Marc Selwan

This commit significantly **expands and refactors the R2 SQL documentation**, transforming the single SQL reference page into a comprehensive multi-page section. It introduces dedicated pages for **scalar functions, aggregate functions, and complex types**, documenting over 160 new functions and their usage. Additionally, it adds detailed documentation for **Common Table Expressions (CTEs)**, including syntax and examples, and updates the `limitations-best-practices.mdx` to reflect their support. This **documentation enhancement** provides users with a much richer and better-organized guide to R2 SQL capabilities, improving discoverability and understanding of available features. The work also includes a **changelog** entry and minor **refactoring** of the documentation structure.

This commit **fixes** inaccuracies within the **pipelines observability metrics documentation**. It updates the example **GraphQL queries** and descriptive text in `src/content/docs/pipelines/observability/metrics.mdx` to reflect the correct dataset names. Specifically, references to `AccountPipelinesXxxAdaptiveGroups` have been changed to `pipelinesXxxAdaptiveGroups`. This **documentation maintenance** ensures that users can accurately query and understand the available metrics, preventing errors caused by outdated dataset identifiers.

This commit delivers significant **usability improvements** and **new capabilities** for **Pipelines**, enhancing both developer experience and observability. It introduces documentation for **typed pipeline bindings**, allowing developers to generate schema-specific TypeScript types for stream records using `wrangler types`, and adds new **user error metrics** (`AccountPipelinesUserErrorsAdaptiveGroups`) with GraphQL examples for monitoring dropped events. The **Pipelines getting-started experience** is also streamlined by automating R2 bucket/catalog creation and simplifying the `wrangler` setup flow with a "Simple mode." These updates collectively make **Pipelines** easier to configure, more robust for development with type safety, and provide better insights into operational health.

This commit **adds comprehensive documentation** for the newly introduced **approximate aggregation functions** within **R2 SQL**. It updates the `sql-reference.mdx` to detail their syntax, usage, and examples, alongside a new changelog entry. This **documentation update** provides users with crucial insights into the algorithms employed by these functions and quantifies their benefits. As a **new capability**, these functions offer more efficient data analysis, and this documentation ensures users can effectively leverage these powerful analytical tools.

This commit **configures code ownership** by adding new entries to the `.github/CODEOWNERS` file. It specifically defines owners for documentation paths related to the **Pipelines**, **R2-SQL**, and **R2 Data Catalog** features. This **maintenance** task enhances the project's contribution workflow by ensuring that future changes to these critical documentation areas receive appropriate review. The update improves **documentation quality** and **codebase accountability** for these key subsystems.

This commit **updates documentation** for **R2 Data Catalog** and **R2 Data Catalog Sink**. It introduces new content for **R2 Local Uploads** within the Data Catalog and clarifies **jurisdiction limitations** for R2 Data Catalog buckets. Furthermore, it adds a new limitation regarding jurisdiction support to the **R2 Data Catalog Sink** documentation, ensuring users are aware of geographical restrictions for these features.

This commit significantly **improves the R2 Data Catalog documentation**, providing clearer and more accurate guidance for users. It updates `deleting-data.mdx` with new sections on **automatic table maintenance** and manual data deletion, while `manage-catalogs.mdx` is enhanced with notes on API token permissions and **Wrangler version requirements** for snapshot expiration. Furthermore, `r2.mdx` now includes references to the new `r2 bucket catalog snapshot-expiration enable` and `disable` Wrangler commands, ensuring users have up-to-date instructions for managing their R2 data. This is a crucial **documentation improvement** that clarifies the process of data management and snapshot expiration within R2.

This commit **introduces comprehensive new documentation** and **enhances existing guides** for the **R2 Data Catalog**. A new page, `deleting-data.mdx`, provides detailed instructions and PySpark examples for data deletion, including insights into Iceberg metadata. Additionally, the `table-maintenance.mdx` documentation has been **reorganized and rewritten** to cover compaction and snapshot expiration, while `manage-catalogs.mdx` now includes sections on enabling/disabling snapshot expiration. These **documentation updates** also include new changelog entries for snapshot expiration and compaction, significantly improving user guidance on critical data management operations within R2 Data Catalog.

This commit introduces **significant new capabilities** to **R2 SQL**, enabling users to perform more advanced data analysis and introspection. It adds comprehensive **aggregation support**, including `GROUP BY` and `HAVING` clauses, alongside new **schema discovery commands**. The work primarily involves **documentation updates** across the `R2 SQL` reference, limitations, and changelog to reflect these enhancements, ensuring users can leverage the expanded query functionality. This broadens the analytical power of **R2 SQL**, allowing for more complex data manipulation and database schema inspection.

This commit primarily **fixes a documentation error** in `src/content/docs/r2-sql/query-data.mdx` by updating the required **R2 API token permissions** for querying data with R2 SQL. This **critical update** ensures users follow the correct programmatic API token example, preventing access issues when attempting to query their data. Additionally, it **enhances the R2 Data Catalog examples** by adding a new guide for connecting Trino, providing more integration options for users. This work improves the accuracy and utility of the **R2 SQL documentation**, making it more reliable and comprehensive.

This commit introduces **comprehensive documentation** for the new **R2 SQL product**, enabling users to understand and utilize its capabilities. It includes a wide array of new guides covering **getting started**, **querying data**, **limitations and best practices**, **pricing**, and an **end-to-end pipeline tutorial**. Furthermore, it integrates R2 SQL-specific **Wrangler commands** and environment variables into the existing Wrangler documentation and announces the **open beta** of R2 SQL in the changelog. This significant **documentation addition** provides essential resources for users to adopt and effectively use R2 SQL.

This commit **introduces and documents the compaction feature** for **R2 Data Catalog**, providing users with enhanced data management capabilities. It includes comprehensive **documentation updates**, adding a new page explaining compaction concepts (`about-compaction.mdx`) and integrating instructions for enabling and disabling it within the `manage-catalogs.mdx` page. Crucially, it also adds documentation for new **Wrangler CLI commands**, `wrangler r2 bucket catalog compaction enable` and `disable`, allowing programmatic control over this functionality. A new changelog entry (`2025-09-25-data-catalog-compaction.mdx`) announces this **new capability**, ensuring users are informed about the enhanced data management options available for R2 Data Catalog.

This commit introduces a **new documentation page** (`trino.mdx`) within the **R2 Data Catalog** examples, providing a comprehensive guide and example configuration. It details the steps required for **connecting Apache Trino** to the R2 Data Catalog, simplifying the integration process for users. This **documentation enhancement** offers a practical, reviewed example to help users leverage Trino as a data source. The addition expands the available integration examples, making it easier for users to connect external tools with their R2 data.