SQL Workspace Tests [[suite_sql_tests: TestSuite]]TestSuite

• locationtests/workspace/
• formatWorkspace with tests/.sql + tests/.expected.json
• test_count41
• implementations[test_sql.py, test-sql.ts, sql.rs]
• is_data_driventrue
• testsQMD.md Format Specification, Parsers

Description [[description: text]]text💡

SQL queries against workspace — verifying that the SQLite database is correctly populated and queries return expected results.

Use SQL tests when:

• Verifying that a workspace loads correctly into SQLite
• Testing SQL queries against objects
• Checking edges (references between objects)
• Testing filtering by __kind, __namespace, __workspace
• Verifying JSON extraction from the data column

Do not use for QMD.md syntax parsing (use parser microtests), LSP features (use LSP microtests), SQL rewrite logic (use SQL rewrite tests), or workspace validation (use workspace tests).

How to Add a New Test

Tests are automatically discovered by test runners only if they are in scanned directories:

• tests/workspace/ — primary path
• tests/workspace/ — alternative path

Do not create tests elsewhere — they will not run automatically.

Step 1: Create a workspace with a tests/ directory

tests/workspace/my-workspace/
├── readme.qmd.md
├── users.qmd.md
└── tests/
    ├── 001-count-users.sql
    └── 001-count-users.expected.json

Step 2: Create the SQL query

tests/001-count-users.sql:

SELECT COUNT(*) as count
FROM objects
WHERE __kind = 'User'

Step 3: Create the expected result

tests/001-count-users.expected.json:

{
  "columns": ["count"],
  "rows": [[2]]
}

Step 4: Run tests

make test

expected.json Format

{
  "columns": ["__id", "name", "age"],
  "rows": [
    ["alice", "Alice", 30],
    ["bob", "Bob", 25]
  ]
}

• columns — array of column names (same order as SELECT)
• rows — array of rows, each row is an array of values

SQLite Schema

Table: objects

Column	Type	Description
__workspace	TEXT NOT NULL	Workspace ID
__namespace	TEXT NOT NULL DEFAULT ''	Namespace ID
__id	TEXT NOT NULL	Unique identifier within workspace/namespace
__global_id	TEXT GENERATED STORED UNIQUE	Globally unique: workspace:namespace:id
__kind	TEXT	Object type (Kind)
__label	TEXT	Human-readable name
__file	TEXT	File path
__parent	TEXT	Parent object ID
__line	INTEGER	Line number in file
__level	INTEGER	Heading level (1–6)
data	TEXT	JSON with user fields

Primary key: (__workspace, __namespace, __id).

Table: edges

Column	Type	Description
source_id	TEXT NOT NULL	__global_id of source object
source_field	TEXT NOT NULL	Field name containing the reference
target_id	TEXT NOT NULL	__global_id of target object
__workspace	TEXT	Workspace ID

JSON Extraction

Use json_extract for the data column:

SELECT json_extract(data, '$.name') as name FROM objects
SELECT json_extract(data, '$.address.city') as city FROM objects