fix: fail safe on missing workspace revisions

This commit is contained in:
2026-08-08 20:03:41 +02:00
parent c3a5621737
commit 1b1213317b
10 changed files with 205 additions and 98 deletions
+29 -56
View File
@@ -1,66 +1,39 @@
language: en
workspace:
schema_version: 3
id: example
name: Example workspace
description: Generic example WorkspaceV3 descriptor.
language: en
dwh:
type: thoth_rest
database:
database: ${THT_DB_NAME}
schema: datawarehouse
endpoint:
base_url: ${THT_DWH_REST_URL}
api_key: ${THT_DWH_API_KEY}
ssl_ca: ${THT_SSL_CA}
engine: postgres
database: postgres
schema: datawarehouse
supported_transports:
- postgres_direct
- rest_api
# Relative logical roots are resolved beneath /data/workspaces/example.
roots:
artifacts: artifacts
indexes: indexes
sessions: sessions
examples:
max_per_column: 10
lsh:
signature_size: 64
n_gram: 3
threshold: 0.5
max_values_per_column: 1000
eligibility:
max_declared_len: 128
max_avg_length: 40
max_sampled_len: 200
ignore_columns: [etl_last_update]
evidence:
source_root: ${THT_DOCS_ROOT}
evidence_dir: evidence
resources:
vector:
# Schema-v3 descriptors publish only the collection identity; the runtime renderer
# supplies this fixed internal Qdrant architecture.
semantic_index:
vector_store:
engine: qdrant
base_url: http://qdrant:6333
collection: example
embeddings:
# Embeddings are fixed to the internal Ollama service for schema-v3 descriptors.
dimensions: 1024
distance: cosine
embedding:
provider: ollama_internal
base_url: http://embedding:11434
model: qwen3-embedding:0.6b
dimensions: 1024
vector:
max_chunk_chars: 4000
llm_policy:
default: zai/glm-5.2
allowed:
- zai/glm-5.2
search:
rrf_k: 60
top_schema_tables: 12
schema_chunk_pool: 150
execution:
allow: [cte_test, explain, preview, aggregate, export]
max_preview_rows: 10
max_export_rows: 100000
statement_timeout_ms: 30000
warn_execution_ms: 5000
max_aggregate_cells: 20
diagnostics:
dwh_rest:
method: GET
path: /health
auth: none
response:
database: database
schema: schema
+31 -32
View File
@@ -1,40 +1,39 @@
language: en
workspace:
schema_version: 3
id: psd-clinical
name: PSD Clinical
description: Example PSD-oriented WorkspaceV3 descriptor.
language: en
dwh:
type: thoth_rest
database:
database: ${THT_DB_NAME}
schema: datawarehouse
endpoint:
base_url: ${THT_DWH_REST_URL}
api_key: ${THT_DWH_API_KEY}
ssl_ca: ${THT_SSL_CA}
engine: postgres
database: postgres
schema: datawarehouse
supported_transports:
- postgres_direct
- rest_api
roots:
artifacts: /data/workspaces/generic/artifacts
indexes: /data/workspaces/generic/indexes
sessions: /data/workspaces/generic/sessions
evidence:
source_root: ${THT_DOCS_ROOT}
evidence_dir: evidence
resources:
vector:
# Schema-v3 descriptors publish only the collection identity; the runtime renderer
# supplies this fixed internal Qdrant architecture.
semantic_index:
vector_store:
engine: qdrant
base_url: http://qdrant:6333
collection: generic
embeddings:
# Embeddings are fixed to the internal Ollama service for schema-v3 descriptors.
collection: psd-clinical
dimensions: 1024
distance: cosine
embedding:
provider: ollama_internal
base_url: http://embedding:11434
model: qwen3-embedding:0.6b
dimensions: 1024
execution:
allow: [cte_test, explain, preview, aggregate, export]
max_preview_rows: 10
max_export_rows: 100000
statement_timeout_ms: 30000
llm_policy:
default: zai/glm-5.2
allowed:
- zai/glm-5.2
diagnostics:
dwh_rest:
method: GET
path: /health
auth: none
response:
database: database
schema: schema