"""TürkPatent DOCX builder — in-memory, returns bytes. Format kuralları reference-scripts/build_turkpatent_separate_docs.py ile aynı: - 20/25/20/20 mm kenar marjları - Times New Roman 12pt, 1.5 satır aralığı - Her 5 satırda sol kenar satır numarası (restart=newPage) - PAGE field alt ortalanmış sayfa numarası """ from __future__ import annotations import io import re from typing import TypedDict from docx import Document from docx.enum.text import WD_ALIGN_PARAGRAPH, WD_LINE_SPACING from docx.oxml import OxmlElement from docx.oxml.ns import qn from docx.shared import Mm, Pt class ClaimDict(TypedDict): number: int type: str # "independent" | "dependent" parent: int | None text: str # --------------------------- Format helpers --------------------------- def _add_line_numbering(section) -> None: """Her 5 satırda sol kenar satır numarası, her sayfada restart.""" sect_pr = section._sectPr ln_num_type = OxmlElement("w:lnNumType") ln_num_type.set(qn("w:countBy"), "5") ln_num_type.set(qn("w:start"), "1") ln_num_type.set(qn("w:restart"), "newPage") ln_num_type.set(qn("w:distance"), "360") sect_pr.append(ln_num_type) def _add_page_number(footer) -> None: """Alt ortalanmış sayfa numarası (Word PAGE field).""" p = footer.paragraphs[0] p.alignment = WD_ALIGN_PARAGRAPH.CENTER run = p.add_run() begin = OxmlElement("w:fldChar") begin.set(qn("w:fldCharType"), "begin") instr = OxmlElement("w:instrText") instr.set(qn("xml:space"), "preserve") instr.text = "PAGE" end = OxmlElement("w:fldChar") end.set(qn("w:fldCharType"), "end") run._r.append(begin) run._r.append(instr) run._r.append(end) run.font.name = "Times New Roman" run.font.size = Pt(11) def _set_base_style(doc) -> None: style = doc.styles["Normal"] style.font.name = "Times New Roman" style.font.size = Pt(12) pf = style.paragraph_format pf.line_spacing_rule = WD_LINE_SPACING.ONE_POINT_FIVE pf.space_before = Pt(0) pf.space_after = Pt(0) def _set_margins(section) -> None: # TürkPatent tarifname/istem/özet: 20/25/20/20 mm section.page_height = Mm(297) section.page_width = Mm(210) section.top_margin = Mm(20) section.left_margin = Mm(25) section.right_margin = Mm(20) section.bottom_margin = Mm(20) section.header_distance = Mm(10) section.footer_distance = Mm(10) def _new_doc(): doc = Document() _set_base_style(doc) section = doc.sections[0] _set_margins(section) _add_line_numbering(section) _add_page_number(section.footer) return doc def _add_heading(doc, text: str, level: int = 1): p = doc.add_paragraph() p.paragraph_format.space_before = Pt(12) p.paragraph_format.space_after = Pt(6) run = p.add_run(text) run.bold = True run.font.name = "Times New Roman" if level == 1: run.font.size = Pt(14) elif level == 2: run.font.size = Pt(13) else: run.font.size = Pt(12) return p def _add_para( doc, text: str, justify: bool = True, center: bool = False, bold: bool = False, size_pt: int = 12, ): p = doc.add_paragraph() if center: p.alignment = WD_ALIGN_PARAGRAPH.CENTER elif justify: p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY pf = p.paragraph_format pf.line_spacing_rule = WD_LINE_SPACING.ONE_POINT_FIVE pf.space_before = Pt(0) pf.space_after = Pt(6) run = p.add_run(text) run.font.name = "Times New Roman" run.font.size = Pt(size_pt) run.bold = bold return p def _clean_inline_markdown(text: str) -> str: """Bold/italic/kod/link markdown'ı düz metne çevirir.""" text = re.sub(r"\*\*([^*]+)\*\*", r"\1", text) text = re.sub(r"(? list[tuple[str, str]]: """Markdown'ı (kind, text) listesine çevir. kind: h1-h6 | p | quote.""" lines = md.splitlines() out: list[tuple[str, str]] = [] in_code = False buffer_para: list[str] = [] def flush_para() -> None: nonlocal buffer_para if buffer_para: t = " ".join(buffer_para).strip() if t: out.append(("p", t)) buffer_para = [] for raw in lines: stripped = raw.strip() if stripped.startswith("```"): flush_para() in_code = not in_code continue if in_code: continue m = re.match(r"^(#{1,6})\s+(.*)", raw) if m: flush_para() level = len(m.group(1)) text = m.group(2).strip() out.append((f"h{min(level, 6)}", text)) continue if not stripped: flush_para() continue if stripped.startswith(("- ", "* ", "• ")): flush_para() bullet = stripped[2:].strip() out.append(("p", "• " + bullet)) continue if stripped.startswith(">"): content = stripped.lstrip("> ").rstrip() if content: out.append(("quote", content)) continue if stripped in ("---", "***"): flush_para() continue buffer_para.append(stripped) flush_para() return out # --------------------------- Public builders --------------------------- def build_description_docx(title: str, markdown: str) -> bytes: """Tarifname.docx — markdown içinden üretir, in-memory bytes döner.""" doc = _new_doc() _add_heading(doc, "TARİFNAME", level=1) _add_para(doc, title, center=True, bold=True, size_pt=13, justify=False) for kind, text in _parse_markdown(markdown): clean = _clean_inline_markdown(text).strip() if not clean: continue if kind.startswith("h"): level = int(kind[1]) _add_heading(doc, clean, level=max(2, min(level, 4))) elif kind in ("p", "quote"): _add_para(doc, clean) buffer = io.BytesIO() doc.save(buffer) return buffer.getvalue() def build_claims_docx(title: str, claims: list[ClaimDict]) -> bytes: """Istemler.docx — yapılandırılmış claims listesinden üretir.""" doc = _new_doc() _add_heading(doc, "İSTEMLER", level=1) _add_para(doc, title, center=True, bold=True, size_pt=13, justify=False) # Bağımsız istemleri (independent) grupla — her grup öncesi bir alt başlık verebiliriz current_independent: int | None = None for claim in claims: if claim["type"] == "independent": current_independent = claim["number"] # İstem metni: "1. " _add_para(doc, f"{claim['number']}. {claim['text']}") _ = current_independent # (gelecekte alt başlık için) buffer = io.BytesIO() doc.save(buffer) return buffer.getvalue() def build_abstract_docx(title: str, abstract: str) -> bytes: """Ozet.docx — tek paragraf, 150 kelime altı olmalı (aramada uyarı değil).""" doc = _new_doc() _add_heading(doc, "ÖZET", level=1) _add_para(doc, title, center=True, bold=True, size_pt=13, justify=False) _add_para(doc, abstract.strip()) buffer = io.BytesIO() doc.save(buffer) return buffer.getvalue()