You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
59 lines
2.0 KiB
Python
59 lines
2.0 KiB
Python
"""Consulting BFrench Jobs adapter — JazzHR single-page table, keyword filter."""
|
|
from __future__ import annotations
|
|
|
|
from loguru import logger
|
|
|
|
from gimme_job.adapters.base import ManifestDrivenAdapter
|
|
from gimme_job.adapters.registry import register
|
|
from gimme_job.models.dto import JobPostingCandidate, RawJobCard
|
|
|
|
_KEYWORDS = ["orthodontist", "orthodontic", "dentist"]
|
|
|
|
|
|
@register("bfrench")
|
|
class BFrenchAdapter(ManifestDrivenAdapter):
|
|
|
|
def collect_cards(self, page, manifest) -> list[RawJobCard]:
|
|
"""Extract all job rows from the table and filter by keywords."""
|
|
raw_data = page.evaluate("""() => {
|
|
const links = document.querySelectorAll('a[href*="/apply/jobs/details/"]');
|
|
return Array.from(links).map(a => {
|
|
const row = a.closest('tr');
|
|
const cells = row ? Array.from(row.querySelectorAll('td')) : [];
|
|
return {
|
|
title: a.innerText.trim(),
|
|
url: a.href,
|
|
location: cells[1]?.innerText?.trim() || null
|
|
};
|
|
}).filter(j => j.title);
|
|
}""")
|
|
|
|
cards = []
|
|
for item in raw_data:
|
|
if not self._matches_keywords(item["title"]):
|
|
continue
|
|
cards.append(RawJobCard(
|
|
title=item["title"],
|
|
company="Consulting BFrench",
|
|
location=item["location"],
|
|
url=item["url"],
|
|
))
|
|
|
|
logger.info(
|
|
f"[bfrench] Keyword filter: {len(cards)}/{len(raw_data)} cards matched"
|
|
)
|
|
return cards
|
|
|
|
def paginate(self, page, page_index: int, config) -> bool:
|
|
return False
|
|
|
|
def normalize(self, raw: RawJobCard) -> JobPostingCandidate:
|
|
return super().normalize(raw)
|
|
|
|
@staticmethod
|
|
def _matches_keywords(title: str | None) -> bool:
|
|
if not title:
|
|
return False
|
|
title_lower = title.lower()
|
|
return any(kw in title_lower for kw in _KEYWORDS)
|