8 nouveaux ATS (Taleo, Njoyn, UltiPro, iCIMS, SuccessFactors, ADP WFN, Digital Recruiters, Workland) + 23 employeurs publics/parapublics et grands employeurs QC

- Secteur public : Santé Québec (emplois.sante.quebec), Emplois Santé Montréal
  (njoyn), CIUSSS de l Est-de-l Île-de-Montréal (taleo), CSSDM (workland),
  villes de Gatineau et Terrebonne, STL, RTC (njoyn via Scrapfly anti-Radware)
- Grands employeurs : Bombardier, Cascades, Domtar (SuccessFactors sitemap
  RSS/urlset + microdonnées), Agnico Eagle, Bell Textron, Bayshore (Taleo REST
  + repli anglophone), Messer Canada (UltiPro), Lassonde, Mitsubishi HC
  Capital, MSI (ADP WFN + repli en_US), CAPREIT, McCarthy Tétrault, Omni,
  EDF Renouvelables (iCIMS + JSON-LD)
- gen_connectors : EXTRA_ATTRS multi-attributs pour les nouveaux ATS
- jobka/connectors/_jsonld.py : extraction schema.org/JobPosting partagée

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Simon-Pierre Boucher committed 2 mo ago (Aug 18, 2026) parent 5a66f48 Browse files

34 changed files +2,093 −0

added data/feeds-public.json +368 −0

++[
++  {
++    "ats": "taleo",
++    "source_id": "agnico_eagle",
++    "employer": "Agnico Eagle Mines",
++    "host": "agnicoeagle",
++    "section": "2",
++    "url": "https://www.agnicoeagle.com",
++    "sectors": [
++      "Mines",
++      "Ingénierie",
++      "Métiers spécialisés"
++    ],
++    "cities": [
++      "Abitibi-Témiscamingue",
++      "Malartic",
++      "Rouyn-Noranda"
++    ]
++  },
++  {
++    "ats": "taleo",
++    "source_id": "ciusss_est_montreal",
++    "employer": "CIUSSS de l'Est-de-l'Île-de-Montréal",
++    "host": "ciusssemtl",
++    "section": "cemtl",
++    "url": "https://ciusss-estmtl.gouv.qc.ca",
++    "sectors": [
++      "Santé",
++      "Services sociaux",
++      "Secteur public"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "taleo",
++    "source_id": "bayshore",
++    "employer": "Bayshore HealthCare",
++    "host": "bayshore",
++    "section": "bs_ex",
++    "url": "https://www.bayshore.ca",
++    "sectors": [
++      "Santé",
++      "Soins à domicile"
++    ],
++    "cities": [
++      "Montréal",
++      "Québec"
++    ]
++  },
++  {
++    "ats": "taleo",
++    "source_id": "bell_textron",
++    "employer": "Bell Textron Canada",
++    "host": "textron",
++    "section": "bell",
++    "url": "https://www.bellflight.com",
++    "sectors": [
++      "Aéronautique",
++      "Fabrication"
++    ],
++    "cities": [
++      "Mirabel"
++    ]
++  },
++  {
++    "ats": "njoyn",
++    "source_id": "sante_montreal",
++    "employer": "Santé Québec — Montréal",
++    "base": "https://emplois.santemontreal.qc.ca",
++    "cl": "CL3",
++    "clid": "54327",
++    "url": "https://santemontreal.ca",
++    "sectors": [
++      "Santé",
++      "Services sociaux",
++      "Secteur public"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "njoyn",
++    "source_id": "ville_gatineau",
++    "employer": "Ville de Gatineau",
++    "base": "https://gatineau.njoyn.com",
++    "cl": "CL2",
++    "clid": "27082",
++    "use_scrapfly": true,
++    "url": "https://www.gatineau.ca",
++    "sectors": [
++      "Municipal",
++      "Secteur public"
++    ],
++    "cities": [
++      "Gatineau"
++    ]
++  },
++  {
++    "ats": "njoyn",
++    "source_id": "ville_terrebonne",
++    "employer": "Ville de Terrebonne",
++    "base": "https://clients.njoyn.com",
++    "cl": "cl4",
++    "clid": "71764",
++    "use_scrapfly": true,
++    "url": "https://www.ville.terrebonne.qc.ca",
++    "sectors": [
++      "Municipal",
++      "Secteur public"
++    ],
++    "cities": [
++      "Terrebonne"
++    ]
++  },
++  {
++    "ats": "njoyn",
++    "source_id": "stl_laval",
++    "employer": "Société de transport de Laval",
++    "base": "https://lavaltransit.njoyn.com",
++    "cl": "CL2",
++    "clid": "60406",
++    "use_scrapfly": true,
++    "url": "https://www.stlaval.ca",
++    "sectors": [
++      "Transport collectif",
++      "Secteur public"
++    ],
++    "cities": [
++      "Laval"
++    ]
++  },
++  {
++    "ats": "njoyn",
++    "source_id": "rtc_quebec",
++    "employer": "Réseau de transport de la Capitale",
++    "base": "https://rtc.njoyn.com",
++    "cl": "CGI",
++    "clid": "23009",
++    "use_scrapfly": true,
++    "url": "https://www.rtcquebec.ca",
++    "sectors": [
++      "Transport collectif",
++      "Secteur public"
++    ],
++    "cities": [
++      "Québec"
++    ]
++  },
++  {
++    "ats": "ultipro",
++    "source_id": "messer_canada",
++    "employer": "Messer Canada",
++    "org": "MES1005MESR",
++    "board": "dbd63926-d756-486f-a254-b2a6ede1d26e",
++    "url": "https://www.messer-ca.com",
++    "sectors": [
++      "Gaz industriels",
++      "Industrie"
++    ],
++    "cities": [
++      "Montréal",
++      "Québec",
++      "Saint-Georges"
++    ]
++  },
++  {
++    "ats": "icims",
++    "source_id": "capreit",
++    "employer": "CAPREIT",
++    "sub": "careers-capreit",
++    "url": "https://www.capreit.ca",
++    "sectors": [
++      "Immobilier",
++      "Gestion immobilière"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "icims",
++    "source_id": "mccarthy_tetrault",
++    "employer": "McCarthy Tétrault",
++    "sub": "careers-mccarthyca",
++    "url": "https://www.mccarthy.ca",
++    "sectors": [
++      "Juridique",
++      "Services professionnels"
++    ],
++    "cities": [
++      "Montréal",
++      "Québec"
++    ]
++  },
++  {
++    "ats": "icims",
++    "source_id": "omni_hotels_horaire",
++    "employer": "Omni Hôtel Mont-Royal",
++    "sub": "externalhourly-omnihotels",
++    "url": "https://www.omnihotels.com/fr/hotels/montreal-mont-royal",
++    "sectors": [
++      "Hôtellerie",
++      "Restauration"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "icims",
++    "source_id": "omni_hotels_gestion",
++    "employer": "Omni Hôtel Mont-Royal",
++    "sub": "externalmanager-omnihotels",
++    "url": "https://www.omnihotels.com/fr/hotels/montreal-mont-royal",
++    "sectors": [
++      "Hôtellerie",
++      "Gestion"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "icims",
++    "source_id": "edf_renouvelables",
++    "employer": "EDF Renouvelables Canada",
++    "sub": "cafrench-edf-re",
++    "url": "https://www.edf-renouvelables.ca",
++    "sectors": [
++      "Énergie",
++      "Ingénierie"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "successfactors",
++    "source_id": "bombardier",
++    "employer": "Bombardier",
++    "base": "https://jobs.bombardier.com",
++    "url": "https://bombardier.com",
++    "sectors": [
++      "Aéronautique",
++      "Ingénierie",
++      "Fabrication"
++    ],
++    "cities": [
++      "Dorval",
++      "Mirabel",
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "successfactors",
++    "source_id": "domtar",
++    "employer": "Domtar",
++    "base": "https://jobs.domtar.com",
++    "url": "https://www.domtar.com",
++    "sectors": [
++      "Pâtes et papiers",
++      "Foresterie",
++      "Fabrication"
++    ],
++    "cities": [
++      "Montréal",
++      "Windsor",
++      "Lachine"
++    ]
++  },
++  {
++    "ats": "successfactors",
++    "source_id": "cascades",
++    "employer": "Cascades",
++    "base": "https://jobs.cascades.com",
++    "url": "https://www.cascades.com",
++    "sectors": [
++      "Pâtes et papiers",
++      "Emballage",
++      "Fabrication"
++    ],
++    "cities": [
++      "Kingsey Falls",
++      "Drummondville",
++      "Candiac"
++    ]
++  },
++  {
++    "ats": "adp",
++    "source_id": "lassonde",
++    "employer": "Industries Lassonde",
++    "cid": "f96ce972-7721-4c7b-99ea-9a970ddc824a",
++    "ccid": "9200604444876_2",
++    "url": "https://www.lassonde.com",
++    "sectors": [
++      "Agroalimentaire",
++      "Fabrication"
++    ],
++    "cities": [
++      "Rougemont",
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "adp",
++    "source_id": "mitsubishi_hc_capital",
++    "employer": "Mitsubishi HC Capital Canada",
++    "cid": "b3ef4f03-f8ff-4ded-80c8-6dd5c5a224f7",
++    "ccid": "9200144510729_2",
++    "url": "https://www.mhccna.com",
++    "sectors": [
++      "Finance",
++      "Financement commercial"
++    ],
++    "cities": [
++      "Laval",
++      "Trois-Rivières"
++    ]
++  },
++  {
++    "ats": "adp",
++    "source_id": "msi_gestion",
++    "employer": "MSI Gestion immobilière",
++    "cid": "47d4e792-26e5-471c-92ac-73c60131bab8",
++    "ccid": "19000101_000001",
++    "url": "https://www.msimmobiliers.com",
++    "sectors": [
++      "Immobilier",
++      "Gestion immobilière"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  },
++  {
++    "ats": "digitalrecruiters",
++    "source_id": "sante_quebec",
++    "employer": "Santé Québec",
++    "base": "https://emplois.sante.quebec",
++    "url": "https://sante.quebec",
++    "sectors": [
++      "Santé",
++      "Secteur public",
++      "Administration"
++    ],
++    "cities": [
++      "Montréal",
++      "Québec"
++    ]
++  },
++  {
++    "ats": "workland",
++    "source_id": "cssdm",
++    "employer": "Centre de services scolaire de Montréal",
++    "list_url": "https://www.cssdm.gouv.qc.ca/travailler-cssdm/offres-emploi/",
++    "url": "https://www.cssdm.gouv.qc.ca",
++    "sectors": [
++      "Éducation",
++      "Secteur public"
++    ],
++    "cities": [
++      "Montréal"
++    ]
++  }
++]

added jobka/connectors/_jsonld.py +160 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/_jsonld.py
++# Rôle    : Extraction du balisage schema.org JobPosting (JSON-LD) — partagé
++#           par les connecteurs dont la source publie des pages détail HTML
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Utilitaires JSON-LD (schema.org/JobPosting).
++
++Beaucoup de sites (iCIMS, Jobillico, Espresso-Jobs, Digital Recruiters,
++Workland/Atlas…) embarquent un bloc ``<script type="application/ld+json">``
++conforme à schema.org sur la page détail de chaque offre. On l'extrait ici de
++façon tolérante (JSON imparfait, @graph, listes) et on le convertit en champs
++standard prêts à verser dans un ``JobPosting``.
++"""
++from __future__ import annotations
++
++import json
++import re
++
++_SCRIPT_RE = re.compile(
++    r'<script[^>]*type=["\']application/ld\+json["\'][^>]*>(.*?)</script>',
++    re.S | re.I)
++
++_UNIT = {"hour": "hour", "hourly": "hour", "day": "day", "week": "week",
++         "month": "month", "year": "year", "annual": "year"}
++
++
++def _iter_nodes(node):
++    """Itère tous les objets JSON-LD (racine, @graph, listes imbriquées)."""
++    if isinstance(node, list):
++        for item in node:
++            yield from _iter_nodes(item)
++    elif isinstance(node, dict):
++        yield node
++        yield from _iter_nodes(node.get("@graph") or [])
++
++
++def extract_jobposting(html: str) -> dict | None:
++    """Retourne le premier objet @type=JobPosting trouvé dans la page."""
++    for m in _SCRIPT_RE.finditer(html or ""):
++        raw = m.group(1).strip()
++        try:
++            data = json.loads(raw)
++        except ValueError:
++            # JSON avec contrôle non échappé (fréquent) : tentative de secours
++            try:
++                data = json.loads(re.sub(r"[\x00-\x1f]", " ", raw))
++            except ValueError:
++                continue
++        for node in _iter_nodes(data):
++            t = node.get("@type")
++            types = t if isinstance(t, list) else [t]
++            if any(str(x).lower() == "jobposting" for x in types if x):
++                return node
++    return None
++
++
++def _first(value):
++    if isinstance(value, list):
++        return value[0] if value else None
++    return value
++
++
++def jobposting_fields(node: dict) -> dict:
++    """Aplati un JobPosting JSON-LD en champs standard Job·Ka (dict sparse)."""
++    out: dict = {}
++    if not node:
++        return out
++    out["title"] = node.get("title") or node.get("name") or ""
++    out["description_html"] = node.get("description") or ""
++    out["date_posted"] = node.get("datePosted") or None
++    out["date_deadline"] = node.get("validThrough") or None
++    et = node.get("employmentType")
++    out["employment_label"] = ", ".join(et) if isinstance(et, list) else (et or "")
++
++    org = _first(node.get("hiringOrganization"))
++    if isinstance(org, dict):
++        out["employer"] = org.get("name") or ""
++    elif isinstance(org, str):
++        out["employer"] = org
++
++    loc = _first(node.get("jobLocation"))
++    if isinstance(loc, dict):
++        addr = loc.get("address") or {}
++        if isinstance(addr, str):
++            out["location_label"] = addr
++        elif isinstance(addr, dict):
++            out["city"] = addr.get("addressLocality") or ""
++            out["region_code"] = addr.get("addressRegion") or ""
++            out["postal_code"] = addr.get("postalCode") or ""
++            out["address"] = addr.get("streetAddress") or ""
++        geo = loc.get("geo") or {}
++        if isinstance(geo, dict) and geo.get("latitude") is not None:
++            try:
++                out["lat"] = float(geo["latitude"])
++                out["lng"] = float(geo["longitude"])
++            except (TypeError, ValueError):
++                pass
++
++    sal = node.get("baseSalary")
++    if isinstance(sal, dict):
++        val = sal.get("value")
++        unit = None
++        lo = hi = None
++        if isinstance(val, dict):
++            lo = val.get("minValue", val.get("value"))
++            hi = val.get("maxValue")
++            unit = val.get("unitText")
++        elif isinstance(val, (int, float)):
++            lo = val
++        if isinstance(lo, str):
++            try:
++                lo = float(lo.replace(",", "."))
++            except ValueError:
++                lo = None
++        if isinstance(hi, str):
++            try:
++                hi = float(hi.replace(",", "."))
++            except ValueError:
++                hi = None
++        if lo:
++            out["salary_min"] = float(lo)
++            out["salary_max"] = float(hi) if hi else None
++            out["salary_unit"] = _UNIT.get(str(unit or "").lower())
++    return out
++
++
++def apply_fields(job, fields: dict, *, override_employer: bool = False) -> None:
++    """Verse les champs extraits dans un JobPosting sans écraser l'existant."""
++    from ..normalize import clean_html
++    if fields.get("description_html") and not job.description:
++        job.description = clean_html(fields["description_html"])
++    if fields.get("title") and not job.title:
++        job.title = fields["title"]
++    if fields.get("employer") and (override_employer or not job.employer):
++        job.employer = fields["employer"]
++    if fields.get("date_posted") and not job.date_posted:
++        job.date_posted = fields["date_posted"]
++    if fields.get("date_deadline") and not job.date_deadline:
++        job.date_deadline = fields["date_deadline"]
++    if fields.get("city") and not job.city:
++        job.city = fields["city"]
++    if fields.get("postal_code") and not job.postal_code:
++        job.postal_code = fields["postal_code"]
++    if fields.get("address") and not job.address:
++        job.address = fields["address"]
++    if fields.get("location_label") and not job.location_label:
++        job.location_label = fields["location_label"]
++    if fields.get("salary_min") is not None and job.salary_min is None:
++        job.salary_min = fields["salary_min"]
++        job.salary_max = fields.get("salary_max")
++        job.salary_unit = fields.get("salary_unit")
++    if fields.get("employment_label"):
++        job.details.setdefault("employment_label", fields["employment_label"])
++    if fields.get("lat") is not None and job.lat is None:
++        job.lat = fields["lat"]
++        job.lng = fields.get("lng")

added jobka/connectors/adp.py +160 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/adp.py
++# Rôle    : Classe de plateforme ADP Workforce Now (centre de carrières) —
++#           API JSON publique job-requisitions, un employeur = une sous-classe
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme ADP Workforce Now (workforcenow.adp.com).
++
++API JSON publique du centre de carrières :
++- liste  : GET /mascsr/default/careercenter/public/events/staffing/v1/
++               job-requisitions?cid=<CID>&ccId=<CCID>&lang=<fr_CA>&$top=&$skip=
++           -> {jobRequisitions:[{itemID, requisitionTitle, postDate,
++               payGradeRange, workLevelCode, requisitionLocations,
++               customFieldGroup}]}
++- détail : GET .../job-requisitions/<itemID>?cid=… -> requisitionDescription
++           (HTML). Visité avec cache BD + budget.
++"""
++from __future__ import annotations
++
++import os
++
++from ..schema import JobPosting, clean_html, is_quebec_location
++from .base import BaseConnector
++
++PAGE_SIZE = 100
++MAX_DETAILS = int(os.environ.get("JOBKA_ADP_DETAIL_LIMIT", "80"))
++
++_API = ("https://workforcenow.adp.com/mascsr/default/careercenter/public/"
++        "events/staffing/v1/job-requisitions")
++
++_SALARY_UNIT = {"AN": "year", "ANNUEL": "year", "YR": "year", "YEAR": "year",
++                "HR": "hour", "HO": "hour", "HORAIRE": "hour", "HOUR": "hour"}
++
++
++class ADPWorkforceNowConnector(BaseConnector):
++    """Base ADP WFN — sous-classes : définir source_id, EMPLOYER, CID, CCID."""
++
++    ats = "adp"
++    request_delay = 1.0
++
++    EMPLOYER = ""
++    CID = ""                       # GUID du centre de carrières
++    CCID = "19000101_000001"       # identifiant du « career center »
++    LANG = "fr_CA"
++    quebec_only = True
++    max_pages = 10
++
++    def _params(self, extra: dict | None = None) -> dict:
++        p = {"cid": self.CID, "ccId": self.CCID, "lang": self.LANG,
++             "locale": self.LANG}
++        p.update(extra or {})
++        return p
++
++    @staticmethod
++    def _locations(req: dict) -> list[dict]:
++        out = []
++        for loc in req.get("requisitionLocations") or []:
++            addr = loc.get("address") or {}
++            out.append({
++                "city": addr.get("cityName") or "",
++                "prov": ((addr.get("countrySubdivisionLevel1") or {})
++                         .get("codeValue") or ""),
++                "postal": addr.get("postalCode") or "",
++                "label": ((loc.get("nameCode") or {}).get("shortName")
++                          or "").strip(),
++            })
++        return out
++
++    def _keep(self, locations: list[dict]) -> bool:
++        if not self.quebec_only:
++            return True
++        return any(l["prov"].upper() == "QC"
++                   or is_quebec_location(f"{l['label']} {l['city']}")
++                   for l in locations)
++
++    def _fetch_detail(self, item_id: str) -> dict:
++        data = self.get(f"{_API}/{item_id}", params=self._params(),
++                        headers={"Accept": "application/json"}).json()
++        return {"description": clean_html(
++            data.get("requisitionDescription") or "")}
++
++    def fetch(self) -> list[JobPosting]:
++        out: list[JobPosting] = []
++        details_used = 0
++        skip = 0
++        for _ in range(self.max_pages):
++            data = self.get(_API, params=self._params(
++                {"$top": str(PAGE_SIZE), "$skip": str(skip)}),
++                headers={"Accept": "application/json"}).json()
++            reqs = data.get("jobRequisitions") or []
++            if not reqs and skip == 0 and self.LANG != "en_US":
++                # certains centres de carrières ne répondent qu'en anglais
++                self.LANG = "en_US"
++                continue
++            if not reqs:
++                break
++            for r in reqs:
++                locations = self._locations(r)
++                if not self._keep(locations):
++                    continue
++                iid = str(r.get("itemID") or "")
++                if not iid:
++                    continue
++                qc = next((l for l in locations
++                           if l["prov"].upper() == "QC"
++                           or is_quebec_location(f"{l['label']} {l['city']}")),
++                          locations[0] if locations else
++                          {"city": "", "postal": "", "label": ""})
++                job = JobPosting(
++                    source=self.source_id, external_id=iid,
++                    url=("https://workforcenow.adp.com/mascsr/default/mdf/"
++                         f"recruitment/recruitment.html?cid={self.CID}"
++                         f"&ccId={self.CCID}&lang={self.LANG}&jobId={iid}"),
++                    employer=self.EMPLOYER,
++                    title=r.get("requisitionTitle") or "",
++                    city=qc["city"], postal_code=qc["postal"],
++                    location_label=qc["label"],
++                    date_posted=r.get("postDate") or None,
++                    ats=self.ats,
++                )
++                pay = r.get("payGradeRange") or {}
++                lo = ((pay.get("minimumRate") or {}).get("amountValue"))
++                hi = ((pay.get("maximumRate") or {}).get("amountValue"))
++                if lo:
++                    unit = None
++                    for c in ((r.get("customFieldGroup") or {})
++                              .get("codeFields") or []):
++                        if ((c.get("nameCode") or {})
++                                .get("codeValue")) == "SalaryType":
++                            unit = _SALARY_UNIT.get(
++                                str(c.get("codeValue") or "").upper()) \
++                                or _SALARY_UNIT.get(
++                                    str(c.get("shortName") or "").upper())
++                    job.salary_min = float(lo)
++                    job.salary_max = float(hi) if hi else None
++                    job.salary_unit = unit or ("year" if float(lo) > 5000
++                                               else "hour")
++                wl = (r.get("workLevelCode") or {}).get("shortName")
++                if wl:
++                    job.details["employment_label"] = wl
++                key = (r.get("postDate") or "") + (job.title or "")[:40]
++                if details_used < MAX_DETAILS:
++                    fresh = [False]
++
++                    def _fn(i=iid, fresh=fresh):
++                        fresh[0] = True
++                        return self._fetch_detail(i)
++
++                    d = self.detail(iid, key, _fn)
++                    if fresh[0]:
++                        details_used += 1
++                    job.description = d.get("description", "")
++                out.append(job)
++            if len(reqs) < PAGE_SIZE:
++                break
++            skip += PAGE_SIZE
++        return out

added jobka/connectors/agnico_eagle.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/agnico_eagle.py
++# Rôle    : Connecteur Agnico Eagle Mines — taleo (host='agnicoeagle', section='2')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .taleo import TaleoConnector
++
++
++class AgnicoEagleConnector(TaleoConnector):
++    source_id = 'agnico_eagle'
++    EMPLOYER = 'Agnico Eagle Mines'
++    HOST = 'agnicoeagle'
++    SECTION = '2'

added jobka/connectors/bayshore.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/bayshore.py
++# Rôle    : Connecteur Bayshore HealthCare — taleo (host='bayshore', section='bs_ex')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .taleo import TaleoConnector
++
++
++class BayshoreConnector(TaleoConnector):
++    source_id = 'bayshore'
++    EMPLOYER = 'Bayshore HealthCare'
++    HOST = 'bayshore'
++    SECTION = 'bs_ex'

added jobka/connectors/bell_textron.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/bell_textron.py
++# Rôle    : Connecteur Bell Textron Canada — taleo (host='textron', section='bell')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .taleo import TaleoConnector
++
++
++class BellTextronConnector(TaleoConnector):
++    source_id = 'bell_textron'
++    EMPLOYER = 'Bell Textron Canada'
++    HOST = 'textron'
++    SECTION = 'bell'

added jobka/connectors/bombardier.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/bombardier.py
++# Rôle    : Connecteur Bombardier — successfactors (base='https://jobs.bombardier.com')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .successfactors import SuccessFactorsConnector
++
++
++class BombardierConnector(SuccessFactorsConnector):
++    source_id = 'bombardier'
++    EMPLOYER = 'Bombardier'
++    BASE = 'https://jobs.bombardier.com'

added jobka/connectors/capreit.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/capreit.py
++# Rôle    : Connecteur CAPREIT — icims (sub='careers-capreit')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .icims import ICIMSConnector
++
++
++class CapreitConnector(ICIMSConnector):
++    source_id = 'capreit'
++    EMPLOYER = 'CAPREIT'
++    SUB = 'careers-capreit'

added jobka/connectors/cascades.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/cascades.py
++# Rôle    : Connecteur Cascades — successfactors (base='https://jobs.cascades.com')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .successfactors import SuccessFactorsConnector
++
++
++class CascadesConnector(SuccessFactorsConnector):
++    source_id = 'cascades'
++    EMPLOYER = 'Cascades'
++    BASE = 'https://jobs.cascades.com'

added jobka/connectors/ciusss_est_montreal.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/ciusss_est_montreal.py
++# Rôle    : Connecteur CIUSSS de l'Est-de-l'Île-de-Montréal — taleo (host='ciusssemtl', section='cemtl')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .taleo import TaleoConnector
++
++
++class CiusssEstMontrealConnector(TaleoConnector):
++    source_id = 'ciusss_est_montreal'
++    EMPLOYER = "CIUSSS de l'Est-de-l'Île-de-Montréal"
++    HOST = 'ciusssemtl'
++    SECTION = 'cemtl'

added jobka/connectors/cssdm.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/cssdm.py
++# Rôle    : Connecteur Centre de services scolaire de Montréal — workland (list_url='https://www.cssdm.gouv.qc.ca/travailler-cssdm/offres-emploi/')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .workland import WorklandConnector
++
++
++class CssdmConnector(WorklandConnector):
++    source_id = 'cssdm'
++    EMPLOYER = 'Centre de services scolaire de Montréal'
++    DEFAULT_CITY = "Montréal"
++    LIST_URL = 'https://www.cssdm.gouv.qc.ca/travailler-cssdm/offres-emploi/'

added jobka/connectors/digitalrecruiters.py +81 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/digitalrecruiters.py
++# Rôle    : Classe de plateforme Cegid Digital Recruiters (sites carrières) —
++#           sitemap + JSON-LD des pages annonce (ex. Santé Québec)
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme Cegid Digital Recruiters.
++
++Les sites carrières Digital Recruiters (ex. emplois.sante.quebec) sont rendus
++côté serveur :
++- liste  : GET <BASE>/<locale>/sitemap.xml -> URLs /annonce/<id>-<slug>
++- détail : chaque page annonce embarque un JSON-LD schema.org/JobPosting
++           complet (description, lieu, salaire, dates) — cache BD + budget.
++"""
++from __future__ import annotations
++
++import os
++import re
++
++from ..schema import JobPosting, is_quebec_location
++from . import _jsonld
++from .base import BaseConnector
++
++MAX_DETAILS = int(os.environ.get("JOBKA_DR_DETAIL_LIMIT", "120"))
++
++_AD_RE = re.compile(r"<loc>([^<]*/annonce/(\d+)[^<]*)</loc>", re.I)
++
++
++class DigitalRecruitersConnector(BaseConnector):
++    """Base Digital Recruiters — sous-classes : définir source_id, EMPLOYER,
++    BASE (ex. https://emplois.sante.quebec) et LOCALE."""
++
++    ats = "digitalrecruiters"
++    request_delay = 0.8
++
++    EMPLOYER = ""
++    BASE = ""
++    LOCALE = "fr_CA"
++    quebec_only = True
++
++    def _fetch_detail(self, url: str) -> dict:
++        html = self.get(url).text
++        node = _jsonld.extract_jobposting(html)
++        return _jsonld.jobposting_fields(node) if node else {}
++
++    def fetch(self) -> list[JobPosting]:
++        xml = self.get(f"{self.BASE}/{self.LOCALE}/sitemap.xml").text
++        out: list[JobPosting] = []
++        details_used = 0
++        seen: set[str] = set()
++        for url, eid in _AD_RE.findall(xml):
++            if eid in seen:
++                continue
++            seen.add(eid)
++            job = JobPosting(source=self.source_id, external_id=eid, url=url,
++                             employer=self.EMPLOYER, title="", ats=self.ats)
++            fields: dict = {}
++            if details_used < MAX_DETAILS:
++                fresh = [False]
++
++                def _fn(u=url, fresh=fresh):
++                    fresh[0] = True
++                    return self._fetch_detail(u)
++
++                fields = self.detail(eid, eid, _fn)
++                if fresh[0]:
++                    details_used += 1
++            if not fields:
++                continue
++            _jsonld.apply_fields(job, fields)
++            if self.quebec_only and job.city and not is_quebec_location(
++                    f"{job.city} {fields.get('region_code', '')}"):
++                if (fields.get("region_code") or "").upper() != "QC":
++                    continue
++            if not job.title:
++                continue
++            out.append(job)
++        return out

added jobka/connectors/domtar.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/domtar.py
++# Rôle    : Connecteur Domtar — successfactors (base='https://jobs.domtar.com')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .successfactors import SuccessFactorsConnector
++
++
++class DomtarConnector(SuccessFactorsConnector):
++    source_id = 'domtar'
++    EMPLOYER = 'Domtar'
++    BASE = 'https://jobs.domtar.com'

added jobka/connectors/edf_renouvelables.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/edf_renouvelables.py
++# Rôle    : Connecteur EDF Renouvelables Canada — icims (sub='cafrench-edf-re')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .icims import ICIMSConnector
++
++
++class EdfRenouvelablesConnector(ICIMSConnector):
++    source_id = 'edf_renouvelables'
++    EMPLOYER = 'EDF Renouvelables Canada'
++    SUB = 'cafrench-edf-re'

added jobka/connectors/icims.py +114 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/icims.py
++# Rôle    : Classe de plateforme iCIMS (portail carrières <org>.icims.com)
++#           — liste iframe paginée + JSON-LD des pages détail
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme iCIMS.
++
++- liste  : GET https://<sub>.icims.com/jobs/search?ss=1&in_iframe=1&pr=<page>
++           (HTML léger, liens /jobs/<id>/<slug>/job) — pagination pr=0,1,2…
++- détail : GET https://<sub>.icims.com/jobs/<id>/<slug>/job?in_iframe=1
++           -> JSON-LD schema.org/JobPosting complet (description, lieu, dates)
++
++Le lieu n'apparaissant pas toujours sur la liste, le filtre Québec est
++appliqué après lecture du détail (avec cache BD : chaque offre n'est visitée
++qu'une fois).
++"""
++from __future__ import annotations
++
++import os
++import re
++
++from ..schema import JobPosting, is_quebec_location
++from . import _jsonld
++from .base import BaseConnector
++
++MAX_DETAILS = int(os.environ.get("JOBKA_ICIMS_DETAIL_LIMIT", "150"))
++
++_LINK_RE = re.compile(r'href="https?://[^"]*?/jobs/(\d+)/([^/"]+)/job[^"]*"')
++
++
++class ICIMSConnector(BaseConnector):
++    """Base iCIMS — sous-classes : définir source_id, EMPLOYER, SUB
++    (sous-domaine complet, ex. « careers-capreit »)."""
++
++    ats = "icims"
++    request_delay = 1.0
++
++    EMPLOYER = ""
++    SUB = ""               # <SUB>.icims.com
++    quebec_only = True
++    max_pages = 15
++
++    @property
++    def _base(self) -> str:
++        return f"https://{self.SUB}.icims.com"
++
++    def _fetch_detail(self, jid: str, slug: str) -> dict:
++        html = self.get(f"{self._base}/jobs/{jid}/{slug}/job",
++                        params={"in_iframe": "1"}).text
++        node = _jsonld.extract_jobposting(html)
++        return _jsonld.jobposting_fields(node) if node else {}
++
++    def fetch(self) -> list[JobPosting]:
++        seen: dict[str, str] = {}
++        for page in range(self.max_pages):
++            html = self.get(f"{self._base}/jobs/search",
++                            params={"ss": "1", "in_iframe": "1",
++                                    "pr": str(page)}).text
++            links = _LINK_RE.findall(html)
++            new = 0
++            for jid, slug in links:
++                if jid not in seen:
++                    seen[jid] = slug
++                    new += 1
++            if not links or new == 0:
++                break
++
++        out: list[JobPosting] = []
++        details_used = 0
++        for jid, slug in seen.items():
++            if details_used >= MAX_DETAILS:
++                # budget épuisé : ne servir que le cache (sans jamais y écrire
++                # un vide) — l'offre sera reprise à la prochaine synchronisation
++                from .. import db
++                if self._detail_con is None:
++                    self._detail_con = db.connect()
++                cached = db.get_cached_detail(self._detail_con, self.source_id,
++                                              jid, jid)
++                if cached is None:
++                    continue
++                fields = cached
++            else:
++                fresh = [False]
++
++                def _fn(j=jid, s=slug, fresh=fresh):
++                    fresh[0] = True
++                    return self._fetch_detail(j, s)
++
++                fields = self.detail(jid, jid, _fn)
++                if fresh[0]:
++                    details_used += 1
++            if not fields:
++                continue
++            place = " ".join(str(fields.get(k) or "") for k in
++                             ("city", "region_code", "postal_code"))
++            if self.quebec_only and not (
++                    (fields.get("region_code") or "").upper() == "QC"
++                    or is_quebec_location(place)):
++                continue
++            job = JobPosting(
++                source=self.source_id, external_id=jid,
++                url=f"{self._base}/jobs/{jid}/{slug}/job",
++                employer=self.EMPLOYER,
++                title="", ats=self.ats,
++            )
++            _jsonld.apply_fields(job, fields)
++            if not job.title:
++                job.title = slug.replace("-", " ").strip().capitalize()
++            out.append(job)
++        return out

added jobka/connectors/lassonde.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/lassonde.py
++# Rôle    : Connecteur Industries Lassonde — adp (cid='f96ce972-7721-4c7b-99ea-9a970ddc824a', ccid='9200604444876_2')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .adp import ADPWorkforceNowConnector
++
++
++class LassondeConnector(ADPWorkforceNowConnector):
++    source_id = 'lassonde'
++    EMPLOYER = 'Industries Lassonde'
++    CID = 'f96ce972-7721-4c7b-99ea-9a970ddc824a'
++    CCID = '9200604444876_2'

added jobka/connectors/mccarthy_tetrault.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/mccarthy_tetrault.py
++# Rôle    : Connecteur McCarthy Tétrault — icims (sub='careers-mccarthyca')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .icims import ICIMSConnector
++
++
++class MccarthyTetraultConnector(ICIMSConnector):
++    source_id = 'mccarthy_tetrault'
++    EMPLOYER = 'McCarthy Tétrault'
++    SUB = 'careers-mccarthyca'

added jobka/connectors/messer_canada.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/messer_canada.py
++# Rôle    : Connecteur Messer Canada — ultipro (org='MES1005MESR', board='dbd63926-d756-486f-a254-b2a6ede1d26e')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .ultipro import UltiProConnector
++
++
++class MesserCanadaConnector(UltiProConnector):
++    source_id = 'messer_canada'
++    EMPLOYER = 'Messer Canada'
++    ORG = 'MES1005MESR'
++    BOARD = 'dbd63926-d756-486f-a254-b2a6ede1d26e'

added jobka/connectors/mitsubishi_hc_capital.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/mitsubishi_hc_capital.py
++# Rôle    : Connecteur Mitsubishi HC Capital Canada — adp (cid='b3ef4f03-f8ff-4ded-80c8-6dd5c5a224f7', ccid='9200144510729_2')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .adp import ADPWorkforceNowConnector
++
++
++class MitsubishiHcCapitalConnector(ADPWorkforceNowConnector):
++    source_id = 'mitsubishi_hc_capital'
++    EMPLOYER = 'Mitsubishi HC Capital Canada'
++    CID = 'b3ef4f03-f8ff-4ded-80c8-6dd5c5a224f7'
++    CCID = '9200144510729_2'

added jobka/connectors/msi_gestion.py +17 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/msi_gestion.py
++# Rôle    : Connecteur MSI Gestion immobilière — adp (cid='47d4e792-26e5-471c-92ac-73c60131bab8', ccid='19000101_000001')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .adp import ADPWorkforceNowConnector
++
++
++class MsiGestionConnector(ADPWorkforceNowConnector):
++    source_id = 'msi_gestion'
++    EMPLOYER = 'MSI Gestion immobilière'
++    CID = '47d4e792-26e5-471c-92ac-73c60131bab8'
++    CCID = '19000101_000001'

added jobka/connectors/njoyn.py +182 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/njoyn.py
++# Rôle    : Classe de plateforme Njoyn (CGI) — très répandu au secteur public
++#           québécois (santé, villes, sociétés d'État). HTML serveur + jeton.
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme Njoyn (CGI).
++
++Les babillards Njoyn servent un HTML rendu côté serveur :
++    <BASE>/<CL>/xweb/xweb.asp?clid=<CLID>&page=joblisting&lang=<n>
++Le serveur redirige en ajoutant un jeton de session (tbtoken) — il suffit de
++suivre la redirection. Chaque offre est un <article> (habillage moderne) ou
++une ligne de tableau (habillage classique) pointant vers
++``page=jobdetails&jobid=J####-####``.
++
++⚠ Les domaines *.njoyn.com sont derrière Radware Bot Manager : pour ces
++tenants, mettre ``USE_SCRAPFLY = True`` (contournement ASP). Les babillards
++servis sur un domaine personnalisé (ex. emplois.santemontreal.qc.ca) se
++consultent directement.
++"""
++from __future__ import annotations
++
++import hashlib
++import html as _html
++import os
++import re
++
++from ..schema import JobPosting, clean_html
++from .base import BaseConnector
++
++MAX_DETAILS = int(os.environ.get("JOBKA_NJOYN_DETAIL_LIMIT", "40"))
++
++_ARTICLE_RE = re.compile(r"<article[^>]*>(.*?)</article>", re.S | re.I)
++_H4_RE = re.compile(r"<h4[^>]*>(.*?)</h4>", re.S | re.I)
++_TIME_RE = re.compile(r"<time[^>]*>(.*?)</time>", re.S | re.I)
++_JOBID_RE = re.compile(r"jobid=(J\d{4}-\d+)", re.I)
++_FIELD_RE = re.compile(
++    r"<li><strong>([^<:]+?)(?:&nbsp;)?\s*:?\s*</strong>\s*(.*?)</li>",
++    re.S | re.I)
++_APPLY_RE = re.compile(
++    r"<a[^>]*class='btn btn-primary'[^>]*href='([^']+)'", re.I)
++_ROW_LINK_RE = re.compile(
++    r'''<a[^>]*href=["'][^"']*jobid=(J\d{4}-\d+)[^"']*["'][^>]*>(.*?)</a>''',
++    re.S | re.I)
++
++
++def _txt(s: str) -> str:
++    return re.sub(r"\s+", " ", _html.unescape(re.sub(r"<[^>]+>", " ", s or ""))).strip()
++
++
++class NjoynConnector(BaseConnector):
++    """Base Njoyn — sous-classes : définir source_id, EMPLOYER, BASE, CL, CLID."""
++
++    ats = "njoyn"
++    request_delay = 1.2
++
++    EMPLOYER = ""
++    BASE = ""              # ex. https://emplois.santemontreal.qc.ca
++    CL = "CL3"             # instance (CL2, CL3, CL4…)
++    CLID = ""              # identifiant client Njoyn
++    LANG = "2"             # 2 = français, 1 = anglais
++    USE_SCRAPFLY = False   # requis pour les domaines *.njoyn.com (Radware)
++
++    def _url(self, page: str, jobid: str = "") -> str:
++        u = (f"{self.BASE}/{self.CL}/xweb/xweb.asp?clid={self.CLID}"
++             f"&page={page}&lang={self.LANG}")
++        if jobid:
++            u += f"&jobid={jobid}"
++        return u
++
++    def _html(self, url: str) -> str:
++        if self.USE_SCRAPFLY:
++            return self.get_scrapfly(url, render_js=False)
++        text = self.get(url).text
++        if "perfdrive" in text[:4000]:   # mur anti-bot : bascule Scrapfly
++            return self.get_scrapfly(url, render_js=False)
++        return text
++
++    def _fetch_detail(self, jobid: str) -> dict:
++        page = self._html(self._url("jobdetails", jobid))
++        page = re.sub(r"<(noscript|nav|script|style)[^>]*>.*?</\1>", " ", page,
++                      flags=re.S | re.I)
++        # habillage moderne : sections <div class="row"><h2>Titre</h2>corps</div>
++        sections = re.findall(
++            r'<div class="row">\s*<h2[^>]*>(.*?)</h2>(.*?)</div>', page,
++            re.S | re.I)
++        if sections:
++            body = "\n\n".join(f"{clean_html(h)}\n{clean_html(b)}"
++                               for h, b in sections)
++            return {"description": body[:20000]}
++        # habillage classique : contenu principal après le <h1>
++        m = re.search(r"<h1[^>]*>.*?</h1>(.*?)(?:<form|<footer|</main)", page,
++                      re.S | re.I)
++        body = m.group(1) if m else ""
++        return {"description": clean_html(body)[:20000]}
++
++    def _parse_articles(self, page: str) -> list[dict]:
++        items = []
++        for art in _ARTICLE_RE.findall(page):
++            m = _JOBID_RE.search(art)
++            if not m:
++                continue
++            fields = {_txt(k).lower(): _txt(v) for k, v in _FIELD_RE.findall(art)}
++            apply_m = _APPLY_RE.search(art)
++            h4 = _H4_RE.search(art)
++            t = _TIME_RE.search(art)
++            items.append({
++                "jobid": m.group(1),
++                "title": _txt(h4.group(1)) if h4 else "",
++                "date": _txt(t.group(1)).lstrip("Le ").strip() if t else "",
++                "fields": fields,
++                "apply_url": apply_m.group(1) if apply_m else "",
++            })
++        return items
++
++    def _parse_rows(self, page: str) -> list[dict]:
++        """Habillage classique : simples liens de tableau vers jobdetails."""
++        items, seen = [], set()
++        for jobid, label in _ROW_LINK_RE.findall(page):
++            title = _txt(label)
++            if not title or title.lower().startswith(("en savoir", "postuler",
++                                                      "apply", "more")):
++                title = ""
++            if jobid in seen:
++                # garder le premier libellé non vide
++                if title:
++                    for it in items:
++                        if it["jobid"] == jobid and not it["title"]:
++                            it["title"] = title
++                continue
++            seen.add(jobid)
++            items.append({"jobid": jobid, "title": title, "date": "",
++                          "fields": {}, "apply_url": ""})
++        return [it for it in items if it["title"]]
++
++    def fetch(self) -> list[JobPosting]:
++        page = self._html(self._url("joblisting"))
++        items = self._parse_articles(page) or self._parse_rows(page)
++        out: list[JobPosting] = []
++        details_used = 0
++        for it in items:
++            f = it["fields"]
++            salary = f.get("salaire", "")
++            if re.match(r"^0\s*\$\s*-\s*0\s*\$", salary):
++                salary = ""
++            job = JobPosting(
++                source=self.source_id, external_id=it["jobid"],
++                url=self._url("jobdetails", it["jobid"]),
++                employer=f.get("établissement") or f.get("etablissement")
++                or self.EMPLOYER,
++                title=it["title"],
++                city=f.get("villes et arrondissements", "").split(",")[0],
++                location_label=f.get("villes et arrondissements", ""),
++                salary_label=salary,
++                date_posted=it["date"] or None,
++                ats=self.ats,
++            )
++            if f.get("statut de l'employé") or f.get("statut de l'employe"):
++                job.details["employment_label"] = (
++                    f.get("statut de l'employé") or f.get("statut de l'employe"))
++            if f.get("niveau de scolarité"):
++                job.requirements["scolarite"] = f["niveau de scolarité"]
++            if it["apply_url"]:
++                job.details["apply_url"] = it["apply_url"]
++            key = hashlib.sha1(
++                f"{it['title']}|{it['date']}".encode()).hexdigest()[:12]
++            if details_used < MAX_DETAILS:
++                fresh = [False]
++
++                def _fn(j=it["jobid"], fresh=fresh):
++                    fresh[0] = True
++                    return self._fetch_detail(j)
++
++                d = self.detail(it["jobid"], key, _fn)
++                if fresh[0]:
++                    details_used += 1
++                job.description = d.get("description", "")
++            out.append(job)
++        return out

added jobka/connectors/omni_hotels_gestion.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/omni_hotels_gestion.py
++# Rôle    : Connecteur Omni Hôtel Mont-Royal — icims (sub='externalmanager-omnihotels')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .icims import ICIMSConnector
++
++
++class OmniHotelsGestionConnector(ICIMSConnector):
++    source_id = 'omni_hotels_gestion'
++    EMPLOYER = 'Omni Hôtel Mont-Royal'
++    SUB = 'externalmanager-omnihotels'

added jobka/connectors/omni_hotels_horaire.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/omni_hotels_horaire.py
++# Rôle    : Connecteur Omni Hôtel Mont-Royal — icims (sub='externalhourly-omnihotels')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .icims import ICIMSConnector
++
++
++class OmniHotelsHoraireConnector(ICIMSConnector):
++    source_id = 'omni_hotels_horaire'
++    EMPLOYER = 'Omni Hôtel Mont-Royal'
++    SUB = 'externalhourly-omnihotels'

added jobka/connectors/rtc_quebec.py +19 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/rtc_quebec.py
++# Rôle    : Connecteur Réseau de transport de la Capitale — njoyn (base='https://rtc.njoyn.com', cl='CGI')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .njoyn import NjoynConnector
++
++
++class RtcQuebecConnector(NjoynConnector):
++    source_id = 'rtc_quebec'
++    EMPLOYER = 'Réseau de transport de la Capitale'
++    BASE = 'https://rtc.njoyn.com'
++    CL = 'CGI'
++    CLID = '23009'
++    USE_SCRAPFLY = True

added jobka/connectors/sante_montreal.py +18 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/sante_montreal.py
++# Rôle    : Connecteur Santé Québec — Montréal — njoyn (base='https://emplois.santemontreal.qc.ca', cl='CL3')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .njoyn import NjoynConnector
++
++
++class SanteMontrealConnector(NjoynConnector):
++    source_id = 'sante_montreal'
++    EMPLOYER = 'Santé Québec — Montréal'
++    BASE = 'https://emplois.santemontreal.qc.ca'
++    CL = 'CL3'
++    CLID = '54327'

added jobka/connectors/sante_quebec.py +16 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/sante_quebec.py
++# Rôle    : Connecteur Santé Québec — digitalrecruiters (base='https://emplois.sante.quebec')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .digitalrecruiters import DigitalRecruitersConnector
++
++
++class SanteQuebecConnector(DigitalRecruitersConnector):
++    source_id = 'sante_quebec'
++    EMPLOYER = 'Santé Québec'
++    BASE = 'https://emplois.sante.quebec'

added jobka/connectors/stl_laval.py +19 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/stl_laval.py
++# Rôle    : Connecteur Société de transport de Laval — njoyn (base='https://lavaltransit.njoyn.com', cl='CL2')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .njoyn import NjoynConnector
++
++
++class StlLavalConnector(NjoynConnector):
++    source_id = 'stl_laval'
++    EMPLOYER = 'Société de transport de Laval'
++    BASE = 'https://lavaltransit.njoyn.com'
++    CL = 'CL2'
++    CLID = '60406'
++    USE_SCRAPFLY = True

added jobka/connectors/successfactors.py +161 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/successfactors.py
++# Rôle    : Classe de plateforme SAP SuccessFactors (sites carrières « Career
++#           Site Builder » / jobs2web) — sitemap RSS ou XML + microdonnées
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme SAP SuccessFactors (Career Site Builder, ex-jobs2web).
++
++Deux formats de « sitemap.xml » selon la génération du site :
++- RSS 2.0 : <item> avec titre « Poste (VILLE, PROV, PAYS[, CP]) » ET la
++  description HTML complète — tout en une requête ;
++- urlset  : URLs /job/<Ville>-<titre>-<Prov>-<CP>/<id>/ + lastmod ; la
++  description est lue sur la page détail (microdonnées itemprop, avec cache
++  BD + budget). Le filtre Québec s'appuie sur la province/le code postal
++  présents dans le slug — aucun détail hors Québec n'est visité.
++"""
++from __future__ import annotations
++
++import html as _html
++import os
++import re
++import urllib.parse
++
++from ..schema import JobPosting, clean_html, is_quebec_location
++from .base import BaseConnector
++
++MAX_DETAILS = int(os.environ.get("JOBKA_SF_DETAIL_LIMIT", "150"))
++
++_ITEM_RE = re.compile(r"<item>(.*?)</item>", re.S | re.I)
++_TAG_RE = {t: re.compile(rf"<{t}>(.*?)</{t}>", re.S | re.I)
++           for t in ("title", "link", "description", "pubDate")}
++_LOC_RE = re.compile(r"<loc>([^<]+)</loc>\s*(?:<lastmod>([^<]+)</lastmod>)?",
++                     re.I)
++_RSS_TITLE_RE = re.compile(r"^(.*)\(([^,()]+),\s*([^,()]+?),\s*([A-Z]{2})"
++                           r"(?:,\s*([^()]+))?\)\s*$", re.S)
++_QC_PROV = {"QC", "QUÉBEC", "QUEBEC"}
++_QC_SLUG_RE = re.compile(r"-(qu[ée]b|qc)-|-[ghj]\d[a-z][- ]?\d[a-z]\d/", re.I)
++_ITEMPROP_RE = re.compile(
++    r'<(?:span|div|p)[^>]*itemprop="description"[^>]*>(.*?)'
++    r"</(?:span|div|p)>", re.S | re.I)
++_META_RE = re.compile(
++    r'itemprop="(datePosted|addressLocality|postalCode|employmentType)"'
++    r'[^>]*content="([^"]*)"', re.I)
++
++
++class SuccessFactorsConnector(BaseConnector):
++    """Base SuccessFactors CSB — sous-classes : définir source_id, EMPLOYER,
++    BASE (URL du site carrières, ex. https://jobs.bombardier.com)."""
++
++    ats = "successfactors"
++    request_delay = 1.0
++
++    EMPLOYER = ""
++    BASE = ""
++    quebec_only = True
++
++    def _cdata(self, s: str) -> str:
++        s = (s or "").strip()
++        if s.startswith("<![CDATA["):
++            s = s[9:]
++            if s.endswith("]]>"):
++                s = s[:-3]
++        return _html.unescape(s.strip())
++
++    # -- format RSS ------------------------------------------------------------
++    def _fetch_rss(self, xml: str) -> list[JobPosting]:
++        out = []
++        for raw in _ITEM_RE.findall(xml):
++            def g(tag: str, raw=raw) -> str:
++                m = _TAG_RE[tag].search(raw)
++                return self._cdata(m.group(1)) if m else ""
++            vals = {t: g(t) for t in _TAG_RE}
++            link = vals["link"]
++            m_id = re.search(r"/(\d+)/?$", link)
++            if not link or not m_id:
++                continue
++            title, city, prov = vals["title"], "", ""
++            m = _RSS_TITLE_RE.match(vals["title"])
++            if m:
++                title = m.group(1).strip()
++                city, prov = m.group(2).strip(), m.group(3).strip()
++            if self.quebec_only and not (
++                    prov.upper() in _QC_PROV
++                    or (city and is_quebec_location(city))):
++                continue
++            job = JobPosting(
++                source=self.source_id, external_id=m_id.group(1),
++                url=link, employer=self.EMPLOYER, title=title,
++                description=clean_html(self._cdata(vals["description"])),
++                city=city.title() if city.isupper() else city,
++                date_posted=vals["pubDate"] or None,
++                ats=self.ats,
++            )
++            out.append(job)
++        return out
++
++    # -- format urlset ---------------------------------------------------------
++    def _fetch_detail(self, url: str) -> dict:
++        html = self.get(url).text
++        d: dict = {}
++        chunks = _ITEMPROP_RE.findall(html)
++        if chunks:
++            d["description"] = clean_html("\n".join(chunks))[:20000]
++        for prop, content in _META_RE.findall(html):
++            d[prop] = content
++        return d
++
++    def _fetch_urlset(self, xml: str) -> list[JobPosting]:
++        out = []
++        details_used = 0
++        for loc, lastmod in _LOC_RE.findall(xml):
++            loc = _html.unescape(loc.strip())
++            m = re.search(r"/job/([^/]+)/(\d+)/?$", loc)
++            if not m:
++                continue
++            slug_raw, eid = m.group(1), m.group(2)
++            slug = urllib.parse.unquote(slug_raw)
++            if self.quebec_only and not _QC_SLUG_RE.search(slug + "/"):
++                continue
++            city = slug.split("-")[0]
++            postal = ""
++            m_cp = re.search(r"([GHJ]\d[A-Z])[- ]?(\d[A-Z]\d)", slug, re.I)
++            if m_cp:
++                postal = f"{m_cp.group(1)}{m_cp.group(2)}".upper()
++            job = JobPosting(
++                source=self.source_id, external_id=eid, url=loc,
++                employer=self.EMPLOYER,
++                title=" ".join(slug.split("-")[1:-3]) or slug,
++                city=city, postal_code=postal,
++                date_posted=lastmod or None,
++                ats=self.ats,
++            )
++            if details_used < MAX_DETAILS:
++                fresh = [False]
++
++                def _fn(u=loc, fresh=fresh):
++                    fresh[0] = True
++                    return self._fetch_detail(u)
++
++                d = self.detail(eid, lastmod or eid, _fn)
++                if fresh[0]:
++                    details_used += 1
++                if d.get("description"):
++                    job.description = d["description"]
++                if d.get("datePosted"):
++                    job.date_posted = d["datePosted"]
++                if d.get("addressLocality"):
++                    job.city = d["addressLocality"]
++                if d.get("employmentType"):
++                    job.details["employment_label"] = d["employmentType"]
++            out.append(job)
++        return out
++
++    def fetch(self) -> list[JobPosting]:
++        xml = self.get(f"{self.BASE}/sitemap.xml").text
++        if "<rss" in xml[:300].lower():
++            return self._fetch_rss(xml)
++        return self._fetch_urlset(xml)

added jobka/connectors/taleo.py +185 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/taleo.py
++# Rôle    : Classe de plateforme Oracle Taleo Enterprise (careersection REST)
++#           — un employeur = une sous-classe (HOST, SECTION, EMPLOYER)
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme Oracle Taleo Enterprise.
++
++Les sections carrières Taleo exposent une API REST publique (celle que le
++frontal jobsearch.ftl consomme lui-même) :
++- amorce : GET  https://<host>.taleo.net/careersection/<section>/jobsearch.ftl
++           (cookies de session + numéro de portail « portal=NNN » dans le HTML)
++- liste  : POST https://<host>.taleo.net/careersection/rest/jobboard/searchjobs
++           ?lang=fr&portal=<NNN>  body JSON {pageNo, sorting…}
++           -> {requisitionList:[{jobId, contestNo, column:[...],
++               locationsColumns:[i]}], pagingData:{totalCount, pageSize}}
++- détail : la page jobdetail.ftl?job=<contestNo> (description HTML rendue
++           côté serveur) — visitée avec cache BD et budget.
++
++Le contenu des colonnes dépend de la configuration du portail : on repère la
++colonne des lieux via `locationsColumns` et on suppose la première colonne
++comme titre (constaté sur tous les portails testés).
++"""
++from __future__ import annotations
++
++import hashlib
++import json
++import os
++import re
++
++from ..schema import JobPosting, clean_html, is_quebec_location
++from .base import BaseConnector
++
++MAX_DETAILS = int(os.environ.get("JOBKA_TALEO_DETAIL_LIMIT", "120"))
++
++_DESC_RE = re.compile(
++    r'requisitionDescriptionInterface[^>]*>', re.I)
++
++
++class TaleoConnector(BaseConnector):
++    """Base Taleo — sous-classes : définir source_id, EMPLOYER, HOST, SECTION
++    (et PORTAL si le numéro n'apparaît pas dans le HTML d'amorce)."""
++
++    ats = "taleo"
++    request_delay = 1.0
++
++    EMPLOYER = ""
++    HOST = ""              # <host>.taleo.net
++    SECTION = "2"          # /careersection/<section>/
++    PORTAL = ""            # numéro de portail (sinon extrait du HTML)
++    LANG = "fr"
++    quebec_only = True
++    max_pages = 40
++
++    @property
++    def _base(self) -> str:
++        return f"https://{self.HOST}.taleo.net/careersection"
++
++    def _careers_url(self) -> str:
++        return f"{self._base}/{self.SECTION}/jobsearch.ftl?lang={self.LANG}"
++
++    def _bootstrap(self) -> str:
++        """Cookies de session + numéro de portail."""
++        html = self.get(self._careers_url()).text
++        if self.PORTAL:
++            return str(self.PORTAL)
++        m = re.search(r"portal=(\d+)", html)
++        if not m:
++            raise RuntimeError(f"taleo {self.HOST}: numéro de portail introuvable")
++        return m.group(1)
++
++    @staticmethod
++    def _parse_locations(raw: str) -> list[str]:
++        """La colonne des lieux est un JSON sérialisé : '["Québec-Malartic"]'."""
++        if not raw:
++            return []
++        try:
++            val = json.loads(raw)
++            if isinstance(val, list):
++                return [str(v) for v in val]
++        except ValueError:
++            pass
++        return [raw]
++
++    def _keep(self, locations: list[str]) -> bool:
++        if not self.quebec_only:
++            return True
++        return any(is_quebec_location(l.replace("-", " ")) for l in locations)
++
++    def _fetch_detail(self, contest_no: str) -> dict:
++        html = self.get(f"{self._base}/{self.SECTION}/jobdetail.ftl",
++                        params={"job": contest_no, "lang": self.LANG}).text
++        # description : contenu des blocs « requisitionDescriptionInterface »
++        # (rendus côté serveur), sinon la zone principale de la page
++        chunks = re.findall(
++            r'<(?:span|div)[^>]*id="requisitionDescriptionInterface[^"]*"[^>]*>'
++            r"(.*?)</(?:span|div)>", html, re.S)
++        text = clean_html("\n".join(c for c in chunks if len(c) > 40))
++        if not text:
++            m = re.search(r'<div[^>]*class="[^"]*mastercontentpanel[^"]*"[^>]*>'
++                          r"(.*?)<!--", html, re.S)
++            text = clean_html(m.group(1)) if m else ""
++        return {"description": text}
++
++    def fetch(self) -> list[JobPosting]:
++        portal = self._bootstrap()
++        url = (f"{self._base}/rest/jobboard/searchjobs"
++               f"?lang={self.LANG}&portal={portal}")
++        body_base = {
++            "multilineEnabled": False,
++            "sortingSelection": {"sortBySelectionParam": "3",
++                                 "ascendingSortingOrder": "false"},
++            "fieldData": {"fields": {"KEYWORD": "", "LOCATION": ""},
++                          "valid": True},
++            "filterSelectionParam": {"searchFilterSelections": [
++                {"id": "POSTING_DATE", "selectedValues": []},
++                {"id": "LOCATION", "selectedValues": []}]},
++            "advancedSearchFiltersSelectionParam":
++                {"searchFilterSelections": []},
++        }
++        out: list[JobPosting] = []
++        details_used = 0
++        page = 1
++        seen: set[str] = set()
++        while page <= self.max_pages:
++            data = self.post(url, json={**body_base, "pageNo": page},
++                             headers={"tz": "GMT-04:00"}).json()
++            reqs = data.get("requisitionList") or []
++            if not reqs and page == 1 and self.LANG != "en":
++                # section carrières anglophone seulement : rebasculer en « en »
++                self.LANG = "en"
++                portal = self._bootstrap()
++                url = (f"{self._base}/rest/jobboard/searchjobs"
++                       f"?lang={self.LANG}&portal={portal}")
++                continue
++            if not reqs:
++                break
++            for r in reqs:
++                cols = r.get("column") or []
++                loc_idx = (r.get("locationsColumns") or [1])[0]
++                title = str(cols[0]) if cols else ""
++                locations = self._parse_locations(
++                    str(cols[loc_idx])) if len(cols) > loc_idx else []
++                date_raw = str(cols[-1]) if len(cols) > 2 else ""
++                contest = str(r.get("contestNo") or r.get("jobId") or "")
++                if not contest or contest in seen or not title:
++                    continue
++                seen.add(contest)
++                if not self._keep(locations):
++                    continue
++                job = JobPosting(
++                    source=self.source_id, external_id=contest,
++                    url=(f"{self._base}/{self.SECTION}/jobdetail.ftl"
++                         f"?job={contest}&lang={self.LANG}"),
++                    employer=self.EMPLOYER,
++                    title=title,
++                    location_label=locations[0].replace("-", ", ")
++                    if locations else "",
++                    date_posted=date_raw or None,
++                    ats=self.ats,
++                )
++                key = hashlib.sha1(
++                    f"{title}|{date_raw}".encode()).hexdigest()[:12]
++                if details_used < MAX_DETAILS:
++                    fresh = [False]
++
++                    def _fn(c=contest, fresh=fresh):
++                        fresh[0] = True
++                        return self._fetch_detail(c)
++
++                    d = self.detail(contest, key, _fn)
++                    if fresh[0]:
++                        details_used += 1
++                    job.description = d.get("description", "")
++                out.append(job)
++            paging = data.get("pagingData") or {}
++            total = int(paging.get("totalCount") or 0)
++            size = int(paging.get("pageSize") or len(reqs) or 25)
++            if total and page * size >= total:
++                break
++            page += 1
++        return out

added jobka/connectors/ultipro.py +162 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/ultipro.py
++# Rôle    : Classe de plateforme UKG Pro Recruiting (UltiPro) — API JSON
++#           publique LoadSearchResults, un employeur = une sous-classe
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme UKG Pro Recruiting (recruiting.ultipro.com).
++
++API JSON publique (celle du frontal) :
++- liste  : POST https://recruiting.ultipro.com/<ORG>/JobBoard/<BOARD>/
++                JobBoardView/LoadSearchResults
++           body {opportunitySearch:{Top, Skip, OrderBy…}}
++           -> {opportunities:[{Id, Title, RequisitionNumber, BriefDescription,
++               FullTime, JobCategoryName, PostedDate, Locations:[{Address}]}],
++               totalCount}
++- détail : POST .../JobBoardView/LoadOpportunity {opportunityId} -> Description
++           complète (HTML). Visité avec cache BD + budget.
++"""
++from __future__ import annotations
++
++import os
++import re
++
++from ..schema import JobPosting, clean_html, is_quebec_location
++from .base import BaseConnector
++
++PAGE_SIZE = 50
++MAX_DETAILS = int(os.environ.get("JOBKA_ULTIPRO_DETAIL_LIMIT", "80"))
++
++
++class UltiProConnector(BaseConnector):
++    """Base UltiPro — sous-classes : définir source_id, EMPLOYER, ORG, BOARD."""
++
++    ats = "ultipro"
++    request_delay = 0.8
++
++    EMPLOYER = ""
++    ORG = ""      # ex. MES1005MESR
++    BOARD = ""    # GUID du JobBoard
++    quebec_only = True
++    max_pages = 20
++
++    @property
++    def _board_url(self) -> str:
++        return f"https://recruiting.ultipro.com/{self.ORG}/JobBoard/{self.BOARD}"
++
++    @staticmethod
++    def _loc_fields(loc: dict) -> tuple[str, str, str, str]:
++        """(ville, province, code postal, libellé) d'un objet Location."""
++        addr = loc.get("Address") or {}
++        city = addr.get("City") or ""
++        state = ((addr.get("State") or {}).get("Code")
++                 if isinstance(addr.get("State"), dict)
++                 else addr.get("State")) or ""
++        postal = addr.get("PostalCode") or ""
++        label = loc.get("LocalizedDescription") or city
++        return city, str(state), postal, label
++
++    def _keep(self, locations: list[dict]) -> bool:
++        if not self.quebec_only:
++            return True
++        for loc in locations:
++            city, state, postal, label = self._loc_fields(loc)
++            if state.upper() == "QC" or re.match(r"^[GHJ]\d[A-Z]", postal.upper()):
++                return True
++            if is_quebec_location(f"{label} {city}"):
++                return True
++        return False
++
++    _DESC_RE = re.compile(r'"Description"\s*:\s*"((?:[^"\\]|\\.)*)"')
++
++    def _fetch_detail(self, opp_id: str) -> dict:
++        """La page OpportunityDetail embarque l'objet JSON de l'offre —
++        on en extrait la description complète (chaîne JSON échappée)."""
++        html = self.get(f"{self._board_url}/OpportunityDetail",
++                        params={"opportunityId": opp_id}).text
++        best = ""
++        for m in self._DESC_RE.finditer(html):
++            try:
++                import json as _json
++                val = _json.loads(f'"{m.group(1)}"')
++            except ValueError:
++                continue
++            if len(val) > len(best):
++                best = val
++        return {"description": clean_html(best)}
++
++    def fetch(self) -> list[JobPosting]:
++        url = f"{self._board_url}/JobBoardView/LoadSearchResults"
++        out: list[JobPosting] = []
++        details_used = 0
++        skip = 0
++        for _ in range(self.max_pages):
++            body = {
++                "opportunitySearch": {
++                    "Top": PAGE_SIZE, "Skip": skip, "QueryString": "",
++                    "OrderBy": [{"Value": "postedDateDesc",
++                                 "PropertyName": "PostedDate",
++                                 "Ascending": False}],
++                    "Filters": [],
++                },
++                "matchCriteria": {"PreferredJobs": [], "Educations": [],
++                                  "LicenseAndCertifications": [], "Skills": [],
++                                  "hasNoLicenses": False, "SkippedSkills": []},
++            }
++            data = self.post(url, json=body,
++                             headers={"Accept": "application/json"}).json()
++            opps = data.get("opportunities") or []
++            if not opps:
++                break
++            for o in opps:
++                locations = o.get("Locations") or []
++                if not self._keep(locations):
++                    continue
++                oid = str(o.get("Id") or "")
++                if not oid:
++                    continue
++                city = label = postal = ""
++                for loc in locations:
++                    c, s, p, lbl = self._loc_fields(loc)
++                    if s.upper() == "QC" or is_quebec_location(f"{lbl} {c}"):
++                        city, postal, label = c or lbl, p, lbl
++                        break
++                job = JobPosting(
++                    source=self.source_id, external_id=oid,
++                    url=f"{self._board_url}/OpportunityDetail?opportunityId={oid}",
++                    employer=self.EMPLOYER,
++                    title=o.get("Title") or "",
++                    city=city, postal_code=postal, location_label=label,
++                    date_posted=o.get("PostedDate") or None,
++                    ats=self.ats,
++                )
++                job.details["requisition"] = o.get("RequisitionNumber") or ""
++                if o.get("JobCategoryName"):
++                    job.details["team"] = o["JobCategoryName"]
++                if o.get("FullTime") is True:
++                    job.details["employment_label"] = "Temps plein"
++                elif o.get("FullTime") is False:
++                    job.details["employment_label"] = "Temps partiel"
++                key = (o.get("PostedDate") or "") + (o.get("Title") or "")[:40]
++                if details_used < MAX_DETAILS:
++                    fresh = [False]
++
++                    def _fn(i=oid, fresh=fresh):
++                        fresh[0] = True
++                        return self._fetch_detail(i)
++
++                    d = self.detail(oid, key, _fn)
++                    if fresh[0]:
++                        details_used += 1
++                    job.description = d.get("description", "")
++                if not job.description:
++                    job.description = clean_html(o.get("BriefDescription") or "")
++                out.append(job)
++            skip += PAGE_SIZE
++            total = int(data.get("totalCount") or 0)
++            if total and skip >= total:
++                break
++        return out

added jobka/connectors/ville_gatineau.py +19 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/ville_gatineau.py
++# Rôle    : Connecteur Ville de Gatineau — njoyn (base='https://gatineau.njoyn.com', cl='CL2')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .njoyn import NjoynConnector
++
++
++class VilleGatineauConnector(NjoynConnector):
++    source_id = 'ville_gatineau'
++    EMPLOYER = 'Ville de Gatineau'
++    BASE = 'https://gatineau.njoyn.com'
++    CL = 'CL2'
++    CLID = '27082'
++    USE_SCRAPFLY = True

added jobka/connectors/ville_terrebonne.py +19 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/ville_terrebonne.py
++# Rôle    : Connecteur Ville de Terrebonne — njoyn (base='https://clients.njoyn.com', cl='cl4')
++#           [généré par scripts/gen_connectors.py, flux vérifié le 2026-08-18]
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++from .njoyn import NjoynConnector
++
++
++class VilleTerrebonneConnector(NjoynConnector):
++    source_id = 'ville_terrebonne'
++    EMPLOYER = 'Ville de Terrebonne'
++    BASE = 'https://clients.njoyn.com'
++    CL = 'cl4'
++    CLID = '71764'
++    USE_SCRAPFLY = True

added jobka/connectors/workland.py +90 −0

++# =============================================================================
++# Job·Ka — Groupe KA
++# Auteur  : Simon-Pierre Boucher
++# Contact : contact@spboucher.ai
++# Fichier : jobka/connectors/workland.py
++# Rôle    : Classe de plateforme Workland / Atlas (ATS québécois utilisé par
++#           les centres de services scolaires, PME…) — liste rendue côté
++#           serveur sur la page carrières de l'employeur
++# Créé    : 2026-08-18   Modifié : 2026-08-18
++# =============================================================================
++"""Plateforme Workland (atlas.workland.com).
++
++Les pages atlas.workland.com sont une SPA Angular sans données côté serveur
++(le bloc JSON-LD reste vide même après rendu) : on s'appuie donc sur la page
++carrières de l'employeur (LIST_URL), rendue côté serveur, qui liste les liens
++https://atlas.workland.com/work/<id>/<slug> — typiquement dans un tableau
++« titre | date limite | lien » (ex. CSSDM). Le titre et la date limite
++proviennent de la ligne du tableau ; à défaut, du slug.
++"""
++from __future__ import annotations
++
++import html as _html
++import re
++
++from ..schema import JobPosting
++from .base import BaseConnector
++
++_WORK_RE = re.compile(
++    r"https?://atlas\.workland\.com/work/(\d+)/([a-z0-9-]+)", re.I)
++_ROW_RE = re.compile(r"<tr[^>]*>(.*?)</tr>", re.S | re.I)
++_TD_RE = re.compile(r"<td[^>]*>(.*?)</td>", re.S | re.I)
++_DATE_RE = re.compile(r"\d{1,2}(?:er)?\s+[a-zû]{3,9}\.?\s+\d{4}|\d{4}-\d{2}-\d{2}",
++                      re.I)
++
++
++def _txt(s: str) -> str:
++    return re.sub(r"\s+", " ",
++                  _html.unescape(re.sub(r"<[^>]+>", " ", s or ""))).strip()
++
++
++class WorklandConnector(BaseConnector):
++    """Base Workland — sous-classes : définir source_id, EMPLOYER, LIST_URL
++    (page carrières de l'employeur qui liste les liens atlas)."""
++
++    ats = "workland"
++    request_delay = 1.0
++
++    EMPLOYER = ""
++    LIST_URL = ""
++    DEFAULT_CITY = ""      # ville par défaut (ex. « Montréal » pour le CSSDM)
++
++    def fetch(self) -> list[JobPosting]:
++        page = self.get(self.LIST_URL).text
++        out: list[JobPosting] = []
++        seen: set[str] = set()
++
++        # 1) lignes de tableau « titre | date limite | lien atlas »
++        for row in _ROW_RE.findall(page):
++            m = _WORK_RE.search(row)
++            if not m or m.group(1) in seen:
++                continue
++            tds = [_txt(td) for td in _TD_RE.findall(row)]
++            title = next((t for t in tds if t and not _DATE_RE.fullmatch(t)), "")
++            deadline = next((t for t in tds if t and _DATE_RE.fullmatch(t)), None)
++            if not title:
++                title = m.group(2).replace("-", " ").capitalize()
++            seen.add(m.group(1))
++            out.append(JobPosting(
++                source=self.source_id, external_id=m.group(1),
++                url=f"https://atlas.workland.com/work/{m.group(1)}/{m.group(2)}",
++                employer=self.EMPLOYER, title=title,
++                city=self.DEFAULT_CITY,
++                date_deadline=deadline,
++                ats=self.ats,
++            ))
++
++        # 2) liens atlas hors tableau (autres habillages)
++        for wid, slug in _WORK_RE.findall(page):
++            if wid in seen:
++                continue
++            seen.add(wid)
++            out.append(JobPosting(
++                source=self.source_id, external_id=wid,
++                url=f"https://atlas.workland.com/work/{wid}/{slug}",
++                employer=self.EMPLOYER,
++                title=slug.replace("-", " ").capitalize(),
++                city=self.DEFAULT_CITY,
++                ats=self.ats,
++            ))
++        return out

modified scripts/gen_connectors.py +39 −0

				`@@ -54,6 +54,20 @@ ORG_ATTR = {"lever": "ORG", "greenhouse": "BOARD",`
54	"smartrecruiters": "COMPANY", "ashby": "ORG", "workable": "ORG",	54	"smartrecruiters": "COMPANY", "ashby": "ORG", "workable": "ORG",
55	"recruitee": "ORG", "breezy": "ORG", "bamboohr": "ORG"}	55	"recruitee": "ORG", "breezy": "ORG", "bamboohr": "ORG"}
56		56
		57	+# ATS à attributs multiples (secteur public/parapublic et grands employeurs) :
		58	+# feed[clé] -> attribut de classe. Voir data/feeds-public.json.
		59	+EXTRA_ATTRS = {
		60	+ "taleo": [("host", "HOST"), ("section", "SECTION"), ("portal", "PORTAL")],
		61	+ "njoyn": [("base", "BASE"), ("cl", "CL"), ("clid", "CLID"),
		62	+ ("use_scrapfly", "USE_SCRAPFLY")],
		63	+ "ultipro": [("org", "ORG"), ("board", "BOARD")],
		64	+ "icims": [("sub", "SUB")],
		65	+ "successfactors": [("base", "BASE")],
		66	+ "adp": [("cid", "CID"), ("ccid", "CCID")],
		67	+ "digitalrecruiters": [("base", "BASE")],
		68	+ "workland": [("list_url", "LIST_URL")],
		69	+}
		70	+
57	CAREERS_URL = {	71	CAREERS_URL = {
58	"workday": "https://{tenant}.{host}.myworkdayjobs.com/{site}",	72	"workday": "https://{tenant}.{host}.myworkdayjobs.com/{site}",
59	"lever": "https://jobs.lever.co/{org}",	73	"lever": "https://jobs.lever.co/{org}",
				`@@ -64,8 +78,28 @@ CAREERS_URL = {`
64	"recruitee": "https://{org}.recruitee.com",	78	"recruitee": "https://{org}.recruitee.com",
65	"breezy": "https://{org}.breezy.hr",	79	"breezy": "https://{org}.breezy.hr",
66	"bamboohr": "https://{org}.bamboohr.com/careers",	80	"bamboohr": "https://{org}.bamboohr.com/careers",
		81	+ "taleo": "https://{host}.taleo.net/careersection/{section}/jobsearch.ftl?lang=fr",
		82	+ "njoyn": "{base}/{cl}/xweb/xweb.asp?clid={clid}&page=joblisting&lang=2",
		83	+ "ultipro": "https://recruiting.ultipro.com/{org}/JobBoard/{board}",
		84	+ "icims": "https://{sub}.icims.com/jobs/search?ss=1",
		85	+ "successfactors": "{base}",
		86	+ "adp": ("https://workforcenow.adp.com/mascsr/default/mdf/recruitment/"
		87	+ "recruitment.html?cid={cid}&ccId={ccid}&lang=fr_CA"),
		88	+ "digitalrecruiters": "{base}",
		89	+ "workland": "{list_url}",
67	}	90	}
68		91
		92	+PLATFORMS.update({
		93	+ "taleo": ("taleo", "TaleoConnector"),
		94	+ "njoyn": ("njoyn", "NjoynConnector"),
		95	+ "ultipro": ("ultipro", "UltiProConnector"),
		96	+ "icims": ("icims", "ICIMSConnector"),
		97	+ "successfactors": ("successfactors", "SuccessFactorsConnector"),
		98	+ "adp": ("adp", "ADPWorkforceNowConnector"),
		99	+ "digitalrecruiters": ("digitalrecruiters", "DigitalRecruitersConnector"),
		100	+ "workland": ("workland", "WorklandConnector"),
		101	+})
		102	+
69		103
70	def slugify(s: str) -> str:	104	def slugify(s: str) -> str:
71	s = unicodedata.normalize("NFD", s)	105	s = unicodedata.normalize("NFD", s)
				`@@ -110,6 +144,11 @@ def gen_one(feed: dict) -> tuple[str, str] \| None:`
110	body = (f" TENANT = {feed['tenant']!r}\n"	144	body = (f" TENANT = {feed['tenant']!r}\n"
111	f" HOST = {feed['host']!r}\n"	145	f" HOST = {feed['host']!r}\n"
112	f" SITE = {feed['site']!r}\n")	146	f" SITE = {feed['site']!r}\n")
		147	+ elif ats in EXTRA_ATTRS:
		148	+ pairs = [(k, a) for k, a in EXTRA_ATTRS[ats]
		149	+ if feed.get(k) not in (None, "")]
		150	+ detail = ", ".join(f"{a.lower()}={feed[k]!r}" for k, a in pairs[:2])
		151	+ body = "".join(f" {a} = {feed[k]!r}\n" for k, a in pairs)
113	else:	152	else:
114	attr = ORG_ATTR[ats]	153	attr = ORG_ATTR[ats]
115	detail = f"org « {feed['org']} »"	154	detail = f"org « {feed['org']} »"
116		155

spb/job-ka

8 nouveaux ATS (Taleo, Njoyn, UltiPro, iCIMS, SuccessFactors, ADP WFN, Digital Recruiters, Workland) + 23 employeurs publics/parapublics et grands employeurs QC

Theme