enh: full description param (#85)

README.md

@@ -4,11 +4,8 @@
 
 **Not technical?** Try out the web scraping tool on our site at [usejobspy.com](https://usejobspy.com).
 
-*Looking to build a data-focused software product?* **[Book a call](https://calendly.com/bunsly/15min)** *to
-work with us.*  
-
-Check out another project we wrote: ***[HomeHarvest](https://github.com/Bunsly/HomeHarvest)** – a Python package
-for real estate scraping*
+*Looking to build a data-focused software product?* **[Book a call](https://bunsly.com/)** *to
+work with us.*
 
 ## Features
 
@@ -62,7 +59,7 @@ zip_recruiter Software Developer                 TEKsystems        Phoenix
 
 ```plaintext
 Required
-├── site_type (List[enum]): linkedin, zip_recruiter, indeed
+├── site_type (List[enum]): linkedin, zip_recruiter, indeed, glassdoor
 └── search_term (str)
 Optional
 ├── location (int)
@@ -70,6 +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)
@@ -107,21 +105,22 @@ The following exceptions may be raised when using JobSpy:
 * `LinkedInException`
 * `IndeedException`
 * `ZipRecruiterException`
+* `GlassdoorException`
 
 ## Supported Countries for Job Searching
 
 ### **LinkedIn**
 
-LinkedIn searches globally & uses only the `location` parameter.
+LinkedIn searches globally & uses only the `location` parameter. You can only fetch 1000 jobs max from the LinkedIn endpoint we're using
 
 ### **ZipRecruiter**
 
 ZipRecruiter searches for jobs in **US/Canada** & uses only the `location` parameter.
 
-### **Indeed**
+### **Indeed / Glassdoor**
 
 Indeed & Glassdoor supports most countries, but the `country_indeed` parameter is required. Additionally, use the `location`
-parameter to narrow down the location, e.g. city & state if necessary.
+parameter to narrow down the location, e.g. city & state if necessary. 
 
 You can specify the following countries when searching on Indeed (use the exact name, * indicates support for Glassdoor):
 
@@ -145,6 +144,7 @@ You can specify the following countries when searching on Indeed (use the exact
 | Venezuela            | Vietnam      |            |                |
 
 
+Glassdoor can only fetch 900 jobs from the endpoint we're using on a given search.
 ## Frequently Asked Questions
 
 ---

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",
 )
 
@@ -28,4 +27,4 @@ print("outputted to jobs.csv")
 # jobs.to_xlsx('jobs.xlsx', index=False)
 
 # 4: display in Jupyter Notebook (1. pip install jupyter 2. jupyter notebook)
-# display(jobs)
+# display(jobs)

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 = "1.1.24"
+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,
     )
@@ -163,6 +165,7 @@ def scrape_jobs(
             "site",
             "title",
             "company",
+            "company_url",
             "location",
             "job_type",
             "date_posted",

src/jobspy/jobs/__init__.py

@@ -55,18 +55,24 @@ class JobType(Enum):
 
 
 class Country(Enum):
-    ARGENTINA = ("argentina", "com.ar")
+    """
+    Gets the subdomain for Indeed and Glassdoor.
+    The second item in the tuple is the subdomain for Indeed
+    The third item in the tuple is the subdomain (and tld if there's a ':' separator) for Glassdoor
+    """
+
+    ARGENTINA = ("argentina", "ar", "com.ar")
     AUSTRALIA = ("australia", "au", "com.au")
     AUSTRIA = ("austria", "at", "at")
     BAHRAIN = ("bahrain", "bh")
-    BELGIUM = ("belgium", "be", "nl:be")
+    BELGIUM = ("belgium", "be", "fr:be")
     BRAZIL = ("brazil", "br", "com.br")
     CANADA = ("canada", "ca", "ca")
     CHILE = ("chile", "cl")
     CHINA = ("china", "cn")
     COLOMBIA = ("colombia", "co")
     COSTARICA = ("costa rica", "cr")
-    CZECHREPUBLIC = ("czech republic", "cz")
+    CZECHREPUBLIC = ("czech republic,czechia", "cz")
     DENMARK = ("denmark", "dk")
     ECUADOR = ("ecuador", "ec")
     EGYPT = ("egypt", "eg")
@@ -112,8 +118,8 @@ class Country(Enum):
     TURKEY = ("turkey", "tr")
     UKRAINE = ("ukraine", "ua")
     UNITEDARABEMIRATES = ("united arab emirates", "ae")
-    UK = ("uk", "uk", "co.uk")
-    USA = ("usa", "www", "com")
+    UK = ("uk,united kingdom", "uk", "co.uk")
+    USA = ("usa,us,united states", "www", "com")
     URUGUAY = ("uruguay", "uy")
     VENEZUELA = ("venezuela", "ve")
     VIETNAM = ("vietnam", "vn")
@@ -121,7 +127,7 @@ class Country(Enum):
     # internal for ziprecruiter
     US_CANADA = ("usa/ca", "www")
 
-    # internal for linkeind
+    # internal for linkedin
     WORLDWIDE = ("worldwide", "www")
 
     @property
@@ -147,7 +153,8 @@ class Country(Enum):
         """Convert a string to the corresponding Country enum."""
         country_str = country_str.strip().lower()
         for country in cls:
-            if country.value[0] == country_str:
+            country_names = country.value[0].split(',')
+            if country_str in country_names:
                 return country
         valid_countries = [country.value for country in cls]
         raise ValueError(
@@ -167,10 +174,13 @@ class Location(BaseModel):
         if self.state:
             location_parts.append(self.state)
         if self.country and self.country not in (Country.US_CANADA, Country.WORLDWIDE):
-            if self.country.value[0] in ("usa", "uk"):
-                location_parts.append(self.country.value[0].upper())
+            country_name = self.country.value[0]
+            if "," in country_name:
+                country_name = country_name.split(",")[0]
+            if country_name in ("usa", "uk"):
+                location_parts.append(country_name.upper())
             else:
-                location_parts.append(self.country.value[0].title())
+                location_parts.append(country_name.title())
         return ", ".join(location_parts)
 
 
@@ -181,6 +191,10 @@ class CompensationInterval(Enum):
     DAILY = "daily"
     HOURLY = "hourly"
 
+    @classmethod
+    def get_interval(cls, pay_period):
+        return cls[pay_period].value if pay_period in cls.__members__ else None
+
 
 class Compensation(BaseModel):
     interval: Optional[CompensationInterval] = None
@@ -196,6 +210,8 @@ class JobPost(BaseModel):
     location: Optional[Location]
 
     description: str | None = None
+    company_url: str | None = None
+
     job_type: list[JobType] | None = None
     compensation: Compensation | None = None
     date_posted: date | None = None

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

@@ -4,17 +4,17 @@ jobspy.scrapers.glassdoor
 
 This module contains routines to scrape Glassdoor.
 """
-import math
-import time
-import re
 import json
-from datetime import datetime, date
-from typing import Optional, Tuple, 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
-from ..utils import count_urgent_words, extract_emails_from_text, create_session
+from ..utils import create_session
 from ...jobs import (
     JobPost,
     Compensation,
@@ -22,7 +22,6 @@ from ...jobs import (
     Location,
     JobResponse,
     JobType,
-    Country,
 )
 
 
@@ -31,7 +30,7 @@ class GlassdoorScraper(Scraper):
         """
         Initializes GlassdoorScraper with the Glassdoor job search url
         """
-        site = Site(Site.ZIP_RECRUITER)
+        site = Site(Site.GLASSDOOR)
         super().__init__(site, proxy=proxy)
 
         self.url = None
@@ -49,15 +48,12 @@ class GlassdoorScraper(Scraper):
     ) -> (list[JobPost], str | None):
         """
         Scrapes a page of Glassdoor for jobs with scraper_input criteria
-        :param scraper_input:
-        :return: jobs found on page
-        :return: cursor for next page
         """
         try:
             payload = self.add_payload(
                 scraper_input, location_id, location_type, page_num, cursor
             )
-            session = create_session(self.proxy, is_tls=False)
+            session = create_session(self.proxy, is_tls=False, has_retry=True)
             response = session.post(
                 f"{self.url}/graph", headers=self.headers(), timeout=10, data=payload
             )
@@ -74,45 +70,70 @@ class GlassdoorScraper(Scraper):
         jobs_data = res_json["data"]["jobListings"]["jobListings"]
 
         jobs = []
-        for i, job in enumerate(jobs_data):
-            job_url = res_json["data"]["jobListings"]["jobListingSeoLinks"][
-                "linkItems"
-            ][i]["url"]
-            job = job["jobview"]
-            title = job["job"]["jobTitleText"]
-            company_name = job["header"]["employerNameFromSearch"]
-            location_name = job["header"].get("locationName", "")
-            location_type = job["header"].get("locationType", "")
-            is_remote = False
-            location = None
-
-            if location_type == "S":
-                is_remote = True
-            else:
-                location = self.parse_location(location_name)
-
-            compensation = self.parse_compensation(job["header"])
-
-            job = JobPost(
-                title=title,
-                company_name=company_name,
-                job_url=job_url,
-                location=location,
-                compensation=compensation,
-                is_remote=is_remote,
-            )
-            jobs.append(job)
+        with ThreadPoolExecutor(max_workers=self.jobs_per_page) as executor:
+            future_to_job_data = {executor.submit(self.process_job, job): job for job in jobs_data}
+            for future in as_completed(future_to_job_data):
+                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:
+            return None
+        self.seen_urls.add(job_url)
+        job = job_data["jobview"]
+        title = job["job"]["jobTitleText"]
+        company_name = job["header"]["employerNameFromSearch"]
+        location_name = job["header"].get("locationName", "")
+        location_type = job["header"].get("locationType", "")
+        age_in_days = job["header"].get("ageInDays")
+        is_remote, location = False, None
+        date_posted = (datetime.now() - timedelta(days=age_in_days)).date() if age_in_days else None
+
+        if location_type == "S":
+            is_remote = True
+        else:
+            location = self.parse_location(location_name)
+
+        compensation = self.parse_compensation(job["header"])
+
+        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,
+            description=description,
+            emails=extract_emails_from_text(description) if description else None,
+            num_urgent_words=count_urgent_words(description) if description else None,
+        )
+        return job_post
+
     def scrape(self, scraper_input: ScraperInput) -> JobResponse:
         """
         Scrapes Glassdoor for jobs with scraper_input criteria.
         :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()
 
@@ -146,6 +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")
@@ -158,15 +216,8 @@ class GlassdoorScraper(Scraper):
         interval = None
         if pay_period == "ANNUAL":
             interval = CompensationInterval.YEARLY
-        elif pay_period == "MONTHLY":
-            interval = CompensationInterval.MONTHLY
-        elif pay_period == "WEEKLY":
-            interval = CompensationInterval.WEEKLY
-        elif pay_period == "DAILY":
-            interval = CompensationInterval.DAILY
-        elif pay_period == "HOURLY":
-            interval = CompensationInterval.HOURLY
-
+        elif pay_period:
+            interval = CompensationInterval.get_interval(pay_period)
         min_amount = int(adjusted_pay.get("p10") // 1)
         max_amount = int(adjusted_pay.get("p90") // 1)
 
@@ -177,17 +228,11 @@ class GlassdoorScraper(Scraper):
             currency=currency,
         )
 
-    def get_job_type_enum(self, job_type_str: str) -> list[JobType] | None:
-        for job_type in JobType:
-            if job_type_str in job_type.value:
-                return [job_type]
-        return None
-
     def get_location(self, location: str, is_remote: bool) -> (int, str):
         if not location or is_remote:
             return "11047", "STATE"  # remote options
         url = f"{self.url}/findPopularLocationAjax.htm?maxLocationsToReturn=10&term={location}"
-        session = create_session(self.proxy)
+        session = create_session(self.proxy, has_retry=True)
         response = session.get(url)
         if response.status_code != 200:
             raise GlassdoorException(
@@ -210,7 +255,7 @@ class GlassdoorScraper(Scraper):
         location_type: str,
         page_num: int,
         cursor: str | None = None,
-    ) -> dict[str, str | Any]:
+    ) -> str:
         payload = {
             "operationName": "JobSearchResultsQuery",
             "variables": {
@@ -240,10 +285,17 @@ class GlassdoorScraper(Scraper):
             payload["variables"]["filterParams"].append(
                 {"filterKey": "jobType", "values": filter_value}
             )
-
         return json.dumps([payload])
 
-    def parse_location(self, location_name: str) -> Location:
+    @staticmethod
+    def get_job_type_enum(job_type_str: str) -> list[JobType] | None:
+        for job_type in JobType:
+            if job_type_str in job_type.value:
+                return [job_type]
+        return None
+
+    @staticmethod
+    def parse_location(location_name: str) -> Location:
         if not location_name or location_name == "Remote":
             return None
         city, _, state = location_name.partition(", ")

src/jobspy/scrapers/indeed/__init__.py

@@ -64,6 +64,7 @@ class IndeedScraper(Scraper):
             "l": scraper_input.location,
             "filter": 0,
             "start": scraper_input.offset + page * 10,
+            "sort": "date"
         }
         if scraper_input.distance:
             params["radius"] = scraper_input.distance
@@ -77,7 +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(),
@@ -139,7 +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")
@@ -150,6 +152,7 @@ class IndeedScraper(Scraper):
                 title=job["normTitle"],
                 description=description,
                 company_name=job["company"],
+                company_url=self.url + job["companyOverviewLink"] if "companyOverviewLink" in job else None,
                 location=Location(
                     city=job.get("jobLocationCity"),
                     state=job.get("jobLocationState"),
@@ -235,31 +238,16 @@ class IndeedScraper(Scraper):
         if response.status_code not in range(200, 400):
             return None
 
-        soup = BeautifulSoup(response.text, "html.parser")
-        script_tag = soup.find(
-            "script", text=lambda x: x and "window._initialData" in x
-        )
-
-        if not script_tag:
-            return None
-
-        script_code = script_tag.string
-        match = re.search(r"window\._initialData\s*=\s*({.*?})\s*;", script_code, re.S)
-
-        if not match:
-            return None
-
-        json_string = match.group(1)
-        data = json.loads(json_string)
         try:
-            job_description = data["jobInfoWrapperModel"]["jobInfoModel"][
+            data = json.loads(response.text)
+            job_description = data["body"]["jobInfoWrapperModel"]["jobInfoModel"][
                 "sanitizedJobDescription"
             ]
         except (KeyError, TypeError, IndexError):
             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
 
@@ -320,7 +308,7 @@ class IndeedScraper(Scraper):
                 raise IndeedException("Could not find mosaic provider job cards data")
         else:
             raise IndeedException(
-                "Could not find a script tag containing mosaic provider data"
+                "Could not find any results for the search"
             )
 
     @staticmethod

src/jobspy/scrapers/linkedin/__init__.py

@@ -4,26 +4,27 @@ jobspy.scrapers.linkedin
 
 This module contains routines to scrape LinkedIn.
 """
+import random
 from typing import Optional
 from datetime import datetime
 
 import requests
 import time
 from requests.exceptions import ProxyError
-from concurrent.futures import ThreadPoolExecutor, as_completed
 from bs4 import BeautifulSoup
 from bs4.element import Tag
 from threading import Lock
+from urllib.parse import urlparse, urlunparse
 
 from .. import Scraper, ScraperInput, Site
-from ..utils import count_urgent_words, extract_emails_from_text, get_enum_from_job_type
 from ..exceptions import LinkedInException
-from ...jobs import JobPost, Location, JobResponse, JobType, Country
+from ..utils import create_session
+from ...jobs import JobPost, Location, JobResponse, JobType, Country, Compensation
+from ..utils import count_urgent_words, extract_emails_from_text, get_enum_from_job_type, currency_parser
 
 
 class LinkedInScraper(Scraper):
-    MAX_RETRIES = 3
-    DELAY = 10
+    DELAY = 3
 
     def __init__(self, proxy: Optional[str] = None):
         """
@@ -57,6 +58,7 @@ class LinkedInScraper(Scraper):
             return mapping.get(job_type_enum, "")
 
         while len(job_list) < scraper_input.results_wanted and page < 1000:
+            session = create_session(is_tls=False, has_retry=True, delay=5)
             params = {
                 "keywords": scraper_input.search_term,
                 "location": scraper_input.location,
@@ -66,87 +68,88 @@ class LinkedInScraper(Scraper):
                 if scraper_input.job_type
                 else None,
                 "pageNum": 0,
-                page: page + scraper_input.offset,
+                "start": page + scraper_input.offset,
                 "f_AL": "true" if scraper_input.easy_apply else None,
             }
 
             params = {k: v for k, v in params.items() if v is not None}
-
-            params = {k: v for k, v in params.items() if v is not None}
-            retries = 0
-            while retries < self.MAX_RETRIES:
-                try:
-                    response = requests.get(
-                        f"{self.url}/jobs-guest/jobs/api/seeMoreJobPostings/search?",
-                        params=params,
-                        allow_redirects=True,
-                        proxies=self.proxy,
-                        timeout=10,
-                    )
-                    response.raise_for_status()
-
-                    break
-                except requests.HTTPError as e:
-                    if hasattr(e, "response") and e.response is not None:
-                        if e.response.status_code == 429:
-                            time.sleep(self.DELAY)
-                            retries += 1
-                            continue
-                        else:
-                            raise LinkedInException(
-                                f"bad response status code: {e.response.status_code}"
-                            )
-                    else:
-                        raise
-                except ProxyError as e:
-                    raise LinkedInException("bad proxy")
-                except Exception as e:
-                    raise LinkedInException(str(e))
-            else:
-                # Raise an exception if the maximum number of retries is reached
-                raise LinkedInException(
-                    "Max retries reached, failed to get a valid response"
+            try:
+                response = session.get(
+                    f"{self.url}/jobs-guest/jobs/api/seeMoreJobPostings/search?",
+                    params=params,
+                    allow_redirects=True,
+                    proxies=self.proxy,
+                    headers=self.headers(),
+                    timeout=10,
                 )
+                response.raise_for_status()
+
+            except requests.HTTPError as e:
+                raise LinkedInException(f"bad response status code: {e.response.status_code}")
+            except ProxyError as e:
+                raise LinkedInException("bad proxy")
+            except Exception as e:
+                raise LinkedInException(str(e))
 
             soup = BeautifulSoup(response.text, "html.parser")
+            job_cards = soup.find_all("div", class_="base-search-card")
+            if len(job_cards) == 0:
+                return JobResponse(jobs=job_list)
 
-            with ThreadPoolExecutor(max_workers=5) as executor:
-                futures = []
-                for job_card in soup.find_all("div", class_="base-search-card"):
-                    job_url = None
-                    href_tag = job_card.find("a", class_="base-card__full-link")
-                    if href_tag and "href" in href_tag.attrs:
-                        href = href_tag.attrs["href"].split("?")[0]
-                        job_id = href.split("-")[-1]
-                        job_url = f"{self.url}/jobs/view/{job_id}"
+            for job_card in job_cards:
+                job_url = None
+                href_tag = job_card.find("a", class_="base-card__full-link")
+                if href_tag and "href" in href_tag.attrs:
+                    href = href_tag.attrs["href"].split("?")[0]
+                    job_id = href.split("-")[-1]
+                    job_url = f"{self.url}/jobs/view/{job_id}"
 
-                    with url_lock:
-                        if job_url in seen_urls:
-                            continue
-                        seen_urls.add(job_url)
+                with url_lock:
+                    if job_url in seen_urls:
+                        continue
+                    seen_urls.add(job_url)
 
-                    futures.append(executor.submit(self.process_job, job_card, job_url))
+                # Call process_job directly without threading
+                try:
+                    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:
+                    raise LinkedInException("Exception occurred while processing jobs")
 
-                for future in as_completed(futures):
-                    try:
-                        job_post = future.result()
-                        if job_post:
-                            job_list.append(job_post)
-                    except Exception as e:
-                        raise LinkedInException(
-                            "Exception occurred while processing jobs"
-                        )
             page += 25
+            time.sleep(random.uniform(LinkedInScraper.DELAY, LinkedInScraper.DELAY + 2))
 
         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
+        if salary_tag:
+            salary_text = salary_tag.get_text(separator=' ').strip()
+            salary_values = [currency_parser(value) for value in salary_text.split('-')]
+            salary_min = salary_values[0]
+            salary_max = salary_values[1]
+            currency = salary_text[0] if salary_text[0] != '$' else 'USD'
+
+            compensation = Compensation(
+                min_amount=int(salary_min),
+                max_amount=int(salary_max),
+                currency=currency,
+            )
+
         title_tag = job_card.find("span", class_="sr-only")
         title = title_tag.get_text(strip=True) if title_tag else "N/A"
 
         company_tag = job_card.find("h4", class_="base-search-card__subtitle")
         company_a_tag = company_tag.find("a") if company_tag else None
+        company_url = (
+            urlunparse(urlparse(company_a_tag.get("href"))._replace(query=""))
+            if company_a_tag and company_a_tag.has_attr("href")
+            else ""
+        )
         company = company_a_tag.get_text(strip=True) if company_a_tag else "N/A"
 
         metadata_card = job_card.find("div", class_="base-search-card__metadata")
@@ -157,7 +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:
@@ -166,18 +169,20 @@ 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
-
-        description, job_type = self.get_job_description(job_url)
+        if full_descr:
+            description, job_type = self.get_job_description(job_url)
 
         return JobPost(
             title=title,
-            description=description,
             company_name=company,
+            company_url=company_url,
             location=location,
             date_posted=date_posted,
             job_url=job_url,
-            job_type=job_type,
+            compensation=compensation,
             benefits=benefits,
+            job_type=job_type,
+            description=description,
             emails=extract_emails_from_text(description) if description else None,
             num_urgent_words=count_urgent_words(description) if description else None,
         )
@@ -191,10 +196,15 @@ class LinkedInScraper(Scraper):
         :return: description or None
         """
         try:
-            response = requests.get(job_page_url, timeout=5, proxies=self.proxy)
+            session = create_session(is_tls=False, has_retry=True)
+            response = session.get(job_page_url, timeout=5, proxies=self.proxy)
             response.raise_for_status()
+        except requests.HTTPError as e:
+            return None, None
         except Exception as e:
             return None, None
+        if response.url == "https://www.linkedin.com/signup":
+            return None, None
 
         soup = BeautifulSoup(response.text, "html.parser")
         div_content = soup.find(
@@ -203,7 +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,
@@ -230,7 +240,7 @@ class LinkedInScraper(Scraper):
                     employment_type = employment_type.lower()
                     employment_type = employment_type.replace("-", "")
 
-            return [get_enum_from_job_type(employment_type)]
+            return [get_enum_from_job_type(employment_type)] if employment_type else []
 
         return description, get_job_type(soup)
 
@@ -254,5 +264,30 @@ class LinkedInScraper(Scraper):
                     state=state,
                     country=Country.from_string(self.country),
                 )
+            elif len(parts) == 3:
+                city, state, country = parts
+                location = Location(
+                    city=city,
+                    state=state,
+                    country=Country.from_string(country),
+                )
 
         return location
+
+    @staticmethod
+    def headers() -> dict:
+        return {
+            'authority': 'www.linkedin.com',
+            'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7',
+            'accept-language': 'en-US,en;q=0.9',
+            'cache-control': 'max-age=0',
+            'sec-ch-ua': '"Not_A Brand";v="8", "Chromium";v="120", "Google Chrome";v="120"',
+            # 'sec-ch-ua-mobile': '?0',
+            # 'sec-ch-ua-platform': '"macOS"',
+            # 'sec-fetch-dest': 'document',
+            # 'sec-fetch-mode': 'navigate',
+            # 'sec-fetch-site': 'none',
+            # 'sec-fetch-user': '?1',
+            'upgrade-insecure-requests': '1',
+            'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'
+        }

src/jobspy/scrapers/utils.py

@@ -1,7 +1,10 @@
 import re
+import numpy as np
 
-import requests
 import tls_client
+import requests
+from requests.adapters import HTTPAdapter, Retry
+
 from ..jobs import JobType
 
 
@@ -26,11 +29,11 @@ def extract_emails_from_text(text: str) -> list[str] | None:
     return email_regex.findall(text)
 
 
-def create_session(proxy: dict | None = None, is_tls: bool = True):
+def create_session(proxy: dict | None = None, is_tls: bool = True, has_retry: bool = False, delay: int = 1) -> requests.Session:
     """
-    Creates a tls client session
+    Creates a requests session with optional tls, proxy, and retry settings.
 
-    :return: A session object with or without proxies.
+    :return: A session object
     """
     if is_tls:
         session = tls_client.Session(
@@ -38,17 +41,21 @@ def create_session(proxy: dict | None = None, is_tls: bool = True):
             random_tls_extension_order=True,
         )
         session.proxies = proxy
-        # TODO multiple proxies
-        # if self.proxies:
-        #     session.proxies = {
-        #         "http": random.choice(self.proxies),
-        #         "https": random.choice(self.proxies),
-        #     }
     else:
         session = requests.Session()
         session.allow_redirects = True
         if proxy:
             session.proxies.update(proxy)
+        if has_retry:
+            retries = Retry(total=3,
+                            connect=3,
+                            status=3,
+                            status_forcelist=[500, 502, 503, 504, 429],
+                            backoff_factor=delay)
+            adapter = HTTPAdapter(max_retries=retries)
+
+            session.mount('http://', adapter)
+            session.mount('https://', adapter)
 
     return session
 
@@ -62,3 +69,19 @@ def get_enum_from_job_type(job_type_str: str) -> JobType | None:
         if job_type_str in job_type.value:
             res = job_type
     return res
+
+def currency_parser(cur_str):
+    # Remove any non-numerical characters
+    # except for ',' '.' or '-' (e.g. EUR)
+    cur_str = re.sub("[^-0-9.,]", '', cur_str)
+    # Remove any 000s separators (either , or .)
+    cur_str = re.sub("[.,]", '', cur_str[:-3]) + cur_str[-3:]
+
+    if '.' in list(cur_str[-3:]):
+        num = float(cur_str)
+    elif ',' in list(cur_str[-3:]):
+        num = float(cur_str.replace(',', '.'))
+    else:
+        num = float(cur_str)
+
+    return np.round(num, 2)

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=False)
-            response = session.get(
+            response = self.session.get(
                 f"https://api.ziprecruiter.com/jobs-app/jobs",
                 headers=self.headers(),
                 params=self.add_params(scraper_input),
-                timeout=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:
@@ -200,7 +206,6 @@ class ZipRecruiterScraper(Scraper):
         """
         return {
             "Host": "api.ziprecruiter.com",
-            "Cookie": "ziprecruiter_browser=018188e0-045b-4ad7-aa50-627a6c3d43aa; ziprecruiter_session=5259b2219bf95b6d2299a1417424bc2edc9f4b38; SplitSV=2016-10-19%3AU2FsdGVkX19f9%2Bx70knxc%2FeR3xXR8lWoTcYfq5QjmLU%3D%0A; __cf_bm=qXim3DtLPbOL83GIp.ddQEOFVFTc1OBGPckiHYxcz3o-1698521532-0-AfUOCkgCZyVbiW1ziUwyefCfzNrJJTTKPYnif1FZGQkT60dMowmSU/Y/lP+WiygkFPW/KbYJmyc+MQSkkad5YygYaARflaRj51abnD+SyF9V; zglobalid=68d49bd5-0326-428e-aba8-8a04b64bc67c.af2d99ff7c03.653d61bb; ziprecruiter_browser=018188e0-045b-4ad7-aa50-627a6c3d43aa; ziprecruiter_session=5259b2219bf95b6d2299a1417424bc2edc9f4b38",
             "accept": "*/*",
             "x-zr-zva-override": "100000000;vid:ZT1huzm_EQlDTVEc",
             "x-pushnotificationid": "0ff4983d38d7fc5b3370297f2bcffcf4b3321c418f5c22dd152a0264707602a0",