Glue

AWS Glue full control plane — Data Catalog, jobs, crawlers, connections, triggers, workflows, schema registry, sessions, ML transforms, data quality. JSON 1.1 protocol.

fakecloud implements all 267 AWS Glue operations at 100% Smithy conformance. The entire control plane is real and persisted — Data Catalog, jobs, crawlers, classifiers, connections, triggers, workflows, blueprints, dev endpoints, the schema registry, interactive sessions, ML transforms, data quality, user-defined functions, usage profiles, column statistics, and tagging. The Data Catalog is the same store Athena reads through: tables created here surface immediately under AwsDataCatalog for Athena's ListDatabases / GetTableMetadata paths.

Status: full control-plane parity. fakecloud is not a Spark engine — job, crawler, ML, and data-quality execution is synthesized (runs transition through real status lifecycles), but every resource is created, persisted, and echoed back exactly as registered.

Supported today

Data Catalog — Databases — CreateDatabase / GetDatabase / GetDatabases / UpdateDatabase / DeleteDatabase. Per-account namespace, name-uniqueness enforcement, parameter passthrough.
Data Catalog — Tables — CreateTable / GetTable / GetTables / UpdateTable / DeleteTable / BatchDeleteTable, plus GetTableVersion / GetTableVersions / DeleteTableVersion / BatchDeleteTableVersion and GetUnfilteredTableMetadata. Full StorageDescriptor round-trip (columns + types, location, input/output format, SerDe, partition keys, parameters, view text, table type).
Data Catalog — Partitions — CreatePartition / BatchCreatePartition / GetPartition / GetPartitions / BatchGetPartition / UpdatePartition / BatchUpdatePartition / DeletePartition / BatchDeletePartition, partition indexes (CreatePartitionIndex / GetPartitionIndexes / DeletePartitionIndex), and GetUnfilteredPartitionMetadata / GetUnfilteredPartitionsMetadata.
- GetPartitions Expression pruning — the Expression filter is parsed and evaluated server-side. Supports = / != / <> / < / <= / > / >= / IN / BETWEEN / LIKE / IS NULL / IS NOT NULL, with AND / OR / NOT and parentheses; type-aware over string / int / bigint / date / timestamp. Unparseable expressions return InvalidInputException.
Data Catalog — Catalogs & settings — CreateCatalog / GetCatalog / GetCatalogs / UpdateCatalog / DeleteCatalog, GetDataCatalogEncryptionSettings / PutDataCatalogEncryptionSettings, GetResourcePolicy / PutResourcePolicy / DeleteResourcePolicy / GetResourcePolicies, and BatchGetCustomEntityTypes / CreateCustomEntityType / GetCustomEntityType / ListCustomEntityTypes / DeleteCustomEntityType.
Jobs & job runs — CreateJob / GetJob / GetJobs / ListJobs / UpdateJob / DeleteJob / BatchGetJobs, StartJobRun / GetJobRun / GetJobRuns / BatchStopJobRun, job bookmarks (GetJobBookmark / ResetJobBookmark), and UpdateSourceControlFromJob / UpdateJobFromSourceControl.
Crawlers — CreateCrawler / GetCrawler / GetCrawlers / ListCrawlers / UpdateCrawler / DeleteCrawler / BatchGetCrawlers, StartCrawler / StopCrawler, crawler schedule (UpdateCrawlerSchedule / StartCrawlerSchedule / StopCrawlerSchedule), and GetCrawlerMetrics. Real READY ↔ RUNNING transitions — a double StartCrawler returns CrawlerRunningException, an idle StopCrawler returns CrawlerNotRunningException.
Classifiers — CreateClassifier / GetClassifier / GetClassifiers / UpdateClassifier / DeleteClassifier (Grok / XML / JSON / CSV).
Connections — CreateConnection / GetConnection / GetConnections / UpdateConnection / DeleteConnection / BatchDeleteConnection, plus TestConnection, GetConnectionType / ListConnectionTypes, and the connection-integration ops.
Triggers — CreateTrigger / GetTrigger / GetTriggers / ListTriggers / UpdateTrigger / DeleteTrigger / BatchGetTriggers, StartTrigger / StopTrigger. Triggers move through CREATED / ACTIVATED / DEACTIVATED.
Workflows — CreateWorkflow / GetWorkflow / ListWorkflows / UpdateWorkflow / DeleteWorkflow / BatchGetWorkflows, runs (StartWorkflowRun / GetWorkflowRun / GetWorkflowRuns / StopWorkflowRun / ResumeWorkflowRun), and run properties (GetWorkflowRunProperties / PutWorkflowRunProperties).
Blueprints — CreateBlueprint / GetBlueprint / ListBlueprints / UpdateBlueprint / DeleteBlueprint / BatchGetBlueprints, runs (StartBlueprintRun / GetBlueprintRun / GetBlueprintRuns).
Dev endpoints — CreateDevEndpoint / GetDevEndpoint / GetDevEndpoints / ListDevEndpoints / UpdateDevEndpoint / DeleteDevEndpoint.
Schema registry — registries (CreateRegistry / GetRegistry / ListRegistries / UpdateRegistry / DeleteRegistry), schemas (CreateSchema / GetSchema / ListSchemas / UpdateSchema / DeleteSchema), versions (RegisterSchemaVersion / GetSchemaVersion / GetSchemaByDefinition / ListSchemaVersions / DeleteSchemaVersions / GetSchemaVersionsDiff / CheckSchemaVersionValidity / QuerySchemaVersionMetadata), version metadata (PutSchemaVersionMetadata / RemoveSchemaVersionMetadata), and UpdateSchemaVersionMetadata.
Interactive sessions — CreateSession / GetSession / ListSessions / DeleteSession / StopSession, statements (RunStatement / GetStatement / ListStatements / CancelStatement). Statement ids auto-increment per session. GetSessionEndpoint returns a synthesized Spark Connect endpoint (URL + short-lived auth token) for an existing session, and GetDashboardUrl returns a synthesized Spark monitoring dashboard URL for a SESSION or JOB.
ML transforms — CreateMLTransform / GetMLTransform / GetMLTransforms / ListMLTransforms / UpdateMLTransform / DeleteMLTransform, task runs (StartMLEvaluationTaskRun / StartMLLabelingSetGenerationTaskRun / GetMLTaskRun / GetMLTaskRuns / CancelMLTaskRun), and GetMLTransform parameter round-trip / StartExportLabelsTaskRun / StartImportLabelsTaskRun.
Data quality — rulesets (CreateDataQualityRuleset / GetDataQualityRuleset / ListDataQualityRulesets / UpdateDataQualityRuleset / DeleteDataQualityRuleset), evaluation runs (StartDataQualityRulesetEvaluationRun / GetDataQualityRulesetEvaluationRun / ListDataQualityRulesetEvaluationRuns), recommendation runs (StartDataQualityRuleRecommendationRun / GetDataQualityRuleRecommendationRun / ListDataQualityRuleRecommendationRuns), and results (GetDataQualityResult / ListDataQualityResults / BatchGetDataQualityResult).
User-defined functions — CreateUserDefinedFunction / GetUserDefinedFunction / GetUserDefinedFunctions / UpdateUserDefinedFunction / DeleteUserDefinedFunction.
Usage profiles — CreateUsageProfile / GetUsageProfile / ListUsageProfiles / UpdateUsageProfile / DeleteUsageProfile.
Column statistics — for tables and partitions (UpdateColumnStatisticsForTable / GetColumnStatisticsForTable / DeleteColumnStatisticsForTable, and the ...ForPartition variants), plus the async stats-task ops (StartColumnStatisticsTaskRun / GetColumnStatisticsTaskRun / ListColumnStatisticsTaskRuns / StopColumnStatisticsTaskRun and the task-settings / schedule ops).
Table optimizers — CreateTableOptimizer / GetTableOptimizer / ListTableOptimizerRuns / UpdateTableOptimizer / DeleteTableOptimizer / BatchGetTableOptimizer.
Security configurations & misc — CreateSecurityConfiguration / GetSecurityConfiguration / GetSecurityConfigurations / DeleteSecurityConfiguration, integrations, materialized-view refresh task runs, the Glue Identity Center configuration ops, and tagging (TagResource / UntagResource / GetTags).

Server-side validation ports the Smithy @length / @range / enum constraints into every handler, so out-of-bounds and malformed inputs return the AWS-shaped InvalidInputException rather than being silently accepted.

Athena integration

The same Data Catalog state powers Athena's catalog reads:

glue:CreateDatabase -> visible in athena:ListDatabases under AwsDataCatalog.
glue:CreateTable -> visible in athena:GetTableMetadata / athena:ListTableMetadata.
Column types and partition keys round-trip end-to-end, so Athena's DESCRIBE / SHOW TABLES return the schema you registered through Glue.

See the Athena docs for the minimal SQL evaluator that runs over this catalog.

Smoke test

fakecloud &

aws --endpoint-url http://localhost:4566 glue create-database \
  --database-input Name=analytics

aws --endpoint-url http://localhost:4566 glue create-table \
  --database-name analytics \
  --table-input 'Name=events,PartitionKeys=[{Name=dt,Type=string}],StorageDescriptor={Columns=[{Name=id,Type=string}],Location=s3://my-bucket/events/}'

# Register a crawler and run it through its lifecycle.
aws --endpoint-url http://localhost:4566 glue create-crawler \
  --name events-crawler \
  --role arn:aws:iam::000000000000:role/glue \
  --database-name analytics \
  --targets 'S3Targets=[{Path=s3://my-bucket/events/}]'

aws --endpoint-url http://localhost:4566 glue start-crawler --name events-crawler
aws --endpoint-url http://localhost:4566 glue get-crawler --name events-crawler \
  --query 'Crawler.State'   # RUNNING

# Prune partitions server-side.
aws --endpoint-url http://localhost:4566 glue get-partitions \
  --database-name analytics --table-name events \
  --expression "dt >= '2026-05-10'"

Introspection

Three IAM-bypass admin endpoints expose Glue state so test assertions don't have to round-trip through the AWS SDK:

GET /_fakecloud/glue/jobs — every Glue Job recorded by CreateJob, across every account.
GET /_fakecloud/glue/job-runs — every JobRun recorded by StartJobRun. Accepts ?job_name=foo to filter to a single job.
GET /_fakecloud/glue/crawlers — every crawler recorded by CreateCrawler, across every account. Returns name, role, databaseName, state, a targetSummary (e.g. "2 S3, 1 JDBC"), schedule, creationTime, lastUpdated. Sorted by account, then name.

All first-party SDKs ship a glue sub-client wrapping these endpoints (getJobs(), getJobRuns(jobName?), getCrawlers()). See reference/introspection for the full endpoint catalog.

Caveats