Convex
This page contains the setup guide and reference information for the Convex source connector.
Get started with Convex at the Convex website. See your data on the Convex dashboard.
Overview
The Convex source connector supports Full Refresh, Incremental Append, and Incremental Dedupe with deletes.
Output schema
This source syncs each Convex table as a separate stream. Check out the list of your tables on the Convex dashboard in the "Data" view.
Types not directly supported by JSON are encoded as described in the JSONSchema for the stream.
For example, the Javascript value new Set(["a", "b"])
is encoded as {"$set": ["a", "b"]}
, as described by the JSONSchema
{"type": "object", "description": "Set", "properties": {"$set": {"type": "array", "items": {"type": "string"}}}}
.
Every record includes the client-defined fields for the table, for example a "messages"
table may contain fields for "author"
and "body"
.
Additionally, each document has system fields:
_id
uniquely identifies the document. It is not changed by.patch
or.replace
operations._creationTime
records a timestamp in milliseconds when the document was initially created. It is not changed by.patch
or.replace
operations._ts
records a timestamp in nanoseconds when the document was last modified. It can be used for ordering operations in Incremental Append mode, and is automatically used in Incremental Dedupe mode._deleted
identifies whether the document was deleted. It can be used to filter deleted documents in Incremental Append mode, and is automatically used to remove documents in Incremental Dedupe mode.
Features
Feature | Supported? |
---|---|
Full Refresh Sync | Yes |
Incremental - Append Sync | Yes |
Incremental - Dedupe Sync | Yes |
Replicate Incremental Deletes | Yes |
Change Data Capture | Yes |
Namespaces | No |
Performance considerations
The Convex connector syncs all documents from the historical log. If you see performance issues due to syncing unnecessary old versions of documents, please reach out to Convex support.