[Docs] Reorganize SeaTunnel onboarding and architecture docs#10785
[Docs] Reorganize SeaTunnel onboarding and architecture docs#10785corgy-w merged 7 commits intoapache:devfrom
Conversation
DanielLeens
left a comment
There was a problem hiding this comment.
Hi @yzeng1618, thank you for putting this documentation reorganization together. I pulled the PR locally and reviewed the full diff against the current dev baseline rather than only reading the PR description.
Local review branch: seatunnel-review-10785
Head reviewed: 350e5dd318e29a20f74bbe9e00e6ba977fb65e7f
Baseline: upstream/dev merge-base ae8b96caeff98ece65c1cd63d0f88c9ccffbfd5c
What this PR improves
Before this PR, a new user or new contributor had to jump between several scattered pages to understand the first-run path, engine choice, job configuration shape, architecture concepts, and developer entry points.
This PR adds and wires a clearer onboarding path across:
- Getting Started overview and job configuration guide
- Architecture overview, core API, CDC pipeline, type system, plugin discovery, and transform system pages
- Flink and Spark translation layer pages
- Source and Sink connector development pages
- REST API and Web UI operational entry page
In plain terms: this makes the docs feel much more like a guided path instead of a set of separate islands.
Runtime / documentation path checked
This PR does not change SeaTunnel runtime code. The real path it changes is the documentation publish and navigation path:
docs source in this PR
-> docs/sidebars.js
-> Architecture adds configuration/core API/CDC/type/plugin discovery entries
-> Getting Started adds overview and job-configuration-guide
-> Zeta Engine adds rest-api-and-web-ui
-> Developer adds contribution-path/source/sink connector development
website sync path
-> tools/documents/sync.sh
-> copies docs/en, docs/zh, and docs/sidebars.js into the Docusaurus website tree
user reading path
-> Getting Started Overview
-> choose Zeta / Flink / Spark
-> Quick Start With SeaTunnel Engine or engine-specific quick starts
-> Job Configuration Guide
-> Source / Transform / Sink docs
-> Architecture and Developer pages when deeper context is needed
I also checked the technical references in the new pages against the repository:
- CDC references such as
IncrementalSource,IncrementalSourceEnumerator, andIncrementalSourceReaderexist underseatunnel-connectors-v2/connector-cdc/connector-cdc-base. - Flink translation references such as
FlinkSource,FlinkSourceReader, andFlinkSourceEnumeratorexist underseatunnel-translation/seatunnel-translation-flink. - Spark translation references such as
SparkSinkandSparkDataSourceWriterexist underseatunnel-translation/seatunnel-translation-spark. - The quick-start use of
FieldMapperis valid becauseseatunnel-transforms-v2.jaris included in the binary distribution assembly.
Local checks performed
git fetch upstream pull/10785/head:seatunnel-review-10785 --force: passed.git diff --stat upstream/dev...HEAD: 67 files,+6425/-259, all in docs/sidebar scope.git diff --check upstream/dev...HEAD: passed, no whitespace errors found.- Sidebar validation: all 110 explicit doc ids in
docs/sidebars.jsresolve underdocs/en. - Modified-doc relative link validation: no missing relative Markdown links found in the changed docs.
- Code-reference spot check: key directories and classes named by the new architecture/developer pages exist in the local source tree.
- GitHub checks observed:
Build,Notify test workflow, andlabelerare all passing.
I did not run a local Docusaurus build because this repository contains the docs source and sync script, not the website package itself. I also did not run Maven tests because this PR is documentation-only and does not touch runtime code.
Review findings
No blocking issue was found.
Compatibility assessment:
- API: no impact
- Configuration options/defaults: no impact
- Protocols: no impact
- Serialization/state format: no impact
- Runtime behavior: no impact
- Historical compatibility: fully compatible
Side-effect assessment:
- No CPU, memory, GC, network, locking, concurrency, retry, idempotency, or resource-release impact, because the PR does not change runtime code.
- The only practical side effect is a larger docs surface area, and the navigation/link checks look healthy.
Merge conclusion
Conclusion: can merge
Blocking items:
- None.
Non-blocking suggestions:
- Please keep the website-side dead-link / Docusaurus build checks in place after merge. The current PR is fine from this review, but broad documentation reorganizations benefit from continuous link validation as the docs evolve.
Overall evaluation:
This is a focused and useful documentation-architecture improvement. It makes the project friendlier for new users and contributors while staying aligned with the actual SeaTunnel Source -> Transform -> Sink model, CDC flow, plugin discovery path, and engine translation modules. Nice work.
Co-authored-by: David Zollo <[email protected]>
davidzollo
left a comment
There was a problem hiding this comment.
+1
Great job. Thank you very much for your contribution to the documentation, which is of great importance to the users and developers of the SeaTunnel community. Thank you again for your significant contribution to the community!
Purpose of this pull request
This PR reorganizes and expands the SeaTunnel documentation for onboarding, architecture learning, engine usage, and contributor guidance.
refer:https://zread.ai/apache/seatunnel/2-quick-start-installing-and-running-your-first-data-sync-job
Does this PR introduce any user-facing change?
How was this patch tested?
Check list
New License Guide
incompatible-changes.mdto describe the incompatibility caused by this PR.