You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
90 lines
1.9 KiB
YAML
90 lines
1.9 KiB
YAML
site_id: indeed
|
|
enabled: true
|
|
repair_needed: false
|
|
|
|
identity:
|
|
label: Indeed Jobs
|
|
category: aggregator
|
|
base_url: https://www.indeed.com
|
|
start_url_template: "https://www.indeed.com/jobs?q={keywords_urlencoded}&fromage=1&sort=date"
|
|
|
|
browser:
|
|
profile_mode: persistent_chrome_profile
|
|
profile_name: JobAgent
|
|
headed_on_learn: true
|
|
headless_on_run: true
|
|
|
|
search:
|
|
keyword_mode: url_param
|
|
location_mode: url_param
|
|
sort_mode: url_param
|
|
date_mode: url_param
|
|
result_list_wait_selector: "div.job_seen_beacon, td.resultContent, div[data-jk]"
|
|
|
|
search_override:
|
|
keywords: null
|
|
location: null
|
|
|
|
pagination:
|
|
mode: next_button
|
|
next_button_selectors:
|
|
- "a[data-testid='pagination-page-next']"
|
|
- "a[aria-label='Next Page']"
|
|
- "a[aria-label*='next']"
|
|
max_pages: 3
|
|
|
|
extract:
|
|
container_selectors:
|
|
- "div.job_seen_beacon"
|
|
- "div[data-jk]"
|
|
- "td.resultContent"
|
|
fields:
|
|
title:
|
|
text:
|
|
- "h2.jobTitle a span[id^='jobTitle']"
|
|
- "h2.jobTitle span"
|
|
- "h2 a span"
|
|
- "h2[class*='jobTitle'] a"
|
|
company:
|
|
text:
|
|
- "span[data-testid='company-name']"
|
|
- ".company"
|
|
- "span.companyName"
|
|
- "[class*='companyName']"
|
|
location:
|
|
text:
|
|
- "div[data-testid='text-location']"
|
|
- ".companyLocation"
|
|
- "div[class*='companyLocation']"
|
|
posted_text:
|
|
text:
|
|
- "span[data-testid='myJobsStateDate']"
|
|
- "span.date"
|
|
- "span[class*='date']"
|
|
url:
|
|
attr:
|
|
selector: "h2.jobTitle a, h2 a[data-jk]"
|
|
name: href
|
|
salary_text:
|
|
text:
|
|
- "div.salary-snippet-container"
|
|
- "div[data-testid='attribute_snippet_testid']"
|
|
- ".salary-snippet"
|
|
|
|
post_filters:
|
|
include_posted_text:
|
|
- "today"
|
|
- "just posted"
|
|
- "1 day"
|
|
- "active"
|
|
|
|
healthcheck:
|
|
min_items_expected: 0
|
|
required_fields:
|
|
- title
|
|
fail_if_zero_when_known_active: false
|
|
|
|
learn:
|
|
last_learned_at: null
|
|
source_url: null
|