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.

80 lines
2.7 KiB
Python

"""Indian Health Service Dentistry adapter — single-page table, keyword post-filter."""
from __future__ import annotations
import re
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
_BASE_URL = "https://www.ihs.gov"
_KEYWORDS = ["orthodontist", "orthodontic", "dentist"]
@register("ihs")
class IHSAdapter(ManifestDrivenAdapter):
def collect_cards(self, page, manifest):
"""Extract all rows via JS, combine city+state into location, filter by keywords."""
try:
page.wait_for_selector("table.cf_grid tbody tr", timeout=15000, state="attached")
except Exception:
logger.debug("[ihs] Table wait timed out")
raw_data = page.evaluate("""() => {
const rows = document.querySelectorAll('table.cf_grid tbody tr');
return Array.from(rows).map(row => {
const titleEl = row.querySelector('td:nth-child(2) a');
const city = (row.querySelector('td:nth-child(3)')?.textContent || '').trim();
const state = (row.querySelector('td:nth-child(4)')?.textContent || '').trim();
return {
title: (titleEl?.textContent || '').trim(),
url: titleEl?.getAttribute('href') || '',
city: city,
state: state
};
}).filter(r => r.title);
}""")
cards = []
for item in raw_data:
title = item["title"]
if not self._matches_keywords(title):
continue
# Strip HPSA score prefix: "(N-HPSA Score) CityName" → "CityName"
city = re.sub(r"^\([^)]+\)\s*", "", item["city"])
location = f"{city}, {item['state']}".strip(", ")
url = item["url"]
if url.startswith("/"):
url = _BASE_URL + url
cards.append(RawJobCard(
title=title,
company="Indian Health Service",
location=location or None,
url=url or None,
))
logger.info(
f"[ihs] Keyword filter: {len(cards)}/{len(raw_data)} cards matched"
)
return cards
def paginate(self, page, page_index: int, config) -> bool:
"""Single page — no 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)