DanielCarter-stack opened a new pull request, #11767:
URL: https://github.com/apache/seatunnel/pull/11767

   ## What Problem Does This PR Solve?
   
   Five V2 connector doc pages were inconsistent with the rest of the connector 
doc set: missing Support Those Engines headers, options tables without a 
description column, broken or undocumented option keys, and only one runnable 
example each. This PR brings all ten files (docs/en + docs/zh) up to the same 
structural standard as the recently-merged connector doc PRs.
   
   ### ClickhouseFile (sink)
   
   - Add the missing Support Those Engines header and a description column on 
the options table.
   - Correct the clickhouse_local_path description: the binary is needed on 
every worker (Spark executor, Flink TaskManager, Zeta worker), not just the 
driver.
   - Add a How it works section that explains why the connector cannot offer 
exactly-once and what to use instead.
   - Add four examples (minimal BATCH, multi-shard password-less SSH, SSH key 
auth, STREAMING with checkpoint).
   
   ### Cloudberry (source)
   
   - Replace the previous key options bullet list with a complete options table 
aligned to the actual JDBC source option keys.
   - Call out partition_lower_bound / partition_upper_bound as the recommended 
pattern for large parallel scans.
   - Add a streaming-poll example and a CDC workaround note.
   
   ### Jira (source)
   
   - Add the missing Support Those Engines header.
   - Add a paginated-read example using cursor-based paging config.
   - Add a POST-with-JQL example that demonstrates content_field.
   - Clarify that the auto-generated Authorization header must not be 
duplicated in headers.
   
   ### Persistiq (source)
   
   - Fix the option-name typo: the docs referred to content_json in the option 
table, but the connector actually exposes content_field. All examples were 
already using content_field.
   - Correct the documented format default from json to TEXT.
   - Add the missing paging, connect_timeout_ms, socket_timeout_ms, 
enable_multi_lines, and json_filed_missed_return_null options.
   - Add a paginated json_field example and a STREAMING poll example.
   
   ### SelectDB Cloud (sink)
   
   - Document two options that exist in the source but were not in the docs: 
sink.label-prefix and sink.flush.queue-size.
   - Add a How CDC works section explaining the three prerequisites for CDC 
ingest: Unique model + batch delete enabled + sink.enable-delete=true.
   - Add a STREAMING-with-checkpoint example, a CDC-from-MySQL-CDC example, and 
a 2PC timeout example.
   
   ## Validation
   
   - 10 files changed, 1406 insertions, 220 deletions.
   - All code fences remain balanced.
   - All HOCON examples use only option keys that exist in the connector source 
code (cross-checked against ClickhouseFileSinkOptions.java, 
SelectDBSinkOptions.java, and HttpSourceOptions.java).
   
   ## Notes
   
   - No support-version rows were added.
   - No connectors touched in the last 7 days (by either merged or open PRs) 
were modified.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to