greenhouse

Greenhouse Data Extractor

With thanks to https://github.com/Siddhant-K-code/greenhouse-data-exporter

Running GHDE

Commands

Records representing the major entities in the Harvest API can be retrieved in a single operation for each entity type. These commands can be modified by the before and after creation date parameters.

The following optional command parameters filter the retrieved records by their creation date. Each requires an additional value specifying the date in ISO-8601 format.

--created_after (-a)
--created_before (-b)

Entity retrieval commands:

• applications
• candidates
• jobs
• offers
• prospect_pools
• scorecards
• sources

e.g. To extract all candidate records created after 1st November 2022:

  python extractor.py candidates -a 2022-11-01

Some commands use the candidate records already extracted into the cache. These should be run after all required candidates have been extracted.

Candidate-dependent commands:

• activity_feeds
• attachments

Cache administration commands:

• check

Cache organisation

The cache is a local folder hierarchy that uses the file system to store one file per entity instance. The top-level folders are named according to the entity type they hold. Within that folder are individual JSON files, each holding one instance whose id is used as the filename.

Each entity folder also has an index.csv file which has one row per entity instance file in that folder. The columns are:

• id : the record id
• moniker : the record name or some useful identifier
• timestamp : the time when the record was extracted from Greenhouse.

The hierarchy is:

├── applications       <- All the applications.
├── candidates         <- All the candidates and their attachments.
├── jobs               <- All the jobs.
├── offers             <- All the offers.
├── prospect_pools     <- All the prospect pools.
├── scorecards         <- All the scorecards.
├── sources            <- All the sources.

Attachments

Within the candidates folder, any candidate having a file attachment has a subfolder named like XXXX-attachments, where XXXX is the candidate id. Within that subfolder are one or more subfolders - one per attachment for that candidate. Each of these subfolders is named according to the attachment type (resume, cover_letter, other) and the date when it was uploaded to Greenhouse. Within that lower subfolder should be two files: the attachment file itself and a completion marker file named complete that indicates the file was completely downloaded from Greenhouse.