laporan-generator 2.6.1

package/.agents/skills/laporan-generator/scripts/docx-pagenum.py

@@ -0,0 +1,192 @@
+#!/usr/bin/env python3
+"""Isi nilai halaman asli ke field PAGEREF di daftar isi DOCX.
+
+Dua-pass:
+  1. Render DOCX (yang sudah di-finalize section-nya) via LibreOffice
+     (soffice --headless) menjadi PDF.
+  2. Baca nomor halaman setiap heading dari teks PDF (footer per halaman),
+     lalu tulis sebagai nilai cache PAGEREF di daftar isi.
+
+Hasil: daftar isi menampilkan nomor halaman asli di aplikasi manapun
+(Word, LibreOffice, Google Docs) tanpa perlu update field manual.
+
+Kalau soffice/pdftotext tidak tersedia, output = input (Word tetap
+memperbarui field saat dibuka karena updateFields=true).
+
+Usage:
+  python3 scripts/docx-pagenum.py input.docx output.docx
+"""
+
+import os
+import re
+import shutil
+import subprocess
+import sys
+import tempfile
+import zipfile
+
+ROMAN = re.compile(r"^[ivxlcdm]+$", re.I)
+DECIMAL = re.compile(r"^\d+$")
+
+
+def page_count(pdf):
+    out = subprocess.check_output(["pdfinfo", pdf], text=True, timeout=15)
+    m = re.search(r"^Pages:\s+(\d+)", out, re.M)
+    return int(m.group(1)) if m else 0
+
+
+def page_texts(pdf):
+    n = page_count(pdf)
+    pages = {}
+    for p in range(1, n + 1):
+        out = subprocess.check_output(
+            ["pdftotext", "-f", str(p), "-l", str(p), "-layout", pdf, "-"],
+            text=True,
+            timeout=15,
+        )
+        lines = [l.strip() for l in out.splitlines() if l.strip()]
+        pages[p] = lines
+    return pages
+
+
+def footer_of(lines):
+    if not lines:
+        return ""
+    return lines[-1]
+
+
+def find_toc_start(pages):
+    for p, lines in sorted(pages.items()):
+        for l in lines:
+            if l == "DAFTAR ISI":
+                return p
+    return None
+
+
+def first_decimal_page(pages, start):
+    for p, lines in sorted(pages.items()):
+        if p <= start:
+            continue
+        if DECIMAL.match(footer_of(lines)):
+            return p
+    return None
+
+
+def find_page(pages, needle, start, end):
+    for p in range(start, end + 1):
+        text = "\n".join(pages.get(p, []))
+        if needle in text:
+            return p
+    return None
+
+
+def parse_entries(doc):
+    entries = []
+    par_pat = re.compile(r"<w:p>(?:(?!</w:p>).)*?</w:p>", re.S)
+    for pm in par_pat.finditer(doc):
+        p = pm.group(0)
+        m = re.search(
+            r'<w:hyperlink w:anchor="([^"]+)"[^>]*>.*?<w:t[^>]*>([^<]*)</w:t>', p, re.S
+        )
+        r = re.search(
+            r"<w:instrText[^>]*>\s*PAGEREF\s+(\S+)\s+\\h\s*</w:instrText>", p
+        )
+        if m and r and m.group(1) == r.group(1):
+            entries.append((p, m.group(1), m.group(2)))
+    return entries
+
+
+def needle_of(entry_text):
+    if entry_text.startswith("BAB "):
+        parts = entry_text.split(" ", 2)
+        return parts[2] if len(parts) == 3 else parts[1]
+    return entry_text
+
+
+def render_pdf(docx):
+    tmp = tempfile.mkdtemp(prefix="lo_pagenum_")
+    profile = os.path.join(tmp, "profile")
+    subprocess.run(
+        [
+            "soffice", "--headless", "--norestore",
+            "-env:UserInstallation=file://" + profile,
+            "--convert-to", "pdf", "--outdir", tmp, docx,
+        ],
+        check=True,
+        capture_output=True,
+        timeout=45,
+    )
+    base = os.path.basename(docx).rsplit(".", 1)[0] + ".pdf"
+    return os.path.join(tmp, base)
+
+
+def main():
+    src, dst = sys.argv[1], sys.argv[2]
+    shutil.copy(src, dst)
+
+    if not shutil.which("soffice") or not shutil.which("pdftotext"):
+        print("[warn] docx-pagenum: soffice/pdftotext tidak ada, nomor halaman "
+              "daftar isi memakai nilai cache (Word tetap update saat dibuka)")
+        return
+
+    with zipfile.ZipFile(src, "r") as zin:
+        names = zin.namelist()
+        items = {n: zin.read(n) for n in names}
+    doc = items["word/document.xml"].decode("utf-8")
+
+    try:
+        pdf = render_pdf(src)
+    except (subprocess.TimeoutExpired, subprocess.SubprocessError) as err:
+        print(f"[warn] docx-pagenum: render LibreOffice gagal/timeout ({err}), nomor halaman memakai nilai cache default.")
+        return
+
+    try:
+        pages = page_texts(pdf)
+    except (subprocess.TimeoutExpired, subprocess.SubprocessError) as err:
+        print(f"[warn] docx-pagenum: ekstraksi teks PDF gagal/timeout ({err}), nomor halaman memakai nilai cache default.")
+        return
+    finally:
+        shutil.rmtree(os.path.dirname(pdf), ignore_errors=True)
+
+    toc_start = find_toc_start(pages)
+    if not toc_start:
+        print("[warn] docx-pagenum: halaman DAFTAR ISI tidak ditemukan di render")
+        return
+    first_dec = first_decimal_page(pages, toc_start)
+    if not first_dec:
+        print("[warn] docx-pagenum: halaman body (decimal) tidak ditemukan")
+        return
+
+    entries = parse_entries(doc)
+    injected = 0
+    for para, pid, text in entries:
+        if pid == "kata-pengantar":
+            page = find_page(pages, "KATA PENGANTAR", 1, toc_start - 1)
+        else:
+            page = find_page(pages, needle_of(text), first_dec, max(pages))
+        if not page:
+            print(f"[warn] docx-pagenum: heading '{text}' tidak ditemukan di render")
+            continue
+        value = footer_of(pages[page])
+        if not (ROMAN.match(value) or DECIMAL.match(value)):
+            print(f"[warn] docx-pagenum: footer halaman {page} tidak valid: {value!r}")
+            continue
+        new_para = re.sub(
+            r'(<w:fldChar w:fldCharType="separate"/></w:r><w:r><w:t[^>]*>)[^<]*(</w:t>)',
+            r"\g<1>" + value + r"\g<2>",
+            para,
+            count=1,
+        )
+        doc = doc.replace(para, new_para)
+        injected += 1
+        print(f"[OK] {text} -> halaman {page} (nomor {value})")
+
+    items["word/document.xml"] = doc.encode("utf-8")
+    with zipfile.ZipFile(dst, "w", zipfile.ZIP_DEFLATED) as zout:
+        for n, content in items.items():
+            zout.writestr(n, content)
+    print(f"[OK] docx-pagenum: {injected} nomor halaman diisi -> {dst}")
+
+
+if __name__ == "__main__":
+    main()

package/.agents/skills/laporan-generator/scripts/finalize-docx.py

@@ -0,0 +1,152 @@
+#!/usr/bin/env python3
+"""Finalisasi document.xml hasil pandoc agar penomoran halaman DOCX
+mengikuti PDF:
+
+  - Cover                  : tanpa nomor halaman (footerReference dihapus)
+  - KATA PENGANTAR + ISI   : i, ii  (lowerRoman, start 1)
+  - BAB I                  : 1, 2, 3 ...  (decimal, start 1)
+  - BAB II dst + PUSTAKA   : lanjut (tanpa restart)
+
+Pandoc otomatis menyisipkan paragraf `w:sectPr` (berisi footerReference
+dari reference.docx) sebelum setiap Heading1. Script ini mengklasifikasikan
+setiap sectPr berdasarkan heading berikutnya dan menyesuaikan isinya.
+
+Usage:
+  python3 scripts/finalize-docx.py input.docx output.docx
+"""
+
+import re
+import sys
+import zipfile
+
+if sys.platform == "win32":
+    try:
+        sys.stdout.reconfigure(encoding="utf-8", errors="replace")
+        sys.stderr.reconfigure(encoding="utf-8", errors="replace")
+    except Exception:
+        pass
+
+SECT_IN_PPR = re.compile(r"<w:pPr><w:sectPr>(.*?)</w:sectPr></w:pPr>", re.S)
+FOOTER_REF = re.compile(r"<w:footerReference[^/]*/>")
+
+
+def next_heading_text(doc, start):
+    nxt = doc[start:]
+    m = re.search(r'<w:pStyle w:val="Heading1"[^>]*/>', nxt)
+    if not m:
+        return ""
+    hdr = nxt[m.end():]
+    t = re.search(r"<w:t[^>]*>([^<]*)", hdr)
+    return t.group(1) if t else ""
+
+
+def is_front_matter_heading(t):
+    norm = t.strip().upper()
+    return norm in (
+        "KATA PENGANTAR",
+        "PRAKATA",
+        "ABSTRAK",
+        "ABSTRACT",
+        "DAFTAR ISI",
+        "DAFTAR TABEL",
+        "DAFTAR GAMBAR",
+        "DAFTAR LAMPIRAN",
+    )
+
+
+def is_chapter_1(t):
+    norm = t.strip().upper()
+    if re.match(r"^BAB\s+(?:I|1|01)\b", norm):
+        return True
+    if re.match(r"^(?:1\.|1\s+|I\.)", norm):
+        return True
+    if norm.startswith("PENDAHULUAN"):
+        return True
+    return False
+
+
+def is_chapter_2(t):
+    norm = t.strip().upper()
+    if re.match(r"^BAB\s+(?:II|2|02)\b", norm):
+        return True
+    if re.match(r"^(?:2\.|2\s+|II\.)", norm):
+        return True
+    if any(
+        norm.startswith(k)
+        for k in (
+            "PEMBAHASAN",
+            "TINJAUAN PUSTAKA",
+            "LANDASAN TEORI",
+            "METODOLOGI",
+            "METODE PENELITIAN",
+        )
+    ):
+        return True
+    return False
+
+
+def classify(text):
+    t = text.strip()
+    if is_front_matter_heading(t):
+        return "cover"
+    if is_chapter_1(t):
+        return "roman"
+    if is_chapter_2(t):
+        return "decimal-start"
+    return "body"
+
+
+def add_pgnum(sect, fmt, start):
+    if "<w:pgNumType" in sect:
+        return sect
+    tag = f'<w:pgNumType w:fmt="{fmt}" w:start="{start}"/>'
+    return sect + tag
+
+
+def process(doc):
+    matches = list(SECT_IN_PPR.finditer(doc))
+    if not matches:
+        return doc, 0
+    out = []
+    last = 0
+    changed = 0
+    for m in matches:
+        out.append(doc[last:m.start()])
+        sect = m.group(1)
+        kind = classify(next_heading_text(doc, m.end()))
+        if kind == "cover":
+            sect = FOOTER_REF.sub("", sect)
+            changed += 1
+        elif kind == "roman":
+            sect = add_pgnum(sect, "lowerRoman", "1")
+            changed += 1
+        elif kind == "decimal-start":
+            sect = add_pgnum(sect, "decimal", "1")
+            changed += 1
+        out.append(f"<w:pPr><w:sectPr>{sect}</w:sectPr></w:pPr>")
+        last = m.end()
+    out.append(doc[last:])
+    return "".join(out), changed
+
+
+def main():
+    src = sys.argv[1]
+    dst = sys.argv[2]
+
+    with zipfile.ZipFile(src, "r") as zin:
+        names = zin.namelist()
+        items = {n: zin.read(n) for n in names}
+
+    doc = items["word/document.xml"].decode("utf-8")
+    new_doc, changed = process(doc)
+    items["word/document.xml"] = new_doc.encode("utf-8")
+
+    with zipfile.ZipFile(dst, "w", zipfile.ZIP_DEFLATED) as zout:
+        for n in items:
+            zout.writestr(n, items[n])
+
+    print(f"[OK] finalize-docx: {changed} sectPr disesuaikan -> {dst}")
+
+
+if __name__ == "__main__":
+    main()