This is because we also have another variable named `router` in the submodule `users`.

If we had imported one after the other, like:

```Python
from .routers.items import router
from .routers.users import router
```

the `router` from `users` would overwrite the one from `items` and we wouldn't be able to use them at the same time.

So, to be able to use both of them in the same file, we import the submodules directly:

Import `File` and `UploadFile` from `fastapi`:

{* ../../docs_src/request_files/tutorial001_an_py39.py hl[3] *}

## Define `File` Parameters { #define-file-parameters }

Create file parameters the same way you would for `Body` or `Form`:

{* ../../docs_src/request_files/tutorial001_an_py39.py hl[9] *}

/// info

`File` is a class that inherits directly from `Form`.

            to.ansiHash = from.ansiHash != null ? Arrays.copyOf(from.ansiHash, from.ansiHash.length) : null;
            to.unicodeHash = from.unicodeHash != null ? Arrays.copyOf(from.unicodeHash, from.unicodeHash.length) : null;
        } else {
            NtlmPasswordAuthenticator.cloneInternal(to, from);
        }
    }

    /**
     * Compares two {@code NtlmPasswordAuthentication} objects for

     * @param to the target authenticator
     * @param from the source authenticator
     */
    public static void cloneInternal(Kerb5Authenticator to, Kerb5Authenticator from) {
        NtlmPasswordAuthenticator.cloneInternal(to, from);
        to.setUser(from.getUser());
        to.setRealm(from.getRealm());
        to.setService(from.getService());
        to.setLifeTime(from.getLifeTime());

        long data4 = bb2.getLong();

        // Construct UUID from components - Java UUID expects big-endian representation
        long mostSig = ((long) data1 << 32) | ((long) (data2 & 0xFFFF) << 16) | (data3 & 0xFFFF);
        long leastSig = data4;

        this.guid = new UUID(mostSig, leastSig);
    }

    /**
     * Create a handle GUID from existing UUID
     * @param uuid the UUID to wrap
     */

        assertEquals(translatedName.sid_type, decodedName.sid_type);
        // The decode process doesn't preserve the original length, it reads from the buffer
        assertNotNull(decodedName.name);
        assertNotNull(decodedName.name.buffer);
        // Buffer length is based on maximum_length/2 from the encoded data
        assertEquals(translatedName.name.maximum_length / 2, decodedName.name.buffer.length);

With this, we get tooling support, from editors and mypy as this code is correct in terms of types, but we also get the data filtering from FastAPI.

How does this work? Let's check that out. 🤓

### Type Annotations and Tooling { #type-annotations-and-tooling }

First let's see how editors, mypy and other tools would see this.

///

### What is "Mounting" { #what-is-mounting }

"Mounting" means adding a complete "independent" application in a specific path, that then takes care of handling all the sub-paths.

        }
    }

    // Test with different from addresses
    public void test_differentFromAddresses() {
        String[] fromAddresses = { "******@****.***", "******@****.***", "******@****.***", "******@****.***" };

        for (String from : fromAddresses) {
            ComponentUtil.setFessConfig(new FessConfig.SimpleImpl() {

    /** Offset from the start of the SMB header to the parameter bytes */
    protected int parameterOffset;
    /** Displacement of these parameter bytes from the start of the total parameter block */
    protected int parameterDisplacement;
    /** Offset from the start of the SMB header to the data bytes */
    protected int dataOffset;
    /** Displacement of these data bytes from the start of the total data block */