Support share/view IDs for Airtable (#3967)

backend/onyx/connectors/airtable/airtable_connector.py

@@ -65,10 +65,25 @@ class AirtableConnector(LoadConnector):
         base_id: str,
         table_name_or_id: str,
         treat_all_non_attachment_fields_as_metadata: bool = False,
+        view_id: str | None = None,
+        share_id: str | None = None,
         batch_size: int = INDEX_BATCH_SIZE,
     ) -> None:
+        """Initialize an AirtableConnector.
+
+        Args:
+            base_id: The ID of the Airtable base to connect to
+            table_name_or_id: The name or ID of the table to index
+            treat_all_non_attachment_fields_as_metadata: If True, all fields except attachments will be treated as metadata.
+                If False, only fields with types in DEFAULT_METADATA_FIELD_TYPES will be treated as metadata.
+            view_id: Optional ID of a specific view to use
+            share_id: Optional ID of a "share" to use for generating record URLs (https://airtable.com/developers/web/api/list-shares)
+            batch_size: Number of records to process in each batch
+        """
         self.base_id = base_id
         self.table_name_or_id = table_name_or_id
+        self.view_id = view_id
+        self.share_id = share_id
         self.batch_size = batch_size
         self._airtable_client: AirtableApi | None = None
         self.treat_all_non_attachment_fields_as_metadata = (
@@ -85,6 +100,39 @@ class AirtableConnector(LoadConnector):
             raise AirtableClientNotSetUpError()
         return self._airtable_client
 
+    @classmethod
+    def _get_record_url(
+        cls,
+        base_id: str,
+        table_id: str,
+        record_id: str,
+        share_id: str | None,
+        view_id: str | None,
+        field_id: str | None = None,
+        attachment_id: str | None = None,
+    ) -> str:
+        """Constructs the URL for a record, optionally including field and attachment IDs
+
+        Full possible structure is:
+
+        https://airtable.com/BASE_ID/SHARE_ID/TABLE_ID/VIEW_ID/RECORD_ID/FIELD_ID/ATTACHMENT_ID
+        """
+        # If we have a shared link, use that view for better UX
+        if share_id:
+            base_url = f"https://airtable.com/{base_id}/{share_id}/{table_id}"
+        else:
+            base_url = f"https://airtable.com/{base_id}/{table_id}"
+
+        if view_id:
+            base_url = f"{base_url}/{view_id}"
+
+        base_url = f"{base_url}/{record_id}"
+
+        if field_id and attachment_id:
+            return f"{base_url}/{field_id}/{attachment_id}?blocks=hide"
+
+        return base_url
+
     def _extract_field_values(
         self,
         field_id: str,
@@ -110,8 +158,10 @@ class AirtableConnector(LoadConnector):
         if field_type == "multipleRecordLinks":
             return []
 
-        # default link to use for non-attachment fields
+        # Get the base URL for this record
-        default_link = f"https://airtable.com/{base_id}/{table_id}/{record_id}"
+        default_link = self._get_record_url(
+            base_id, table_id, record_id, self.share_id, self.view_id or view_id
+        )
 
         if field_type == "multipleAttachments":
             attachment_texts: list[tuple[str, str]] = []
@@ -165,16 +215,15 @@ class AirtableConnector(LoadConnector):
                             extension=file_ext,
                         )
                         if attachment_text:
-                            # slightly nicer loading experience if we can specify the view ID
+                            # Use the helper method to construct attachment URLs
-                            if view_id:
+                            attachment_link = self._get_record_url(
-                                attachment_link = (
+                                base_id,
-                                    f"https://airtable.com/{base_id}/{table_id}/{view_id}/{record_id}"
+                                table_id,
-                                    f"/{field_id}/{attachment_id}?blocks=hide"
+                                record_id,
-                                )
+                                self.share_id,
-                            else:
+                                self.view_id or view_id,
-                                attachment_link = (
+                                field_id,
-                                    f"https://airtable.com/{base_id}/{table_id}/{record_id}"
+                                attachment_id,
-                                    f"/{field_id}/{attachment_id}?blocks=hide"
                             )
                             attachment_texts.append(
                                 (f"{filename}:\n{attachment_text}", attachment_link)

backend/tests/daily/connectors/airtable/test_airtable_basic.py

@@ -9,6 +9,8 @@ from onyx.connectors.airtable.airtable_connector import AirtableConnector
 from onyx.connectors.models import Document
 from onyx.connectors.models import Section
 
+BASE_VIEW_ID = "viwVUEJjWPd8XYjh8"
+
 
 class AirtableConfig(BaseModel):
     base_id: str
@@ -46,6 +48,8 @@ def create_test_document(
     days_since_status_change: int | None,
     attachments: list[tuple[str, str]] | None = None,
     all_fields_as_metadata: bool = False,
+    share_id: str | None = None,
+    view_id: str | None = None,
 ) -> Document:
     base_id = os.environ.get("AIRTABLE_TEST_BASE_ID")
     table_id = os.environ.get("AIRTABLE_TEST_TABLE_ID")
@@ -60,7 +64,13 @@ def create_test_document(
             f"Required environment variables not set: {', '.join(missing_vars)}. "
             "These variables are required to run Airtable connector tests."
         )
-    link_base = f"https://airtable.com/{base_id}/{table_id}"
+    link_base = f"https://airtable.com/{base_id}"
+    if share_id:
+        link_base = f"{link_base}/{share_id}"
+    link_base = f"{link_base}/{table_id}"
+    if view_id:
+        link_base = f"{link_base}/{view_id}"
+
     sections = []
 
     if not all_fields_as_metadata:
@@ -214,6 +224,7 @@ def test_airtable_connector_basic(
             assignee="Chris Weaver (chris@onyx.app)",
             submitted_by="Chris Weaver (chris@onyx.app)",
             all_fields_as_metadata=False,
+            view_id=BASE_VIEW_ID,
         ),
         create_test_document(
             id="reccSlIA4pZEFxPBg",
@@ -234,6 +245,7 @@ def test_airtable_connector_basic(
                 )
             ],
             all_fields_as_metadata=False,
+            view_id=BASE_VIEW_ID,
         ),
     ]
 
@@ -285,6 +297,81 @@ def test_airtable_connector_all_metadata(
                 )
             ],
             all_fields_as_metadata=True,
+            view_id=BASE_VIEW_ID,
+        ),
+    ]
+
+    # Compare documents using the utility function
+    compare_documents(doc_batch, expected_docs)
+
+
+def test_airtable_connector_with_share_and_view(
+    mock_get_unstructured_api_key: MagicMock, airtable_config: AirtableConfig
+) -> None:
+    """Test behavior when using share_id and view_id for URL generation."""
+    SHARE_ID = "shrkfjEzDmLaDtK83"
+
+    connector = AirtableConnector(
+        base_id=airtable_config.base_id,
+        table_name_or_id=airtable_config.table_identifier,
+        treat_all_non_attachment_fields_as_metadata=False,
+        share_id=SHARE_ID,
+        view_id=BASE_VIEW_ID,
+    )
+    connector.load_credentials(
+        {
+            "airtable_access_token": airtable_config.access_token,
+        }
+    )
+    doc_batch_generator = connector.load_from_state()
+    doc_batch = next(doc_batch_generator)
+    with pytest.raises(StopIteration):
+        next(doc_batch_generator)
+
+    assert len(doc_batch) == 2
+
+    expected_docs = [
+        create_test_document(
+            id="rec8BnxDLyWeegOuO",
+            title="Slow Internet",
+            description="The internet connection is very slow.",
+            priority="Medium",
+            status="In Progress",
+            ticket_id="2",
+            created_time="2024-12-24T21:02:49.000Z",
+            status_last_changed="2024-12-24T21:02:49.000Z",
+            days_since_status_change=0,
+            assignee="Chris Weaver (chris@onyx.app)",
+            submitted_by="Chris Weaver (chris@onyx.app)",
+            all_fields_as_metadata=False,
+            share_id=SHARE_ID,
+            view_id=BASE_VIEW_ID,
+        ),
+        create_test_document(
+            id="reccSlIA4pZEFxPBg",
+            title="Printer Issue",
+            description="The office printer is not working.",
+            priority="High",
+            status="Open",
+            ticket_id="1",
+            created_time="2024-12-24T21:02:49.000Z",
+            status_last_changed="2024-12-24T21:02:49.000Z",
+            days_since_status_change=0,
+            assignee="Chris Weaver (chris@onyx.app)",
+            submitted_by="Chris Weaver (chris@onyx.app)",
+            attachments=[
+                (
+                    "Test.pdf:\ntesting!!!",
+                    (
+                        f"https://airtable.com/{airtable_config.base_id}/{SHARE_ID}/"
+                        f"{os.environ['AIRTABLE_TEST_TABLE_ID']}/{BASE_VIEW_ID}/reccSlIA4pZEFxPBg/"
+                        "fld1u21zkJACIvAEF/attlj2UBWNEDZngCc?blocks=hide"
+                    ),
+                )
+            ],
+            all_fields_as_metadata=False,
+            share_id=SHARE_ID,
+            view_id=BASE_VIEW_ID,
         ),
     ]
 

web/src/lib/connectors/connectors.tsx

@@ -1115,7 +1115,24 @@ For example, specifying .*-support.* as a "channel" will cause the connector to
         optional: false,
       },
     ],
-    advanced_values: [],
+    advanced_values: [
+      {
+        type: "text",
+        label: "View ID",
+        name: "view_id",
+        optional: true,
+        description:
+          "If you need to link to a specific View, put that ID here e.g. viwVUEJjWPd8XYjh8.",
+      },
+      {
+        type: "text",
+        label: "Share ID",
+        name: "share_id",
+        optional: true,
+        description:
+          "If you need to link to a specific Share, put that ID here e.g. shrkfjEzDmLaDtK83.",
+      },
+    ],
     overrideDefaultFreq: 60 * 60 * 24,
   },
 };