import io import logging import re import sys from datetime import date from django.core.exceptions import ObjectDoesNotExist from django.core.files.base import ContentFile from django.template import Context, Template from django.utils import timezone from docx import Document from docx.shared import Pt logger = logging.getLogger(__name__) def _competencies_for_resume(profile): from portfolio.competencies_catalog import COMPETENCIES_BY_CODE, VALID_CODES from portfolio.models import StudentCompetencies try: rec = profile.competencies_record except ObjectDoesNotExist: rec = None if not rec: return { "compet_personal_qualities": "", "compet_ok_lines": [], "compet_pk_lines": [], "compet_additional": "", "has_competencies_section": False, } codes = rec.selected_competency_codes or [] valid = [c for c in codes if c in VALID_CODES] ok_lines = [COMPETENCIES_BY_CODE[c] for c in valid if c.startswith("OK")] pk_lines = [COMPETENCIES_BY_CODE[c] for c in valid if c.startswith("PK")] pq = (rec.personal_qualities or "").strip() extra = (rec.additional_competencies or "").strip() has = bool(pq or ok_lines or pk_lines or extra) return { "compet_personal_qualities": pq, "compet_ok_lines": ok_lines, "compet_pk_lines": pk_lines, "compet_additional": extra, "has_competencies_section": has, } def _serialize_profile(profile, request=None): photo_url = "" if profile.photo: url = profile.photo.url if request is not None: photo_url = request.build_absolute_uri(url) else: photo_url = url parts = [ profile.last_name or "", profile.first_name or "", profile.patronymic or "", ] full_name = " ".join(p for p in parts if p).strip() return { "first_name": profile.first_name or "", "last_name": profile.last_name or "", "patronymic": profile.patronymic or "", "full_name": full_name, "phone": profile.phone or "", "email": profile.user.email if profile.user_id else "", "bio": profile.bio or "", "student_group": profile.student_group or "", "specialty": profile.specialty or "", "average_grade": str(profile.average_grade) if profile.average_grade is not None else "", "photo_url": photo_url, } def build_resume_context(profile, keywords, match_result, request=None): """Контекст для шаблонов: ORM-объекты в списках + актуальные ключи для адаптации.""" kw_skills = [str(k).lower() for k in (keywords.get("skills") or [])] kw_tech = [str(k).lower() for k in (keywords.get("technologies") or [])] kw_all_set = set(kw_skills) | set(kw_tech) soft = list(keywords.get("soft_skills") or []) student = _serialize_profile(profile, request=request) all_education = list(profile.educations.all().order_by("-start_year")) all_experience = list(profile.experiences.all().order_by("-start_date")) all_projects = list(profile.projects.all().order_by("-end_date", "-start_date", "-id")) all_skills = list(profile.skills.all()) all_achievements = list(profile.achievements.all().order_by("-date")) matched_skill_ids = {s.id for s in match_result["skills"]} relevant_skills = [s for s in all_skills if s.id in matched_skill_ids] other_skills = [s for s in all_skills if s.id not in matched_skill_ids] relevant_skills.sort(key=lambda s: s.name.lower()) other_skills.sort(key=lambda s: s.name.lower()) sorted_skills = relevant_skills + other_skills matched_projects = list(match_result["projects"]) matched_pids = {p.id for p in matched_projects} projects_with_score = [] for project in all_projects: blob = f"{project.title} {project.description} {project.technologies or ''}".lower() score = sum(1 for kw in kw_all_set if kw and kw in blob) projects_with_score.append((project, score)) projects_with_score.sort(key=lambda x: x[1], reverse=True) sorted_all_projects = [p[0] for p in projects_with_score] order = {p.id: i for i, p in enumerate(matched_projects)} projects_highlight = sorted( matched_projects, key=lambda p: order.get(p.id, 999) ) projects_other = [p for p in sorted_all_projects if p.id not in matched_pids] def ach_score(a): blob = f"{a.title} {a.description or ''} {a.organization or ''}".lower() return sum(1 for kw in kw_all_set if kw and kw in blob) + sum( 1 for s in soft if s.lower() in blob ) sorted_achievements = sorted(all_achievements, key=ach_score, reverse=True) def edu_score(edu): blob = f"{edu.institution} {edu.specialty} {edu.degree or ''}".lower() return sum(1 for kw in kw_all_set if kw and kw in blob) educations_sorted = sorted(all_education, key=edu_score, reverse=True) def exp_score(exp): blob = f"{exp.company} {exp.position} {exp.description}".lower() return sum(1 for kw in kw_all_set if kw and kw in blob) + sum( 1 for s in soft if s.lower() in blob ) experiences_sorted = sorted(all_experience, key=exp_score, reverse=True) ctx = { "student": student, "education": educations_sorted, "experience": experiences_sorted, "projects": sorted_all_projects, "projects_highlight": projects_highlight, "projects_other": projects_other, "skills": sorted_skills, "skills_highlight": relevant_skills, "skills_other": other_skills, "soft_skills": soft, "achievements": sorted_achievements, "keywords": keywords, "generated_date": date.today().isoformat(), "generated_at": timezone.now().strftime("%d.%m.%Y"), "vacancy_summary": keywords.get("vacancy_summary", ""), } ctx.update(_competencies_for_resume(profile)) return ctx def render_resume_html(template_obj, context_dict): css = (template_obj.css_styles or "").strip() body_tpl = template_obj.html_template or "" looks_full = body_tpl.lstrip().lower().startswith("]", body_tpl[:500], re.I) ) if looks_full: html = body_tpl if css and "" in html.lower(): head_part = re.split(r"", html, maxsplit=1, flags=re.I)[0] if "", f"", html, count=1, flags=re.I, ) elif css: html = f"" + html tpl = Template(html) return tpl.render(Context(context_dict)) full_html = f""" {body_tpl} """ template = Template(full_html) return template.render(Context(context_dict)) def _generate_pdf_weasyprint(html_content, base_url=None): """Высокое качество; на Windows часто требует GTK (libgobject).""" try: from weasyprint import HTML except (ImportError, OSError) as e: logger.info("WeasyPrint недоступен при импорте: %s", e) return None try: return HTML(string=html_content, base_url=base_url).write_pdf() except Exception as e: logger.warning("WeasyPrint: ошибка рендера PDF: %s", e) return None def _generate_pdf_xhtml2pdf(html_content): """Чистый Python (ReportLab), работает на Windows без GTK; качество проще, чем у WeasyPrint.""" try: from io import BytesIO from xhtml2pdf import pisa except ImportError: return None try: out = BytesIO() src = BytesIO((html_content or "").encode("utf-8")) doc = pisa.pisaDocument(src, dest=out, encoding="utf-8") if getattr(doc, "err", 0): logger.warning("xhtml2pdf: err-флаг при генерации PDF") return None data = out.getvalue() return data if data else None except Exception as e: logger.warning("xhtml2pdf: %s", e) return None def generate_pdf_fallback(html_content, base_url=None): """ PDF: сначала WeasyPrint, затем xhtml2pdf (fallback без нативных DLL на Windows). Возвращает (bytes | None, сообщение_об_ошибке | None). """ pdf = _generate_pdf_weasyprint(html_content, base_url=base_url) if pdf: return pdf, None pdf = _generate_pdf_xhtml2pdf(html_content) if pdf: logger.info("PDF сформирован резервным движком xhtml2pdf (WeasyPrint недоступен или не справился).") return pdf, None msg = ( "PDF не создан: WeasyPrint недоступен (часто отсутствует libgobject / GTK на Windows) " "и резервный xhtml2pdf не смог обработать HTML. " "См. docs/WINDOWS_PDF_SETUP.md. Доступны экспорт HTML и DOCX." ) if sys.platform.startswith("win"): msg += " На Windows можно установить GTK Runtime для WeasyPrint или использовать HTML/DOCX." return None, msg def html_to_pdf_bytes(html_string, base_url=None): pdf, _err = generate_pdf_fallback(html_string, base_url=base_url) return pdf def generate_docx_from_html(html_content): """Генерирует DOCX из HTML (fallback без оформления WeasyPrint, работает на Windows).""" try: from bs4 import BeautifulSoup except ImportError: return None try: document = Document() style = document.styles["Normal"] style.font.name = "Calibri" style.font.size = Pt(11) soup = BeautifulSoup(html_content or "", "html.parser") for tag in soup.find_all(["h1", "h2", "h3", "p", "li"]): text = tag.get_text(" ", strip=True) if not text: continue name = tag.name.lower() if name == "h1": document.add_heading(text, level=0) elif name == "h2": document.add_heading(text, level=1) elif name == "h3": document.add_heading(text, level=2) elif name == "li": document.add_paragraph(text, style="List Bullet") else: document.add_paragraph(text) buf = io.BytesIO() document.save(buf) buf.seek(0) return buf.read() except Exception: return None def build_docx_from_context(context_dict): """DOCX напрямую из структурированного контекста (предпочтительнее HTML-парсинга).""" doc = Document() style = doc.styles["Normal"] style.font.name = "Calibri" style.font.size = Pt(11) s = context_dict["student"] doc.add_heading(s.get("full_name") or "Резюме", 0) if s.get("email"): doc.add_paragraph(f"Email: {s['email']}") if s.get("phone"): doc.add_paragraph(f"Телефон: {s['phone']}") if s.get("student_group"): doc.add_paragraph(f"Группа: {s['student_group']}") if s.get("specialty"): doc.add_paragraph(f"Специальность: {s['specialty']}") if s.get("average_grade"): doc.add_paragraph(f"Средний балл: {s['average_grade']}") if s.get("bio"): doc.add_paragraph(s["bio"]) if context_dict.get("has_competencies_section"): doc.add_heading("Личные качества и компетенции", level=1) if context_dict.get("compet_personal_qualities"): doc.add_paragraph( "Личные качества: " + context_dict["compet_personal_qualities"] ) oks = context_dict.get("compet_ok_lines") or [] if oks: doc.add_heading("Общие компетенции (ОК)", level=2) for line in oks: doc.add_paragraph(line, style="List Bullet") pks = context_dict.get("compet_pk_lines") or [] if pks: doc.add_heading("Профессиональные компетенции (ПК)", level=2) for line in pks: doc.add_paragraph(line, style="List Bullet") if context_dict.get("compet_additional"): doc.add_paragraph( "Дополнительные компетенции: " + context_dict["compet_additional"] ) def add_skills_block(title, skill_list): if not skill_list: return doc.add_heading(title, level=1) for sk in skill_list: doc.add_paragraph(f"• {sk.name} — {sk.get_level_display()}") add_skills_block("Ключевые навыки (по вакансии)", context_dict.get("skills_highlight")) add_skills_block("Прочие навыки", context_dict.get("skills_other")) educations = context_dict.get("education") or [] if educations: doc.add_heading("Образование", level=1) for e in educations: years = f"{e.start_year}—" if e.is_current: years += "н.в." else: years += str(e.end_year or "н.в.") line = f"{e.institution}, {e.specialty} ({years})" if e.degree: line += f" — {e.degree}" doc.add_paragraph(line) experiences = context_dict.get("experience") or [] if experiences: doc.add_heading("Опыт", level=1) for x in experiences: doc.add_paragraph(f"{x.position}, {x.company}", style="Heading 3") end = "н.в." if x.is_current else (str(x.end_date) if x.end_date else "н.в.") doc.add_paragraph(f"{x.start_date} — {end}") doc.add_paragraph(x.description or "") for label, key in ( ("Проекты (релевантные)", "projects_highlight"), ("Другие проекты", "projects_other"), ): plist = context_dict.get(key) or [] if not plist: continue doc.add_heading(label, level=1) for p in plist: doc.add_paragraph(p.title, style="Heading 3") doc.add_paragraph(p.get_project_type_display()) doc.add_paragraph(p.description or "") if p.technologies: doc.add_paragraph(f"Технологии: {p.technologies}") achievements = context_dict.get("achievements") or [] if achievements: doc.add_heading("Достижения", level=1) for a in achievements: doc.add_paragraph( f"{a.title} — {a.result} ({a.date})" + (f", {a.organization}" if a.organization else "") ) buf = io.BytesIO() doc.save(buf) buf.seek(0) return buf.read() def save_generated_files(instance, html_string, context_dict, base_url=None): """Сохраняет HTML, PDF (если WeasyPrint доступен) и DOCX. Возвращает ошибку PDF или None.""" instance.html_content = html_string or "" pdf_bytes, pdf_err = generate_pdf_fallback(html_string, base_url=base_url) if pdf_bytes: name_pdf = f"resume_{instance.id or 'new'}.pdf" instance.pdf_file.save(name_pdf, ContentFile(pdf_bytes), save=False) elif pdf_err: logger.warning("WeasyPrint / PDF пропущен: %s", pdf_err) docx_bytes = None try: docx_bytes = build_docx_from_context(context_dict) except Exception: logger.exception("DOCX из контекста не сформирован, пробуем HTML") docx_bytes = None if not docx_bytes: docx_bytes = generate_docx_from_html(html_string) if docx_bytes: name_docx = f"resume_{instance.id or 'new'}.docx" instance.docx_file.save(name_docx, ContentFile(docx_bytes), save=False) # Обязательно сохранить запись: иначе html_content и файлы не попадают в БД/хранилище instance.save() return pdf_err