enh: full description param (#85)

README.md

@@ -67,6 +67,7 @@ Optional
 ├── job_type (enum): fulltime, parttime, internship, contract
 ├── proxy (str): in format 'http://user:pass@host:port' or [https, socks]
 ├── is_remote (bool)
+├── full_description (bool): fetches full description for Indeed / LinkedIn (much slower)
 ├── results_wanted (int): number of job results to retrieve for each site specified in 'site_type'
 ├── easy_apply (bool): filters for jobs that are hosted on LinkedIn
 ├── country_indeed (enum): filters the country on Indeed (see below for correct spelling)

examples/JobSpy_AllSites.py

@@ -2,12 +2,11 @@ from jobspy import scrape_jobs
 import pandas as pd
 
 jobs: pd.DataFrame = scrape_jobs(
-    site_name=["indeed", "linkedin", "zip_recruiter"],
+    site_name=["indeed", "linkedin", "zip_recruiter", "glassdoor"],
     search_term="software engineer",
     location="Dallas, TX",
-    results_wanted=50,  # be wary the higher it is, the more likey you'll get blocked (rotating proxy should work tho)
+    results_wanted=25,  # be wary the higher it is, the more likey you'll get blocked (rotating proxy can help tho)
     country_indeed="USA",
-    offset=25  # start jobs from an offset (use if search failed and want to continue)
     # proxy="http://jobspy:5a4vpWtj8EeJ2hoYzk@ca.smartproxy.com:20001",
 )
 

examples/JobSpy_LongScrape.py

@@ -0,0 +1,77 @@
+from jobspy import scrape_jobs
+import pandas as pd
+import os
+import time
+
+# creates csv a new filename if the jobs.csv already exists.
+csv_filename = "jobs.csv"
+counter = 1
+while os.path.exists(csv_filename):
+    csv_filename = f"jobs_{counter}.csv"
+    counter += 1
+
+# results wanted and offset
+results_wanted = 1000
+offset = 0
+
+all_jobs = []
+
+# max retries
+max_retries = 3
+
+# nuumber of results at each iteration
+results_in_each_iteration = 30
+
+while len(all_jobs) < results_wanted:
+    retry_count = 0
+    while retry_count < max_retries:
+        print("Doing from", offset, "to", offset + results_in_each_iteration, "jobs")
+        try:
+            jobs = scrape_jobs(
+                site_name=["indeed"],
+                search_term="software engineer",
+                # New York, NY
+                # Dallas, TX
+
+                # Los Angeles, CA
+                location="Los Angeles, CA",
+                results_wanted=min(results_in_each_iteration, results_wanted - len(all_jobs)),
+                country_indeed="USA",
+                offset=offset,
+                # proxy="http://jobspy:5a4vpWtj8EeJ2hoYzk@ca.smartproxy.com:20001",
+            )
+
+            # Add the scraped jobs to the list
+            all_jobs.extend(jobs.to_dict('records'))
+
+            # Increment the offset for the next page of results
+            offset += results_in_each_iteration
+
+            # Add a delay to avoid rate limiting (you can adjust the delay time as needed)
+            print(f"Scraped {len(all_jobs)} jobs")
+            print("Sleeping secs", 100 * (retry_count + 1))
+            time.sleep(100 * (retry_count + 1))  # Sleep for 2 seconds between requests
+
+            break  # Break out of the retry loop if successful
+        except Exception as e:
+            print(f"Error: {e}")
+            retry_count += 1
+            print("Sleeping secs before retry", 100 * (retry_count + 1))
+            time.sleep(100 * (retry_count + 1))
+            if retry_count >= max_retries:
+                print("Max retries reached. Exiting.")
+                break
+
+# DataFrame from the collected job data
+jobs_df = pd.DataFrame(all_jobs)
+
+# Formatting
+pd.set_option("display.max_columns", None)
+pd.set_option("display.max_rows", None)
+pd.set_option("display.width", None)
+pd.set_option("display.max_colwidth", 50)
+
+print(jobs_df)
+
+jobs_df.to_csv(csv_filename, index=False)
+print(f"Outputted to {csv_filename}")

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "python-jobspy"
-version = "1.1.33"
+version = "1.1.35"
 description = "Job scraper for LinkedIn, Indeed, Glassdoor & ZipRecruiter"
 authors = ["Zachary Hampton <zachary@bunsly.com>", "Cullen Watson <cullen@bunsly.com>"]
 homepage = "https://github.com/Bunsly/JobSpy"

src/jobspy/__init__.py

@@ -40,6 +40,7 @@ def scrape_jobs(
     country_indeed: str = "usa",
     hyperlinks: bool = False,
     proxy: Optional[str] = None,
+    full_description: Optional[bool] = False,
     offset: Optional[int] = 0,
 ) -> pd.DataFrame:
     """
@@ -74,6 +75,7 @@ def scrape_jobs(
         is_remote=is_remote,
         job_type=job_type,
         easy_apply=easy_apply,
+        full_description=full_description,
         results_wanted=results_wanted,
         offset=offset,
     )

src/jobspy/scrapers/__init__.py

@@ -19,6 +19,7 @@ class ScraperInput(BaseModel):
     is_remote: bool = False
     job_type: Optional[JobType] = None
     easy_apply: bool = None  # linkedin
+    full_description: bool = False
     offset: int = 0
 
     results_wanted: int = 15

src/jobspy/scrapers/glassdoor/__init__.py

@@ -5,8 +5,12 @@ jobspy.scrapers.glassdoor
 This module contains routines to scrape Glassdoor.
 """
 import json
-from typing import Optional, Any
+import requests
+from bs4 import BeautifulSoup
+from typing import Optional
 from datetime import datetime, timedelta
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from ..utils import count_urgent_words, extract_emails_from_text
 
 from .. import Scraper, ScraperInput, Site
 from ..exceptions import GlassdoorException
@@ -66,14 +70,29 @@ class GlassdoorScraper(Scraper):
         jobs_data = res_json["data"]["jobListings"]["jobListings"]
 
         jobs = []
-        for i, job in enumerate(jobs_data):
+        with ThreadPoolExecutor(max_workers=self.jobs_per_page) as executor:
-            job_url = res_json["data"]["jobListings"]["jobListingSeoLinks"][
+            future_to_job_data = {executor.submit(self.process_job, job): job for job in jobs_data}
-                "linkItems"
+            for future in as_completed(future_to_job_data):
-            ][i]["url"]
+                job_data = future_to_job_data[future]
+                try:
+                    job_post = future.result()
+                    if job_post:
+                        jobs.append(job_post)
+                except Exception as exc:
+                    raise GlassdoorException(f'Glassdoor generated an exception: {exc}')
+
+        return jobs, self.get_cursor_for_page(
+            res_json["data"]["jobListings"]["paginationCursors"], page_num + 1
+        )
+
+    def process_job(self, job_data):
+        """Processes a single job and fetches its description."""
+        job_id = job_data["jobview"]["job"]["listingId"]
+        job_url = f'{self.url}/job-listing/?jl={job_id}'
         if job_url in self.seen_urls:
-                continue
+            return None
         self.seen_urls.add(job_url)
-            job = job["jobview"]
+        job = job_data["jobview"]
         title = job["job"]["jobTitleText"]
         company_name = job["header"]["employerNameFromSearch"]
         location_name = job["header"].get("locationName", "")
@@ -89,20 +108,24 @@ class GlassdoorScraper(Scraper):
 
         compensation = self.parse_compensation(job["header"])
 
-            job = JobPost(
+        try:
+            description = self.fetch_job_description(job_id)
+        except Exception as e :
+            description = None
+
+        job_post = JobPost(
             title=title,
             company_name=company_name,
             date_posted=date_posted,
             job_url=job_url,
             location=location,
             compensation=compensation,
-                is_remote=is_remote
+            is_remote=is_remote,
-            )
+            description=description,
-            jobs.append(job)
+            emails=extract_emails_from_text(description) if description else None,
-
+            num_urgent_words=count_urgent_words(description) if description else None,
-        return jobs, self.get_cursor_for_page(
-            res_json["data"]["jobListings"]["paginationCursors"], page_num + 1
         )
+        return job_post
 
     def scrape(self, scraper_input: ScraperInput) -> JobResponse:
         """
@@ -110,6 +133,7 @@ class GlassdoorScraper(Scraper):
         :param scraper_input: Information about job search criteria.
         :return: JobResponse containing a list of jobs.
         """
+        scraper_input.results_wanted = min(900, scraper_input.results_wanted)
         self.country = scraper_input.country
         self.url = self.country.get_url()
 
@@ -143,6 +167,43 @@ class GlassdoorScraper(Scraper):
 
         return JobResponse(jobs=all_jobs)
 
+    def fetch_job_description(self, job_id):
+        """Fetches the job description for a single job ID."""
+        url = f"{self.url}/graph"
+        body = [
+            {
+                "operationName": "JobDetailQuery",
+                "variables": {
+                    "jl": job_id,
+                    "queryString": "q",
+                    "pageTypeEnum": "SERP"
+                },
+                "query": """
+                query JobDetailQuery($jl: Long!, $queryString: String, $pageTypeEnum: PageTypeEnum) {
+                    jobview: jobView(
+                        listingId: $jl
+                        contextHolder: {queryString: $queryString, pageTypeEnum: $pageTypeEnum}
+                    ) {
+                        job {
+                            description
+                            __typename
+                        }
+                        __typename
+                    }
+                }
+                """
+            }
+        ]
+        response = requests.post(url, json=body, headers=GlassdoorScraper.headers())
+        if response.status_code != 200:
+            return None
+        data = response.json()[0]
+        desc = data['data']['jobview']['job']['description']
+        soup = BeautifulSoup(desc, 'html.parser')
+        description = soup.get_text(separator='\n')
+
+        return description
+
     @staticmethod
     def parse_compensation(data: dict) -> Optional[Compensation]:
         pay_period = data.get("payPeriod")

src/jobspy/scrapers/indeed/__init__.py

@@ -78,7 +78,7 @@ class IndeedScraper(Scraper):
         if sc_values:
             params["sc"] = "0kf:" + "".join(sc_values) + ";"
         try:
-            session = create_session(self.proxy, is_tls=True)
+            session = create_session(self.proxy)
             response = session.get(
                 f"{self.url}/jobs",
                 headers=self.get_headers(),
@@ -140,7 +140,8 @@ class IndeedScraper(Scraper):
             date_posted = datetime.fromtimestamp(timestamp_seconds)
             date_posted = date_posted.strftime("%Y-%m-%d")
 
-            description = self.get_description(job_url)
+            description = self.get_description(job_url) if scraper_input.full_description else None
+
             with io.StringIO(job["snippet"]) as f:
                 soup_io = BeautifulSoup(f, "html.parser")
                 li_elements = soup_io.find_all("li")
@@ -246,7 +247,7 @@ class IndeedScraper(Scraper):
             return None
 
         soup = BeautifulSoup(job_description, "html.parser")
-        text_content = " ".join(soup.get_text(separator=" ").split()).strip()
+        text_content = "\n".join(soup.stripped_strings)
 
         return text_content
 

src/jobspy/scrapers/linkedin/__init__.py

@@ -111,7 +111,7 @@ class LinkedInScraper(Scraper):
 
                 # Call process_job directly without threading
                 try:
-                    job_post = self.process_job(job_card, job_url)
+                    job_post = self.process_job(job_card, job_url, scraper_input.full_description)
                     if job_post:
                         job_list.append(job_post)
                 except Exception as e:
@@ -123,7 +123,7 @@ class LinkedInScraper(Scraper):
         job_list = job_list[: scraper_input.results_wanted]
         return JobResponse(jobs=job_list)
 
-    def process_job(self, job_card: Tag, job_url: str) -> Optional[JobPost]:
+    def process_job(self, job_card: Tag, job_url: str, full_descr: bool) -> Optional[JobPost]:
         salary_tag = job_card.find('span', class_='job-search-card__salary-info')
 
         compensation = None
@@ -160,7 +160,7 @@ class LinkedInScraper(Scraper):
             if metadata_card
             else None
         )
-        date_posted = None
+        date_posted = description = job_type = None
         if datetime_tag and "datetime" in datetime_tag.attrs:
             datetime_str = datetime_tag["datetime"]
             try:
@@ -169,9 +169,8 @@ class LinkedInScraper(Scraper):
                 date_posted = None
         benefits_tag = job_card.find("span", class_="result-benefits__text")
         benefits = " ".join(benefits_tag.get_text().split()) if benefits_tag else None
-
+        if full_descr:
-        # removed to speed up scraping
+            description, job_type = self.get_job_description(job_url)
-        # description, job_type = self.get_job_description(job_url)
 
         return JobPost(
             title=title,
@@ -182,10 +181,10 @@ class LinkedInScraper(Scraper):
             job_url=job_url,
             compensation=compensation,
             benefits=benefits,
-            # job_type=job_type,
+            job_type=job_type,
-            # description=description,
+            description=description,
-            # emails=extract_emails_from_text(description) if description else None,
+            emails=extract_emails_from_text(description) if description else None,
-            # num_urgent_words=count_urgent_words(description) if description else None,
+            num_urgent_words=count_urgent_words(description) if description else None,
         )
 
     def get_job_description(
@@ -214,7 +213,7 @@ class LinkedInScraper(Scraper):
 
         description = None
         if div_content:
-            description = " ".join(div_content.get_text().split()).strip()
+            description = "\n".join(line.strip() for line in div_content.get_text(separator="\n").splitlines() if line.strip())
 
         def get_job_type(
             soup_job_type: BeautifulSoup,

src/jobspy/scrapers/ziprecruiter/__init__.py

@@ -10,6 +10,7 @@ import re
 from datetime import datetime, date
 from typing import Optional, Tuple, Any
 
+import requests
 from bs4 import BeautifulSoup
 from concurrent.futures import ThreadPoolExecutor
 
@@ -26,6 +27,8 @@ class ZipRecruiterScraper(Scraper):
         """
         site = Site(Site.ZIP_RECRUITER)
         self.url = "https://www.ziprecruiter.com"
+        self.session = create_session(proxy)
+        self.get_cookies()
         super().__init__(site, proxy=proxy)
 
         self.jobs_per_page = 20
@@ -44,12 +47,10 @@ class ZipRecruiterScraper(Scraper):
         if continue_token:
             params["continue"] = continue_token
         try:
-            session = create_session(self.proxy, is_tls=True)
+            response = self.session.get(
-            response = session.get(
                 f"https://api.ziprecruiter.com/jobs-app/jobs",
                 headers=self.headers(),
                 params=self.add_params(scraper_input),
-                timeout_seconds=10,
             )
             if response.status_code != 200:
                 raise ZipRecruiterException(
@@ -108,7 +109,7 @@ class ZipRecruiterScraper(Scraper):
 
         description = BeautifulSoup(
             job.get("job_description", "").strip(), "html.parser"
-        ).get_text()
+        ).get_text(separator="\n")
 
         company = job["hiring_company"].get("name") if "hiring_company" in job else None
         country_value = "usa" if job.get("job_country") == "US" else "canada"
@@ -156,6 +157,11 @@ class ZipRecruiterScraper(Scraper):
             num_urgent_words=count_urgent_words(description) if description else None,
         )
 
+    def get_cookies(self):
+        url="https://api.ziprecruiter.com/jobs-app/event"
+        data="event_type=session&logged_in=false&number_of_retry=1&property=model%3AiPhone&property=os%3AiOS&property=locale%3Aen_us&property=app_build_number%3A4734&property=app_version%3A91.0&property=manufacturer%3AApple&property=timestamp%3A2024-01-12T12%3A04%3A42-06%3A00&property=screen_height%3A852&property=os_version%3A16.6.1&property=source%3Ainstall&property=screen_width%3A393&property=device_model%3AiPhone%2014%20Pro&property=brand%3AApple"
+        self.session.post(url, data=data, headers=ZipRecruiterScraper.headers())
+
     @staticmethod
     def get_job_type_enum(job_type_str: str) -> list[JobType] | None:
         for job_type in JobType:
@@ -195,12 +201,16 @@ class ZipRecruiterScraper(Scraper):
     @staticmethod
     def headers() -> dict:
         """
-        Returns headers needed for ZipRecruiter API requests
+        Returns headers needed for requests
         :return: dict - Dictionary containing headers
         """
         return {
-            'Host': 'api.ziprecruiter.com',
+            "Host": "api.ziprecruiter.com",
-            'accept': '*/*',
+            "accept": "*/*",
-            'authorization': 'Basic YTBlZjMyZDYtN2I0Yy00MWVkLWEyODMtYTI1NDAzMzI0YTcyOg==',
+            "x-zr-zva-override": "100000000;vid:ZT1huzm_EQlDTVEc",
-            'Cookie': '__cf_bm=DZ7eJOw6lka.Bwy5jLeDqWanaZ8BJlVAwaXrmcbYnxM-1701505132-0-AfGaVIfTA2kJlmleK14o722vbVwpZ+4UxFznsWv+guvzXSpD9KVEy/+pNzvEZUx88yaEShJwGt3/EVjhHirX/ASustKxg47V/aXRd2XIO2QN; zglobalid=61f94830-1990-4130-b222-d9d0e09c7825.57da9ea9581c.656ae86b; ziprecruiter_browser=018188e0-045b-4ad7-aa50-627a6c3d43aa; ziprecruiter_session=5259b2219bf95b6d2299a1417424bc2edc9f4b38; zva=100000000%3Bvid%3AZWroa0x_F1KEeGeU'
+            "x-pushnotificationid": "0ff4983d38d7fc5b3370297f2bcffcf4b3321c418f5c22dd152a0264707602a0",
+            "x-deviceid": "D77B3A92-E589-46A4-8A39-6EF6F1D86006",
+            "user-agent": "Job Search/87.0 (iPhone; CPU iOS 16_6_1 like Mac OS X)",
+            "authorization": "Basic YTBlZjMyZDYtN2I0Yy00MWVkLWEyODMtYTI1NDAzMzI0YTcyOg==",
+            "accept-language": "en-US,en;q=0.9",
         }