refactor!: Remove Request.id field

docs/03_concepts/code/03_rq.py

@@ -20,13 +20,10 @@ async def main() -> None:
 
         # If you try to add an existing request again, it will not do anything
         add_request_info = await queue.add_request(
-            Request.from_url('http://different-example.com/5')
+            Request.from_url('http://example.com/5')
         )
         Actor.log.info(f'Add request info: {add_request_info}')
 
-        processed_request = await queue.get_request(add_request_info.id)
-        Actor.log.info(f'Processed request: {processed_request}')
-
         # Finally, process the queue until all requests are handled
         while not await queue.is_finished():
             # Fetch the next unhandled request in the queue

docs/04_upgrading/upgrading_to_v3.md

@@ -1,6 +1,6 @@
 ---
-id: upgrading-to-v2
-title: Upgrading to v2
+id: upgrading-to-v3
+title: Upgrading to v3
 ---
 
 This page summarizes the breaking changes between Apify Python SDK v2.x and v3.0.

src/apify/scrapy/requests.py

@@ -122,7 +122,7 @@ def to_scrapy_request(apify_request: ApifyRequest, spider: Spider) -> ScrapyRequ
 
         # Update the meta field with the meta field from the apify_request
         meta = scrapy_request.meta or {}
-        meta.update({'apify_request_id': apify_request.id, 'apify_request_unique_key': apify_request.unique_key})
+        meta.update({'apify_request_unique_key': apify_request.unique_key})
         # scrapy_request.meta is a property, so we have to set it like this
         scrapy_request._meta = meta  # noqa: SLF001
 
@@ -134,7 +134,6 @@ def to_scrapy_request(apify_request: ApifyRequest, spider: Spider) -> ScrapyRequ
             url=apify_request.url,
             method=apify_request.method,
             meta={
-                'apify_request_id': apify_request.id,
                 'apify_request_unique_key': apify_request.unique_key,
             },
         )

src/apify/storage_clients/_apify/_models.py

@@ -94,8 +94,8 @@ class CachedRequest(BaseModel):
     Only internal structure.
     """
 
-    id: str
-    """The ID of the request."""
+    unique_key: str
+    """Unique key of the request."""
 
     was_already_handled: bool
     """Whether the request was already handled."""

tests/integration/test_actor_request_queue.py

@@ -85,6 +85,7 @@ async def test_force_cloud(
 ) -> None:
     request_queue_id = (await apify_named_rq.get_metadata()).id
     request_info = await apify_named_rq.add_request(Request.from_url('http://example.com'))
+    assert request_info.id is not None
     request_queue_client = apify_client_async.request_queue(request_queue_id)
 
     request_queue_details = await request_queue_client.get()

tests/integration/test_request_queue.py

@@ -399,38 +399,35 @@ async def main() -> None:
     assert run_result.status == 'SUCCEEDED'
 
 
-async def test_get_request_by_id(
+async def test_get_request_by_unique_key(
     make_actor: MakeActorFunction,
     run_actor: RunActorFunction,
 ) -> None:
-    """Test retrieving specific requests by their ID."""
+    """Test retrieving specific requests by their unique_key."""
 
     async def main() -> None:
         async with Actor:
             rq = await Actor.open_request_queue()
             Actor.log.info('Request queue opened')
 
-            # Add a request and get its ID
+            # Add a request and get its unique_key
             add_result = await rq.add_request('https://example.com/test')
-            request_id = add_result.id
-            Actor.log.info(f'Request added with ID: {request_id}')
+            request_unique_key = add_result.unique_key
+            Actor.log.info(f'Request added with unique_key: {request_unique_key}')
 
-            # Retrieve the request by ID
-            retrieved_request = await rq.get_request(request_id)
+            # Retrieve the request by unique_key
+            retrieved_request = await rq.get_request(request_unique_key)
             assert retrieved_request is not None, f'retrieved_request={retrieved_request}'
             assert retrieved_request.url == 'https://example.com/test', f'retrieved_request.url={retrieved_request.url}'
-            assert retrieved_request.id == request_id, (
-                f'retrieved_request.id={retrieved_request.id}',
-                f'request_id={request_id}',
-            )
-            Actor.log.info('Request retrieved successfully by ID')
+            assert retrieved_request.unique_key == request_unique_key, (f'{request_unique_key=}',)
+            Actor.log.info('Request retrieved successfully by unique_key')
 
-            # Test with non-existent ID
-            non_existent_request = await rq.get_request('non-existent-id')
+            # Test with non-existent unique_key
+            non_existent_request = await rq.get_request('non-existent-unique_key')
             assert non_existent_request is None, f'non_existent_request={non_existent_request}'
-            Actor.log.info('Non-existent ID correctly returned None')
+            Actor.log.info('Non-existent unique_key correctly returned None')
 
-    actor = await make_actor(label='rq-get-by-id-test', main_func=main)
+    actor = await make_actor(label='rq-get-by-unique-key-test', main_func=main)
     run_result = await run_actor(actor)
     assert run_result.status == 'SUCCEEDED'
 

tests/unit/scrapy/requests/test_to_apify_request.py

@@ -66,7 +66,6 @@ def test_with_id_and_unique_key(spider: Spider) -> None:
         url='https://example.com',
         method='GET',
         meta={
-            'apify_request_id': 'abc123',
             'apify_request_unique_key': 'https://example.com',
             'userData': {'some_user_data': 'hello'},
         },
@@ -77,7 +76,6 @@ def test_with_id_and_unique_key(spider: Spider) -> None:
 
     assert apify_request.url == 'https://example.com'
     assert apify_request.method == 'GET'
-    assert apify_request.id == 'abc123'
     assert apify_request.unique_key == 'https://example.com'
 
     user_data = apify_request.user_data

tests/unit/scrapy/requests/test_to_scrapy_request.py

@@ -36,7 +36,6 @@ def test_without_reconstruction(spider: Spider) -> None:
     assert isinstance(scrapy_request, Request)
     assert apify_request.url == scrapy_request.url
     assert apify_request.method == scrapy_request.method
-    assert apify_request.id == scrapy_request.meta.get('apify_request_id')
     assert apify_request.unique_key == scrapy_request.meta.get('apify_request_unique_key')
 
 
@@ -56,7 +55,6 @@ def test_without_reconstruction_with_optional_fields(spider: Spider) -> None:
     assert isinstance(scrapy_request, Request)
     assert apify_request.url == scrapy_request.url
     assert apify_request.method == scrapy_request.method
-    assert apify_request.id == scrapy_request.meta.get('apify_request_id')
     assert apify_request.unique_key == scrapy_request.meta.get('apify_request_unique_key')
 
     scrapy_request_headers = scrapy_request.headers.get('authorization')
@@ -82,7 +80,6 @@ def test_with_reconstruction(spider: Spider) -> None:
     assert isinstance(scrapy_request, Request)
     assert apify_request.url == scrapy_request.url
     assert apify_request.method == scrapy_request.method
-    assert apify_request.id == scrapy_request.meta.get('apify_request_id')
     assert apify_request.unique_key == scrapy_request.meta.get('apify_request_unique_key')
     assert apify_request.user_data == scrapy_request.meta.get('userData')
 
@@ -106,7 +103,6 @@ def test_with_reconstruction_with_optional_fields(spider: Spider) -> None:
     assert isinstance(scrapy_request, Request)
     assert apify_request.url == scrapy_request.url
     assert apify_request.method == scrapy_request.method
-    assert apify_request.id == scrapy_request.meta.get('apify_request_id')
     assert apify_request.unique_key == scrapy_request.meta.get('apify_request_unique_key')
 
     scrapy_request_headers = scrapy_request.headers.get('authorization')

tests/unit/storage_clients/test_apify_request_queue_client.py

@@ -0,0 +1,38 @@
+import pytest
+
+from apify.storage_clients._apify._request_queue_client import unique_key_to_request_id
+
+
+def test_unique_key_to_request_id_length() -> None:
+    unique_key = 'exampleKey123'
+    request_id = unique_key_to_request_id(unique_key, request_id_length=15)
+    assert len(request_id) == 15, 'Request ID should have the correct length.'
+
+
+def test_unique_key_to_request_id_consistency() -> None:
+    unique_key = 'consistentKey'
+    request_id_1 = unique_key_to_request_id(unique_key)
+    request_id_2 = unique_key_to_request_id(unique_key)
+    assert request_id_1 == request_id_2, 'The same unique key should generate consistent request IDs.'
+
+
+@pytest.mark.parametrize(
+    ('unique_key', 'expected_request_id'),
+    [
+        ('abc', 'ungWv48BzpBQUDe'),
+        ('uniqueKey', 'xiWPs083cree7mH'),
+        ('', '47DEQpj8HBSaTIm'),
+        ('测试中文', 'lKPdJkdvw8MXEUp'),
+        ('test+/=', 'XZRQjhoG0yjfnYD'),
+    ],
+    ids=[
+        'basic_abc',
+        'keyword_uniqueKey',
+        'empty_string',
+        'non_ascii_characters',
+        'url_unsafe_characters',
+    ],
+)
+def test_unique_key_to_request_id_matches_known_values(unique_key: str, expected_request_id: str) -> None:
+    request_id = unique_key_to_request_id(unique_key)
+    assert request_id == expected_request_id, f'Unique key "{unique_key}" should produce the expected request ID.'