feat(api): add the Glue Data Catalog to the cloud explorer
CI / checks (pull_request) Successful in 28m15s

Create and delete Glue databases, and describe a folder of CSV, JSON-lines or
Parquet files in S3 as a table: its columns, and for CSV the delimiter and
whether the first line is a header. Tables are written with the storage
settings Athena expects, so TheCloud's Athena can query one as soon as it
exists. Glue jobs and crawlers are not offered: the runtime stores them but
never runs them.
This commit is contained in:
Zakaria
2026-10-25 17:51:16 -04:00
parent 26b635db0b
commit 69d80c60a3
23 changed files with 918 additions and 1 deletions
+1
View File
@@ -90,6 +90,7 @@ cd packages/api && bun run scripts/service-matrix.ts
| Analytics | MSK / Managed Streaming for Kafka | Yes (legacy page) | No | No |
| Analytics | MSK Connect / Managed Streaming for Kafka Connect | Yes (legacy page) | No | No |
| Analytics | Kinesis / Kinesis Data Streams | Yes (list, create, inspect, delete) | No | No |
| Analytics | Glue Data Catalog / Data Catalog | Yes (list, create, delete, inspect) | No | No |
| Provisioning | CloudFormation / Infrastructure as Code | Yes (list, create, update, delete, inspect) | No | No |
| Provisioning | AppConfig / Configuration | Yes (list, create, delete, inspect) | No | No |
| Security | Identity | Yes (list, create, delete, inspect) | No | No |