Create a Workspace from DDL

A DDL-based workspace derives database context from object definitions you provide. It is useful when the target database is unreachable, credentials cannot be shared, or the analysis scope is small and well defined.

Goal

Build a self-contained PawSQL workspace from table, view, constraint, and index definitions without a live database connection.

What to include

CREATE TABLE statements for every referenced table;
column names, data types, nullability, and defaults;
primary keys, unique constraints, and relevant foreign keys;
existing index definitions and key order;
view definitions used by the query;
schema-qualified object names where needed;
relevant partitioning or distribution definitions.

Table and column names may be enough for basic parsing, but metadata-dependent rewrites and index analysis often require types, constraints, and existing indexes.

Example

CREATE TABLE orders (
    order_id      BIGINT       NOT NULL,
    customer_id   BIGINT       NOT NULL,
    order_status  VARCHAR(32)  NOT NULL,
    created_at    TIMESTAMP    NOT NULL,
    PRIMARY KEY (order_id)
);

CREATE INDEX idx_orders_customer_created
    ON orders (customer_id, created_at);

Use syntax valid for the actual target engine and version.

Create the workspace

With DDL as the source, create the workspace in these steps:

Create a workspace

Open workspace management in the correct organization or project, then enter a name and purpose.

Select engine and version

Match the DDL and target SQL.

Choose DDL as the source

Open the DDL input or upload flow.

Provide the definitions

Paste or upload current, sanitized DDL.

Review parsing

Inspect recognized schemas, objects, constraints, indexes, and failures.

Multiple schemas

When identical object names exist in multiple schemas:

qualify names in DDL;
set the appropriate default schema;
preserve the schema relationships used in production SQL;
test unqualified names explicitly.

Expected Result

After parsing completes, the workspace contains the recognized schemas, tables, columns, constraints, and indexes defined in the DDL, ready for SQL review and optimization.

Verification

Confirm the workspace was created correctly by checking that the DDL parsed without failures, referenced objects are resolved, and the recognized objects match the supplied definitions.

Limitations

A DDL-only workspace generally cannot provide live statistics, data distribution, database-generated plans, runtime validation, or objects omitted from the supplied definitions.

Do not invent a primary key, uniqueness constraint, or index to make analysis succeed. Incorrect metadata can change rewrite safety and recommendation quality.

Troubleshooting

Symptom	Check first
DDL does not parse	Engine, version, encoding, delimiters, vendor syntax
Query object is unresolved	Missing DDL, schema, case, quoting
Recommendation duplicates an existing index	Existing index definition missing or incomplete

Create a Workspace from DDL

Goal

What to include

Example

Create the workspace

Multiple schemas

Expected Result

Verification

Limitations

Troubleshooting

Next steps

SQL optimization

Create from a database

​Goal

​What to include

​Example

​Create the workspace

​Multiple schemas

​Expected Result

​Verification

​Limitations

​Troubleshooting

​Next steps

SQL optimization

Create from a database

Goal

What to include

Example

Create the workspace

Multiple schemas

Expected Result

Verification

Limitations

Troubleshooting

Next steps