# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/werkliv.py : connecteur Werkliv (logement étudiant)
# werkliv.com est le site corporatif ; la location de ses immeubles passe
# par sa plateforme University Apartments (universityapartments.ca —
# WordPress + FacetWP rendu serveur). Immeubles montréalais : Palay
# (2025 rue Peel, centre-ville) et Le Mojave (3476 rue Saint-Dominique,
# Plateau/Milton-Parc). Une annonce par typologie (1-BEDROOM, 4-BEDROOM...),
# loyer par personne (colocation étudiante meublée).
# -----------------------------------------------------------------------------
from __future__ import annotations
import hashlib
import re
from bs4 import BeautifulSoup
from ..schema import Listing
from .base import BaseConnector
BASE = "https://universityapartments.ca"
LIST_URL = f"{BASE}/apartment-listings/?_listings_city_en=montreal"
# Immeuble (nom affiché) -> (slug fiche immeuble, secteur)
BUILDINGS = {
"palay": ("palay", "Centre-ville (Ville-Marie)"),
"le mojave": ("le-mojave", "Milton-Parc / Plateau-Mont-Royal"),
"mojave": ("le-mojave", "Milton-Parc / Plateau-Mont-Royal"),
}
ADDR_RE = re.compile(
r"\d{2,5}[^<>\"|]{2,60}?(?:Montr[ée]al)[,\s]+QC(?:\s+[A-Z]\d[A-Z]\s?\d[A-Z]\d)?")
# Émojis/puces en tête des items de commodités des fiches
_EMOJI_PREFIX_RE = re.compile(r"^[\W_]+", re.UNICODE)
def _unit_type(label: str) -> str:
"""'1-BEDROOM' -> 3½, '4-BEDROOM' -> 6½, 'STUDIO' -> Studio."""
s = (label or "").lower()
if "studio" in s:
return "Studio"
m = re.search(r"(\d+)", s)
if m:
n = int(m.group(1))
return "Studio" if n == 0 else f"{n + 2}½"
return label.strip()
class WerklivConnector(BaseConnector):
source_id = "werkliv"
request_delay = 0.6
max_details = 20 # garde-fou
def fetch(self) -> list[Listing]:
listings: list[Listing] = []
try:
html = self.get(LIST_URL).text
except Exception:
return listings
soup = BeautifulSoup(html, "html.parser")
addresses: dict[str, str] = {} # slug immeuble -> adresse
seen: set[str] = set()
for card in soup.select(".lcl-card"):
try:
a = card.select_one('a.lcl-link[href*="/listings/"]') or \
card.select_one('a[href*="/listings/"]')
if not a:
continue
url = (a.get("href") or "").split("?")[0]
m = re.search(r"/listings/([^/]+)/?$", url)
if not m or m.group(1) in seen:
continue
slug = m.group(1)
seen.add(slug)
city_el = card.select_one(".lcl-city")
city_raw = city_el.get_text(" ", strip=True) if city_el else ""
if "montreal" not in city_raw.lower():
continue # hors Montréal (Halifax, PEI...)
typo_el = card.select_one(".lcl-title .h4, .lcl-title")
typology = typo_el.get_text(" ", strip=True) if typo_el else ""
prop_el = card.select_one(".lcl-property")
building = prop_el.get_text(" ", strip=True) if prop_el else ""
price_el = card.select_one(".lcl-price")
price_label = price_el.get_text(" ", strip=True) \
if price_el else ""
price = None
pm = re.search(r"\$\s*([\d,]+)", price_label)
if pm:
try:
v = float(pm.group(1).replace(",", ""))
price = v if 100 <= v <= 20000 else None
except ValueError:
pass
avail = ""
av_el = card.select_one(".lcl-available span")
if av_el:
avail = re.sub(r"\s+", " ",
av_el.get_text(" ", strip=True))
amenities = ["Logement étudiant"]
for chip in card.select(".lcl-chip"):
amenities.append(re.sub(r"\s+", " ",
chip.get_text(" ", strip=True)))
bslug, sector = BUILDINGS.get(building.strip().lower(),
("", ""))
# adresse depuis la fiche de l'immeuble (mise en cache)
address = ""
if bslug:
if bslug not in addresses:
addresses[bslug] = self._building_address(bslug)
address = addresses[bslug]
images = []
header = card.select_one("[data-bg]")
if header and header.get("data-bg", "").startswith("http"):
images.append(header["data-bg"])
listings.append(Listing(
source=self.source_id,
external_id=slug,
url=url,
title=f"{building} — {typology} (par chambre)",
address=address,
sector=sector,
city="Montréal",
unit_type=_unit_type(typology),
price=price,
price_label=f"{price_label} (par personne)"
if price_label else "",
availability=avail,
amenities=list(dict.fromkeys(amenities)),
images=images,
))
except Exception:
continue
# Fiches détaillées (photos, description, dispo ACF, bail, commodités)
# via le cache BD : 1 vraie requête par annonce et par changement.
self._detail_requests = 0
for i, lst in enumerate(listings):
if i >= self.max_details:
break
key = hashlib.sha1(
f"{lst.title}|{lst.price_label}|{lst.availability}"
f"|{'|'.join(lst.amenities)}".encode("utf-8")).hexdigest()
def _fetch(url=lst.url) -> dict:
if self._detail_requests >= self.max_details:
return {}
self._detail_requests += 1
return self._fetch_detail(url)
try:
payload = self.detail(lst.external_id, key, _fetch) or {}
except Exception:
payload = {}
self._apply_detail(lst, payload)
return listings
# -- fiche immeuble (adresse) --------------------------------------------------
def _building_address(self, slug: str) -> str:
try:
html = self.get(f"{BASE}/properties/{slug}/").text
except Exception:
return ""
m = ADDR_RE.search(html.replace("+", " "))
if not m:
return ""
addr = re.sub(r"\s+", " ", m.group(0)).strip()
return addr.replace("Montreal", "Montréal")
# -- fiche annonce ---------------------------------------------------------------
def _fetch_detail(self, url: str) -> dict:
"""Fiche d'une annonce : photos, description, sidebar ACF structuré
(Available, Lease Terms) et commodités (BUILDING AMENITIES, cuisine).
"""
html = self.get(url).text
soup = BeautifulSoup(html, "html.parser")
imgs = re.findall(
r'https://universityapartments\.ca/wp-content/uploads/'
r'[^"\s\\\)]+\.(?:jpg|jpeg|png|webp)', html)
imgs = [u for u in dict.fromkeys(imgs)
if not re.search(r"logo|icon|favicon|chrome|-\d{2,3}x\d{2,3}\.",
u, re.I)]
description = ""
og = soup.find("meta", attrs={"property": "og:description"})
if og and og.get("content"):
description = og["content"].strip()[:600]
else:
p = soup.select_one(".fl-rich-text p, article p")
if p:
description = p.get_text(" ", strip=True)[:600]
# Sidebar ACF : - Available
- ...
available = ""
for dl in soup.select("dl.acf-data"):
dts = [d.get_text(" ", strip=True) for d in dl.select("dt")]
dds = [d.get_text(" ", strip=True) for d in dl.select("dd")]
for k, v in zip(dts, dds):
if k.lower().startswith("available") and v:
available = v
lease_terms = [t.get_text(" ", strip=True)
for t in soup.select(".rental-term")]
# Commodités de l'immeuble : items à émoji après le titre
# « BUILDING AMENITIES » (même bloc rich-text).
amenities: list[str] = []
head = soup.find(string=re.compile(r"BUILDING AMENITIES", re.I))
if head:
h = head.find_parent(["h1", "h2", "h3", "h4", "strong"]) \
or head.parent
node = h.find_parent(["h1", "h2", "h3", "h4"]) or h
for sib in node.find_next_siblings():
if sib.name in ("h1", "h2", "h3"):
break
text = _EMOJI_PREFIX_RE.sub("", sib.get_text(" ", strip=True))
text = re.sub(r"\s+", " ", text).strip()
if not text or text.startswith("***") or "Disclaimer" in text:
break
if len(text) > 90:
continue
# « Wi-Fi ($) » = payant : ne pas laisser la normalisation
# le classer « internet inclus » (lacune générique notée)
if text.endswith("($)"):
continue
# stationnement à vélo ≠ stationnement auto
if re.search(r"\bbike (?:parking|storage)\b", text, re.I):
text = "Espace vélos (sous-sol)"
if text not in amenities:
amenities.append(text)
# Électroménagers de la cuisine partagée (ligne explicite de la fiche)
kitchen = soup.find(string=re.compile(
r"appliances,? including a fridge", re.I))
if kitchen:
amenities.append(_EMOJI_PREFIX_RE.sub(
"", re.sub(r"\s+", " ", str(kitchen)).strip()))
return {"images": imgs, "description": description,
"available": available, "lease_terms": lease_terms,
"amenities": amenities}
def _apply_detail(self, lst: Listing, payload: dict) -> None:
if not payload:
return
if payload.get("images"):
lst.images = list(dict.fromkeys(
lst.images + payload["images"]))[:40]
if payload.get("description"):
lst.description = payload["description"]
if not lst.availability and payload.get("available"):
lst.availability = f"Available {payload['available']}"
if payload.get("lease_terms"):
lst.amenities.append(
"Lease terms: " + ", ".join(payload["lease_terms"]))
for a in payload.get("amenities") or []:
if a not in lst.amenities:
lst.amenities.append(a)