.*?

\s*

(.*?)

', ] # Curated descriptions for pages with no extractable lead paragraph. # Keyed by page path (""=home) then folder lang code (en/mk/si). # MK/SI copy is AI-drafted — flag for native-speaker review. DESC_OVERRIDES = { "": { "en": "The first official Macedonian student organisation in Slovenia — support, events, community and a voice for Macedonian students from application to graduation.", "mk": "Првата официјална македонска студентска организација во Словенија — поддршка, настани, заедница и глас за македонските студенти од пријава до дипломирање.", "si": "Prva uradna makedonska študentska organizacija v Sloveniji — podpora, dogodki, skupnost in glas makedonskih študentov od prijave do diplome.", }, "become-a-member": { "en": "Join MSOS and find your people in Slovenia. Meet students, attend events, get support and help create projects across the Macedonian student community.", "mk": "Придружете се на МСОС и најдете ги вашите луѓе во Словенија. Запознајте студенти, посетувајте настани, добијте поддршка и создавајте проекти во македонската студентска заедница.", "si": "Pridružite se MSOS in najdite svoje ljudi v Sloveniji. Spoznajte študente, obiskujte dogodke, prejmite podporo in soustvarjajte projekte makedonske študentske skupnosti.", }, "faq": { "en": "Short answers to the questions Macedonian students ask most about studying and living in Slovenia — applications, residence, work, housing and more.", "mk": "Кратки одговори на прашањата што најчесто ги поставуваат македонските студенти за студирање и живеење во Словенија — пријави, престој, работа, сместување и повеќе.", "si": "Kratki odgovori na vprašanja, ki jih makedonski študenti najpogosteje postavljajo o študiju in življenju v Sloveniji — prijave, prebivanje, delo, nastanitev in več.", }, "gallery": { "en": "A look back at the events, gatherings and moments that make the MSOS community — browse photo albums from Macedonian student life in Slovenia.", "mk": "Поглед наназад на настаните, дружењата и моментите што ја градат заедницата на МСОС — прелистајте фото-албуми од македонскиот студентски живот во Словенија.", "si": "Pogled nazaj na dogodke, druženja in trenutke, ki gradijo skupnost MSOS — prebrskajte foto-albume iz makedonskega študentskega življenja v Sloveniji.", }, "news": { "en": "Everything new at MSOS in one place — events and projects, blog stories, announcements and official updates for Macedonian students in Slovenia.", "mk": "Сè ново кај МСОС на едно место — настани и проекти, блог приказни, соопштенија и официјални новости за македонските студенти во Словенија.", "si": "Vse novo pri MSOS na enem mestu — dogodki in projekti, blog zgodbe, obvestila in uradne novice za makedonske študente v Sloveniji.", }, "student-welcome-guide/my-route": { "en": "Your personal study route to Slovenia — follow the Student Welcome Guide steps for Macedonian students, from choosing a programme to enrolment and residence.", "mk": "Вашиот личен студиски пат до Словенија — следете ги чекорите од Водичот за добредојде за македонските студенти, од избор на програма до упис и престој.", "si": "Vaša osebna študijska pot v Slovenijo — sledite korakom Vodnika za bruce za makedonske študente, od izbire programa do vpisa in prebivanja.", }, } def extract_description(content, lang=None, path=None): if path is not None and path in DESC_OVERRIDES: ov = DESC_OVERRIDES[path].get(lang) if ov: return ov for pat in DESC_PATTERNS: m = re.search(pat, content, re.DOTALL) if m: txt = clean_text(m.group(1)) if len(txt) >= 40: return truncate(txt) return None HERO_PATTERNS = [ r'

.*?]*?src="(.*?)"', r'

\s*]*?src="(.*?)"', ] def extract_og_image(content): for pat in HERO_PATTERNS: m = re.search(pat, content, re.DOTALL) if m: src = m.group(1) # normalise ../../../images/x -> /images/x idx = src.find("images/") if idx != -1: return DOMAIN + "/" + src[idx:] return DEFAULT_OG_IMAGE def title_text(content): m = TITLE_RE.search(content) if not m: return SITE_NAME["en"] return clean_text(m.group(1)) H1_RE = re.compile(r"]*>(.*?)", re.DOTALL) def extract_h1(content): m = H1_RE.search(content) return clean_text(m.group(1)) if m else None def canonical_for(lang, path): return f"{DOMAIN}/{lang}/" + (f"{path}/" if path else "") def article_schema(lang, path, content, canon, desc, ogimg): """Article + BreadcrumbList JSON-LD for 2-segment 'section/slug' pages.""" parts = path.split("/") if len(parts) != 2 or parts[0] not in SECTION_NAMES: return [] section, _ = parts headline = extract_h1(content) or title_text(content).split(" - ")[0] org = {"@type": "Organization", "name": SITE_NAME[lang], "url": f"{DOMAIN}/{lang}/"} article = { "@context": "https://schema.org", "@type": "Article", "headline": headline, "description": desc or "", "image": ogimg, "inLanguage": HREF[lang], "author": org, "publisher": {"@type": "Organization", "name": SITE_NAME[lang], "logo": {"@type": "ImageObject", "url": f"{DOMAIN}/images/1-logo.png"}}, "mainEntityOfPage": canon, } if not article["description"]: del article["description"] dates = ARTICLE_DATES.get(path) if dates: pub = dates.get("published") or dates.get("modified") mod = dates.get("modified") or dates.get("published") article["datePublished"] = pub article["dateModified"] = mod crumbs = { "@context": "https://schema.org", "@type": "BreadcrumbList", "itemListElement": [ {"@type": "ListItem", "position": 1, "name": HOME_CRUMB[lang], "item": f"{DOMAIN}/{lang}/"}, {"@type": "ListItem", "position": 2, "name": SECTION_NAMES[section][lang], "item": f"{DOMAIN}/{lang}/{section}/"}, {"@type": "ListItem", "position": 3, "name": headline, "item": canon}, ], } dump = lambda o: ' " return [dump(article), dump(crumbs)] def build_block(lang, path, content): canon = canonical_for(lang, path) title = title_text(content) desc = extract_description(content, lang, path) ogimg = extract_og_image(content) lines = [START] lines.extend(GA_LINES) if desc: lines.append(f' ') lines.append(f' ') # hreflang alternates (only langs that actually have this path) have = [L for L in LANGS if path in TREES[L]] for L in have: lines.append(f' ') xdef = "en" if "en" in have else have[0] lines.append(f' ') # Open Graph lines.append(f' ') lines.append(f' ') lines.append(f' ') if desc: lines.append(f' ') lines.append(f' ') lines.append(f' ') lines.append(f' ') for L in have: if L != lang: lines.append(f' ') # Twitter lines.append(f' ') lines.append(f' ') if desc: lines.append(f' ') lines.append(f' ') # Organization JSON-LD (site-wide) org = ( ' ' ) lines.append(org) lines.extend(article_schema(lang, path, content, canon, desc, ogimg)) lines.extend(FAVICON_LINES) lines.append(" " + END) return "\n".join(lines) def inject(content, lang, path): block = build_block(lang, path, content) # remove any existing block first (idempotent) content = BLOCK_RE.sub("", content) nl = "\r\n" if "\r\n" in content else "\n" block = block.replace("\n", nl) def repl(m): return m.group(1) + nl + " " + block return TITLE_RE.sub(repl, content, count=1) def page_path_of(fp, lang): base = os.path.join(ROOT, lang) rel = os.path.relpath(os.path.dirname(fp), base).replace("\\", "/") return "" if rel == "." else rel def all_pages(): for L in LANGS: base = os.path.join(ROOT, L) for root, _, files in os.walk(base): if "index.html" in files: yield L, os.path.join(root, "index.html") PILOT = [ ("en", os.path.join(ROOT, "en", "index.html")), ("en", os.path.join(ROOT, "en", "blog", "how-to-open-slovenian-bank-account", "index.html")), ("mk", os.path.join(ROOT, "mk", "blog", "how-to-open-slovenian-bank-account", "index.html")), ("si", os.path.join(ROOT, "si", "blog", "how-to-open-slovenian-bank-account", "index.html")), ] def write_sitemap(): urls = [] for L, fp in all_pages(): path = page_path_of(fp, L) urls.append((L, path)) # group by path so we can emit xhtml:link alternates from collections import defaultdict bypath = defaultdict(list) for L, path in urls: bypath[path].append(L) lines = ['', ''] lines[1] = lines[1].replace("www.sitemap.org", "www.sitemaps.org") for path in sorted(bypath): have = [L for L in LANGS if L in bypath[path]] for L in have: lines.append(" ") lines.append(f" {canonical_for(L, path)}") for A in have: lines.append(f' ') xdef = "en" if "en" in have else have[0] lines.append(f' ') lines.append(" ") lines.append("") with open(os.path.join(ROOT, "sitemap.xml"), "w", encoding="utf-8", newline="\n") as f: f.write("\n".join(lines) + "\n") robots = f"User-agent: *\nAllow: /\n\nSitemap: {DOMAIN}/sitemap.xml\n" with open(os.path.join(ROOT, "robots.txt"), "w", encoding="utf-8", newline="\n") as f: f.write(robots) print(f"sitemap.xml: {sum(len(v) for v in bypath.values())} urls; robots.txt written") def main(): args = sys.argv[1:] if "--all" in args: n = 0 for L, fp in all_pages(): with open(fp, "r", encoding="utf-8", newline="") as f: content = f.read() path = page_path_of(fp, L) out = inject(content, L, path) with open(fp, "w", encoding="utf-8", newline="") as f: f.write(out) n += 1 print(f"injected SEO block into {n} pages") if "--sitemap" in args: write_sitemap() else: # pilot: print blocks only for L, fp in PILOT: with open(fp, "r", encoding="utf-8", newline="") as f: content = f.read() path = page_path_of(fp, L) print(f"\n===== {L} /{path or '(home)'} =====") print(build_block(L, path, content)) if __name__ == "__main__": main()