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.
95 lines
3.7 KiB
Python
95 lines
3.7 KiB
Python
"""NATIVE HEALTH adapter — SmartRecruiters career page, Show More AJAX, 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
|
|
|
|
_MAX_SHOW_MORE = 10 # max "Show more jobs" clicks to prevent infinite loops
|
|
_KEYWORDS = ["orthodontist", "orthodontic", "dentist"]
|
|
|
|
|
|
@register("nativehealth")
|
|
class NativeHealthAdapter(ManifestDrivenAdapter):
|
|
|
|
def collect_cards(self, page, manifest) -> list[RawJobCard]:
|
|
"""Click all 'Show more jobs' links, then extract and keyword-filter all jobs."""
|
|
try:
|
|
page.wait_for_selector(
|
|
"a[href*='smartrecruiters.com/NATIVEHEALTH/']",
|
|
timeout=15000,
|
|
state="attached",
|
|
)
|
|
except Exception:
|
|
logger.debug("[nativehealth] Initial jobs wait timed out")
|
|
|
|
# Click "Show more jobs" links until none remain (AJAX loads more into same page)
|
|
for i in range(_MAX_SHOW_MORE):
|
|
show_more = page.query_selector("a:text('Show more jobs')")
|
|
if not show_more:
|
|
break
|
|
|
|
prev_count = len(
|
|
page.query_selector_all("a[href*='smartrecruiters.com/NATIVEHEALTH/']")
|
|
)
|
|
show_more.click()
|
|
|
|
try:
|
|
page.wait_for_function(
|
|
f"document.querySelectorAll(\"a[href*='smartrecruiters.com/NATIVEHEALTH/']\").length > {prev_count}",
|
|
timeout=10000,
|
|
)
|
|
logger.debug(f"[nativehealth] Show more {i + 1}: loaded more jobs")
|
|
except Exception:
|
|
logger.debug(f"[nativehealth] Show more {i + 1}: no new jobs appeared")
|
|
break
|
|
|
|
# Extract all jobs via JS, getting location from h3 heading of each group
|
|
raw_data = page.evaluate("""() => {
|
|
const links = document.querySelectorAll("a[href*='smartrecruiters.com/NATIVEHEALTH/']");
|
|
return Array.from(links).map(a => {
|
|
// Walk up: a → li → ul (job list) → location group div → h3
|
|
const locationDiv = a.closest('li')?.parentElement?.parentElement;
|
|
const location = locationDiv?.querySelector('h3')?.textContent?.trim() || null;
|
|
return {
|
|
title: (a.querySelector('h4')?.textContent || '').trim(),
|
|
employment: (a.querySelector('p')?.textContent || '').trim(),
|
|
url: a.href,
|
|
location
|
|
};
|
|
}).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="NATIVE HEALTH",
|
|
location=item["location"] or None,
|
|
employment_type=item["employment"] or None,
|
|
url=item["url"] or None,
|
|
))
|
|
|
|
logger.info(
|
|
f"[nativehealth] Keyword filter: {len(cards)}/{len(raw_data)} cards matched"
|
|
)
|
|
return cards
|
|
|
|
def paginate(self, page, page_index: int, config) -> bool:
|
|
"""Single page after Show More clicks — no further pagination."""
|
|
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)
|