34282926a1
## What Generates version 2.0 of the Airbyte platform documentation using Docusaurus's built-in versioning system. This creates a frozen snapshot of the current documentation that users can reference. Requested by ian.alton@airbyte.io via [Slack thread](https://airbytehq-team.slack.com/archives/D08FX8EC9L0/p1760490197805979?thread_ts=1760490197.805979). Link to Devin run: https://app.devin.ai/sessions/689693593bac44f4903f476aa17b872e ## How - Ran `pnpm run docusaurus docs:version:platform 2.0` in the docusaurus directory - This automatically: - Created `platform_versioned_docs/version-2.0/` containing a snapshot of all current platform docs - Created `platform_versioned_sidebars/version-2.0-sidebars.json` with the sidebar navigation structure - Updated `platform_versions.json` to add "2.0" to the version list - Ran prettier to format the JSON files - Verified the documentation builds successfully locally (build completed in ~3 minutes with only pre-existing broken anchor warnings) ## Review guide 1. **Verify timing**: Confirm this is the correct time to release version 2.0 of the documentation 2. **Version order**: Check `docusaurus/platform_versions.json` - verify "2.0" is first in the array (newest version first) 3. **Build verification**: Ensure CI/Vercel builds pass without errors 4. **Spot check**: Optionally review 2-3 files in `docusaurus/platform_versioned_docs/version-2.0/` to ensure content looks reasonable Note: This is a standard Docusaurus versioning operation that creates a frozen snapshot of the current "next" documentation. The generated files are extensive (500+ files) but follow Docusaurus conventions. ## User Impact Users will see version 2.0 available in the version dropdown on docs.airbyte.com. This provides a stable reference point for platform documentation at this point in time. Existing versions (1.6, 1.7, 1.8) remain unchanged. ## Can this PR be safely reverted and rolled back? - [x] YES 💚 This is an additive change that doesn't modify existing versioned docs. Reverting would simply remove version 2.0 from the version list and delete the associated documentation files. Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: ian.alton@airbyte.io <ian.alton@airbyte.io>
1.3 KiB
products
| products |
|---|
| all |
import Tabs from "@theme/Tabs"; import TabItem from "@theme/TabItem"; import DocCardList from '@theme/DocCardList';
Connections and streams
A connection is the relationship between a source connector and a destination connector. When you create a connection, you create an automated data pipeline that syncs data from a source to a destination. Connections define things like:
- What data Airbyte should replicate
- How Airbyte should read and write data
- When Airbyte should initiate a data sync
- Where Airbyte should write replicated data
- How Airbyte should handle schema drift
Each iteration of a connection is called a sync. In most cases, you run syncs on an automated schedule, but you can also run them manually.
A stream is a group of related records within a connection. Depending on the destination, a stream might be a table, file, multiple files, or blob. Airbyte uses the term stream to generalize the flow of data to destinations of all types and formats. You can configure each stream if you need granular control over specifics parts of your data. Examples of streams are:
- A table in a relational database
- A resource or API endpoint for a REST API
- The records from a directory containing files in a filesystem