docs(integrations): add integration development best practices guide#6141
Draft
alexfrancoeur wants to merge 9 commits into
Draft
docs(integrations): add integration development best practices guide#6141alexfrancoeur wants to merge 9 commits into
alexfrancoeur wants to merge 9 commits into
Conversation
Adds a new guide at /integrations/building-integrations for ISVs, connector builders, and tool vendors integrating with ClickHouse. Covers: protocol selection (HTTP/JDBC/ODBC/native), authentication, schema discovery via system.* tables, data type mapping (unsigned integers, Nullable, LowCardinality, DateTime timezone handling, complex types), query parameterization, output formats, batch writes, idempotent inserts, client/query identification for system.query_log, error handling and streaming error detection, ClickHouse Cloud specifics, testing guidance, and implementation examples for both HTTP API (Python) and JDBC (Java). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
3 Skipped Deployments
|
|
Alex Francoeur seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account. You have signed the CLA already but the status is still pending? Let us recheck it. |
…pecifiers to bare code blocks Resolves markdownlint failures (custom-anchor-headings, MD040) from the docs CI check. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Covers: when to choose MCP over JDBC/HTTP API, using ClickHouse Cloud's built-in remote MCP server, building custom remote MCP servers (transport selection, OAuth vs bearer token auth, read-only enforcement, resource limits, tool description design, prompt injection mitigation, observability via query_id/log_comment). Adds MCP items to checklist. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- HTTP 200 does not guarantee success — explain mid-stream error injection and the X-ClickHouse-Exception-Code header check requirement - Remove session_timezone recommendation (experimental, can break partition pruning) - Explain Too many parts failure mode and 1-2 inserts/sec limit - Add async insert gotchas: error-on-flush with wait_for_async_insert=0, SETTINGS must precede VALUES - Add JDBC PreparedStatement batch reuse regression (driver 0.8.6+) - Add materialized view row count inflation warning for ETL validation - Add mutations-are-async section with deduplication checksum caveat - Add JSON insert DEFAULT expression gotcha and input_format_defaults_for_omitted_fields - Add case sensitivity section (identifiers, function names) - Add Int64/UInt64 JSON precision loss for JavaScript consumers - Note INFORMATION_SCHEMA.TABLE_ROWS is always NULL, FK tables are empty - Note system.query_log has ~7.5s flush delay — do not query immediately Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Split integration development docs into dedicated pages: - ingestion.md: batch fundamentals, formats, async insert, deduplication, schema evolution, CDC/upsert patterns, MV row count gotcha - consumption.md: schema discovery, type mapping, streaming queries, pagination, connection pooling, HTTP 200 error detection - jdbc.md: end-to-end JDBC connector walkthrough — Maven setup, HikariCP, schema introspection, full type table, batch inserts, V1→V2 migration - python.md: clickhouse-connect connector guide — streaming, InsertContext, type mapping, Windows tzdata note - go.md: clickhouse-go connector guide — native and database/sql APIs, type mapping, batch inserts, context cancellation gotcha - javascript.md: @clickhouse/client guide — Int64 precision problem, streaming, web client limitations, CORS guidance Updated index.md routing table and expanded sidebars.js to a category with ingestion/consumption and a nested language guides section. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Extract Remote MCP section from index.md into dedicated mcp.md page - mcp.md adds consumption-focused best practices: schema discovery tool design, result sizing for LLM context, per-tool observability query, focused MCP checklist - Move "Identifying your integration" to position 3 in index.md (right after authentication) — it is foundational setup, not an afterthought - Add MCP to routing table in index.md overview - Wire mcp.md into sidebar between consumption and language guides Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
AlexF Disclaimer
Using Claude to draft some best practices for integration development in our docs
Summary
Adds a new guide at
/integrations/building-integrationsfor ISVs, connector builders, and tool vendors integrating with ClickHouse (BI connectors, ETL pipelines, data catalog adapters, IDE plugins, etc.).This is distinct from the existing language client docs — it targets the developer building on top of ClickHouse on behalf of end users, not an application developer writing queries directly.
What's covered
system.tables,system.columns,system.databasesqueries; when to useINFORMATION_SCHEMAvs system tablesNullable/LowCardinalityunwrapping,FixedStringzero-byte handling,DateTime64timezone semantics, complex types{name:type}syntax + JDBC),query_idfor tracing and idempotent retries, output format selection, compression, pagination, timeoutsexecuteBatch(), batch sizing (10k–100k rows),insert_deduplication_tokenfor retries, no-transaction designUser-Agentheader,log_comment,system.query_logquerysystem.query_logverificationTest plan
yarn start)/integrations/java,/interfaces/http, etc.)🤖 Generated with Claude Code