Update pyproject.toml

.pre-commit-config.yaml

@@ -0,0 +1,7 @@
+repos:
+- repo: https://github.com/psf/black
+  rev: 24.2.0
+  hooks:
+  - id: black
+    language_version: python
+    args: [--line-length=88, --quiet]

README.md

@@ -11,7 +11,7 @@ work with us.*
 
 - Scrapes job postings from **LinkedIn**, **Indeed**, **Glassdoor**, & **ZipRecruiter** simultaneously
 - Aggregates the job postings in a Pandas DataFrame
-- Proxy support (HTTP/S, SOCKS)
+- Proxy support
 
 [Video Guide for JobSpy](https://www.youtube.com/watch?v=RuP1HrAZnxs&pp=ygUgam9icyBzY3JhcGVyIGJvdCBsaW5rZWRpbiBpbmRlZWQ%3D) -
 Updated for release v1.1.3
@@ -21,7 +21,7 @@ Updated for release v1.1.3
 ### Installation
 
 ```
-pip install python-jobspy
+pip install -U python-jobspy
 ```
 
 _Python version >= [3.10](https://www.python.org/downloads/release/python-3100/) required_
@@ -37,8 +37,9 @@ jobs = scrape_jobs(
     search_term="software engineer",
     location="Dallas, TX",
     results_wanted=20,
-    hours_old=72, # (only linkedin is hour specific, others round up to days old)
+    hours_old=72, # (only Linkedin/Indeed is hour specific, others round up to days old)
-    country_indeed='USA'  # only needed for indeed / glassdoor
+    country_indeed='USA',  # only needed for indeed / glassdoor
+    # linkedin_fetch_description=True # get full description and direct job url for linkedin (slower)
 )
 print(f"Found {len(jobs)} jobs")
 print(jobs.head())
@@ -48,7 +49,7 @@ jobs.to_csv("jobs.csv", quoting=csv.QUOTE_NONNUMERIC, escapechar="\\", index=Fal
 ### Output
 
 ```
-SITE           TITLE                             COMPANY_NAME      CITY          STATE  JOB_TYPE  INTERVAL  MIN_AMOUNT  MAX_AMOUNT  JOB_URL                                            DESCRIPTION
+SITE           TITLE                             COMPANY           CITY          STATE  JOB_TYPE  INTERVAL  MIN_AMOUNT  MAX_AMOUNT  JOB_URL                                            DESCRIPTION
 indeed         Software Engineer                 AMERICAN SYSTEMS  Arlington     VA     None      yearly    200000      150000      https://www.indeed.com/viewjob?jk=5e409e577046...  THIS POSITION COMES WITH A 10K SIGNING BONUS!...
 indeed         Senior Software Engineer          TherapyNotes.com  Philadelphia  PA     fulltime  yearly    135000      110000      https://www.indeed.com/viewjob?jk=da39574a40cb...  About Us TherapyNotes is the national leader i...
 linkedin       Software Engineer - Early Career  Lockheed Martin   Sunnyvale     CA     fulltime  yearly    None        None        https://www.linkedin.com/jobs/view/3693012711      Description:By bringing together people that u...
@@ -60,22 +61,24 @@ zip_recruiter Software Developer                 TEKsystems        Phoenix
 ### Parameters for `scrape_jobs()`
 
 ```plaintext
-Required
-├── site_type (List[enum]): linkedin, zip_recruiter, indeed, glassdoor
-└── search_term (str)
 Optional
-├── location (int)
+├── site_name (list|str): linkedin, zip_recruiter, indeed, glassdoor (default is all four)
-├── distance (int): in miles
+├── search_term (str)
-├── job_type (enum): fulltime, parttime, internship, contract
+├── location (str)
-├── proxy (str): in format 'http://user:pass@host:port' or [https, socks]
+├── distance (int): in miles, default 50
+├── job_type (str): fulltime, parttime, internship, contract
+├── proxy (str): in format 'http://user:pass@host:port'
 ├── is_remote (bool)
-├── full_description (bool): fetches full description for LinkedIn (slower)
+├── results_wanted (int): number of job results to retrieve for each site specified in 'site_name'
-├── 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 the job board site (LinkedIn & Indeed do not allow pairing this with hours_old)
-├── easy_apply (bool): filters for jobs that are hosted on the job board site
+├── linkedin_fetch_description (bool): fetches full description and direct job url for LinkedIn (slower)
-├── linkedin_company_ids (list[int): searches for linkedin jobs with specific company ids
+├── linkedin_company_ids (list[int]): searches for linkedin jobs with specific company ids
-├── country_indeed (enum): filters the country on Indeed (see below for correct spelling)
+├── description_format (str): markdown, html (Format type of the job descriptions. Default is markdown.)
-├── offset (num): starts the search from an offset (e.g. 25 will start the search from the 25th result)
+├── country_indeed (str): filters the country on Indeed (see below for correct spelling)
-├── hours_old (int): filters jobs by the number of hours since the job was posted (all but LinkedIn rounds up to next day)
+├── offset (int): starts the search from an offset (e.g. 25 will start the search from the 25th result)
+├── hours_old (int): filters jobs by the number of hours since the job was posted (ZipRecruiter and Glassdoor round up to next day. If you use this on Indeed, it will not filter by job_type/is_remote/easy_apply)
+├── verbose (int) {0, 1, 2}: Controls the verbosity of the runtime printouts (0 prints only errors, 1 is errors+warnings, 2 is all logs. Default is 2.)
+├── hyperlinks (bool): Whether to turn `job_url`s into hyperlinks. Default is false.
 ```
 
 ### JobPost Schema
@@ -99,24 +102,26 @@ JobPost
 │   └── currency (enum)
 └── date_posted (date)
 └── emails (str)
-└── num_urgent_words (int)
 └── is_remote (bool)
+
+Indeed specific
+├── company_country (str)
+└── company_addresses (str)
+└── company_industry (str)
+└── company_employees_label (str)
+└── company_revenue_label (str)
+└── company_description (str)
+└── ceo_name (str)
+└── ceo_photo_url (str)
+└── logo_photo_url (str)
+└── banner_photo_url (str)
 ```
 
-### Exceptions
-
-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. You can only fetch 1000 jobs max from the LinkedIn endpoint we're using
+LinkedIn searches globally & uses only the `location` parameter. 
 
 ### **ZipRecruiter**
 
@@ -146,10 +151,14 @@ You can specify the following countries when searching on Indeed (use the exact
 | South Korea          | Spain*       | Sweden     | Switzerland*   |
 | Taiwan               | Thailand     | Turkey     | Ukraine        |
 | United Arab Emirates | UK*          | USA*       | Uruguay        |
-| Venezuela            | Vietnam      |            |                |
+| Venezuela            | Vietnam*     |            |                |
 
 
-Glassdoor can only fetch 900 jobs from the endpoint we're using on a given search.
+## Notes
+* Indeed is the best scraper currently with no rate limiting.  
+* All the job board endpoints are capped at around 1000 jobs on a given search.  
+* LinkedIn is the most restrictive and usually rate limits around the 10th page.
+
 ## Frequently Asked Questions
 
 ---
@@ -167,7 +176,3 @@ persist, [submit an issue](https://github.com/Bunsly/JobSpy/issues).
 - Trying a VPN or proxy to change your IP address.
 
 ---
-
-
-
-  

examples/JobSpy_LongScrape.py

@@ -32,17 +32,18 @@ while len(all_jobs) < results_wanted:
                 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)),
+                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'))
+            all_jobs.extend(jobs.to_dict("records"))
 
             # Increment the offset for the next page of results
             offset += results_in_each_iteration

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "python-jobspy"
-version = "1.1.44"
+version = "1.1.53"
 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"
@@ -13,17 +13,24 @@ packages = [
 [tool.poetry.dependencies]
 python = "^3.10"
 requests = "^2.31.0"
-tls-client = "*"
 beautifulsoup4 = "^4.12.2"
 pandas = "^2.1.0"
 NUMPY = "1.24.2"
 pydantic = "^2.3.0"
+tls-client = "^1.0.1"
+markdownify = "^0.11.6"
+regex = "^2024.4.28"
 
 
 [tool.poetry.group.dev.dependencies]
 pytest = "^7.4.1"
 jupyter = "^1.0.0"
+black = "*"
+pre-commit = "*"
 
 [build-system]
 requires = ["poetry-core"]
 build-backend = "poetry.core.masonry.api"
+
+[tool.black]
+line-length = 88

src/jobspy/__init__.py

@@ -1,8 +1,11 @@
+from __future__ import annotations
+
 import pandas as pd
 from typing import Tuple
 from concurrent.futures import ThreadPoolExecutor, as_completed
 
 from .jobs import JobType, Location
+from .scrapers.utils import logger, set_logger_level
 from .scrapers.indeed import IndeedScraper
 from .scrapers.ziprecruiter import ZipRecruiterScraper
 from .scrapers.glassdoor import GlassdoorScraper
@@ -15,23 +18,12 @@ from .scrapers.exceptions import (
     GlassdoorException,
 )
 
-SCRAPER_MAPPING = {
-    Site.LINKEDIN: LinkedInScraper,
-    Site.INDEED: IndeedScraper,
-    Site.ZIP_RECRUITER: ZipRecruiterScraper,
-    Site.GLASSDOOR: GlassdoorScraper,
-}
-
-
-def _map_str_to_site(site_name: str) -> Site:
-    return Site[site_name.upper()]
-
 
 def scrape_jobs(
     site_name: str | list[str] | Site | list[Site] | None = None,
     search_term: str | None = None,
     location: str | None = None,
-    distance: int | None = None,
+    distance: int | None = 50,
     is_remote: bool = False,
     job_type: str | None = None,
     easy_apply: bool | None = None,
@@ -39,16 +31,28 @@ def scrape_jobs(
     country_indeed: str = "usa",
     hyperlinks: bool = False,
     proxy: str | None = None,
-    full_description: bool | None = False,
+    description_format: str = "markdown",
+    linkedin_fetch_description: bool | None = False,
     linkedin_company_ids: list[int] | None = None,
     offset: int | None = 0,
     hours_old: int = None,
+    verbose: int = 2,
     **kwargs,
 ) -> pd.DataFrame:
     """
     Simultaneously scrapes job data from multiple job sites.
-    :return: results_wanted: pandas dataframe containing job data
+    :return: pandas dataframe containing job data
     """
+    SCRAPER_MAPPING = {
+        Site.LINKEDIN: LinkedInScraper,
+        Site.INDEED: IndeedScraper,
+        Site.ZIP_RECRUITER: ZipRecruiterScraper,
+        Site.GLASSDOOR: GlassdoorScraper,
+    }
+    set_logger_level(verbose)
+
+    def map_str_to_site(site_name: str) -> Site:
+        return Site[site_name.upper()]
 
     def get_enum_from_value(value_str):
         for job_type in JobType:
@@ -61,12 +65,12 @@ def scrape_jobs(
     def get_site_type():
         site_types = list(Site)
         if isinstance(site_name, str):
-            site_types = [_map_str_to_site(site_name)]
+            site_types = [map_str_to_site(site_name)]
         elif isinstance(site_name, Site):
             site_types = [site_name]
         elif isinstance(site_name, list):
             site_types = [
-                _map_str_to_site(site) if isinstance(site, str) else site
+                map_str_to_site(site) if isinstance(site, str) else site
                 for site in site_name
             ]
         return site_types
@@ -82,32 +86,21 @@ def scrape_jobs(
         is_remote=is_remote,
         job_type=job_type,
         easy_apply=easy_apply,
-        full_description=full_description,
+        description_format=description_format,
+        linkedin_fetch_description=linkedin_fetch_description,
         results_wanted=results_wanted,
         linkedin_company_ids=linkedin_company_ids,
         offset=offset,
-        hours_old=hours_old
+        hours_old=hours_old,
     )
 
     def scrape_site(site: Site) -> Tuple[str, JobResponse]:
         scraper_class = SCRAPER_MAPPING[site]
         scraper = scraper_class(proxy=proxy)
-
-        try:
         scraped_data: JobResponse = scraper.scrape(scraper_input)
-        except (LinkedInException, IndeedException, ZipRecruiterException) as lie:
+        cap_name = site.value.capitalize()
-            raise lie
+        site_name = "ZipRecruiter" if cap_name == "Zip_recruiter" else cap_name
-        except Exception as e:
+        logger.info(f"{site_name} finished scraping")
-            if site == Site.LINKEDIN:
-                raise LinkedInException(str(e))
-            if site == Site.INDEED:
-                raise IndeedException(str(e))
-            if site == Site.ZIP_RECRUITER:
-                raise ZipRecruiterException(str(e))
-            if site == Site.GLASSDOOR:
-                raise GlassdoorException(str(e))
-            else:
-                raise e
         return site.value, scraped_data
 
     site_to_jobs_dict = {}
@@ -130,9 +123,8 @@ def scrape_jobs(
     for site, job_response in site_to_jobs_dict.items():
         for job in job_response.jobs:
             job_data = job.dict()
-            job_data[
+            job_url = job_data["job_url"]
-                "job_url_hyper"
+            job_data["job_url_hyper"] = f'<a href="{job_url}">{job_url}</a>'
-            ] = f'<a href="{job_data["job_url"]}">{job_data["job_url"]}</a>'
             job_data["site"] = site
             job_data["company"] = job_data["company_name"]
             job_data["job_type"] = (
@@ -168,13 +160,20 @@ def scrape_jobs(
             jobs_dfs.append(job_df)
 
     if jobs_dfs:
-        jobs_df = pd.concat(jobs_dfs, ignore_index=True)
+        # Step 1: Filter out all-NA columns from each DataFrame before concatenation
-        desired_order: list[str] = [
+        filtered_dfs = [df.dropna(axis=1, how="all") for df in jobs_dfs]
-            "job_url_hyper" if hyperlinks else "job_url",
+
+        # Step 2: Concatenate the filtered DataFrames
+        jobs_df = pd.concat(filtered_dfs, ignore_index=True)
+
+        # Desired column order
+        desired_order = [
+            "id",
             "site",
+            "job_url_hyper" if hyperlinks else "job_url",
+            "job_url_direct",
             "title",
             "company",
-            "company_url",
             "location",
             "job_type",
             "date_posted",
@@ -183,13 +182,30 @@ def scrape_jobs(
             "max_amount",
             "currency",
             "is_remote",
-            "num_urgent_words",
-            "benefits",
             "emails",
             "description",
+            "company_url",
+            "company_url_direct",
+            "company_addresses",
+            "company_industry",
+            "company_num_employees",
+            "company_revenue",
+            "company_description",
+            "logo_photo_url",
+            "banner_photo_url",
+            "ceo_name",
+            "ceo_photo_url",
         ]
-        jobs_formatted_df = jobs_df[desired_order]
-    else:
-        jobs_formatted_df = pd.DataFrame()
 
-    return jobs_formatted_df.sort_values(by=['site', 'date_posted'], ascending=[True, False])
+        # Step 3: Ensure all desired columns are present, adding missing ones as empty
+        for column in desired_order:
+            if column not in jobs_df.columns:
+                jobs_df[column] = None  # Add missing columns as empty
+
+        # Reorder the DataFrame according to the desired order
+        jobs_df = jobs_df[desired_order]
+
+        # Step 4: Sort the DataFrame as required
+        return jobs_df.sort_values(by=["site", "date_posted"], ascending=[True, False])
+    else:
+        return pd.DataFrame()

src/jobspy/jobs/__init__.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from typing import Optional
 from datetime import date
 from enum import Enum
@@ -57,7 +59,7 @@ class JobType(Enum):
 class Country(Enum):
     """
     Gets the subdomain for Indeed and Glassdoor.
-    The second item in the tuple is the subdomain for Indeed
+    The second item in the tuple is the subdomain (and API country code if there's a ':' separator) for Indeed
     The third item in the tuple is the subdomain (and tld if there's a ':' separator) for Glassdoor
     """
 
@@ -118,11 +120,11 @@ class Country(Enum):
     TURKEY = ("turkey", "tr")
     UKRAINE = ("ukraine", "ua")
     UNITEDARABEMIRATES = ("united arab emirates", "ae")
-    UK = ("uk,united kingdom", "uk", "co.uk")
+    UK = ("uk,united kingdom", "uk:gb", "co.uk")
-    USA = ("usa,us,united states", "www", "com")
+    USA = ("usa,us,united states", "www:us", "com")
     URUGUAY = ("uruguay", "uy")
     VENEZUELA = ("venezuela", "ve")
-    VIETNAM = ("vietnam", "vn")
+    VIETNAM = ("vietnam", "vn", "com")
 
     # internal for ziprecruiter
     US_CANADA = ("usa/ca", "www")
@@ -132,7 +134,10 @@ class Country(Enum):
 
     @property
     def indeed_domain_value(self):
-        return self.value[1]
+        subdomain, _, api_country_code = self.value[1].partition(":")
+        if subdomain and api_country_code:
+            return subdomain, api_country_code.upper()
+        return self.value[1], self.value[1].upper()
 
     @property
     def glassdoor_domain_value(self):
@@ -145,7 +150,7 @@ class Country(Enum):
         else:
             raise Exception(f"Glassdoor is not available for {self.name}")
 
-    def get_url(self):
+    def get_glassdoor_url(self):
         return f"https://{self.glassdoor_domain_value}/"
 
     @classmethod
@@ -153,7 +158,7 @@ class Country(Enum):
         """Convert a string to the corresponding Country enum."""
         country_str = country_str.strip().lower()
         for country in cls:
-            country_names = country.value[0].split(',')
+            country_names = country.value[0].split(",")
             if country_str in country_names:
                 return country
         valid_countries = [country.value for country in cls]
@@ -163,7 +168,7 @@ class Country(Enum):
 
 
 class Location(BaseModel):
-    country: Country | None = None
+    country: Country | str | None = None
     city: Optional[str] = None
     state: Optional[str] = None
 
@@ -173,7 +178,12 @@ class Location(BaseModel):
             location_parts.append(self.city)
         if self.state:
             location_parts.append(self.state)
-        if self.country and self.country not in (Country.US_CANADA, Country.WORLDWIDE):
+        if isinstance(self.country, str):
+            location_parts.append(self.country)
+        elif self.country and self.country not in (
+            Country.US_CANADA,
+            Country.WORLDWIDE,
+        ):
             country_name = self.country.value[0]
             if "," in country_name:
                 country_name = country_name.split(",")[0]
@@ -210,23 +220,39 @@ class Compensation(BaseModel):
     currency: Optional[str] = "USD"
 
 
+class DescriptionFormat(Enum):
+    MARKDOWN = "markdown"
+    HTML = "html"
+
+
 class JobPost(BaseModel):
+    id: str | None = None
     title: str
-    company_name: str
+    company_name: str | None
     job_url: str
+    job_url_direct: str | None = None
     location: Optional[Location]
 
     description: str | None = None
     company_url: str | None = None
+    company_url_direct: str | None = None
 
     job_type: list[JobType] | None = None
     compensation: Compensation | None = None
     date_posted: date | None = None
-    benefits: str | None = None
     emails: list[str] | None = None
-    num_urgent_words: int | None = None
     is_remote: bool | None = None
-    # company_industry: str | None = None
+
+    # indeed specific
+    company_addresses: str | None = None
+    company_industry: str | None = None
+    company_num_employees: str | None = None
+    company_revenue: str | None = None
+    company_description: str | None = None
+    ceo_name: str | None = None
+    ceo_photo_url: str | None = None
+    logo_photo_url: str | None = None
+    banner_photo_url: str | None = None
 
 
 class JobResponse(BaseModel):

src/jobspy/scrapers/__init__.py

@@ -1,4 +1,15 @@
-from ..jobs import Enum, BaseModel, JobType, JobResponse, Country
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+
+from ..jobs import (
+    Enum,
+    BaseModel,
+    JobType,
+    JobResponse,
+    Country,
+    DescriptionFormat,
+)
 
 
 class Site(Enum):
@@ -18,17 +29,19 @@ class ScraperInput(BaseModel):
     is_remote: bool = False
     job_type: JobType | None = None
     easy_apply: bool | None = None
-    full_description: bool = False
     offset: int = 0
+    linkedin_fetch_description: bool = False
     linkedin_company_ids: list[int] | None = None
+    description_format: DescriptionFormat | None = DescriptionFormat.MARKDOWN
 
     results_wanted: int = 15
     hours_old: int | None = None
 
 
-class Scraper:
+class Scraper(ABC):
     def __init__(self, site: Site, proxy: list[str] | None = None):
         self.site = site
         self.proxy = (lambda p: {"http": p, "https": p} if p else None)(proxy)
 
+    @abstractmethod
     def scrape(self, scraper_input: ScraperInput) -> JobResponse: ...

src/jobspy/scrapers/glassdoor/__init__.py

@@ -4,16 +4,24 @@ jobspy.scrapers.glassdoor
 
 This module contains routines to scrape Glassdoor.
 """
+
+from __future__ import annotations
+
+import re
 import json
 import requests
-from typing import Optional
+from typing import Optional, Tuple
 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 ..utils import extract_emails_from_text
 from ..exceptions import GlassdoorException
-from ..utils import create_session
+from ..utils import (
+    create_session,
+    markdown_converter,
+    logger,
+)
 from ...jobs import (
     JobPost,
     Compensation,
@@ -21,6 +29,7 @@ from ...jobs import (
     Location,
     JobResponse,
     JobType,
+    DescriptionFormat,
 )
 
 
@@ -32,73 +41,142 @@ class GlassdoorScraper(Scraper):
         site = Site(Site.GLASSDOOR)
         super().__init__(site, proxy=proxy)
 
-        self.url = None
+        self.base_url = None
         self.country = None
         self.session = None
+        self.scraper_input = None
         self.jobs_per_page = 30
+        self.max_pages = 30
         self.seen_urls = set()
 
-    def fetch_jobs_page(
+    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.
+        """
+        self.scraper_input = scraper_input
+        self.scraper_input.results_wanted = min(900, scraper_input.results_wanted)
+        self.base_url = self.scraper_input.country.get_glassdoor_url()
+
+        self.session = create_session(self.proxy, is_tls=True, has_retry=True)
+        token = self._get_csrf_token()
+        self.headers["gd-csrf-token"] = token if token else self.fallback_token
+
+        location_id, location_type = self._get_location(
+            scraper_input.location, scraper_input.is_remote
+        )
+        if location_type is None:
+            logger.error("Glassdoor: location not parsed")
+            return JobResponse(jobs=[])
+        all_jobs: list[JobPost] = []
+        cursor = None
+
+        range_start = 1 + (scraper_input.offset // self.jobs_per_page)
+        tot_pages = (scraper_input.results_wanted // self.jobs_per_page) + 2
+        range_end = min(tot_pages, self.max_pages + 1)
+        for page in range(range_start, range_end):
+            logger.info(f"Glassdoor search page: {page}")
+            try:
+                jobs, cursor = self._fetch_jobs_page(
+                    scraper_input, location_id, location_type, page, cursor
+                )
+                all_jobs.extend(jobs)
+                if not jobs or len(all_jobs) >= scraper_input.results_wanted:
+                    all_jobs = all_jobs[: scraper_input.results_wanted]
+                    break
+            except Exception as e:
+                logger.error(f"Glassdoor: {str(e)}")
+                break
+        return JobResponse(jobs=all_jobs)
+
+    def _fetch_jobs_page(
         self,
         scraper_input: ScraperInput,
         location_id: int,
         location_type: str,
         page_num: int,
         cursor: str | None,
-    ) -> (list[JobPost], str | None):
+    ) -> Tuple[list[JobPost], str | None]:
         """
         Scrapes a page of Glassdoor for jobs with scraper_input criteria
         """
+        jobs = []
+        self.scraper_input = scraper_input
         try:
-            payload = self.add_payload(
+            payload = self._add_payload(location_id, location_type, page_num, cursor)
-                scraper_input, location_id, location_type, page_num, cursor
-            )
             response = self.session.post(
-                f"{self.url}/graph", headers=self.headers(), timeout=10, data=payload
+                f"{self.base_url}/graph",
+                headers=self.headers,
+                timeout_seconds=15,
+                data=payload,
             )
             if response.status_code != 200:
-                raise GlassdoorException(
+                exc_msg = f"bad response status code: {response.status_code}"
-                    f"bad response status code: {response.status_code}"
+                raise GlassdoorException(exc_msg)
-                )
             res_json = response.json()[0]
             if "errors" in res_json:
                 raise ValueError("Error encountered in API response")
-        except Exception as e:
+        except (
-            raise GlassdoorException(str(e))
+            requests.exceptions.ReadTimeout,
+            GlassdoorException,
+            ValueError,
+            Exception,
+        ) as e:
+            logger.error(f"Glassdoor: {str(e)}")
+            return jobs, None
 
         jobs_data = res_json["data"]["jobListings"]["jobListings"]
 
-        jobs = []
         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}
+            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):
                 try:
                     job_post = future.result()
                     if job_post:
                         jobs.append(job_post)
                 except Exception as exc:
-                    raise GlassdoorException(f'Glassdoor generated an exception: {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):
+    def _get_csrf_token(self):
-        """Processes a single job and fetches its description."""
+        """
+        Fetches csrf token needed for API by visiting a generic page
+        """
+        res = self.session.get(
+            f"{self.base_url}/Job/computer-science-jobs.htm", headers=self.headers
+        )
+        pattern = r'"token":\s*"([^"]+)"'
+        matches = re.findall(pattern, res.text)
+        token = None
+        if matches:
+            token = matches[0]
+        return token
+
+    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/j?jl={job_id}'
+        job_url = f"{self.base_url}job-listing/j?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"]
-        company_id = job_data['jobview']['header']['employer']['id']
+        company_id = job_data["jobview"]["header"]["employer"]["id"]
         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 is not None else None
+        date_diff = (datetime.now() - timedelta(days=age_in_days)).date()
+        date_posted = date_diff if age_in_days is not None else None
 
         if location_type == "S":
             is_remote = True
@@ -106,15 +184,15 @@ class GlassdoorScraper(Scraper):
             location = self.parse_location(location_name)
 
         compensation = self.parse_compensation(job["header"])
-
         try:
-            description = self.fetch_job_description(job_id)
+            description = self._fetch_job_description(job_id)
         except:
             description = None
-
+        company_url = f"{self.base_url}Overview/W-EI_IE{company_id}.htm"
-        job_post = JobPost(
+        return JobPost(
+            id=str(job_id),
             title=title,
-            company_url=f"{self.url}Overview/W-EI_IE{company_id}.htm" if company_id else None,
+            company_url=company_url if company_id else None,
             company_name=company_name,
             date_posted=date_posted,
             job_url=job_url,
@@ -123,62 +201,20 @@ class GlassdoorScraper(Scraper):
             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:
+    def _fetch_job_description(self, job_id):
         """
-        Scrapes Glassdoor for jobs with scraper_input criteria.
+        Fetches the job description for a single job ID.
-        :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)
+        url = f"{self.base_url}/graph"
-        self.country = scraper_input.country
-        self.url = self.country.get_url()
-
-        location_id, location_type = self.get_location(
-            scraper_input.location, scraper_input.is_remote
-        )
-        all_jobs: list[JobPost] = []
-        cursor = None
-        max_pages = 30
-        self.session = create_session(self.proxy, is_tls=False, has_retry=True)
-        self.session.get(self.url)
-
-        try:
-            for page in range(
-                1 + (scraper_input.offset // self.jobs_per_page),
-                min(
-                    (scraper_input.results_wanted // self.jobs_per_page) + 2,
-                    max_pages + 1,
-                ),
-            ):
-                try:
-                    jobs, cursor = self.fetch_jobs_page(
-                        scraper_input, location_id, location_type, page, cursor
-                    )
-                    all_jobs.extend(jobs)
-                    if len(all_jobs) >= scraper_input.results_wanted:
-                        all_jobs = all_jobs[: scraper_input.results_wanted]
-                        break
-                except Exception as e:
-                    raise GlassdoorException(str(e))
-        except Exception as e:
-            raise GlassdoorException(str(e))
-
-        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"
+                    "pageTypeEnum": "SERP",
                 },
                 "query": """
                 query JobDetailQuery($jl: Long!, $queryString: String, $pageTypeEnum: PageTypeEnum) {
@@ -193,22 +229,90 @@ class GlassdoorScraper(Scraper):
                         __typename
                     }
                 }
-                """
+                """,
             }
         ]
-        response = requests.post(url, json=body, headers=GlassdoorScraper.headers())
+        res = requests.post(url, json=body, headers=self.headers)
-        if response.status_code != 200:
+        if res.status_code != 200:
             return None
-        data = response.json()[0]
+        data = res.json()[0]
-        desc = data['data']['jobview']['job']['description']
+        desc = data["data"]["jobview"]["job"]["description"]
+        if self.scraper_input.description_format == DescriptionFormat.MARKDOWN:
+            desc = markdown_converter(desc)
         return desc
 
+    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.base_url}/findPopularLocationAjax.htm?maxLocationsToReturn=10&term={location}"
+        session = create_session(self.proxy, has_retry=True)
+        res = self.session.get(url, headers=self.headers)
+        if res.status_code != 200:
+            if res.status_code == 429:
+                err = f"429 Response - Blocked by Glassdoor for too many requests"
+                logger.error(err)
+                return None, None
+            else:
+                err = f"Glassdoor response status code {res.status_code}"
+                err += f" - {res.text}"
+                logger.error(f"Glassdoor response status code {res.status_code}")
+                return None, None
+        items = res.json()
+
+        if not items:
+            raise ValueError(f"Location '{location}' not found on Glassdoor")
+        location_type = items[0]["locationType"]
+        if location_type == "C":
+            location_type = "CITY"
+        elif location_type == "S":
+            location_type = "STATE"
+        elif location_type == "N":
+            location_type = "COUNTRY"
+        return int(items[0]["locationId"]), location_type
+
+    def _add_payload(
+        self,
+        location_id: int,
+        location_type: str,
+        page_num: int,
+        cursor: str | None = None,
+    ) -> str:
+        fromage = None
+        if self.scraper_input.hours_old:
+            fromage = max(self.scraper_input.hours_old // 24, 1)
+        filter_params = []
+        if self.scraper_input.easy_apply:
+            filter_params.append({"filterKey": "applicationType", "values": "1"})
+        if fromage:
+            filter_params.append({"filterKey": "fromAge", "values": str(fromage)})
+        payload = {
+            "operationName": "JobSearchResultsQuery",
+            "variables": {
+                "excludeJobListingIds": [],
+                "filterParams": filter_params,
+                "keyword": self.scraper_input.search_term,
+                "numJobsToShow": 30,
+                "locationType": location_type,
+                "locationId": int(location_id),
+                "parameterUrlInput": f"IL.0,12_I{location_type}{location_id}",
+                "pageNumber": page_num,
+                "pageCursor": cursor,
+                "fromage": fromage,
+                "sort": "date",
+            },
+            "query": self.query_template,
+        }
+        if self.scraper_input.job_type:
+            payload["variables"]["filterParams"].append(
+                {"filterKey": "jobType", "values": self.scraper_input.job_type.value[0]}
+            )
+        return json.dumps([payload])
+
     @staticmethod
     def parse_compensation(data: dict) -> Optional[Compensation]:
         pay_period = data.get("payPeriod")
         adjusted_pay = data.get("payPeriodAdjustedPay")
         currency = data.get("payCurrency", "USD")
-
         if not pay_period or not adjusted_pay:
             return None
 
@@ -219,7 +323,6 @@ class GlassdoorScraper(Scraper):
             interval = CompensationInterval.get_interval(pay_period)
         min_amount = int(adjusted_pay.get("p10") // 1)
         max_amount = int(adjusted_pay.get("p90") // 1)
-
         return Compensation(
             interval=interval,
             min_amount=min_amount,
@@ -227,59 +330,44 @@ class GlassdoorScraper(Scraper):
             currency=currency,
         )
 
-    def get_location(self, location: str, is_remote: bool) -> (int, str):
+    @staticmethod
-        if not location or is_remote:
+    def get_job_type_enum(job_type_str: str) -> list[JobType] | None:
-            return "11047", "STATE"  # remote options
+        for job_type in JobType:
-        url = f"{self.url}/findPopularLocationAjax.htm?maxLocationsToReturn=10&term={location}"
+            if job_type_str in job_type.value:
-        session = create_session(self.proxy, has_retry=True)
+                return [job_type]
-        response = session.get(url)
-        if response.status_code != 200:
-            raise GlassdoorException(
-                f"bad response status code: {response.status_code}"
-            )
-        items = response.json()
-        if not items:
-            raise ValueError(f"Location '{location}' not found on Glassdoor")
-        location_type = items[0]["locationType"]
-        if location_type == "C":
-            location_type = "CITY"
-        elif location_type == "S":
-            location_type = "STATE"
-        elif location_type == 'N':
-            location_type = "COUNTRY"
-        return int(items[0]["locationId"]), location_type
 
     @staticmethod
-    def add_payload(
+    def parse_location(location_name: str) -> Location | None:
-        scraper_input,
+        if not location_name or location_name == "Remote":
-        location_id: int,
+            return
-        location_type: str,
+        city, _, state = location_name.partition(", ")
-        page_num: int,
+        return Location(city=city, state=state)
-        cursor: str | None = None,
+
-    ) -> str:
+    @staticmethod
-        # `fromage` is the posting time filter in days
+    def get_cursor_for_page(pagination_cursors, page_num):
-        fromage = max(scraper_input.hours_old // 24, 1) if scraper_input.hours_old else None
+        for cursor_data in pagination_cursors:
-        filter_params = []
+            if cursor_data["pageNumber"] == page_num:
-        if scraper_input.easy_apply:
+                return cursor_data["cursor"]
-            filter_params.append({"filterKey": "applicationType", "values": "1"})
+
-        if fromage:
+    fallback_token = "Ft6oHEWlRZrxDww95Cpazw:0pGUrkb2y3TyOpAIqF2vbPmUXoXVkD3oEGDVkvfeCerceQ5-n8mBg3BovySUIjmCPHCaW0H2nQVdqzbtsYqf4Q:wcqRqeegRUa9MVLJGyujVXB7vWFPjdaS1CtrrzJq-ok"
-            filter_params.append({"filterKey": "fromAge", "values": str(fromage)})
+    headers = {
-        payload = {
+        "authority": "www.glassdoor.com",
-            "operationName": "JobSearchResultsQuery",
+        "accept": "*/*",
-            "variables": {
+        "accept-language": "en-US,en;q=0.9",
-                "excludeJobListingIds": [],
+        "apollographql-client-name": "job-search-next",
-                "filterParams": filter_params,
+        "apollographql-client-version": "4.65.5",
-                "keyword": scraper_input.search_term,
+        "content-type": "application/json",
-                "numJobsToShow": 30,
+        "origin": "https://www.glassdoor.com",
-                "locationType": location_type,
+        "referer": "https://www.glassdoor.com/",
-                "locationId": int(location_id),
+        "sec-ch-ua": '"Chromium";v="118", "Google Chrome";v="118", "Not=A?Brand";v="99"',
-                "parameterUrlInput": f"IL.0,12_I{location_type}{location_id}",
+        "sec-ch-ua-mobile": "?0",
-                "pageNumber": page_num,
+        "sec-ch-ua-platform": '"macOS"',
-                "pageCursor": cursor,
+        "sec-fetch-dest": "empty",
-                "fromage": fromage,
+        "sec-fetch-mode": "cors",
-                "sort": "date"
+        "sec-fetch-site": "same-origin",
-            },
+        "user-agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36",
-            "query": """
+    }
+    query_template = """
             query JobSearchResultsQuery(
                 $excludeJobListingIds: [Long!], 
                 $keyword: String, 
@@ -445,54 +533,3 @@ class GlassdoorScraper(Scraper):
                 __typename
             }
     """
-        }
-
-        if scraper_input.job_type:
-            payload["variables"]["filterParams"].append(
-                {"filterKey": "jobType", "values": scraper_input.job_type.value[0]}
-            )
-        return json.dumps([payload])
-
-    @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]
-
-    @staticmethod
-    def parse_location(location_name: str) -> Location | None:
-        if not location_name or location_name == "Remote":
-            return
-        city, _, state = location_name.partition(", ")
-        return Location(city=city, state=state)
-
-    @staticmethod
-    def get_cursor_for_page(pagination_cursors, page_num):
-        for cursor_data in pagination_cursors:
-            if cursor_data["pageNumber"] == page_num:
-                return cursor_data["cursor"]
-
-    @staticmethod
-    def headers() -> dict:
-        """
-        Returns headers needed for requests
-        :return: dict - Dictionary containing headers
-        """
-        return {
-            "authority": "www.glassdoor.com",
-            "accept": "*/*",
-            "accept-language": "en-US,en;q=0.9",
-            "apollographql-client-name": "job-search-next",
-            "apollographql-client-version": "4.65.5",
-            "content-type": "application/json",
-            "gd-csrf-token": "Ft6oHEWlRZrxDww95Cpazw:0pGUrkb2y3TyOpAIqF2vbPmUXoXVkD3oEGDVkvfeCerceQ5-n8mBg3BovySUIjmCPHCaW0H2nQVdqzbtsYqf4Q:wcqRqeegRUa9MVLJGyujVXB7vWFPjdaS1CtrrzJq-ok",
-            "origin": "https://www.glassdoor.com",
-            "referer": "https://www.glassdoor.com/",
-            "sec-ch-ua": '"Chromium";v="118", "Google Chrome";v="118", "Not=A?Brand";v="99"',
-            "sec-ch-ua-mobile": "?0",
-            "sec-ch-ua-platform": '"macOS"',
-            "sec-fetch-dest": "empty",
-            "sec-fetch-mode": "cors",
-            "sec-fetch-site": "same-origin",
-            "user-agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36",
-        }

src/jobspy/scrapers/indeed/__init__.py

@@ -4,24 +4,22 @@ jobspy.scrapers.indeed
 
 This module contains routines to scrape Indeed.
 """
-import re
-import math
-import json
-import requests
-from typing import Any
-from datetime import datetime
 
-from bs4 import BeautifulSoup
+from __future__ import annotations
-from bs4.element import Tag
+
+import math
+from typing import Tuple
+from datetime import datetime
 from concurrent.futures import ThreadPoolExecutor, Future
 
-from ..exceptions import IndeedException
+import requests
+
+from .. import Scraper, ScraperInput, Site
 from ..utils import (
-    count_urgent_words,
     extract_emails_from_text,
-    create_session,
     get_enum_from_job_type,
-    logger
+    markdown_converter,
+    logger,
 )
 from ...jobs import (
     JobPost,
@@ -30,353 +28,355 @@ from ...jobs import (
     Location,
     JobResponse,
     JobType,
+    DescriptionFormat,
 )
-from .. import Scraper, ScraperInput, Site
 
 
 class IndeedScraper(Scraper):
     def __init__(self, proxy: str | None = None):
         """
-        Initializes IndeedScraper with the Indeed job search url
+        Initializes IndeedScraper with the Indeed API url
         """
-        self.url = None
+        self.scraper_input = None
-        self.country = None
+        self.jobs_per_page = 100
+        self.num_workers = 10
+        self.seen_urls = set()
+        self.headers = None
+        self.api_country_code = None
+        self.base_url = None
+        self.api_url = "https://apis.indeed.com/graphql"
         site = Site(Site.INDEED)
         super().__init__(site, proxy=proxy)
 
-        self.jobs_per_page = 25
-        self.seen_urls = set()
-
-    def scrape_page(
-        self, scraper_input: ScraperInput, page: int
-    ) -> list[JobPost]:
-        """
-        Scrapes a page of Indeed for jobs with scraper_input criteria
-        :param scraper_input:
-        :param page:
-        :return: jobs found on page, total number of jobs found for search
-        """
-        job_list = []
-        self.country = scraper_input.country
-        domain = self.country.indeed_domain_value
-        self.url = f"https://{domain}.indeed.com"
-
-        try:
-            session = create_session(self.proxy)
-            response = session.get(
-                f"{self.url}/m/jobs",
-                headers=self.get_headers(),
-                params=self.add_params(scraper_input, page),
-                allow_redirects=True,
-                timeout_seconds=10,
-            )
-            if response.status_code not in range(200, 400):
-                raise IndeedException(
-                    f"bad response with status code: {response.status_code}"
-                )
-        except Exception as e:
-            if "Proxy responded with" in str(e):
-                logger.error(f'Indeed: Bad proxy')
-            else:
-                logger.error(f'Indeed: {str(e)}')
-            return job_list
-
-        soup = BeautifulSoup(response.content, "html.parser")
-        if "did not match any jobs" in response.text:
-            return job_list
-
-        jobs = IndeedScraper.parse_jobs(
-            soup
-        )  #: can raise exception, handled by main scrape function
-
-        if (
-            not jobs.get("metaData", {})
-            .get("mosaicProviderJobCardsModel", {})
-            .get("results")
-        ):
-            raise IndeedException("No jobs found.")
-
-        def process_job(job: dict, job_detailed: dict) -> JobPost | None:
-            job_url = f'{self.url}/m/jobs/viewjob?jk={job["jobkey"]}'
-            job_url_client = f'{self.url}/viewjob?jk={job["jobkey"]}'
-            if job_url in self.seen_urls:
-                return None
-            self.seen_urls.add(job_url)
-            description = job_detailed['description']['html']
-
-
-            job_type = IndeedScraper.get_job_type(job)
-            timestamp_seconds = job["pubDate"] / 1000
-            date_posted = datetime.fromtimestamp(timestamp_seconds)
-            date_posted = date_posted.strftime("%Y-%m-%d")
-
-            job_post = JobPost(
-                title=job["normTitle"],
-                description=description,
-                company_name=job["company"],
-                company_url=f"{self.url}{job_detailed['employer']['relativeCompanyPageUrl']}" if job_detailed['employer'] else None,
-                location=Location(
-                    city=job.get("jobLocationCity"),
-                    state=job.get("jobLocationState"),
-                    country=self.country,
-                ),
-                job_type=job_type,
-                compensation=self.get_compensation(job, job_detailed),
-                date_posted=date_posted,
-                job_url=job_url_client,
-                emails=extract_emails_from_text(description) if description else None,
-                num_urgent_words=count_urgent_words(description)
-                if description
-                else None,
-                is_remote=IndeedScraper.is_job_remote(job, job_detailed, description)
-
-            )
-            return job_post
-
-        workers = 10
-        jobs = jobs["metaData"]["mosaicProviderJobCardsModel"]["results"]
-        job_keys = [job['jobkey'] for job in jobs]
-        jobs_detailed = self.get_job_details(job_keys)
-
-        with ThreadPoolExecutor(max_workers=workers) as executor:
-            job_results: list[Future] = [
-                executor.submit(process_job, job, job_detailed['job']) for job, job_detailed in zip(jobs, jobs_detailed)
-            ]
-
-        job_list = [result.result() for result in job_results if result.result()]
-
-        return job_list
-
     def scrape(self, scraper_input: ScraperInput) -> JobResponse:
         """
         Scrapes Indeed for jobs with scraper_input criteria
         :param scraper_input:
         :return: job_response
         """
-        job_list = self.scrape_page(scraper_input, 0)
+        self.scraper_input = scraper_input
-        pages_processed = 1
+        domain, self.api_country_code = self.scraper_input.country.indeed_domain_value
+        self.base_url = f"https://{domain}.indeed.com"
+        self.headers = self.api_headers.copy()
+        self.headers["indeed-co"] = self.scraper_input.country.indeed_domain_value
+        job_list = []
+        page = 1
+
+        cursor = None
+        offset_pages = math.ceil(self.scraper_input.offset / 100)
+        for _ in range(offset_pages):
+            logger.info(f"Indeed skipping search page: {page}")
+            __, cursor = self._scrape_page(cursor)
+            if not __:
+                logger.info(f"Indeed found no jobs on page: {page}")
+                break
 
         while len(self.seen_urls) < scraper_input.results_wanted:
-            pages_to_process = math.ceil((scraper_input.results_wanted - len(self.seen_urls)) / self.jobs_per_page)
+            logger.info(f"Indeed search page: {page}")
-            new_jobs = False
+            jobs, cursor = self._scrape_page(cursor)
-
+            if not jobs:
-            with ThreadPoolExecutor(max_workers=10) as executor:
+                logger.info(f"Indeed found no jobs on page: {page}")
-                futures: list[Future] = [
+                break
-                    executor.submit(self.scrape_page, scraper_input, page + pages_processed)
-                    for page in range(pages_to_process)
-                ]
-
-                for future in futures:
-                    jobs = future.result()
-                    if jobs:
             job_list += jobs
-                        new_jobs = True
+            page += 1
-                    if len(self.seen_urls) >= scraper_input.results_wanted:
+        return JobResponse(jobs=job_list[: scraper_input.results_wanted])
-                        break
 
-            pages_processed += pages_to_process
+    def _scrape_page(self, cursor: str | None) -> Tuple[list[JobPost], str | None]:
-            if not new_jobs:
+        """
-                break
+        Scrapes a page of Indeed for jobs with scraper_input criteria
+        :param cursor:
+        :return: jobs found on page, next page cursor
+        """
+        jobs = []
+        new_cursor = None
+        filters = self._build_filters()
+        search_term = self.scraper_input.search_term.replace('"', '\\"') if self.scraper_input.search_term else ""
+        query = self.job_search_query.format(
+            what=(
+                f'what: "{search_term}"'
+                if search_term
+                else ""
+            ),
+            location=(
+                f'location: {{where: "{self.scraper_input.location}", radius: {self.scraper_input.distance}, radiusUnit: MILES}}'
+                if self.scraper_input.location
+                else ""
+            ),
+            dateOnIndeed=self.scraper_input.hours_old,
+            cursor=f'cursor: "{cursor}"' if cursor else "",
+            filters=filters,
+        )
+        payload = {
+            "query": query,
+        }
+        api_headers = self.api_headers.copy()
+        api_headers["indeed-co"] = self.api_country_code
+        response = requests.post(
+            self.api_url,
+            headers=api_headers,
+            json=payload,
+            proxies=self.proxy,
+            timeout=10,
+        )
+        if response.status_code != 200:
+            logger.info(
+                f"Indeed responded with status code: {response.status_code} (submit GitHub issue if this appears to be a bug)"
+            )
+            return jobs, new_cursor
+        data = response.json()
+        jobs = data["data"]["jobSearch"]["results"]
+        new_cursor = data["data"]["jobSearch"]["pageInfo"]["nextCursor"]
 
+        with ThreadPoolExecutor(max_workers=self.num_workers) as executor:
+            job_results: list[Future] = [
+                executor.submit(self._process_job, job["job"]) for job in jobs
+            ]
+        job_list = [result.result() for result in job_results if result.result()]
+        return job_list, new_cursor
 
-        if len(self.seen_urls) > scraper_input.results_wanted:
+    def _build_filters(self):
-            job_list = job_list[:scraper_input.results_wanted]
+        """
+        Builds the filters dict for job type/is_remote. If hours_old is provided, composite filter for job_type/is_remote is not possible.
+        IndeedApply: filters: { keyword: { field: "indeedApplyScope", keys: ["DESKTOP"] } }
+        """
+        filters_str = ""
+        if self.scraper_input.hours_old:
+            filters_str = """
+            filters: {{
+                date: {{
+                  field: "dateOnIndeed",
+                  start: "{start}h"
+                }}
+            }}
+            """.format(
+                start=self.scraper_input.hours_old
+            )
+        elif self.scraper_input.easy_apply:
+            filters_str = """
+            filters: {
+                keyword: {
+                  field: "indeedApplyScope",
+                  keys: ["DESKTOP"]
+                }
+            }
+            """
+        elif self.scraper_input.job_type or self.scraper_input.is_remote:
+            job_type_key_mapping = {
+                JobType.FULL_TIME: "CF3CP",
+                JobType.PART_TIME: "75GKK",
+                JobType.CONTRACT: "NJXCK",
+                JobType.INTERNSHIP: "VDTG7",
+            }
 
-        return JobResponse(jobs=job_list)
+            keys = []
+            if self.scraper_input.job_type:
+                key = job_type_key_mapping[self.scraper_input.job_type]
+                keys.append(key)
+
+            if self.scraper_input.is_remote:
+                keys.append("DSQF7")
+
+            if keys:
+                keys_str = '", "'.join(keys)  # Prepare your keys string
+                filters_str = f"""
+                filters: {{
+                  composite: {{
+                    filters: [{{
+                      keyword: {{
+                        field: "attributes",
+                        keys: ["{keys_str}"]
+                      }}
+                    }}]
+                  }}
+                }}
+                """
+        return filters_str
+
+    def _process_job(self, job: dict) -> JobPost | None:
+        """
+        Parses the job dict into JobPost model
+        :param job: dict to parse
+        :return: JobPost if it's a new job
+        """
+        job_url = f'{self.base_url}/viewjob?jk={job["key"]}'
+        if job_url in self.seen_urls:
+            return
+        self.seen_urls.add(job_url)
+        description = job["description"]["html"]
+        if self.scraper_input.description_format == DescriptionFormat.MARKDOWN:
+            description = markdown_converter(description)
+
+        job_type = self._get_job_type(job["attributes"])
+        timestamp_seconds = job["datePublished"] / 1000
+        date_posted = datetime.fromtimestamp(timestamp_seconds).strftime("%Y-%m-%d")
+        employer = job["employer"].get("dossier") if job["employer"] else None
+        employer_details = employer.get("employerDetails", {}) if employer else {}
+        rel_url = job["employer"]["relativeCompanyPageUrl"] if job["employer"] else None
+        return JobPost(
+            id=str(job["key"]),
+            title=job["title"],
+            description=description,
+            company_name=job["employer"].get("name") if job.get("employer") else None,
+            company_url=(f"{self.base_url}{rel_url}" if job["employer"] else None),
+            company_url_direct=(
+                employer["links"]["corporateWebsite"] if employer else None
+            ),
+            location=Location(
+                city=job.get("location", {}).get("city"),
+                state=job.get("location", {}).get("admin1Code"),
+                country=job.get("location", {}).get("countryCode"),
+            ),
+            job_type=job_type,
+            compensation=self._get_compensation(job),
+            date_posted=date_posted,
+            job_url=job_url,
+            job_url_direct=(
+                job["recruit"].get("viewJobUrl") if job.get("recruit") else None
+            ),
+            emails=extract_emails_from_text(description) if description else None,
+            is_remote=self._is_job_remote(job, description),
+            company_addresses=(
+                employer_details["addresses"][0]
+                if employer_details.get("addresses")
+                else None
+            ),
+            company_industry=(
+                employer_details["industry"]
+                .replace("Iv1", "")
+                .replace("_", " ")
+                .title()
+                if employer_details.get("industry")
+                else None
+            ),
+            company_num_employees=employer_details.get("employeesLocalizedLabel"),
+            company_revenue=employer_details.get("revenueLocalizedLabel"),
+            company_description=employer_details.get("briefDescription"),
+            ceo_name=employer_details.get("ceoName"),
+            ceo_photo_url=employer_details.get("ceoPhotoUrl"),
+            logo_photo_url=(
+                employer["images"].get("squareLogoUrl")
+                if employer and employer.get("images")
+                else None
+            ),
+            banner_photo_url=(
+                employer["images"].get("headerImageUrl")
+                if employer and employer.get("images")
+                else None
+            ),
+        )
 
     @staticmethod
-    def get_job_type(job: dict) -> list[JobType] | None:
+    def _get_job_type(attributes: list) -> list[JobType]:
         """
-        Parses the job to get list of job types
+        Parses the attributes to get list of job types
-        :param job:
+        :param attributes:
-        :return:
+        :return: list of JobType
         """
         job_types: list[JobType] = []
-        for taxonomy in job["taxonomyAttributes"]:
+        for attribute in attributes:
-            if taxonomy["label"] == "job-types":
+            job_type_str = attribute["label"].replace("-", "").replace(" ", "").lower()
-                for i in range(len(taxonomy["attributes"])):
-                    label = taxonomy["attributes"][i].get("label")
-                    if label:
-                        job_type_str = label.replace("-", "").replace(" ", "").lower()
             job_type = get_enum_from_job_type(job_type_str)
             if job_type:
                 job_types.append(job_type)
         return job_types
 
     @staticmethod
-    def get_compensation(job: dict, job_detailed: dict) -> Compensation:
+    def _get_compensation(job: dict) -> Compensation | None:
         """
-        Parses the job to get
+        Parses the job to get compensation
+        :param job:
         :param job:
-        :param job_detailed:
         :return: compensation object
         """
-        comp = job_detailed['compensation']['baseSalary']
+        comp = job["compensation"]["baseSalary"]
-        if comp:
+        if not comp:
-            interval = IndeedScraper.get_correct_interval(comp['unitOfWork'])
+            return None
-            if interval:
+        interval = IndeedScraper._get_compensation_interval(comp["unitOfWork"])
+        if not interval:
+            return None
+        min_range = comp["range"].get("min")
+        max_range = comp["range"].get("max")
         return Compensation(
             interval=interval,
-                    min_amount=round(comp['range'].get('min'), 2) if comp['range'].get('min') is not None else None,
+            min_amount=round(min_range, 2) if min_range is not None else None,
-                    max_amount=round(comp['range'].get('max'), 2) if comp['range'].get('max') is not None else None,
+            max_amount=round(max_range, 2) if max_range is not None else None,
-                    currency=job_detailed['compensation']['currencyCode']
+            currency=job["compensation"]["currencyCode"],
-                )
-
-        extracted_salary = job.get("extractedSalary")
-        compensation = None
-        if extracted_salary:
-            salary_snippet = job.get("salarySnippet")
-            currency = salary_snippet.get("currency") if salary_snippet else None
-            interval = (extracted_salary.get("type"),)
-            if isinstance(interval, tuple):
-                interval = interval[0]
-
-            interval = interval.upper()
-            if interval in CompensationInterval.__members__:
-                compensation = Compensation(
-                    interval=CompensationInterval[interval],
-                    min_amount=int(extracted_salary.get("min")),
-                    max_amount=int(extracted_salary.get("max")),
-                    currency=currency,
-                )
-        return compensation
-
-    @staticmethod
-    def parse_jobs(soup: BeautifulSoup) -> dict:
-        """
-        Parses the jobs from the soup object
-        :param soup:
-        :return: jobs
-        """
-
-        def find_mosaic_script() -> Tag | None:
-            """
-            Finds jobcards script tag
-            :return: script_tag
-            """
-            script_tags = soup.find_all("script")
-
-            for tag in script_tags:
-                if (
-                    tag.string
-                    and "mosaic.providerData" in tag.string
-                    and "mosaic-provider-jobcards" in tag.string
-                ):
-                    return tag
-            return None
-
-        script_tag = find_mosaic_script()
-
-        if script_tag:
-            script_str = script_tag.string
-            pattern = r'window.mosaic.providerData\["mosaic-provider-jobcards"\]\s*=\s*({.*?});'
-            p = re.compile(pattern, re.DOTALL)
-            m = p.search(script_str)
-            if m:
-                jobs = json.loads(m.group(1).strip())
-                return jobs
-            else:
-                raise IndeedException("Could not find mosaic provider job cards data")
-        else:
-            raise IndeedException(
-                "Could not find any results for the search"
         )
 
     @staticmethod
-    def get_headers():
+    def _is_job_remote(job: dict, description: str) -> bool:
-        return {
+        """
-          'Host': 'www.indeed.com',
+        Searches the description, location, and attributes to check if job is remote
-          'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
+        """
-          'sec-fetch-site': 'same-origin',
+        remote_keywords = ["remote", "work from home", "wfh"]
-          'sec-fetch-dest': 'document',
-          'accept-language': 'en-US,en;q=0.9',
-          'sec-fetch-mode': 'navigate',
-          'user-agent': 'Mozilla/5.0 (iPhone; CPU iPhone OS 16_6_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Mobile/15E148 Indeed App 192.0',
-          'referer': 'https://www.indeed.com/m/jobs?q=software%20intern&l=Dallas%2C%20TX&from=serpso&rq=1&rsIdx=3',
-        }
-
-    @staticmethod
-    def add_params(scraper_input: ScraperInput, page: int) -> dict[str, str | Any]:
-        # `fromage` is the posting time filter in days
-        fromage = max(scraper_input.hours_old // 24, 1) if scraper_input.hours_old else None
-        params = {
-            "q": scraper_input.search_term,
-            "l": scraper_input.location if scraper_input.location else scraper_input.country.value[0].split(',')[-1],
-            "filter": 0,
-            "start": scraper_input.offset + page * 10,
-            "sort": "date",
-            "fromage": fromage,
-        }
-        if scraper_input.distance:
-            params["radius"] = scraper_input.distance
-
-        sc_values = []
-        if scraper_input.is_remote:
-            sc_values.append("attr(DSQF7)")
-        if scraper_input.job_type:
-            sc_values.append("jt({})".format(scraper_input.job_type.value[0]))
-
-        if sc_values:
-            params["sc"] = "0kf:" + "".join(sc_values) + ";"
-
-        if scraper_input.easy_apply:
-            params['iafilter'] = 1
-
-        return params
-
-    @staticmethod
-    def is_job_remote(job: dict, job_detailed: dict, description: str) -> bool:
-        remote_keywords = ['remote', 'work from home', 'wfh']
         is_remote_in_attributes = any(
-            any(keyword in attr['label'].lower() for keyword in remote_keywords)
+            any(keyword in attr["label"].lower() for keyword in remote_keywords)
-            for attr in job_detailed['attributes']
+            for attr in job["attributes"]
+        )
+        is_remote_in_description = any(
+            keyword in description.lower() for keyword in remote_keywords
         )
-        is_remote_in_description = any(keyword in description.lower() for keyword in remote_keywords)
         is_remote_in_location = any(
-            keyword in job_detailed['location']['formatted']['long'].lower()
+            keyword in job["location"]["formatted"]["long"].lower()
             for keyword in remote_keywords
         )
-        is_remote_in_taxonomy = any(
+        return (
-            taxonomy["label"] == "remote" and len(taxonomy["attributes"]) > 0
+            is_remote_in_attributes or is_remote_in_description or is_remote_in_location
-            for taxonomy in job.get("taxonomyAttributes", [])
         )
-        return is_remote_in_attributes or is_remote_in_description or is_remote_in_location or is_remote_in_taxonomy
 
-    def get_job_details(self, job_keys: list[str]) -> dict:
+    @staticmethod
-        """
+    def _get_compensation_interval(interval: str) -> CompensationInterval:
-        Queries the GraphQL endpoint for detailed job information for the given job keys.
+        interval_mapping = {
-        """
+            "DAY": "DAILY",
-        url = "https://apis.indeed.com/graphql"
+            "YEAR": "YEARLY",
-        headers = {
+            "HOUR": "HOURLY",
-            'Host': 'apis.indeed.com',
+            "WEEK": "WEEKLY",
-            'content-type': 'application/json',
+            "MONTH": "MONTHLY",
-            'indeed-api-key': '161092c2017b5bbab13edb12461a62d5a833871e7cad6d9d475304573de67ac8',
-            'accept': 'application/json',
-            'indeed-locale': 'en-US',
-            'accept-language': 'en-US,en;q=0.9',
-            'user-agent': 'Mozilla/5.0 (iPhone; CPU iPhone OS 16_6_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Mobile/15E148 Indeed App 193.1',
-            'indeed-app-info': 'appv=193.1; appid=com.indeed.jobsearch; osv=16.6.1; os=ios; dtype=phone',
-            'indeed-co': 'US',
         }
+        mapped_interval = interval_mapping.get(interval.upper(), None)
+        if mapped_interval and mapped_interval in CompensationInterval.__members__:
+            return CompensationInterval[mapped_interval]
+        else:
+            raise ValueError(f"Unsupported interval: {interval}")
 
-        job_keys_gql = '[' + ', '.join(f'"{key}"' for key in job_keys) + ']'
+    api_headers = {
-
+        "Host": "apis.indeed.com",
-        payload = {
+        "content-type": "application/json",
-            "query": f"""
+        "indeed-api-key": "161092c2017b5bbab13edb12461a62d5a833871e7cad6d9d475304573de67ac8",
+        "accept": "application/json",
+        "indeed-locale": "en-US",
+        "accept-language": "en-US,en;q=0.9",
+        "user-agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 16_6_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Mobile/15E148 Indeed App 193.1",
+        "indeed-app-info": "appv=193.1; appid=com.indeed.jobsearch; osv=16.6.1; os=ios; dtype=phone",
+    }
+    job_search_query = """
         query GetJobData {{
-              jobData(input: {{
+          jobSearch(
-                jobKeys: {job_keys_gql}
+            {what}
-              }}) {{
+            {location}
+            includeSponsoredResults: NONE
+            limit: 100
+            sort: DATE
+            {cursor}
+            {filters}
+          ) {{
+            pageInfo {{
+              nextCursor
+            }}
             results {{
+              trackingKey
               job {{
                 key
                 title
+                datePublished
+                dateOnIndeed
                 description {{
                   html
                 }}
                 location {{
                   countryName
                   countryCode
+                  admin1Code
                   city
                   postalCode
                   streetAddress
@@ -398,10 +398,30 @@ class IndeedScraper(Scraper):
                   currencyCode
                 }}
                 attributes {{
+                  key
                   label
                 }}
                 employer {{
                   relativeCompanyPageUrl
+                  name
+                  dossier {{
+                      employerDetails {{
+                        addresses
+                        industry
+                        employeesLocalizedLabel
+                        revenueLocalizedLabel
+                        briefDescription
+                        ceoName
+                        ceoPhotoUrl
+                      }}
+                      images {{
+                            headerImageUrl
+                            squareLogoUrl
+                      }}
+                      links {{
+                        corporateWebsite
+                    }}
+                  }}
                 }}
                 recruit {{
                   viewJobUrl
@@ -413,24 +433,3 @@ class IndeedScraper(Scraper):
           }}
         }}
         """
-        }
-        response = requests.post(url, headers=headers, json=payload, proxies=self.proxy)
-        if response.status_code == 200:
-            return response.json()['data']['jobData']['results']
-        else:
-            return {}
-
-    @staticmethod
-    def get_correct_interval(interval: str) -> CompensationInterval:
-        interval_mapping = {
-            "DAY": "DAILY",
-            "YEAR": "YEARLY",
-            "HOUR": "HOURLY",
-            "WEEK": "WEEKLY",
-            "MONTH": "MONTHLY"
-        }
-        mapped_interval = interval_mapping.get(interval.upper(), None)
-        if mapped_interval and mapped_interval in CompensationInterval.__members__:
-            return CompensationInterval[mapped_interval]
-        else:
-            raise ValueError(f"Unsupported interval: {interval}")

src/jobspy/scrapers/linkedin/__init__.py

@@ -4,13 +4,16 @@ jobspy.scrapers.linkedin
 
 This module contains routines to scrape LinkedIn.
 """
+
+from __future__ import annotations
+
 import time
 import random
+import regex as re
+import urllib.parse
 from typing import Optional
 from datetime import datetime
 
-import requests
-from requests.exceptions import ProxyError
 from threading import Lock
 from bs4.element import Tag
 from bs4 import BeautifulSoup
@@ -25,27 +28,32 @@ from ...jobs import (
     JobResponse,
     JobType,
     Country,
-    Compensation
+    Compensation,
+    DescriptionFormat,
 )
 from ..utils import (
-    count_urgent_words,
+    logger,
     extract_emails_from_text,
     get_enum_from_job_type,
-    currency_parser
+    currency_parser,
+    markdown_converter,
 )
 
 
 class LinkedInScraper(Scraper):
-    DELAY = 3
+    base_url = "https://www.linkedin.com"
+    delay = 3
+    band_delay = 4
+    jobs_per_page = 25
 
     def __init__(self, proxy: Optional[str] = None):
         """
         Initializes LinkedInScraper with the LinkedIn job search url
         """
-        site = Site(Site.LINKEDIN)
+        super().__init__(Site(Site.LINKEDIN), proxy=proxy)
+        self.scraper_input = None
         self.country = "worldwide"
-        self.url = "https://www.linkedin.com"
+        self.job_url_direct_regex = re.compile(r'(?<=\?url=)[^"]+')
-        super().__init__(site, proxy=proxy)
 
     def scrape(self, scraper_input: ScraperInput) -> JobResponse:
         """
@@ -53,67 +61,68 @@ class LinkedInScraper(Scraper):
         :param scraper_input:
         :return: job_response
         """
+        self.scraper_input = scraper_input
         job_list: list[JobPost] = []
         seen_urls = set()
         url_lock = Lock()
         page = scraper_input.offset // 25 + 25 if scraper_input.offset else 0
-
         seconds_old = (
-            scraper_input.hours_old * 3600
+            scraper_input.hours_old * 3600 if scraper_input.hours_old else None
-            if scraper_input.hours_old
+        )
-            else None
+        continue_search = (
+            lambda: len(job_list) < scraper_input.results_wanted and page < 1000
         )
-
-        def job_type_code(job_type_enum):
-            mapping = {
-                JobType.FULL_TIME: "F",
-                JobType.PART_TIME: "P",
-                JobType.INTERNSHIP: "I",
-                JobType.CONTRACT: "C",
-                JobType.TEMPORARY: "T",
-            }
-
-            return mapping.get(job_type_enum, "")
-
-        continue_search = lambda: len(job_list) < scraper_input.results_wanted and page < 1000
-
         while continue_search():
+            logger.info(f"LinkedIn search page: {page // 25 + 1}")
             session = create_session(is_tls=False, has_retry=True, delay=5)
             params = {
                 "keywords": scraper_input.search_term,
                 "location": scraper_input.location,
                 "distance": scraper_input.distance,
                 "f_WT": 2 if scraper_input.is_remote else None,
-                "f_JT": job_type_code(scraper_input.job_type)
+                "f_JT": (
+                    self.job_type_code(scraper_input.job_type)
                     if scraper_input.job_type
-                else None,
+                    else None
+                ),
                 "pageNum": 0,
                 "start": page + scraper_input.offset,
                 "f_AL": "true" if scraper_input.easy_apply else None,
-                "f_C": ','.join(map(str, scraper_input.linkedin_company_ids)) if scraper_input.linkedin_company_ids else None,
+                "f_C": (
-                "f_TPR": f"r{seconds_old}",
+                    ",".join(map(str, scraper_input.linkedin_company_ids))
+                    if scraper_input.linkedin_company_ids
+                    else None
+                ),
             }
+            if seconds_old is not None:
+                params["f_TPR"] = f"r{seconds_old}"
 
             params = {k: v for k, v in params.items() if v is not None}
             try:
                 response = session.get(
-                    f"{self.url}/jobs-guest/jobs/api/seeMoreJobPostings/search?",
+                    f"{self.base_url}/jobs-guest/jobs/api/seeMoreJobPostings/search?",
                     params=params,
                     allow_redirects=True,
                     proxies=self.proxy,
-                    headers=self.headers(),
+                    headers=self.headers,
                     timeout=10,
                 )
-                response.raise_for_status()
+                if response.status_code not in range(200, 400):
-
+                    if response.status_code == 429:
-            except requests.HTTPError as e:
+                        err = (
-                raise LinkedInException(
+                            f"429 Response - Blocked by LinkedIn for too many requests"
-                    f"bad response status code: {e.response.status_code}"
                         )
-            except ProxyError as e:
+                    else:
-                raise LinkedInException("bad proxy")
+                        err = f"LinkedIn response status code {response.status_code}"
+                        err += f" - {response.text}"
+                    logger.error(err)
+                    return JobResponse(jobs=job_list)
             except Exception as e:
-                raise LinkedInException(str(e))
+                if "Proxy responded with" in str(e):
+                    logger.error(f"LinkedIn: Bad proxy")
+                else:
+                    logger.error(f"LinkedIn: {str(e)}")
+                return JobResponse(jobs=job_list)
 
             soup = BeautifulSoup(response.text, "html.parser")
             job_cards = soup.find_all("div", class_="base-search-card")
@@ -126,30 +135,33 @@ class LinkedInScraper(Scraper):
                 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}"
+                    job_url = f"{self.base_url}/jobs/view/{job_id}"
 
                 with url_lock:
                     if job_url in seen_urls:
                         continue
                     seen_urls.add(job_url)
-
-                # Call process_job directly without threading
                 try:
-                    job_post = self.process_job(job_card, job_url, scraper_input.full_description)
+                    fetch_desc = scraper_input.linkedin_fetch_description
+                    job_post = self._process_job(job_card, job_url, fetch_desc)
                     if job_post:
                         job_list.append(job_post)
+                    if not continue_search():
+                        break
                 except Exception as e:
-                    raise LinkedInException("Exception occurred while processing jobs")
+                    raise LinkedInException(str(e))
 
             if continue_search():
-                time.sleep(random.uniform(LinkedInScraper.DELAY, LinkedInScraper.DELAY + 2))
+                time.sleep(random.uniform(self.delay, self.delay + self.band_delay))
-                page += 25
+                page += self.jobs_per_page
 
         job_list = job_list[: scraper_input.results_wanted]
         return JobResponse(jobs=job_list)
 
-    def process_job(self, job_card: Tag, job_url: str, full_descr: bool) -> Optional[JobPost]:
+    def _process_job(
-        salary_tag = job_card.find('span', class_='job-search-card__salary-info')
+        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:
@@ -178,26 +190,26 @@ class LinkedInScraper(Scraper):
         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")
-        location = self.get_location(metadata_card)
+        location = self._get_location(metadata_card)
 
         datetime_tag = (
             metadata_card.find("time", class_="job-search-card__listdate")
             if metadata_card
             else None
         )
-        date_posted = description = job_type = None
+        date_posted = None
         if datetime_tag and "datetime" in datetime_tag.attrs:
             datetime_str = datetime_tag["datetime"]
             try:
                 date_posted = datetime.strptime(datetime_str, "%Y-%m-%d")
-            except Exception as e:
+            except:
                 date_posted = None
-        benefits_tag = job_card.find("span", class_="result-benefits__text")
+        job_details = {}
-        benefits = " ".join(benefits_tag.get_text().split()) if benefits_tag else None
         if full_descr:
-            description, job_type = self.get_job_description(job_url)
+            job_details = self._get_job_details(job_url)
 
         return JobPost(
+            id=self._get_id(job_url),
             title=title,
             company_name=company,
             company_url=company_url,
@@ -205,31 +217,39 @@ class LinkedInScraper(Scraper):
             date_posted=date_posted,
             job_url=job_url,
             compensation=compensation,
-            benefits=benefits,
+            job_type=job_details.get("job_type"),
-            job_type=job_type,
+            description=job_details.get("description"),
-            description=description,
+            job_url_direct=job_details.get("job_url_direct"),
-            emails=extract_emails_from_text(description) if description else None,
+            emails=extract_emails_from_text(job_details.get("description")),
-            num_urgent_words=count_urgent_words(description) if description else None,
+            logo_photo_url=job_details.get("logo_photo_url"),
         )
 
-    def get_job_description(
+    def _get_id(self, url: str):
-        self, job_page_url: str
-    ) -> tuple[None, None] | tuple[str | None, tuple[str | None, JobType | None]]:
         """
-        Retrieves job description by going to the job page url
+        Extracts the job id from the job url
+        :param url:
+        :return: str
+        """
+        if not url:
+            return None
+        return url.split("/")[-1]
+
+    def _get_job_details(self, job_page_url: str) -> dict:
+        """
+        Retrieves job description and other job details by going to the job page url
         :param job_page_url:
-        :return: description or None
+        :return: dict
         """
         try:
             session = create_session(is_tls=False, has_retry=True)
-            response = session.get(job_page_url, timeout=5, proxies=self.proxy)
+            response = session.get(
+                job_page_url, headers=self.headers, timeout=5, proxies=self.proxy
+            )
             response.raise_for_status()
-        except requests.HTTPError as e:
+        except:
-            return None, None
+            return {}
-        except Exception as e:
-            return None, None
         if response.url == "https://www.linkedin.com/signup":
-            return None, None
+            return {}
 
         soup = BeautifulSoup(response.text, "html.parser")
         div_content = soup.find(
@@ -237,6 +257,7 @@ class LinkedInScraper(Scraper):
         )
         description = None
         if div_content is not None:
+
             def remove_attributes(tag):
                 for attr in list(tag.attrs):
                     del tag[attr]
@@ -244,37 +265,18 @@ class LinkedInScraper(Scraper):
 
             div_content = remove_attributes(div_content)
             description = div_content.prettify(formatter="html")
+            if self.scraper_input.description_format == DescriptionFormat.MARKDOWN:
+                description = markdown_converter(description)
+        return {
+            "description": description,
+            "job_type": self._parse_job_type(soup),
+            "job_url_direct": self._parse_job_url_direct(soup),
+            "logo_photo_url": soup.find("img", {"class": "artdeco-entity-image"}).get(
+                "data-delayed-url"
+            ),
+        }
 
-        def get_job_type(
+    def _get_location(self, metadata_card: Optional[Tag]) -> Location:
-            soup_job_type: BeautifulSoup,
-        ) -> list[JobType] | None:
-            """
-            Gets the job type from job page
-            :param soup_job_type:
-            :return: JobType
-            """
-            h3_tag = soup_job_type.find(
-                "h3",
-                class_="description__job-criteria-subheader",
-                string=lambda text: "Employment type" in text,
-            )
-
-            employment_type = None
-            if h3_tag:
-                employment_type_span = h3_tag.find_next_sibling(
-                    "span",
-                    class_="description__job-criteria-text description__job-criteria-text--criteria",
-                )
-                if employment_type_span:
-                    employment_type = employment_type_span.get_text(strip=True)
-                    employment_type = employment_type.lower()
-                    employment_type = employment_type.replace("-", "")
-
-            return [get_enum_from_job_type(employment_type)] if employment_type else []
-
-        return description, get_job_type(soup)
-
-    def get_location(self, metadata_card: Optional[Tag]) -> Location:
         """
         Extracts the location data from the job metadata card.
         :param metadata_card
@@ -296,28 +298,67 @@ class LinkedInScraper(Scraper):
                 )
             elif len(parts) == 3:
                 city, state, country = parts
-                location = Location(
+                country = Country.from_string(country)
-                    city=city,
+                location = Location(city=city, state=state, country=country)
-                    state=state,
-                    country=Country.from_string(country),
-                )
-
         return location
 
     @staticmethod
-    def headers() -> dict:
+    def _parse_job_type(soup_job_type: BeautifulSoup) -> list[JobType] | None:
+        """
+        Gets the job type from job page
+        :param soup_job_type:
+        :return: JobType
+        """
+        h3_tag = soup_job_type.find(
+            "h3",
+            class_="description__job-criteria-subheader",
+            string=lambda text: "Employment type" in text,
+        )
+        employment_type = None
+        if h3_tag:
+            employment_type_span = h3_tag.find_next_sibling(
+                "span",
+                class_="description__job-criteria-text description__job-criteria-text--criteria",
+            )
+            if employment_type_span:
+                employment_type = employment_type_span.get_text(strip=True)
+                employment_type = employment_type.lower()
+                employment_type = employment_type.replace("-", "")
+
+        return [get_enum_from_job_type(employment_type)] if employment_type else []
+
+    def _parse_job_url_direct(self, soup: BeautifulSoup) -> str | None:
+        """
+        Gets the job url direct from job page
+        :param soup:
+        :return: str
+        """
+        job_url_direct = None
+        job_url_direct_content = soup.find("code", id="applyUrl")
+        if job_url_direct_content:
+            job_url_direct_match = self.job_url_direct_regex.search(
+                job_url_direct_content.decode_contents().strip()
+            )
+            if job_url_direct_match:
+                job_url_direct = urllib.parse.unquote(job_url_direct_match.group())
+
+        return job_url_direct
+
+    @staticmethod
+    def job_type_code(job_type_enum: JobType) -> str:
         return {
+            JobType.FULL_TIME: "F",
+            JobType.PART_TIME: "P",
+            JobType.INTERNSHIP: "I",
+            JobType.CONTRACT: "C",
+            JobType.TEMPORARY: "T",
+        }.get(job_type_enum, "")
+
+    headers = {
         "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

@@ -1,35 +1,48 @@
+from __future__ import annotations
+
 import re
 import logging
-import numpy as np
-
-import tls_client
 import requests
+import tls_client
+import numpy as np
+from markdownify import markdownify as md
 from requests.adapters import HTTPAdapter, Retry
 
 from ..jobs import JobType
 
 logger = logging.getLogger("JobSpy")
+logger.propagate = False
 if not logger.handlers:
-    logger.setLevel(logging.ERROR)
+    logger.setLevel(logging.INFO)
     console_handler = logging.StreamHandler()
-    console_handler.setLevel(logging.ERROR)
+    format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
-    formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+    formatter = logging.Formatter(format)
     console_handler.setFormatter(formatter)
     logger.addHandler(console_handler)
 
 
-def count_urgent_words(description: str) -> int:
+def set_logger_level(verbose: int = 2):
     """
-    Count the number of urgent words or phrases in a job description.
+    Adjusts the logger's level. This function allows the logging level to be changed at runtime.
-    """
-    urgent_patterns = re.compile(
-        r"\burgen(t|cy)|\bimmediate(ly)?\b|start asap|\bhiring (now|immediate(ly)?)\b",
-        re.IGNORECASE,
-    )
-    matches = re.findall(urgent_patterns, description)
-    count = len(matches)
 
-    return count
+    Parameters:
+    - verbose: int {0, 1, 2} (default=2, all logs)
+    """
+    if verbose is None:
+        return
+    level_name = {2: "INFO", 1: "WARNING", 0: "ERROR"}.get(verbose, "INFO")
+    level = getattr(logging, level_name.upper(), None)
+    if level is not None:
+        logger.setLevel(level)
+    else:
+        raise ValueError(f"Invalid log level: {level_name}")
+
+
+def markdown_converter(description_html: str):
+    if description_html is None:
+        return None
+    markdown = md(description_html)
+    return markdown.strip()
 
 
 def extract_emails_from_text(text: str) -> list[str] | None:
@@ -39,17 +52,18 @@ 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, has_retry: bool = False, delay: int = 1) -> requests.Session:
+def create_session(
+    proxy: dict | None = None,
+    is_tls: bool = True,
+    has_retry: bool = False,
+    delay: int = 1,
+) -> requests.Session:
     """
     Creates a requests session with optional tls, proxy, and retry settings.
-
     :return: A session object
     """
     if is_tls:
-        session = tls_client.Session(
+        session = tls_client.Session(random_tls_extension_order=True)
-            client_identifier="chrome112",
-            random_tls_extension_order=True,
-        )
         session.proxies = proxy
     else:
         session = requests.Session()
@@ -57,16 +71,17 @@ def create_session(proxy: dict | None = None, is_tls: bool = True, has_retry: bo
         if proxy:
             session.proxies.update(proxy)
         if has_retry:
-            retries = Retry(total=3,
+            retries = Retry(
+                total=3,
                 connect=3,
                 status=3,
                 status_forcelist=[500, 502, 503, 504, 429],
-                            backoff_factor=delay)
+                backoff_factor=delay,
+            )
             adapter = HTTPAdapter(max_retries=retries)
 
-            session.mount('http://', adapter)
+            session.mount("http://", adapter)
-            session.mount('https://', adapter)
+            session.mount("https://", adapter)
-
     return session
 
 
@@ -84,17 +99,15 @@ def get_enum_from_job_type(job_type_str: str) -> JobType | None:
 def currency_parser(cur_str):
     # Remove any non-numerical characters
     # except for ',' '.' or '-' (e.g. EUR)
-    cur_str = re.sub("[^-0-9.,]", '', cur_str)
+    cur_str = re.sub("[^-0-9.,]", "", cur_str)
     # Remove any 000s separators (either , or .)
-    cur_str = re.sub("[.,]", '', cur_str[:-3]) + cur_str[-3:]
+    cur_str = re.sub("[.,]", "", cur_str[:-3]) + cur_str[-3:]
 
-    if '.' in list(cur_str[-3:]):
+    if "." in list(cur_str[-3:]):
         num = float(cur_str)
-    elif ',' in list(cur_str[-3:]):
+    elif "," in list(cur_str[-3:]):
-        num = float(cur_str.replace(',', '.'))
+        num = float(cur_str.replace(",", "."))
     else:
         num = float(cur_str)
 
     return np.round(num, 2)
-
-

src/jobspy/scrapers/ziprecruiter/__init__.py

@@ -4,35 +4,80 @@ jobspy.scrapers.ziprecruiter
 
 This module contains routines to scrape ZipRecruiter.
 """
+
+from __future__ import annotations
+
 import math
 import time
-from datetime import datetime, timezone
+from datetime import datetime
 from typing import Optional, Tuple, Any
 
 from concurrent.futures import ThreadPoolExecutor
 
 from .. import Scraper, ScraperInput, Site
-from ..exceptions import ZipRecruiterException
+from ..utils import (
-from ...jobs import JobPost, Compensation, Location, JobResponse, JobType, Country
+    logger,
-from ..utils import count_urgent_words, extract_emails_from_text, create_session
+    extract_emails_from_text,
+    create_session,
+    markdown_converter,
+)
+from ...jobs import (
+    JobPost,
+    Compensation,
+    Location,
+    JobResponse,
+    JobType,
+    Country,
+    DescriptionFormat,
+)
 
 
 class ZipRecruiterScraper(Scraper):
+    base_url = "https://www.ziprecruiter.com"
+    api_url = "https://api.ziprecruiter.com"
+
     def __init__(self, proxy: Optional[str] = None):
         """
         Initializes ZipRecruiterScraper with the ZipRecruiter job search url
         """
-        site = Site(Site.ZIP_RECRUITER)
+        self.scraper_input = None
-        self.url = "https://www.ziprecruiter.com"
         self.session = create_session(proxy)
-        self.get_cookies()
+        self._get_cookies()
-        super().__init__(site, proxy=proxy)
+        super().__init__(Site.ZIP_RECRUITER, proxy=proxy)
 
+        self.delay = 5
         self.jobs_per_page = 20
         self.seen_urls = set()
-        self.delay = 5
 
-    def find_jobs_in_page(
+    def scrape(self, scraper_input: ScraperInput) -> JobResponse:
+        """
+        Scrapes ZipRecruiter for jobs with scraper_input criteria.
+        :param scraper_input: Information about job search criteria.
+        :return: JobResponse containing a list of jobs.
+        """
+        self.scraper_input = scraper_input
+        job_list: list[JobPost] = []
+        continue_token = None
+
+        max_pages = math.ceil(scraper_input.results_wanted / self.jobs_per_page)
+        for page in range(1, max_pages + 1):
+            if len(job_list) >= scraper_input.results_wanted:
+                break
+            if page > 1:
+                time.sleep(self.delay)
+            logger.info(f"ZipRecruiter search page: {page}")
+            jobs_on_page, continue_token = self._find_jobs_in_page(
+                scraper_input, continue_token
+            )
+            if jobs_on_page:
+                job_list.extend(jobs_on_page)
+            else:
+                break
+            if not continue_token:
+                break
+        return JobResponse(jobs=job_list[: scraper_input.results_wanted])
+
+    def _find_jobs_in_page(
         self, scraper_input: ScraperInput, continue_token: str | None = None
     ) -> Tuple[list[JobPost], Optional[str]]:
         """
@@ -41,73 +86,54 @@ class ZipRecruiterScraper(Scraper):
         :param continue_token:
         :return: jobs found on page
         """
-        params = self.add_params(scraper_input)
+        jobs_list = []
+        params = self._add_params(scraper_input)
         if continue_token:
             params["continue_from"] = continue_token
         try:
-            response = self.session.get(
+            res = self.session.get(
-                f"https://api.ziprecruiter.com/jobs-app/jobs",
+                f"{self.api_url}/jobs-app/jobs", headers=self.headers, params=params
-                headers=self.headers(),
-                params=params
-            )
-            if response.status_code != 200:
-                raise ZipRecruiterException(
-                    f"bad response status code: {response.status_code}"
             )
+            if res.status_code not in range(200, 400):
+                if res.status_code == 429:
+                    err = "429 Response - Blocked by ZipRecruiter for too many requests"
+                else:
+                    err = f"ZipRecruiter response status code {res.status_code}"
+                    err += f" with response: {res.text}"  # ZipRecruiter likely not available in EU
+                logger.error(err)
+                return jobs_list, ""
         except Exception as e:
-            if "Proxy responded with non 200 code" in str(e):
+            if "Proxy responded with" in str(e):
-                raise ZipRecruiterException("bad proxy")
+                logger.error(f"Indeed: Bad proxy")
-            raise ZipRecruiterException(str(e))
+            else:
-
+                logger.error(f"Indeed: {str(e)}")
-        response_data = response.json()
+            return jobs_list, ""
-        jobs_list = response_data.get("jobs", [])
-        next_continue_token = response_data.get("continue", None)
 
+        res_data = res.json()
+        jobs_list = res_data.get("jobs", [])
+        next_continue_token = res_data.get("continue", None)
         with ThreadPoolExecutor(max_workers=self.jobs_per_page) as executor:
-            job_results = [executor.submit(self.process_job, job) for job in jobs_list]
+            job_results = [executor.submit(self._process_job, job) for job in jobs_list]
 
         job_list = list(filter(None, (result.result() for result in job_results)))
         return job_list, next_continue_token
 
-    def scrape(self, scraper_input: ScraperInput) -> JobResponse:
+    def _process_job(self, job: dict) -> JobPost | None:
         """
-        Scrapes ZipRecruiter for jobs with scraper_input criteria.
+        Processes an individual job dict from the response
-        :param scraper_input: Information about job search criteria.
-        :return: JobResponse containing a list of jobs.
         """
-        job_list: list[JobPost] = []
-        continue_token = None
-
-        max_pages = math.ceil(scraper_input.results_wanted / self.jobs_per_page)
-
-        for page in range(1, max_pages + 1):
-            if len(job_list) >= scraper_input.results_wanted:
-                break
-
-            if page > 1:
-                time.sleep(self.delay)
-
-            jobs_on_page, continue_token = self.find_jobs_in_page(
-                scraper_input, continue_token
-            )
-            if jobs_on_page:
-                job_list.extend(jobs_on_page)
-
-            if not continue_token:
-                break
-
-        return JobResponse(jobs=job_list[: scraper_input.results_wanted])
-
-    def process_job(self, job: dict) -> JobPost | None:
-        """Processes an individual job dict from the response"""
         title = job.get("name")
-        job_url = f"https://www.ziprecruiter.com/jobs//j?lvk={job['listing_key']}"
+        job_url = f"{self.base_url}/jobs//j?lvk={job['listing_key']}"
         if job_url in self.seen_urls:
             return
         self.seen_urls.add(job_url)
 
         description = job.get("job_description", "").strip()
-
+        description = (
+            markdown_converter(description)
+            if self.scraper_input.description_format == DescriptionFormat.MARKDOWN
+            else description
+        )
         company = job.get("hiring_company", {}).get("name")
         country_value = "usa" if job.get("job_country") == "US" else "canada"
         country_enum = Country.from_string(country_value)
@@ -115,80 +141,66 @@ class ZipRecruiterScraper(Scraper):
         location = Location(
             city=job.get("job_city"), state=job.get("job_state"), country=country_enum
         )
-        job_type = ZipRecruiterScraper.get_job_type_enum(
+        job_type = self._get_job_type_enum(
             job.get("employment_type", "").replace("_", "").lower()
         )
-        date_posted = datetime.fromisoformat(job['posted_time'].rstrip("Z")).date()
+        date_posted = datetime.fromisoformat(job["posted_time"].rstrip("Z")).date()
-
+        comp_interval = job.get("compensation_interval")
+        comp_interval = "yearly" if comp_interval == "annual" else comp_interval
+        comp_min = int(job["compensation_min"]) if "compensation_min" in job else None
+        comp_max = int(job["compensation_max"]) if "compensation_max" in job else None
+        comp_currency = job.get("compensation_currency")
         return JobPost(
+            id=str(job['listing_key']),
             title=title,
             company_name=company,
             location=location,
             job_type=job_type,
             compensation=Compensation(
-                interval="yearly"
+                interval=comp_interval,
-                if job.get("compensation_interval") == "annual"
+                min_amount=comp_min,
-                else job.get("compensation_interval"),
+                max_amount=comp_max,
-                min_amount=int(job["compensation_min"])
+                currency=comp_currency,
-                if "compensation_min" in job
-                else None,
-                max_amount=int(job["compensation_max"])
-                if "compensation_max" in job
-                else None,
-                currency=job.get("compensation_currency"),
             ),
             date_posted=date_posted,
             job_url=job_url,
             description=description,
             emails=extract_emails_from_text(description) if description else None,
-            num_urgent_words=count_urgent_words(description) if description else None,
         )
 
-    def get_cookies(self):
+    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"
-        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"
+        url = f"{self.api_url}/jobs-app/event"
-        self.session.post(url, data=data, headers=ZipRecruiterScraper.headers())
+        self.session.post(url, data=data, headers=self.headers)
 
     @staticmethod
-    def get_job_type_enum(job_type_str: str) -> list[JobType] | None:
+    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 add_params(scraper_input) -> dict[str, str | Any]:
+    def _add_params(scraper_input) -> dict[str, str | Any]:
         params = {
             "search": scraper_input.search_term,
             "location": scraper_input.location,
         }
         if scraper_input.hours_old:
-            fromage = max(scraper_input.hours_old // 24, 1) if scraper_input.hours_old else None
+            params["days"] = max(scraper_input.hours_old // 24, 1)
-            params['days'] = fromage
+        job_type_map = {JobType.FULL_TIME: "full_time", JobType.PART_TIME: "part_time"}
-        job_type_map = {
-            JobType.FULL_TIME: 'full_time',
-            JobType.PART_TIME: 'part_time'
-        }
         if scraper_input.job_type:
-            params['employment_type'] = job_type_map[scraper_input.job_type] if scraper_input.job_type in job_type_map else scraper_input.job_type.value[0]
+            job_type = scraper_input.job_type
+            params["employment_type"] = job_type_map.get(job_type, job_type.value[0])
         if scraper_input.easy_apply:
-            params['zipapply'] = 1
+            params["zipapply"] = 1
         if scraper_input.is_remote:
             params["remote"] = 1
         if scraper_input.distance:
             params["radius"] = scraper_input.distance
+        return {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}
+    headers = {
-
-        return params
-
-    @staticmethod
-    def headers() -> dict:
-        """
-        Returns headers needed for requests
-        :return: dict - Dictionary containing headers
-        """
-        return {
         "Host": "api.ziprecruiter.com",
         "accept": "*/*",
         "x-zr-zva-override": "100000000;vid:ZT1huzm_EQlDTVEc",