## Tool: `generate_ddl_suggestion`

Suggests Data Definition Language (DDL) statements for an input query. For example, `CREATE TABLE` or `CREATE VIEW`. The generated DDL provides schema definitions for tables and views that are used in the query. To get DDL suggestions, call this tool, and then use the `fetch_ddl_suggestion` tool with the returned suggestion ID to poll its state until it is `SUCCEEDED` or `FAILED` and retrieve the DDL. The suggestion typically takes 8 to 15 seconds. Wait 8 to 10 seconds before the first check and at least 5 seconds between checks. While the state is `RUNNING`, keep polling instead of stopping or calling this tool again. You can then prepend the retrieved DDL to the original input query and translate it again to improve translation quality. Use this as a fallback when no metadata .ZIP file from the source system is available. The suggested columns and types are inferred only from the columns referenced in the query or from a placeholder column when none are referenced. They are approximate and may not cover the full table schema. Prefer a metadata .ZIP file passed in `metadata_file_path` of `translate_query` for exact results. Ask the user for one before falling back to suggestions. Prepend the suggested DDL to the input query and re-translate. Then, remove the prepended DDL statements such as `CREATE TABLE` or `CREATE VIEW` from the final query keeping `SELECT *` if the inferred DDL expanded `*` into partial or placeholder columns. Report the `RelationNotFound` or `AttributeNotFound` warning and the suggested DDL to the user for review.

The following code sample shows how to use `curl` to call the `generate_ddl_suggestion` MCP tool.

| Curl Request |
|---|
| ```bash curl --location 'https://bigquerymigration.googleapis.com/mcp' \ --header 'content-type: application/json' \ --header 'accept: application/json, text/event-stream' \ --data '{ "method": "tools/call", "params": { "name": "generate_ddl_suggestion", "arguments": { // Provide these details according to the MCP tool specification. } }, "jsonrpc": "2.0", "id": 1 }' ``` |

## Input Schema

Request message for `GenerateDdlSuggestion`.

### GenerateDdlSuggestionRequest

| JSON representation |
|---|
| ``` { "projectNumber": string, "location": string, "inputQuery": string, "sourceDialect": string, "targetDialect": string } ``` |

| Fields ||
|---|---|
| `projectNumber` | `string` Required. The Google Cloud project number. |
| `location` | `string` Required. The location. |
| `inputQuery` | `string` Required. The query string for which the DDL suggestion is generated. |
| `sourceDialect` | `string` Required. The dialect of the source query. The following source to target dialect pairs are supported: source: Teradata, Bteq, Redshift, Oracle, HiveQL, Impala, SparkSQL, Snowflake, Netezza, AzureSynapse, Vertica, SQLServer, Presto, MySQL, Postgresql, Db2, SQLite, Greenplum, BigQuery; target: BigQuery. |
| `targetDialect` | `string` Required. The dialect of the target query. See list of supported pairs in source_dialect. |

## Output Schema

Response message for `GenerateDdlSuggestion`.

### GenerateDdlSuggestionResponse

| JSON representation |
|---|
| ``` { "suggestion": string, "suggestionState": string, "logs": [ { object (`Log`) } ], "errorInfo": { object (`ErrorInfo`) } } ``` |

| Fields ||
|---|---|
| `suggestion` | `string` The ID of the DDL suggestion. Use this ID for the `suggestion` field in `fetch_ddl_suggestion` tool. |
| `suggestionState` | `string` The current state of the DDL suggestion, for example, `RUNNING`, `SUCCEEDED` or `FAILED`. |
| `logs[]` | ``object (`Log`)`` A list of logs generated during the DDL suggestion process. |
| `errorInfo` | ``object (`ErrorInfo`)`` The error information. |

### Log

| JSON representation |
|---|
| ``` { "severity": string, "category": string, "message": string, "action": string, "effect": string, "impactedObject": string } ``` |

| Fields ||
|---|---|
| `severity` | `string` Severity of the translation record, for example, `INFO`, `WARNING`, or `ERROR`. |
| `category` | `string` Category of the error or warning, for example, `SyntaxError`. |
| `message` | `string` Detailed message of the record. |
| `action` | `string` Recommended action to address the log. |
| `effect` | `string` The effect or impact of the issue noted in the log. Effect can be one of the following values: `CORRECTNESS`: Errors with this effect indicate that the translation service couldn't meaningfully process the translation. This is caused by issues in the user's input such as incorrect language or formatting, or using an unsupported file type. `COMPLETENESS`: Errors with this effect indicate that the translation service doesn't have sufficient information to complete the translation. This can be caused by missing information in the user's input such as missing metadata for name resolution. `COMPATIBILITY`: Errors with this effect indicate that the translation service encountered compatibility issues when it processed the translation. This can happen when the target platform doesn't support a feature used in the input script, and the translation service tries to make a semantic approximation for the target platform. `NONE`: Errors with this effect are purely informational messages that have no effect on the output. Effects are ordered by their stage in the translation process. For example, `CORRECTNESS` issues are identified before `COMPLETENESS` issues. |
| `impactedObject` | `string` Name of the object that is impacted by the log message. |

### ErrorInfo

| JSON representation |
|---|
| ``` { "reason": string, "domain": string, "metadata": { string: string, ... } } ``` |

| Fields ||
|---|---|
| `reason` | `string` The reason for the error. This is a constant value that identifies the proximate cause of the error. Error reasons are unique within a particular domain of errors. This should be at most 63 characters and match a regular expression of `[A-Z][A-Z0-9_]+[A-Z0-9]`, which represents UPPER_SNAKE_CASE. |
| `domain` | `string` The logical grouping to which the "reason" belongs. The error domain is typically the registered service name of the tool or product that generates the error. Example: "pubsub.googleapis.com". If the error is generated by some common infrastructure, the error domain must be a globally unique value that identifies the infrastructure. For Google API infrastructure, the error domain is "googleapis.com". |
| `metadata` | `map (key: string, value: string)` Additional structured details about this error. Keys must match a regular expression of `[a-z][a-zA-Z0-9-_]+` but should ideally be lowerCamelCase. Also, they must be limited to 64 characters in length. When identifying the current value of an exceeded limit, the units should be contained in the key, not the value. For example, rather than `{"instanceLimit": "100/request"}`, should be returned as, `{"instanceLimitPerRequest": "100"}`, if the client exceeds the number of instances that can be created in a single (batch) request. An object containing a list of `"key": value` pairs. Example: `{ "name": "wrench", "mass": "1.3kg", "count": "3" }`. |

### MetadataEntry

| JSON representation |
|---|
| ``` { "key": string, "value": string } ``` |

| Fields ||
|---|---|
| `key` | `string` |
| `value` | `string` |

### Tool Annotations


[Tool annotations](https://modelcontextprotocol.io/specification/latest/schema#toolannotations) are sent to MCP clients to describe the basic risk of a given tool. Most clients treat these hints as untrusted, but they can be used to decide when a confirmation prompt might be sent to a user.


Along with the title string, the following boolean hints are defined as follows:

- `readOnlyHint`: If true, the tool doesn't modify its environment. Default: false.
- `destructiveHint`: If true, then the tool can perform destructive actions. If false, then the tool can only perform additive actions. Default: true.
- `idempotentHint`: If true, then calling the tool repeatedly with the same arguments will have no additional effect on its environment. Default: false.
- `openWorldHint`: If true, then the tool can interact with an 'open world' of external entities. If false, then the tool can only interact with internal entities. For example, a web search tool would be open world, while a memory tool would not be open world.

Destructive Hint: ❌ \| Idempotent Hint: ❌ \| Read Only Hint: ❌ \| Open World Hint: ❌