Replication Service 📦

info

Click here for the OpenAPI

Purpose

The server lets users to replicate vector data of MapColonies via a REST API, it allows both getting a single vector layer snapshot and keep track of changes in MapColonies vector DB for each layer.

Terminology

There are two types of replicas:

1. Snapshot - A complete layer (e.g buildings, roads, sites) snapshot.
2. Delta - A time-related replica that describes what has changed in a single layer from time T1 to T2.

Usage

You can query for replicas via the replication service API.

caution

You will need an API token as part of the service authentication  

The API in general is split to two parts:

1. Metadata - you can query the service what layers are available for replication.
2. replicas - you can query what replicas are available to download.

Files structure

Snapsho

When receiving a response from the replication service API for snapshot, The response will contain an array of files to download.

Each downloaded file is a zip.

Each zip will contain one or more GeoJSON files.

Each GeoJSON file is a FeatureCollection

File hierarchy for snapshot

snapshot.zip/
├─ uuid1.geojson
├─ uuid2.geojson
├─ uuid3.geojson

Delta

When receiving a response from the replication service API for delta, The response will contain an array of files to download.

1. Each downloaded file is a zip.

2. Each zip will contain:

   1. zero or more GeoJSON files.

      1. Each GeoJSON file is a FeatureCollection

   2. zero or more deleted_uuid.json files.

File hierarchy for delta

delta.zip/
├─ uuid1.geojson
├─ uuid2.geojson
├─ uuid3.geojson
├─ deleted_uuid1.json
├─ deleted_uuid2.json
├─ deleted_uuid3.json

Delta GeoJSON file

The GeoJSON file is following the official specification of GeoJSON (RFC-7946).

Each feature in the feature collection will have a "diff" key, the "diff" key purpose is to present the changes of the feature from its previous version.

The structure (JSON schema) of the "diff" key is as follows:

JSON Schema of Response

{
    "definitions": {
        "diff": {
            "type": "object",
            "additionalProperties": false,
            "properties": {
                "isGeometryChanged": {
                    "type": "boolean"
                },
                "changedTags": {
                    "type": "array",
                    "items": {
                        "type": "string"
                    }
                }
            },
            "required": [
                "changedTags",
                "isGeometryChanged"
            ]
        }
    }
}

If no deleted_uuid.json files are present -> no features were deleted.

if no uuid.geojson files are present -> no changes occurred.

Delta deleted_uuid.json files

Each file is a valid JSON object.

deleted_uuid.json files are indicating what features have been deleted from the layer.

The structure(JSON Schema) of the deleted_uuid.json files is:

{
     "type": "array",
     "items": {
         "type": "string"
     }
 }

Each string is the unique identifier of the deleted feature.

Example of deleted_uuid.json:

["{6b973f22-46d9-4540-908f-fe5395ac0ca5}", "{0e4bcbb5-aa99-4bf6-a818-fc9df28666aa}", "{5652897d-e661-4e14-83fc-969bab027bb2}"]

Important Notes

1. The data is in WGS84 (following the GeoJSON spec).
2. We aim that zip would not exceed a 500MB size threshold.