# Changelog of Welcome to the Jungle Jobs Scraper — Full Job & Company Data (`teodor_banea/wttj-jobs-scraper`) Actor

- **URL**: https://apify.com/teodor_banea/wttj-jobs-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/teodor_banea/wttj-jobs-scraper.md

## Changelog

### 0.1.13 (2026-10-22)

- Removed the only-new-since-last-run option and its named key-value store; the actor no longer leaves any storage
  in your account. Use the date filter (`postedWithinDays`) for scheduled runs.

### 0.1.12 (2026-10-20)

- Multi-office jobs: every location field (city, state, country, coordinates, address, postcode) now comes from the
  job page's primary office, which is also listed first in `locations`.
- Titles, summaries and apply links no longer contain HTML tags or entities (also double-encoded ones).
- Only-new feeds sorted by newest: after a run that collected every matching job, later runs of the same search stop
  paging where that run left off (cheap frequent schedules without a date filter; searches with `postedWithinDays` are already small and keep
  paging normally).
- Clearer messages: the cap on unlimited runs in the start line, skipped invalid job URLs in the final status.
- README: timing for large runs and what happens at the run timeout.

### 0.1.0 (2026-10-19)

- First version: Welcome to the Jungle jobs in three modes — search (keyword + filters), company jobs (all open
  jobs of listed companies, by slug or company URL) and job URLs (single job pages, archived jobs included).
- Filters applied by the site's own search: country, city, contract type, remote policy, experience band,
  education level, company sector (group or single), job category (category or sub-category), salary published,
  minimum yearly salary + currency, posted within N days.
- Every match, not just the first 1,000: relevance order for the first 1,000 jobs, the rest after; "newest
  first" is ordered across all matches. Rows are pushed in result order even though details load in parallel.
- 69 fields per row: full description, profile, recruitment process, skills, tools, start date, experience band,
  education, contract duration, salary numbers + readable text, all office locations with coordinates and
  address, benefits, key missions, company website / LinkedIn / size / founding year / headquarters / average
  age / open jobs / labels / commitments / gender-equality index, sectors and job category.
- English or French labels and page links; job texts stay in their original language.
- A job whose details cannot be read is still delivered from its search data (`detailsFetched: false`) and
  listed in ERRORS; deleted jobs and invalid URLs are listed in ERRORS.
- "Only jobs not delivered by earlier runs" (named store `wttj-jobs-seen`, per saved task).
- Apify datacenter proxy built in (no proxy input); in-memory request queue; logs, status and ERRORS show only
  user-level facts.