diff --git a/api/core/scrapers/__init__.py b/api/core/scrapers/__init__.py index 35de67b..401f9bc 100644 --- a/api/core/scrapers/__init__.py +++ b/api/core/scrapers/__init__.py @@ -26,8 +26,9 @@ class ScraperInput(BaseModel): class Scraper: - def __init__(self, site: Site): + def __init__(self, site: Site, url: str): self.site = site + self.url = url def scrape(self, scraper_input: ScraperInput) -> JobResponse: ... diff --git a/api/core/scrapers/indeed/__init__.py b/api/core/scrapers/indeed/__init__.py index f04432a..4bfaa26 100644 --- a/api/core/scrapers/indeed/__init__.py +++ b/api/core/scrapers/indeed/__init__.py @@ -3,6 +3,7 @@ import json from typing import Optional, Tuple, List import tls_client +import urllib.parse from bs4 import BeautifulSoup from bs4.element import Tag from fastapi import status @@ -25,9 +26,8 @@ class IndeedScraper(Scraper): Initializes IndeedScraper with the Indeed job search url """ site = Site(Site.INDEED) - super().__init__(site) - self.url = "https://www.indeed.com/jobs" - self.job_url = "https://www.indeed.com/viewjob?jk=" + url = "https://www.indeed.com" + super().__init__(site, url) self.jobs_per_page = 15 self.seen_urls = set() @@ -60,7 +60,7 @@ class IndeedScraper(Scraper): if sc_values: params["sc"] = "0kf:" + "".join(sc_values) + ";" - response = session.get(self.url, params=params) + response = session.get(self.url + "/jobs", params=params) if ( response.status_code != status.HTTP_200_OK @@ -82,10 +82,10 @@ class IndeedScraper(Scraper): ): raise Exception("No jobs found.") - for job in jobs["metaData"]["mosaicProviderJobCardsModel"]["results"]: - job_url = f'{self.job_url}{job["jobkey"]}' + def process_job(job) -> Optional[JobPost]: + job_url = f'{self.url}/jobs/viewjob?jk={job["jobkey"]}' if job_url in self.seen_urls: - continue + return None snippet_html = BeautifulSoup(job["snippet"], "html.parser") @@ -110,11 +110,8 @@ class IndeedScraper(Scraper): job_type = IndeedScraper.get_job_type(job) timestamp_seconds = job["pubDate"] / 1000 date_posted = datetime.fromtimestamp(timestamp_seconds) - li_elements = snippet_html.find_all("li") - if li_elements: - description = " ".join(li.text for li in li_elements) - else: - description = None + description = self.get_description(job_url, session) + first_li = snippet_html.find("li") job_post = JobPost( title=job["normTitle"], @@ -131,6 +128,10 @@ class IndeedScraper(Scraper): date_posted=date_posted, job_url=job_url, ) + return job_post + + for job in jobs["metaData"]["mosaicProviderJobCardsModel"]["results"]: + job_post = process_job(job) job_list.append(job_post) return job_list, total_num_jobs @@ -190,6 +191,27 @@ class IndeedScraper(Scraper): ) return job_response + def get_description(self, job_page_url: str, session: tls_client.Session) -> str: + """ + Retrieves job description by going to the job page url + :param job_page_url: + :param session: + :return: description + """ + parsed_url = urllib.parse.urlparse(job_page_url) + params = urllib.parse.parse_qs(parsed_url.query) + jk_value = params.get("jk", [None])[0] + formatted_url = f"{self.url}/viewjob?jk={jk_value}&spa=1" + + response = session.get(formatted_url, allow_redirects=True) + + raw_description = response.json()["body"]["jobInfoWrapperModel"][ + "jobInfoModel" + ]["sanitizedJobDescription"] + soup = BeautifulSoup(raw_description, "html.parser") + text_content = " ".join(soup.get_text().split()).strip() + return text_content + @staticmethod def get_job_type(job: dict) -> Optional[JobType]: """ diff --git a/api/core/scrapers/linkedin/__init__.py b/api/core/scrapers/linkedin/__init__.py index c0deb78..7e12143 100644 --- a/api/core/scrapers/linkedin/__init__.py +++ b/api/core/scrapers/linkedin/__init__.py @@ -15,10 +15,8 @@ class LinkedInScraper(Scraper): Initializes LinkedInScraper with the LinkedIn job search url """ site = Site(Site.LINKEDIN) - super().__init__(site) - - self.url = "https://www.linkedin.com/jobs/search/" - self.job_url = "https://www.linkedin.com/jobs/view/" + url = "https://www.linkedin.com" + super().__init__(site, url) def scrape(self, scraper_input: ScraperInput) -> JobResponse: """ @@ -57,7 +55,9 @@ class LinkedInScraper(Scraper): params = {k: v for k, v in params.items() if v is not None} print(params) - response = session.get(self.url, params=params, allow_redirects=True) + response = session.get( + f"{self.url}/jobs/search", params=params, allow_redirects=True + ) if response.status_code != status.HTTP_200_OK: return JobResponse( @@ -82,7 +82,7 @@ class LinkedInScraper(Scraper): job_id = ( data_entity_urn.split(":")[-1] if data_entity_urn else "N/A" ) - job_url = f"{self.job_url}{job_id}" + job_url = f"{self.url}/jobs/view/{job_id}" if job_url in seen_urls: continue seen_urls.add(job_url) diff --git a/api/core/scrapers/ziprecruiter/__init__.py b/api/core/scrapers/ziprecruiter/__init__.py index 695d700..0bedc03 100644 --- a/api/core/scrapers/ziprecruiter/__init__.py +++ b/api/core/scrapers/ziprecruiter/__init__.py @@ -5,6 +5,7 @@ from urllib.parse import urlparse, parse_qs import tls_client from fastapi import status from bs4 import BeautifulSoup +from bs4.element import Tag from concurrent.futures import ThreadPoolExecutor, Future from api.core.jobs import JobPost @@ -14,14 +15,13 @@ import math class ZipRecruiterScraper(Scraper): - url = "https://www.ziprecruiter.com" - def __init__(self): """ Initializes LinkedInScraper with the ZipRecruiter job search url """ site = Site(Site.ZIP_RECRUITER) - super().__init__(site) + url = "https://www.ziprecruiter.com" + super().__init__(site, url) self.jobs_per_page = 20 self.seen_urls = set() @@ -84,11 +84,11 @@ class ZipRecruiterScraper(Scraper): job_posts = soup.find_all("div", {"class": "job_content"}) def process_job(job: Tag) -> Optional[JobPost]: - ''' + """ Parses a job from the job content tag :param job: BeautifulSoup Tag for one job post :return JobPost - ''' + """ job_url = job.find("a", {"class": "job_link"})["href"] if job_url in self.seen_urls: return None @@ -201,7 +201,7 @@ class ZipRecruiterScraper(Scraper): job_description_div = soup_job.find("div", {"class": "job_description"}) if job_description_div: - return job_description_div.text.strip("\n"), response.url + return job_description_div.text.strip(), response.url return None, response.url @staticmethod