Skip to content

Data Schema

Schema

honestroles.schema centralizes column-name constants and a TypedDict for job rows. Use these symbols to keep downstream pipelines consistent.

Modules

  • schema.py: Column constants, required/known column lists, and JobsCurrentRow.

Public API reference

Column name constants

These constants are strings used as DataFrame column names:

  • JOB_KEY, COMPANY, SOURCE, JOB_ID
  • TITLE, TEAM
  • LOCATION_RAW, REMOTE_FLAG, REMOTE_TYPE
  • EMPLOYMENT_TYPE, POSTED_AT, UPDATED_AT
  • APPLY_URL
  • DESCRIPTION_HTML, DESCRIPTION_TEXT
  • INGESTED_AT, CONTENT_HASH, LAST_SEEN
  • SALARY_TEXT, SALARY_MIN, SALARY_MAX, SALARY_CURRENCY, SALARY_INTERVAL
  • CITY, REGION, COUNTRY
  • SKILLS, LANGUAGES, BENEFITS, VISA_SPONSORSHIP

REQUIRED_COLUMNS

REQUIRED_COLUMNS: set[str] is the minimal column set required by I/O validation:

  • JOB_KEY, COMPANY, SOURCE, JOB_ID, TITLE, LOCATION_RAW, APPLY_URL, INGESTED_AT, CONTENT_HASH

ALL_COLUMNS

ALL_COLUMNS: list[str] is a stable list of all known columns, in canonical order.

JobsCurrentRow

JobsCurrentRow is a TypedDict describing the jobs_current row shape. All fields are optional (total=False), but values are typed to match the schema.

Usage examples

from honestroles import schema

required = schema.REQUIRED_COLUMNS
all_columns = schema.ALL_COLUMNS

print(schema.TITLE)
print(schema.DESCRIPTION_TEXT)

from honestroles.schema import JobsCurrentRow

row: JobsCurrentRow = {
    "job_key": "acme::greenhouse::123",
    "company": "acme",
    "source": "greenhouse",
    "job_id": "123",
    "title": "Senior Software Engineer",
    "location_raw": "New York, NY",
    "apply_url": "https://example.com/apply",
    "ingested_at": "2025-01-01T00:00:00Z",
    "content_hash": "abc123",
}

Design notes

  • Use constants for column names to avoid typos and to simplify refactors.
  • Validation in honestroles.io.validate_dataframe defaults to REQUIRED_COLUMNS.