Conversion workspaces let you fix conversion issues, modify your objects with the
interactive editor, and apply converted schema to your destination database.
This page provides an overview of all actions you can perform with
conversion workspaces.

## Review conversion status

[![Conversion workspace screen with the Conversion Overview tab where
you can see the number of converted objects, conversion issues,
and Gemini-assisted conversion enhancements.](https://docs.cloud.google.com/static/database-migration/docs/images/htm/conversion_workspace_overview.png)](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces#lightbox-trigger) **Figure 1.** Conversion workspace overview screen. (click to enlarge) ![Conversion workspace screen with the Conversion Overview tab where
you can see the number of converted objects, conversion issues,
and Gemini-assisted conversion enhancements.](https://docs.cloud.google.com/static/database-migration/docs/images/htm/conversion_workspace_overview.png)

In the **Conversion overview** section, you can view all information
regarding your schema conversion: the number and type of converted objects,
their conversion status (including open or resolved issues), as well as
the list of enhancements provided by
[Gemini-assisted auto-conversion](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/code-conversion-with-gemini.md.txt#auto-conversion).


You can also use the filters in each category to narrow the number of objects
displayed in the tree view by their type, issue severity, outstanding or resolved
issues. The **Conversion overview** is available only for
[upgraded conversion workspaces](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#ws-upgrade).

To view detailed information about the progress of your code
and schema conversion, do the following:

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. In the **Oracle** tab, use the schema tree view to do one of the following:
   - Select **Oracle** to view the information for all the objects in all the converted schemas in your workspace.
   - Select a specific schema or object to view conversion details for a subset of objects.
3. In the **Conversion overview** tab, explore the information in the **Converted object summary** section. Use the **Show objects** button to apply any filter to the tree view in the **Oracle** tab.
4. Use the **Conversion issues** tab to start working through the conversion issues. For more information, see [Fix conversion issues](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#fix-conversion-issues).

## Remove objects from the source schema

Remove objects from the source schema to exclude them from conversion.
You can later [add removed objects](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#add-source-objects) if needed.

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. In the **Oracle** tab, next to the object you want to remove, select **More** \> **Remove**.
3. In the **Remove object** confirmation dialog, click **Remove**.
4. Refresh your converted schema. Click **Convert source** . Database Migration Service now analyzes your changes and generates the updated PostgreSQL schema.

   You can inspect the converted schema in the
   **AlloyDB for PostgreSQL draft** tab.

## Add objects to the source schema

You can add objects previously removed from the source schema to add them to
the conversion.

> [!NOTE]
> **Note:** This operation only includes objects that were included in the source snapshot. If your source database contains new objects that you want to add to the workspace, [pull the source snapshot again](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#pull-snapshot).

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. In the **Oracle** tab, click **Add objects**.
3. In the **Add objects** section, use the source list to select all entities you want add to the source schema.

   You can use the **Filter objects** button to reduce the number of displayed objects.
   See [Filter objects in the source schema view](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#filter-objects)
4. Click **Add objects**.
5. Refresh your converted schema. Click **Convert source** . Database Migration Service now analyzes your changes and generates the updated PostgreSQL schema.

   You can inspect the converted schema in the
   **AlloyDB for PostgreSQL draft** tab.

## Pull source schema and code snapshot

You can pull schema and code from your source database into the conversion
workspace at any moment. Pulling the source gives you an opportunity to add new,
or update existing objects in the conversion.

Pulling updated database schema and code doesn't reset any custom mappings
that exist in the conversion workspace unless you explicitly choose to remove
custom mappings. Additionally, pulling the source doesn't override SQL changes
on your code objects. You can reset these changes directly at the object level.

If you initialized your conversion workspace with JSON metadata files uploaded
to Cloud Storage, new objects can't be detected automatically since the files
are static representations of your database objects. To refresh the source
schema and preserve your existing custom mappings and SQL modifications, perform
the extraction again and upload the updated schema files to the same folder in
your Cloud Storage bucket (overwriting the previous files) before pulling the
snapshot again. For more information, see
[Extract schema metadata with Python scripts](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#extract-schema-files). If you instead switch
the workspace to a different Cloud Storage folder or to a source connection
profile using **Edit source configuration**, Database Migration Service resets the
workspace and removes all selected objects, draft schemas, custom mappings, and
DDL modifications.

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. Click **Edit configuration** \> **Pull source schema snapshot again**.
3. In the dialog:
   1. Optional: Select **Remove any custom mapping** to remove existing custom mappings. Pulling the source schema doesn't reset SQL code modifications on your objects.
   2. Click **Pull schema snapshot** .

      Database Migration Service pulls the new snapshot from your configured source.
4. Refresh your converted schema. Click **Convert source** . Database Migration Service now analyzes your changes and generates the updated PostgreSQL schema.

   You can inspect the converted schema in the
   **AlloyDB for PostgreSQL draft** tab.

## Use custom conversion mappings

You can customize the conversion logic with a *conversion mapping file*.
The conversion mapping file is a text file contains precise instructions
(referred to as *conversion directives*) for how your Oracle
objects should be converted into PostgreSQL objects.

### Write a custom mapping file

To write your custom mapping files:

1. [Review the list of supported conversion directives](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/about-conversion-workspaces.md.txt#conversion-directives-reference).

2. Use the [example configuration file](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/about-conversion-workspaces.md.txt#sample-mapping-file) as a point of reference.

3. Write your custom conversion mappings in a text editor and upload it
   to the conversion workspace.

### Add custom mapping files

To add a custom conversion mapping file to your workspace, do the following:

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. On the conversion editor page, select **Edit configuration** \> **Customize mapping with a configuration file**.
3. In the **Configuration file** box, click **Browse** and use the system file picker to select your configuration file.
4. Click **Add file** \> **Create mapping**.

### Remove custom mapping files

To remove a custom conversion mapping file from your workspace, do the following:

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. On the conversion editor page, select **Edit configuration** \> **Remove custom mapping**.

## Review conversion results

After you perform the source conversion, you can review the conversion results
and possible issues for every individual converted object in the workspace
editor area. You can also use Google Cloud CLI to save all results and issues
in bulk to a text file.

### Console

1. In the Google Cloud console, go to **Conversion workspaces**.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. Click the display name of the conversion workspace that you want to work with.

   Conversion workspace editor opens.
3. Select the **Oracle** tab, and locate the object for which you want to
   review conversion results in the tree view table.

4. Select the object. Use the **SQL** and **Conversion issues** tabs to
   review the conversion.

5. If you determine that your issue is a warning and doesn't warrant any action,
   you can [mark it as resolved](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#mark-cw-issues-resolved). Otherwise, you can investigate further
   and [fix it in the conversion editor](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#fix-conversion-issues).

### gcloud

With Google Cloud CLI you can print all conversion results or issues to
the terminal. Redirect the output to a file for more convenient
bulk object reviews.

#### Get a list of conversion results

gcloud CLI displays conversion results in the terminal in the form
of *Data Definition Language* (DDL) statements. To save conversion results to
a file, execute the following command:

```bash
  gcloud database-migration conversion-workspaces describe-ddls \
  CONVERSION_WORKSPACE_ID \
  --region=REGION_ID \
  > OUTPUT_FILE_PATH
```

Replace:

- `CONVERSION_WORKSPACE_ID` with the conversion workspace identifier. For information on how to retrieve conversion workspace identifiers, see [View conversion workspace details](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/manage-conversion-workspace).
- `REGION_ID` with the name of the region where the conversion workspace is located.
- `OUTPUT_FILE_PATH` with the path where to the text file where you want
  to save the output.

  Example:

  ```bash
  gcloud database-migration conversion-workspaces describe-ddls \
  my-conversion-workspace \
  --region=us-central1 \
  > ./my-ddls-list.txt
  ```

  Result:

  Your schema conversion results are saved in a text format where the first
  line says `DDLs` and subsequent lines are occupied by SQL statements:

      DDLs
      CREATE SCHEMA IF NOT EXISTS "SCHEMA1";
      ALTER TABLE "SCHEMA1"."EMPLOYEES" ADD CONSTRAINT PK_ID PRIMARY KEY ("ID");
      CREATE OR REPLACE FUNCTION mockschema.func_test_datatype(str1 VARCHAR(65000))
        RETURNS DECIMAL
        LANGUAGE plpgsql
        AS $$
          DECLARE
            str2 VARCHAR(100);
          BEGIN
            SELECT
                employees.first_name
              INTO STRICT
                STR2
              FROM
                mockschema.employees
              WHERE employees.employee_id = CAST(FUNC_TEST_DATATYPE.str1 as DECIMAL)
            ;
            RAISE NOTICE '%', concat('Input : ', FUNC_TEST_DATATYPE.str1, ' Output : ', str2);
            RETURN 0;
          END;
        $$;
      CREATE OR REPLACE PROCEDURE greetings AS BEGIN dbms_output.put_line('Hello World!'); END;
      CREATE SYNONYM TABLE "SCHEMA1"."SYNONYM1" ON "SCHEMA1"."EMPLOYEES";
      CREATE OR REPLACE VIEW "SCHEMA1"."VIEW1" AS SELECT * FROM JOBS;

#### Get a list of conversion issues

To save conversion issues to a file, execute the following command:

```bash
  gcloud database-migration conversion-workspaces describe-issues \
  CONVERSION_WORKSPACE_ID \
  --region=REGION_ID \
  > OUTPUT_FILE_PATH
```

Replace:

- `CONVERSION_WORKSPACE_ID` with the conversion workspace identifier. For information on how to retrieve conversion workspace identifiers, see [View conversion workspace details](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/manage-conversion-workspace).
- `REGION_ID` with the name of the region where the conversion workspace is located.
- `OUTPUT_FILE_PATH` with the path where to the text file where you want to save the output.

<br />

Example:

```bash
  gcloud database-migration conversion-workspaces describe-issues \
  my-conversion-workspace \
  --region=us-central1 \
  > ./my-conversion-issues.txt
```

Result:

All the conversion issues contained in your workspace are saved in a text
format where the first line contains column headers and each subsequent
line contains a separate conversion issue:

      PARENT   NAME               ENTITY_TYPE       ISSUE_TYPE  ISSUE_SEVERITY  ISSUE_CODE  ISSUE_MESSAGE
      SCHEMA1  EMPLOYEES          TABLE             DDL         ERROR           500         unable to parse DDL.
      SCHEMA1  EMPLOYEES          TABLE             CONVERT     WARNING         206         no conversion done.
      SCHEMA1  STORED_PROCEDURE1  STORED_PROCEDURE  DDL         ERROR           500         invalid DDL.
      SCHEMA1  SYNONYM1           SYNONYM           CONVERT     WARNING         206         synonym warning message.

We also recommend that you run *Gemini quality assessment*
reports on your converted objects. These assessments analyze the converted code,
providing feedback on its correctness and behavior compared to the source code.
For more information, see
[Gemini quality assessments](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/code-conversion-with-gemini.md.txt#quality-assessments).

## Fix conversion issues

Database Migration Service might not be able to automatically convert your entire source.
For most Oracle objects, you can use the conversion editor directly
in Database Migration Service to adjust the generated SQL. For others, you might need to
change the object directly in your source database and then
[pull the source snapshot again](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#pull-snapshot).

For a complete list of objects that Database Migration Service supports for editing
directly in the conversion workspace, see
[Editable Oracle objects](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/about-conversion-workspaces.md.txt#editable-objects).

To fix the conversion issues found in your schema, do the following:

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. [Review the conversion results](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#review-results) and identify possible issues.

   You can use the Google Cloud console for reviewing individual objects
   or gcloud CLI for reviewing all objects in bulk.
3. Depending on the type of your issue, you can fix it directly in the
   workspace editor, or you might need to provide a customized
   conversion mapping file. Expand the following sections for more information.

   Regardless of what type of issue you're working with, you can try the
   the Gemini-powered conversion assistant to find a solution.
   For more information, see
   [Use Gemini conversion assistant](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/code-conversion-with-gemini.md.txt#conversion-assistant).
   If you determine that your issue is a warning and doesn't warrant any action,
   you can [mark it as resolved](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#mark-cw-issues-resolved).

   #### Fix issues with objects supported by the built-in editor


   To fix issues encountered with objects that are supported in the workspace
   editor, do the following:
   1. Find your issue in the **Oracle** tree view, and then select the **Code** tab.
   2. Use the SQL editor to update the generated code.Database Migration Service doesn't validate the correctness of SQL code you add in the editor.

   <br />

   #### Fix issues with objects that aren't
   supported by the built-in editor

   To fix issues encountered with objects that aren't supported in the workspace
   editor, perform one of the following:

   Update faulty objects in your source
   :   For some issues, you might need to change the Oracle statement
       directly in your source database. Do the following:
       1. Update the schema and code directly in your source database.
       2. [Pull the updated sources](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#pull-snapshot) to Database Migration Service.
       3. [Convert the sources again](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/convert-sql.md.txt) and check if the issues persist.

   Provide additional mappings with a conversion mapping file

   :   You can use a conversion mapping file to provide precise definitions
       for how Database Migration Service should convert certain PostgreSQL
       objects. To use a conversion mapping file, do the following:

       1. [Write a conversion mapping file](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#write-custom-mappings).
       2. [Add your configuration file to the workspace](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#add-custom-mappings).
       3. [Convert the sources again](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/convert-sql.md.txt) and check if the issues persist.

   Try removing the object from the sources tab
   :   Database Migration Service lets you convert .
       To unblock the conversion process, you can decide to look closer at the
       faulty object at a later time.
       [Remove the object from the sources tab](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#remove-source-objects).
4. [Test your schema](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#test-apply-to-dest) to verify Database Migration Service can correctly apply your code to the destination database.
5. If the tests are successful, we recommend that you use the
   **Conversion issues** tab to mark your issues as **Resolved**.
   This way you can filter out these issues from the conversion overviews
   and pick up later where you left off.

## Mark issues as resolved

As you review and fix conversion issues, you might want to
keep track of which issues you already fixed (or determined that they won't impact
your migration and don't warrant any action). You can use the **Mark as resolved**
feature to filter out such issues to make the issue list easier to navigate.

To mark an issue as resolved, do the following:

1. In the Google Cloud console, go to **Conversion workspaces** and select your workspace.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. Use the **Source tree** view to select the object that has conversion issues.
3. In the **Conversion issues** tab, next to your issue, click **Mark as resolved**.
   Now you can filter out these issues from conversion overview.

   [![Issue resolution screen where you can mark conversion issues
   as resolved to filter them out in the conversion overview screen.](https://docs.cloud.google.com/static/database-migration/docs/images/htm/conversion_issues_mark_resolved.png)](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces#lightbox-trigger) **Figure 2.** Conversion issues screen where you can mark issues as resolved. (click to enlarge) ![Issue resolution screen where you can mark conversion issues
   as resolved to filter them out in the conversion overview screen.](https://docs.cloud.google.com/static/database-migration/docs/images/htm/conversion_issues_mark_resolved.png)

## Test your schema in the destination

Before you apply the schema to the destination database, you can first perform
a test run to proactively check for possible issues.
To perform the test, Database Migration Service creates a temporary
database. The test run doesn't impact your destination AlloyDB for PostgreSQL cluster.

1. Make sure your dedicated migration user has the `CREATEDB`
   permission. For more information, see
   [Create and configure your destination AlloyDB for PostgreSQL cluster](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/configure-your-destination-postgresql-database).

2. In the Google Cloud console, go to **Conversion workspaces**.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
3. Click the display name of the conversion workspace that you want to work with.

   Conversion workspace editor opens.
4. Click **Apply to destination** \> **Test (recommended)**.

   The wizard for applying schema to destination database appears.
5. In the **Define destination** section, select the connection profile that
   points to your destination database.

6. Click **Define and continue**.

7. In the **Select objects and test application** section, select
   the schemas of database entities you want to test for your destination database.

   You can use the **Filter objects** button to reduce the number of displayed objects.
   See [Filter objects in the source schema view](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#filter-objects).
8. Click **Test application**.

   You can review the application status in the **AlloyDB for PostgreSQL** tab.

## Apply schema to destination

When the schema you would like to use in the destination database is converted
according to your requirements and mappings, you can apply the results to the
destination database. Applying schema in the destination doesn't alter any data
on the source database.

> [!CAUTION]
> **Caution:** Applying your schema to the destination cluster impacts your AlloyDB for PostgreSQL cluster. Database Migration Service first cleans the destination database by running the SQL `DROP` command, and then re-creates the objects.

1. In the Google Cloud console, go to **Conversion workspaces**.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. Click the display name of the conversion workspace that you want to work with.

   Conversion workspace editor opens.
3. Click **Apply to destination** \> **Apply**.

   The wizard for applying schema to destination database appears.
4. In the **Define destination** section, select the connection profile that
   points to your destination database.

5. Click **Define and continue**.

6. In the **Review objects and apply conversion to destination** section, select
   the schemas of database entities you want to create in your destination database.

   You can use the **Filter objects** button to reduce the number of displayed objects.
   See [Filter objects in the source schema view](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#filter-objects).
7. Click **Apply to destination**.

   You can review the application status in the **AlloyDB for PostgreSQL** tab.

## Create a migration job from the conversion editor

You can create a migration job that uses your conversion workspace directly
from the conversion editor interface.

1. In the Google Cloud console, go to **Conversion workspaces**.

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. Click the display name of the conversion workspace that you want to work with.

   Conversion workspace editor opens.
3. Click **Create migration job**.

4. Proceed with the standard migration job steps, as outlined in
   [Create a migration job](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/create-migration-job.md.txt).

## Filter objects in the source schema view

Database schemas often contain thousands of objects, making it challenging
to partition conversion work. When you add objects from the schema snapshot
to the source schema view, you can use filters to limit the number of displayed
objects. Filters let you add objects in a more granular fashion and focus
on converting a select subset of your schema.

Use the filtered view when you add objects to the source schema view:

1. In the **Filter** field, use one of the following filtering methods:
2. From the **Properties** list, select one of the following auto-suggested options:
   1. **Object name** to filter the objects by name, for example, `ADMIN`.
   2. **Object type** to filter the objects by several types, such as **Function** or **Table**.
   3. **Conversion status** to filter the objects by the conversion status, such as **Action required** or **No issues**.
   4. **Augmented by Gemini** to filter objects that were converted with [Gemini auto-conversion](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/code-conversion-with-gemini.md.txt#auto-conversion).

   You can combine filter properties with logical operators.
3. Enter your query. For example: `type=table`.

   For more information on the filtering syntax, see
   [Supported filtering syntax](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#filter-syntax).
4. Select the objects that you want to add to the source schema view.

   > [!IMPORTANT]
   > **Important:** Object selections don't persist through filter changes. If you want to add objects that meet different criteria, use a more complex query or add them in separate attempts.

### Supported text filtering syntax

You can filter objects by name with basic free text search, or use a dedicated `type` property.
Both approaches support the
[Google API formal specification for filtering](https://google.aip.dev/160),
meaning you can use literals with wildcards, as well as logical and comparison
operators.

Free text filtering

:   Use free text to filter the objects by name. This approach is case-sensitive
    and supports wildcards.

    **Example**:

    The `*JOB*` query uses wildcards to search for entities that
    contain the `JOB` substring. The filtered view returns some tables
    and one stored procedure:
    ![Screenshot of example results for filtering objects by name.](https://docs.cloud.google.com/static/database-migration/docs/images/filter_name_example.png)

Filter by using the `type` property

:   You can filter objects by all standard types supported in Database Migration Service.

    The `type` property supports the following literals
    with the equality (`=`) and inequality (`!=`) operators:
    `database`, `schema`, `table`, `column`,
    `index`, `sequence`, `stored_procedure`,
    `function`, `view`, `synonym`,
    `materialized_view`, `udt`, `constraint`,
    `database_package`, `trigger`, and `event_trigger`.

    **Example**:

    The `type=table` filter returns only tables present in your
    schema:
    ![Screenshot of example results for filtering objects by type property.](https://docs.cloud.google.com/static/database-migration/docs/images/filter_type_example.png)

### Combine text filtering conditions

You can specify multiple conditions by combining them with logical operators.

For example, to search exclusively for tables whose names contain the
`JOB` or `EMPLOYEE` substrings, use this query:

```bash
  type=table AND (*JOB* OR *EMPLOYEE*)
```

As a result, the filter displays all matching tables:
![Screenshot of example results for filtering objects by a combination of
type and name properties.](https://docs.cloud.google.com/static/database-migration/docs/images/filter_combined_query_example.png)

## Extract schema metadata with Python scripts

When you create a conversion workspace, you can choose to create the initial
schema snapshot with Database Migration Service or manually with Google-provided
Python extraction scripts. If your source schema changes or you want to
include additional objects, you must run the extraction again to refresh the
conversion workspace. Follow these steps:

1. In the Google Cloud console, go to **Conversion workspaces** .

   [Go to Conversion workspaces](https://console.cloud.google.com/dbmigration/conversion-workspaces)
2. Click **Set up workspace**.
3. On the **Create a conversion workspace** page:
   1. Enter the workspace name, identifier, and region.
   2. From the **Source database engine** drop-down menu, select **Oracle**.
   3. From the **Destination database engine** drop-down menu, select **AlloyDB for PostgreSQL**.
   4. *Optional*: Enable Gemini settings for your workspace and click **Continue**. If you don't plan to use this workspace for conversion, and only want to retrieve the Python extraction scripts, you can skip these settings.
4. On the **Define source and pull schema snapshot** page, in the **Source connection method** section, select **Cloud Storage**.
5. In the **Before selecting a folder** section, next to the **Oracle metadata extractor** heading, click **Open**.
6. Follow the instructions in the side panel to download and run the extraction utility against your source database to produce the schema metadata JSON files. This utility requires that you are able to establish a network connection with your source database.
7. Upload the extracted metadata JSON files to the same folder in your dedicated Cloud Storage bucket, overwriting the previous files.

You can now refresh your conversion workspace with a new source schema snapshot.
See [Pull source schema and code snapshot](https://docs.cloud.google.com/database-migration/docs/oracle-to-alloydb/work-with-conversion-workspaces.md.txt#pull-snapshot).