╠╣╦╩╬]")
+    manual_num_regex = re.compile(r"^#{2,4}\s+[0-9]+(\.[0-9]+)+")
+
+    for mf in markdown_files:
+        with open(mf, "r", encoding="utf-8", errors="ignore") as f:
+            for line_no, line in enumerate(f, 1):
+                if box_draw_regex.search(line):
+                    anomalies.append((os.path.basename(mf), line_no, "Karakter box-drawing terdeteksi (gunakan tabel/ASCII)"))
+                if manual_num_regex.search(line):
+                    anomalies.append((os.path.basename(mf), line_no, f"Manual numbering di heading: '{line.strip()}'"))
+    return anomalies
+
+
+def main():
+    root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+    os.chdir(root)
+
+    print(f"{CYAN}{BOLD}")
+    print("  ========================================================")
+    print("             LAPORAN GENERATOR - DOCTOR & AUDIT SUITE     ")
+    print("  ========================================================")
+    print(f"{NC}")
+
+    errors = 0
+    warnings = 0
+
+    # 1. Dependensi
+    print(f"  {BLUE}{BOLD}[1/4] Audit Toolchain & Dependensi Sistem{NC}")
+    deps, has_im = check_dependencies()
+    for cmd, name, req, found in deps:
+        if "ImageMagick" in name:
+            continue
+        if found:
+            print(f"    {GREEN}[OK]{NC} {name:<26}: Terpasang ({cmd})")
+        else:
+            if req:
+                print(f"    {RED}[FAIL]{NC} {name:<24}: TIDAK DITEMUKAN (Wajib)")
+                errors += 1
+            else:
+                print(f"    {YELLOW}[INFO]{NC} {name:<24}: Tidak ditemukan (Opsional)")
+
+    if has_im:
+        print(f"    {GREEN}[OK]{NC} {'ImageMagick':<26}: Terpasang")
+    else:
+        print(f"    {YELLOW}[INFO]{NC} {'ImageMagick':<24}: Tidak ditemukan (Opsional untuk kompresi gambar)")
+    print("")
+
+    # 2. Berkas & Struktur
+    print(f"  {BLUE}{BOLD}[2/4] Audit Struktur Direktori & Berkas Wajib{NC}")
+    req_files = [
+        "metadata.yml", "template.typ", "reference.docx",
+        "references.bib", "docx.lua", "cover.md"
+    ]
+    for rf in req_files:
+        if os.path.isfile(rf):
+            print(f"    {GREEN}[OK]{NC} Berkas {rf}")
+        else:
+            print(f"    {RED}[FAIL]{NC} Berkas {rf} TIDAK DITEMUKAN")
+            errors += 1
+
+    chapter_files = sorted(glob.glob("chapters/*.md"))
+    if chapter_files:
+        print(f"    {GREEN}[OK]{NC} Direktori chapters/ ({len(chapter_files)} bab ditemukan)")
+    else:
+        print(f"    {RED}[FAIL]{NC} Tidak ada berkas markdown di chapters/")
+        errors += 1
+    print("")
+
+    # 3. Sitasi & Referensi
+    print(f"  {BLUE}{BOLD}[3/4] Audit Sitasi & Integritas Bibliografi{NC}")
+    bib_keys, cited_keys, missing_keys, orphan_keys = check_citations("references.bib", chapter_files)
+    print(f"    * Total entri .bib     : {len(bib_keys)}")
+    print(f"    * Total sitasi di teks : {len(cited_keys)}")
+
+    if missing_keys:
+        for k, locs in missing_keys.items():
+            loc_str = ", ".join([f"{f}:{l}" for f, l in locs])
+            print(f"    {RED}[ERROR]{NC} Sitasi '@{k}' TIDAK DITEMUKAN di references.bib ({loc_str})")
+            errors += 1
+    else:
+        print(f"    {GREEN}[OK]{NC} Semua sitasi dalam teks valid terdaftar di references.bib")
+
+    if orphan_keys:
+        print(f"    {YELLOW}[WARN]{NC} {len(orphan_keys)} entri di references.bib belum disitir dalam teks:")
+        for ok in list(orphan_keys)[:5]:
+            print(f"      - @{ok}")
+        if len(orphan_keys) > 5:
+            print(f"      - ... dan {len(orphan_keys) - 5} lainnya")
+        warnings += 1
+    print("")
+
+    # 4. Gambar & Format Teks
+    print(f"  {BLUE}{BOLD}[4/4] Audit Media Citra & Anomali Tipografi{NC}")
+    total_imgs, broken_imgs = check_images(chapter_files)
+    if broken_imgs:
+        for p, f, l in broken_imgs:
+            print(f"    {RED}[ERROR]{NC} Berkas gambar tidak ditemukan: '{p}' (di {f}:{l})")
+            errors += 1
+    else:
+        print(f"    {GREEN}[OK]{NC} Semua tautan gambar ({total_imgs} gambar) valid dan ada di disk")
+
+    anomalies = check_style_anomalies(chapter_files)
+    if anomalies:
+        for f, l, msg in anomalies:
+            print(f"    {YELLOW}[WARN]{NC} {f}:{l} - {msg}")
+            warnings += 1
+    else:
+        print(f"    {GREEN}[OK]{NC} Bebas dari manual heading numbering dan box-drawing characters")
+    print("")
+
+    # Skor Kesehatan
+    print("  ========================================================")
+    if errors == 0 and warnings == 0:
+        print(f"  {GREEN}{BOLD}STATUS: SEHAT SEMPURNA (100/100) - SIAP DIPUBLIKASIKAN!{NC}")
+    elif errors == 0:
+        print(f"  {YELLOW}{BOLD}STATUS: SEHAT DENGAN CATATAN (95/100) - {warnings} Peringatan Terdeteksi{NC}")
+    else:
+        print(f"  {RED}{BOLD}STATUS: DITEMUKAN MASALAH - {errors} Error, {warnings} Warning{NC}")
+    print("  ========================================================")
+
+    return 1 if errors > 0 else 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

package/scripts/report-stats.py

@@ -0,0 +1,149 @@
+#!/usr/bin/env python3
+"""report-stats.py: Menghitung statistik komprehensif dokumen laporan akademik."""
+
+import glob
+import os
+import re
+import sys
+
+GREEN = "\033[0;32m"
+BLUE = "\033[0;34m"
+CYAN = "\033[0;36m"
+YELLOW = "\033[1;33m"
+BOLD = "\033[1m"
+NC = "\033[0m"
+
+
+def count_file_stats(filepath):
+    with open(filepath, "r", encoding="utf-8", errors="ignore") as f:
+        content = f.read()
+
+    clean_text = re.sub(r"```.*?```", "", content, flags=re.S)
+    clean_text = re.sub(r"<!--.*?-->", "", clean_text, flags=re.S)
+    clean_text = re.sub(r"[#*_`~>|]", " ", clean_text)
+    words = clean_text.split()
+
+    h1 = len(re.findall(r"^#[^#].*$", content, re.M))
+    h2 = len(re.findall(r"^##[^#].*$", content, re.M))
+    h3 = len(re.findall(r"^###[^#].*$", content, re.M))
+
+    images = len(re.findall(r"!\[.*?\]\(.*?\)", content))
+    tables = len(re.findall(r"^\|.*\|[ \t]*$", content, re.M))
+    equations = len(re.findall(r"\$\$.*?\$\$", content, re.S)) + len(
+        re.findall(r"(?<!\$)\$(?!\$).+?(?<!\$)\$(?!\$)", content)
+    )
+    citations = re.findall(r"@([a-zA-Z0-9_:-]+)", content)
+
+    return {
+        "words": len(words),
+        "chars": len(content),
+        "h1": h1,
+        "h2": h2,
+        "h3": h3,
+        "images": images,
+        "tables": max(0, tables // 3),
+        "equations": equations,
+        "citations": citations,
+    }
+
+
+def main():
+    root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+    os.chdir(root)
+
+    title = "Laporan Akademik"
+    preset = "standard"
+
+    if os.path.isfile("metadata.yml"):
+        with open("metadata.yml", "r", encoding="utf-8") as f:
+            for line in f:
+                if line.startswith("title:"):
+                    title = line.split(":", 1)[1].strip().strip('"\'')
+                elif line.startswith("preset:"):
+                    preset = line.split(":", 1)[1].strip().strip('"\'')
+
+    chapter_files = sorted(glob.glob("chapters/*.md"))
+    all_files = (["cover.md"] if os.path.isfile("cover.md") else []) + chapter_files
+
+    total_words = 0
+    total_chars = 0
+    total_h1 = 0
+    total_h2 = 0
+    total_h3 = 0
+    total_images = 0
+    total_tables = 0
+    total_equations = 0
+    all_citations = set()
+
+    file_breakdown = []
+
+    for f in all_files:
+        if os.path.isfile(f):
+            st = count_file_stats(f)
+            total_words += st["words"]
+            total_chars += st["chars"]
+            total_h1 += st["h1"]
+            total_h2 += st["h2"]
+            total_h3 += st["h3"]
+            total_images += st["images"]
+            total_tables += st["tables"]
+            total_equations += st["equations"]
+            all_citations.update(st["citations"])
+            file_breakdown.append((os.path.basename(f), st["words"]))
+
+    bib_count = 0
+    if os.path.isfile("references.bib"):
+        with open("references.bib", "r", encoding="utf-8") as f:
+            bib_count = len(re.findall(r"@\w+\s*\{", f.read()))
+
+    est_pages = max(1, round(total_words / 280))
+    read_mins = max(1, round(total_words / 200))
+
+    pdf_size = "-"
+    if os.path.isfile("Laporan.pdf"):
+        sz = os.path.getsize("Laporan.pdf")
+        pdf_size = f"{sz / 1024:.1f} KB"
+
+    docx_size = "-"
+    if os.path.isfile("Laporan.docx"):
+        sz = os.path.getsize("Laporan.docx")
+        docx_size = f"{sz / 1024:.1f} KB"
+
+    print(f"{CYAN}{BOLD}")
+    print("  ========================================================")
+    print("                 LAPORAN GENERATOR - STATISTIK DOKUMEN    ")
+    print("  ========================================================")
+    print(f"{NC}")
+    print(f"  {BOLD}Judul Laporan    :{NC} {title}")
+    print(f"  {BOLD}Preset Aktif     :{NC} {preset}")
+    print(f"  {BOLD}Total Berkas Bab :{NC} {len(chapter_files)} berkas")
+    print("")
+    print(f"  {BLUE}{BOLD}[Metrik Konten]{NC}")
+    print(f"  * Total Kata            : {GREEN}{BOLD}{total_words:,}{NC} kata")
+    print(f"  * Total Karakter        : {total_chars:,} karakter")
+    print(f"  * Estimasi Halaman      : {CYAN}{BOLD}~{est_pages}{NC} halaman (standar A4 1.5 spasi)")
+    print(f"  * Estimasi Durasi Baca  : ~{read_mins} menit")
+    print("")
+    print(f"  {BLUE}{BOLD}[Struktur & Elemen Akademik]{NC}")
+    print(f"  * Bab Utama (Heading 1) : {total_h1}")
+    print(f"  * Sub-Bab (Heading 2)   : {total_h2}")
+    print(f"  * Sub-Sub-Bab (Heading 3): {total_h3}")
+    print(f"  * Gambar / Diagram      : {total_images}")
+    print(f"  * Tabel Terdeteksi      : {total_tables}")
+    print(f"  * Persamaan Matematika  : {total_equations}")
+    print(f"  * Sitasi Digunakan      : {len(all_citations)} sitasi unik")
+    print(f"  * Entri Daftar Pustaka  : {bib_count} referensi di .bib")
+    print("")
+    print(f"  {BLUE}{BOLD}[Ukuran Berkas Terkompilasi]{NC}")
+    print(f"  * Laporan.pdf           : {pdf_size}")
+    print(f"  * Laporan.docx          : {docx_size}")
+    print("")
+    print(f"  {BLUE}{BOLD}[Rincian Kata per Berkas]{NC}")
+    for fname, cnt in file_breakdown:
+        bar = "■" * max(1, min(25, cnt // 100))
+        print(f"  {fname:<28} : {cnt:>5} kata {CYAN}{bar}{NC}")
+    print("  ========================================================")
+
+
+if __name__ == "__main__":
+    main()

package/scripts/scan-preset.py

@@ -0,0 +1,311 @@
+#!/usr/bin/env python3
+"""scan-preset.py: Ekstraktor otomatis aturan format & preset kampus dari berkas PDF.
+
+Mengekstrak aturan tata letak (margin, font, spasi baris, penomoran heading,
+dan nama institusi) dari dokumen PDF Buku Pedoman Penulisan Skripsi / Tugas Akhir.
+
+Usage:
+  python3 scripts/scan-preset.py <input.pdf> [options]
+
+Options:
+  --preset-id <id>      ID unik preset (default: slug nama kampus/file)
+  --name <name>         Nama lengkap preset/institusi
+  --output-dir <dir>    Direktori output (default: presets/)
+  --non-interactive     Jalankan tanpa konfirmasi interaktif
+  --apply               Otomatis terapkan ke metadata.yml setelah scan
+"""
+
+import argparse
+import os
+import re
+import subprocess
+import sys
+
+
+def extract_text_from_pdf(pdf_path):
+    """Ekstraksi teks dari PDF menggunakan pdftotext (poppler-utils) atau fallback."""
+    if not os.path.exists(pdf_path):
+        print(f"[ERROR] Berkas '{pdf_path}' tidak ditemukan.", file=sys.stderr)
+        sys.exit(1)
+
+    # Jika file teks/mock markdown diberikan langsung
+    if pdf_path.endswith(".txt") or pdf_path.endswith(".md"):
+        with open(pdf_path, "r", encoding="utf-8", errors="ignore") as f:
+            return f.read()
+
+    # Cek pdftotext
+    if subprocess.run(["which", "pdftotext"], capture_output=True).returncode == 0:
+        try:
+            # Ambil maksimal 40 halaman pertama (bab pedoman format biasanya di bagian awal)
+            res = subprocess.run(
+                ["pdftotext", "-f", "1", "-l", "40", "-layout", pdf_path, "-"],
+                capture_output=True,
+                text=True,
+                check=True,
+            )
+            return res.stdout
+        except Exception as e:
+            print(f"[WARN] pdftotext gagal: {e}", file=sys.stderr)
+
+    # Fallback jika file binary PDF tidak bisa diurai via pdftotext
+    print("[WARN] pdftotext tidak tersedia, membaca teks mentah...", file=sys.stderr)
+    try:
+        with open(pdf_path, "rb") as f:
+            raw = f.read().decode("latin1", errors="ignore")
+            # Ekstraksi string teks dalam kurung PDF (Tj / TJ)
+            strings = re.findall(r"\(([^\)]+)\)\s*Tj", raw)
+            return "\n".join(strings) if strings else raw
+    except Exception as e:
+        print(f"[ERROR] Gagal membaca berkas: {e}", file=sys.stderr)
+        sys.exit(1)
+
+
+def detect_institution(text, filename):
+    """Deteksi nama universitas / institut / politeknik."""
+    patterns = [
+        r"(UNIVERSITAS\s+[A-Z0-9\s]+)",
+        r"(INSTITUT\s+[A-Z0-9\s]+)",
+        r"(POLITEKNIK\s+[A-Z0-9\s]+)",
+        r"(SEKOLAH\s+TINGGI\s+[A-Z0-9\s]+)",
+        r"(Universitas\s+[A-Za-z0-9\s]+)",
+        r"(Institut\s+[A-Za-z0-9\s]+)",
+        r"(Politeknik\s+[A-Za-z0-9\s]+)",
+    ]
+    for pat in patterns:
+        m = re.search(pat, text)
+        if m:
+            found = m.group(1).strip()
+            # Bersihkan baris berlebih
+            found = found.split("\n")[0].strip()
+            if len(found) > 8 and len(found) < 60:
+                return found
+
+    # Fallback dari nama file (contoh: pedoman-its.pdf -> ITS)
+    base = os.path.basename(pdf_slug(filename)).replace("-", " ").title()
+    return base
+
+
+def pdf_slug(path):
+    """Buat slug ID dari nama berkas atau nama institusi."""
+    base = os.path.basename(path).rsplit(".", 1)[0]
+    base = re.sub(r"^(pedoman|buku|panduan|skripsi|ta|format)[-_]?", "", base, flags=re.I)
+    base = re.sub(r"[^a-zA-Z0-9]+", "-", base).strip("-").lower()
+    return base if base else "custom-kampus"
+
+
+def detect_margins(text):
+    """Deteksi batas margin (top, bottom, left, right)."""
+    # Pola shorthand 4-4-3-3 atau 4 4 3 3 (kiri-atas-kanan-bawah atau atas-kiri-bawah-kanan)
+    if re.search(r"4\s*[-–,]\s*4\s*[-–,]\s*3\s*[-–,]\s*3", text):
+        return {"top": "4cm", "bottom": "3cm", "left": "4cm", "right": "3cm"}
+    if re.search(r"3\s*[-–,]\s*3\s*[-–,]\s*3\s*[-–,]\s*3", text):
+        return {"top": "3cm", "bottom": "3cm", "left": "3cm", "right": "3cm"}
+
+    margins = {"top": "2.5cm", "bottom": "2.5cm", "left": "2.5cm", "right": "2.5cm"}
+
+    # Pola spesifik bahasa Indonesia
+    m_left = re.search(r"(?:kiri|left)\s*[:=]?\s*([0-9]+(?:[.,][0-9]+)?)\s*cm", text, re.I)
+    m_top = re.search(r"(?:atas|top)\s*[:=]?\s*([0-9]+(?:[.,][0-9]+)?)\s*cm", text, re.I)
+    m_right = re.search(r"(?:kanan|right)\s*[:=]?\s*([0-9]+(?:[.,][0-9]+)?)\s*cm", text, re.I)
+    m_bottom = re.search(r"(?:bawah|bottom)\s*[:=]?\s*([0-9]+(?:[.,][0-9]+)?)\s*cm", text, re.I)
+
+    if m_left:
+        margins["left"] = f"{m_left.group(1).replace(',', '.')}cm"
+    if m_top:
+        margins["top"] = f"{m_top.group(1).replace(',', '.')}cm"
+    if m_right:
+        margins["right"] = f"{m_right.group(1).replace(',', '.')}cm"
+    if m_bottom:
+        margins["bottom"] = f"{m_bottom.group(1).replace(',', '.')}cm"
+
+    # Deteksi format skripsi jilid umum jika terdeteksi teks jilid
+    if re.search(r"(jilid|hardcover|penjilidan)", text, re.I) and not m_left:
+        margins["left"] = "4cm"
+        margins["top"] = "4cm"
+        margins["right"] = "3cm"
+        margins["bottom"] = "3cm"
+
+    return margins
+
+
+def detect_typography(text):
+    """Deteksi font family, size, line spacing, dan indent."""
+    typo = {
+        "font_family": "Libertinus Serif",
+        "font_size": "12pt",
+        "line_spacing": "0.75em",  # 0.75em = 1.5 spasi di Typst
+        "first_line_indent": "1.25cm",
+    }
+
+    # Deteksi Font
+    if re.search(r"Arial", text, re.I):
+        typo["font_family"] = "Arial"
+    elif re.search(r"Calibri", text, re.I):
+        typo["font_family"] = "Calibri"
+    elif re.search(r"Times\s*New\s*Roman", text, re.I):
+        typo["font_family"] = "Libertinus Serif"
+
+    # Deteksi Ukuran Font
+    m_sz = re.search(r"(?:ukuran|font\s*size|size)\s*[:=]?\s*([0-9]+)\s*(?:pt|point)?", text, re.I)
+    if m_sz:
+        typo["font_size"] = f"{m_sz.group(1)}pt"
+
+    # Deteksi Spasi Baris
+    if re.search(r"(?:spasi|line\s*spacing)\s*[:=]?\s*(?:ganda|2(?:\.0)?|double)", text, re.I):
+        typo["line_spacing"] = "1.0em"  # double spacing
+    elif re.search(r"(?:spasi|line\s*spacing)\s*[:=]?\s*1[,.]15", text, re.I):
+        typo["line_spacing"] = "0.55em"
+    elif re.search(r"(?:spasi|line\s*spacing)\s*[:=]?\s*(?:1[,.]5|satu\s*setengah)", text, re.I):
+        typo["line_spacing"] = "0.75em"
+
+    # Deteksi Indentasi
+    m_ind = re.search(r"(?:indentasi|alinea|awal\s*paragraf)\s*[:=]?\s*([0-9]+(?:[.,][0-9]+)?)\s*cm", text, re.I)
+    if m_ind:
+        typo["first_line_indent"] = f"{m_ind.group(1).replace(',', '.')}cm"
+
+    return typo
+
+
+def detect_headings(text):
+    """Deteksi format penomoran bab dan sub-bab."""
+    headings = {
+        "chapter_prefix": "BAB ",
+        "chapter_num_format": "roman",
+        "sub_dot": True,
+        "subsub_dot": False,
+    }
+
+    # Cek BAB 1 vs BAB I
+    if re.search(r"BAB\s+[0-9]+\b", text) and not re.search(r"BAB\s+[IVXLCDM]+\b", text):
+        headings["chapter_num_format"] = "arabic"
+
+    # Cek sub-bab trailing dot (1.1. vs 1.1)
+    if re.search(r"\b1\.1\s+[A-Z]", text):
+        headings["sub_dot"] = False
+    elif re.search(r"\b1\.1\.\s+[A-Z]", text):
+        headings["sub_dot"] = True
+
+    return headings
+
+
+def detect_cover(text):
+    """Deteksi opsi sampul depan (apakah dosen pembimbing dicantumkan)."""
+    cover = {
+        "show_lecturer": True,
+        "logo_width": "4cm",
+    }
+    # Jika ada klausul eksplisit bahwa sampul luar tidak mencantumkan pembimbing (seperti ITB)
+    if re.search(r"sampul\s*(?:luar|depan)?\s*tanpa\s*(?:nama)?\s*(?:dosen|pembimbing)", text, re.I):
+        cover["show_lecturer"] = False
+    return cover
+
+
+def build_preset_yaml(preset_id, name, margins, typo, headings, cover):
+    """Hasilkan string YAML preset standar."""
+    hide_lecturer_yaml = ""
+    if not cover["show_lecturer"]:
+        hide_lecturer_yaml = "\ncover_hide_lecturer: true"
+
+    return f"""# Preset: {name}
+# Dihasilkan secara otomatis oleh scan-preset.py
+preset_id: "{preset_id}"
+name: "{name}"
+description: "Format otomatis hasil pemindaian pedoman akademik {name}."
+
+# 1. Geometri & Margin Halaman
+margin_top: {margins['top']}
+margin_bottom: {margins['bottom']}
+margin_left: {margins['left']}
+margin_right: {margins['right']}
+
+# 2. Tipografi & Spasi
+font_family: "{typo['font_family']}"
+font_size: {typo['font_size']}
+line_spacing: {typo['line_spacing']}
+first_line_indent: {typo['first_line_indent']}
+
+# 3. Penomoran Heading
+heading_chapter_prefix: "{headings['chapter_prefix']}"
+heading_chapter_num_format: "{headings['chapter_num_format']}"
+heading_sub_dot: {str(headings['sub_dot']).lower()}
+heading_subsub_dot: {str(headings['subsub_dot']).lower()}
+
+# 4. Pengaturan Halaman Sampul
+cover_show_lecturer: {str(cover['show_lecturer']).lower()}{hide_lecturer_yaml}
+cover_logo_width: {cover['logo_width']}
+"""
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Scan pedoman penulisan PDF dan buat preset kampus otomatis.")
+    parser.add_argument("pdf_path", help="Path ke berkas PDF pedoman penulisan kampus")
+    parser.add_argument("--preset-id", help="ID unik preset (contoh: unpad-skripsi)")
+    parser.add_argument("--name", help="Nama lengkap institusi / preset")
+    parser.add_argument("--output-dir", default="presets", help="Direktori output penyimpanan preset")
+    parser.add_argument("--non-interactive", action="store_true", help="Jalankan otomatis tanpa prompt interaktif")
+    parser.add_argument("--apply", action="store_true", help="Terapkan preset langsung ke metadata.yml")
+
+    args = parser.parse_args()
+
+    print(f"\n[1/3] Membaca dan menganalisis berkas: {args.pdf_path}...")
+    text = extract_text_from_pdf(args.pdf_path)
+    print(f"      Total teks terdeteksi: {len(text.split())} kata")
+
+    print("[2/3] Mengekstraksi aturan penulisan akademik...")
+    inst_name = args.name or detect_institution(text, args.pdf_path)
+    pid = args.preset_id or pdf_slug(inst_name or args.pdf_path)
+    margins = detect_margins(text)
+    typo = detect_typography(text)
+    headings = detect_headings(text)
+    cover = detect_cover(text)
+
+    print("\n" + "=" * 56)
+    print("        HASIL DETEKSI ATURAN FORMAT KAMPUS")
+    print("=" * 56)
+    print(f"  * ID Preset       : {pid}")
+    print(f"  * Nama Institusi  : {inst_name}")
+    print(f"  * Margin Halaman  : Atas {margins['top']}, Bawah {margins['bottom']}, Kiri {margins['left']}, Kanan {margins['right']}")
+    print(f"  * Tipografi       : {typo['font_family']} {typo['font_size']} (Spasi: {typo['line_spacing']}, Indent: {typo['first_line_indent']})")
+    print(f"  * Penomoran Bab   : {headings['chapter_prefix']}{'I' if headings['chapter_num_format'] == 'roman' else '1'} (Sub-bab dot: {headings['sub_dot']})")
+    print(f"  * Sampul Depan    : Dosen={'Ya' if cover['show_lecturer'] else 'Tidak'}, Lebar Logo={cover['logo_width']}")
+    print("=" * 56 + "\n")
+
+    if not args.non_interactive:
+        confirm = input("Simpan konfigurasi preset di atas? [Y/n]: ").strip().lower()
+        if confirm in ("n", "no"):
+            print("[INFO] Pembuatan preset dibatalkan.")
+            sys.exit(0)
+
+    os.makedirs(args.output_dir, exist_ok=True)
+    out_file = os.path.join(args.output_dir, f"{pid}.yml")
+    yaml_content = build_preset_yaml(pid, inst_name, margins, typo, headings, cover)
+
+    with open(out_file, "w", encoding="utf-8") as f:
+        f.write(yaml_content)
+
+    print(f"[3/3] [OK] Preset berhasil disimpan ke: {out_file}")
+
+    if args.apply:
+        meta_file = "metadata.yml"
+        if os.path.exists(meta_file):
+            with open(meta_file, "r", encoding="utf-8") as f:
+                content = f.read()
+            if re.search(r"^\s*preset:", content, re.M):
+                content = re.sub(r"^\s*preset:.*", f'preset: "{pid}"', content, flags=re.M)
+            elif re.search(r"^\s*margin_preset:", content, re.M):
+                content = re.sub(r"^\s*margin_preset:.*", f'preset: "{pid}"', content, flags=re.M)
+            elif "..." in content:
+                content = content.replace("...", f'preset: "{pid}"\n...')
+            else:
+                content += f'\npreset: "{pid}"\n'
+            with open(meta_file, "w", encoding="utf-8") as f:
+                f.write(content)
+            print(f"[OK] Preset '{pid}' telah diterapkan ke metadata.yml!")
+
+    print(f"\nUntuk menggunakan preset ini, jalankan:")
+    print(f"  ./laporan preset apply {pid}")
+    print(f"  ./laporan build\n")
+
+
+if __name__ == "__main__":
+    main()