Skip to content

data-reset.md

Latest commit

 

History

History
40 lines (29 loc) · 1.68 KB

data-reset.md

File metadata and controls

40 lines (29 loc) · 1.68 KB
Raw
title Data Reset
description How mutated data gets restored from the backup twins
audience
developer
architect

Wiki Home › Endpoint Data

Data Reset

Writes to the API persist to disk, so the data drifts as people experiment. Every API file has a pristine .json.backup twin, and the reset routes restore from it:

  • GET /resetit/all — restore every API
  • GET /resetit/:api — restore one API (JSON 404 if the name is unknown)
sequenceDiagram
    participant U as Caller
    participant S as /resetit/all
    participant B as *.json.backup
    participant F as *.json
    U->>S: GET
    S->>B: read each backup
    B->>F: copy over live file
    Note over F: Next data request reads<br/>the restored file — no restart
Loading

The restore is a straight file copy. Because the JSON router reads its file fresh from disk on every request, the reset takes effect immediately — no server restart.

In production the data is reset on a regular cadence (weekly, per the site's disclaimer), and also whenever a new endpoint ships. Want a change to survive resets? Contribute it to the repo so it lands in the backup itself — that is the loop that keeps datasets community-maintained. The reasoning behind this design lives in Why Persistent Writes + Resets.

Key files

Related