_OpenContext/databricks.yml at main · rbarok/_OpenContext · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
# Asset Bundle — kg-agentic-rag
# Deploy-time overrides: .env → BUNDLE_VAR_* (scripts/lib/common.sh).
# Never commit SP application UUIDs, workspace hosts, or personal paths.
#
# Identity / workspace lookups (NO defaults — set in .env):
#   JOBS_SP_NAME, SQL_WAREHOUSE_NAME
# Product defaults below are safe for Clock B job runtime (catalog/models).

bundle:
  name: kg-agentic-rag

include:
  - resources/*.yml

variables:
  uc_catalog:
    description: Unity Catalog name (override via UC_CATALOG / BUNDLE_VAR_uc_catalog)
    default: kg_agentic_rag

  sql_warehouse_name:
    description: SQL warehouse display name — REQUIRED via SQL_WAREHOUSE_NAME / BUNDLE_VAR (no committed default)

  sql_warehouse_id:
    description: Resolved at deploy/runtime from sql_warehouse_name — do not hardcode IDs
    lookup:
      warehouse: ${var.sql_warehouse_name}

  vs_endpoint_name:
    description: Vector Search endpoint base name (override via VS_ENDPOINT_BASE)
    default: kg-agentic-rag-vs

  vs_endpoint_full_name:
    description: Full VS endpoint for the app (override via VS_ENDPOINT_NAME; usually base-target)
    default: ${var.vs_endpoint_name}-${bundle.target}

  jobs_sp_name:
    description: Clock B jobs SP display name — REQUIRED via JOBS_SP_NAME (never commit application UUID)

  jobs_sp_application_id:
    description: Application ID of Clock B jobs SP — lookup from jobs_sp_name (never commit UUIDs)
    lookup:
      service_principal: ${var.jobs_sp_name}

  lakebase_project_id:
    description: Lakebase project id (override via LAKEBASE_PROJECT_ID)
    default: kg-agentic-rag-chat

  lakebase_branch_id:
    description: Branch within project (override via LAKEBASE_BRANCH_ID)
    default: production

  lakebase_database_id:
    description: Database resource id within branch (override via LAKEBASE_DATABASE_ID)
    default: databricks-postgres

  lakebase_branch_path:
    description: Full branch resource path
    default: projects/${var.lakebase_project_id}/branches/${var.lakebase_branch_id}

  lakebase_database_path:
    description: Full database resource path
    default: projects/${var.lakebase_project_id}/branches/${var.lakebase_branch_id}/databases/${var.lakebase_database_id}

  model_id:
    description: Foundation model id (override via MODEL_ID)
    default: system.ai.meta-llama-3-3-70b-instruct

  embedding_model:
    description: Embedding endpoint name (override via EMBEDDING_MODEL)
    default: databricks-gte-large-en

  embedding_dimension:
    description: Embedding vector size (override via EMBEDDING_DIMENSION)
    default: "1024"

  upload_max_bytes:
    description: Max upload size in bytes (override via UPLOAD_MAX_BYTES)
    default: "10485760"

  allowed_mime_types:
    description: Comma-separated MIME allowlist (override via ALLOWED_MIME_TYPES)
    default: application/pdf,text/plain,text/markdown

targets:
  dev:
    default: true
    mode: development
    # workspace.profile: pass at CLI — `databricks bundle deploy -t dev --profile <name>`