turtlebasket
/
JobSpy
mirror of https://github.com/Bunsly/JobSpy
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix: put num urgent words in output

pull/56/head
Cullen Watson 2023-10-10 10:23:50 -05:00
parent 8ae50a40de
commit 4aa832d3e2
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/jobspy/__init__.py
View File

@ -1,7 +1,7 @@
import pandas as pd import pandas as pd
import concurrent.futures import concurrent.futures
from concurrent.futures import ThreadPoolExecutor from concurrent.futures import ThreadPoolExecutor
from typing import List, Tuple, Optional from typing import Tuple, Optional
from .jobs import JobType, Location from .jobs import JobType, Location
from .scrapers.indeed import IndeedScraper from .scrapers.indeed import IndeedScraper
@ -26,7 +26,7 @@ def _map_str_to_site(site_name: str) -> Site:
def scrape_jobs( def scrape_jobs(
site_name: str | List[str] | Site | List[Site], site_name: str | list[str] | Site | list[Site],
search_term: str, search_term: str,
location: str = "", location: str = "",
distance: int = None, distance: int = None,
@ -110,7 +110,7 @@ def scrape_jobs(
site_value, scraped_data = future.result() site_value, scraped_data = future.result()
site_to_jobs_dict[site_value] = scraped_data site_to_jobs_dict[site_value] = scraped_data
jobs_dfs: List[pd.DataFrame] = [] jobs_dfs: list[pd.DataFrame] = []
for site, job_response in site_to_jobs_dict.items(): for site, job_response in site_to_jobs_dict.items():
for job in job_response.jobs: for job in job_response.jobs:
@ -146,7 +146,7 @@ def scrape_jobs(
if jobs_dfs: if jobs_dfs:
jobs_df = pd.concat(jobs_dfs, ignore_index=True) jobs_df = pd.concat(jobs_dfs, ignore_index=True)
desired_order: List[str] = [ desired_order: list[str] = [
"job_url_hyper" if hyperlinks else "job_url", "job_url_hyper" if hyperlinks else "job_url",
"site", "site",
"title", "title",
@ -155,10 +155,12 @@ def scrape_jobs(
"job_type", "job_type",
"date_posted", "date_posted",
"interval", "interval",
"benefits",
"min_amount", "min_amount",
"max_amount", "max_amount",
"currency", "currency",
"is_remote",
"num_urgent_words",
"benefits",
"emails", "emails",
"description", "description",
] ]