fix: disable advanced mode in read file components on cloud (#11124)

* disable advanced mode

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes (attempt 2/3)

* fix ruff errors

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes (attempt 2/3)

* add a helper method to handle cloud env

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes (attempt 2/3)

* [autofix.ci] apply automated fixes (attempt 3/3)

* fix ruff errors

* [autofix.ci] apply automated fixes

* disable docling components on cloud

* fix ruff errors

* [autofix.ci] apply automated fixes

* fix ruff errors

* template updates

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes (attempt 2/3)

---------

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

src/backend/tests/unit/components/files_and_knowledge/test_file_component.py

@@ -644,3 +644,77 @@ class TestFileComponentToolMode:
         )
         new_temp_files = temp_files_after - temp_files_before
         assert len(new_temp_files) == 0, f"Temp files not cleaned up: {new_temp_files}"
+
+
+class TestFileComponentCloudEnvironment:
+    """Test FileComponent behavior in cloud environments."""
+
+    def test_advanced_mode_disabled_in_cloud(self, monkeypatch):
+        """Test that advanced_mode and all Docling fields are disabled when ASTRA_CLOUD_DISABLE_COMPONENT is set."""
+        # Set the environment variable to simulate cloud environment
+        monkeypatch.setenv("ASTRA_CLOUD_DISABLE_COMPONENT", "true")
+
+        component = FileComponent()
+        build_config = {
+            "advanced_mode": {"show": True, "value": False},
+            "pipeline": {"show": False},
+            "ocr_engine": {"show": False},
+            "doc_key": {"show": False},
+            "md_image_placeholder": {"show": False},
+            "md_page_break_placeholder": {"show": False},
+            "path": {"file_path": ["document.pdf"]},
+        }
+
+        result = component.update_build_config(build_config, ["document.pdf"], "path")
+
+        # In cloud, advanced_mode should be hidden regardless of file type
+        assert result["advanced_mode"]["show"] is False, "advanced_mode should be hidden in cloud"
+        assert result["advanced_mode"]["value"] is False, "advanced_mode value should be False in cloud"
+        # All related fields should be hidden
+        assert result["pipeline"]["show"] is False
+        assert result["ocr_engine"]["show"] is False
+        assert result["ocr_engine"]["value"] == "None"
+        assert result["doc_key"]["show"] is False
+        assert result["md_image_placeholder"]["show"] is False
+        assert result["md_page_break_placeholder"]["show"] is False
+
+    def test_advanced_mode_toggle_disabled_in_cloud(self, monkeypatch):
+        """Test that toggling advanced_mode in cloud doesn't show Docling fields."""
+        monkeypatch.setenv("ASTRA_CLOUD_DISABLE_COMPONENT", "true")
+
+        component = FileComponent()
+        build_config = {
+            "advanced_mode": {"show": True, "value": True},
+            "pipeline": {"show": False},
+            "ocr_engine": {"show": False},
+            "doc_key": {"show": False},
+            "md_image_placeholder": {"show": False},
+            "md_page_break_placeholder": {"show": False},
+        }
+
+        result = component.update_build_config(build_config, field_value=True, field_name="advanced_mode")
+
+        # Even if advanced_mode is toggled to True, it should be disabled in cloud
+        assert result["advanced_mode"]["show"] is False
+        assert result["advanced_mode"]["value"] is False
+        # All Docling fields should remain hidden
+        assert result["pipeline"]["show"] is False
+        assert result["ocr_engine"]["show"] is False
+        assert result["ocr_engine"]["value"] == "None"
+
+    def test_pipeline_change_disabled_in_cloud(self, monkeypatch):
+        """Test that changing pipeline in cloud doesn't show OCR engine."""
+        monkeypatch.setenv("ASTRA_CLOUD_DISABLE_COMPONENT", "true")
+
+        component = FileComponent()
+        build_config = {
+            "advanced_mode": {"show": False, "value": False},
+            "pipeline": {"show": False},
+            "ocr_engine": {"show": False, "value": "easyocr"},
+        }
+
+        result = component.update_build_config(build_config, "standard", "pipeline")
+
+        # Even if pipeline is set to "standard", OCR engine should be disabled in cloud
+        assert result["ocr_engine"]["show"] is False
+        assert result["ocr_engine"]["value"] == "None"

src/lfx/src/lfx/components/docling/__init__.py

@@ -3,35 +3,69 @@ from __future__ import annotations
 from typing import TYPE_CHECKING, Any
 
 from lfx.components._importing import import_mod
+from lfx.utils.validate_cloud import is_astra_cloud_environment
 
 if TYPE_CHECKING:
-    from .chunk_docling_document import ChunkDoclingDocumentComponent
-    from .docling_inline import DoclingInlineComponent
-    from .docling_remote import DoclingRemoteComponent
-    from .export_docling_document import ExportDoclingDocumentComponent
+    from .chunk_docling_document import ChunkDoclingDocumentComponent  # noqa: F401
+    from .docling_inline import DoclingInlineComponent  # noqa: F401
+    from .docling_remote import DoclingRemoteComponent  # noqa: F401
+    from .export_docling_document import ExportDoclingDocumentComponent  # noqa: F401
 
-_dynamic_imports = {
-    "ChunkDoclingDocumentComponent": "chunk_docling_document",
-    "DoclingInlineComponent": "docling_inline",
-    "DoclingRemoteComponent": "docling_remote",
-    "ExportDoclingDocumentComponent": "export_docling_document",
-}
-
-__all__ = [
+_all_components = [
     "ChunkDoclingDocumentComponent",
     "DoclingInlineComponent",
     "DoclingRemoteComponent",
     "ExportDoclingDocumentComponent",
 ]
 
+_all_dynamic_imports = {
+    "ChunkDoclingDocumentComponent": "chunk_docling_document",
+    "DoclingInlineComponent": "docling_inline",
+    "DoclingRemoteComponent": "docling_remote",
+    "ExportDoclingDocumentComponent": "export_docling_document",
+}
+
+# Components that require local Docling/EasyOCR dependencies (disabled in cloud)
+_cloud_disabled_components = {
+    "ChunkDoclingDocumentComponent",
+    "DoclingInlineComponent",
+    "ExportDoclingDocumentComponent",
+}
+
+
+def _get_available_components() -> list[str]:
+    """Get list of available components, filtering out cloud-disabled ones."""
+    if is_astra_cloud_environment():
+        # Only show DoclingRemoteComponent (Docling Serve) in cloud
+        return [comp for comp in _all_components if comp not in _cloud_disabled_components]
+    return _all_components
+
+
+def _get_dynamic_imports() -> dict[str, str]:
+    """Get dynamic imports dict, filtering out cloud-disabled ones."""
+    if is_astra_cloud_environment():
+        # Only allow DoclingRemoteComponent (Docling Serve) in cloud
+        return {k: v for k, v in _all_dynamic_imports.items() if k not in _cloud_disabled_components}
+    return _all_dynamic_imports
+
+
+# Dynamically set __all__ and _dynamic_imports based on cloud environment
+__all__: list[str] = _get_available_components()  # noqa: PLE0605
+_dynamic_imports: dict[str, str] = _get_dynamic_imports()
+
 
 def __getattr__(attr_name: str) -> Any:
     """Lazily import docling components on attribute access."""
-    if attr_name not in _dynamic_imports:
+    # Check if component is available (not disabled in cloud)
+    if is_astra_cloud_environment() and attr_name in _cloud_disabled_components:
+        msg = f"module '{__name__}' has no attribute '{attr_name}'"
+        raise AttributeError(msg)
+
+    if attr_name not in _all_dynamic_imports:
         msg = f"module '{__name__}' has no attribute '{attr_name}'"
         raise AttributeError(msg)
     try:
-        result = import_mod(attr_name, _dynamic_imports[attr_name], __spec__.parent)
+        result = import_mod(attr_name, _all_dynamic_imports[attr_name], __spec__.parent)
     except (ModuleNotFoundError, ImportError, AttributeError) as e:
         msg = f"Could not import '{attr_name}' from '{__name__}': {e}"
         raise AttributeError(msg) from e
@@ -40,4 +74,4 @@ def __getattr__(attr_name: str) -> Any:
 
 
 def __dir__() -> list[str]:
-    return list(__all__)
+    return _get_available_components()

src/lfx/src/lfx/components/files_and_knowledge/file.py

@@ -189,7 +189,8 @@ class FileComponent(BaseFileComponent):
                 "Enable advanced document processing and export with Docling for PDFs, images, and office documents. "
                 "Note that advanced document processing can consume significant resources."
             ),
-            show=True,
+            # Disabled in cloud
+            show=not is_astra_cloud_environment(),
         ),
         DropdownInput(
             name="pipeline",
@@ -345,6 +346,20 @@ class FileComponent(BaseFileComponent):
         """Return the list of currently selected file paths from the template."""
         return template.get("path", {}).get("file_path", [])
 
+    def _disable_docling_fields_in_cloud(self, build_config: dict[str, Any]) -> None:
+        """Disable all Docling-related fields in cloud environments."""
+        if "advanced_mode" in build_config:
+            build_config["advanced_mode"]["show"] = False
+            build_config["advanced_mode"]["value"] = False
+        # Hide all Docling-related fields
+        docling_fields = ("pipeline", "ocr_engine", "doc_key", "md_image_placeholder", "md_page_break_placeholder")
+        for field in docling_fields:
+            if field in build_config:
+                build_config[field]["show"] = False
+        # Also disable OCR engine specifically
+        if "ocr_engine" in build_config:
+            build_config["ocr_engine"]["value"] = "None"
+
     def update_build_config(
         self,
         build_config: dict[str, Any],
@@ -422,25 +437,50 @@ class FileComponent(BaseFileComponent):
         if field_name == "path":
             paths = self._path_value(build_config)
 
-            # If all files can be processed by docling, do so
-            allow_advanced = all(not file_path.endswith((".csv", ".xlsx", ".parquet")) for file_path in paths)
-            build_config["advanced_mode"]["show"] = allow_advanced
-            if not allow_advanced:
-                build_config["advanced_mode"]["value"] = False
-                for f in ("pipeline", "ocr_engine", "doc_key", "md_image_placeholder", "md_page_break_placeholder"):
-                    if f in build_config:
-                        build_config[f]["show"] = False
+            # Disable in cloud environments
+            if is_astra_cloud_environment():
+                self._disable_docling_fields_in_cloud(build_config)
+            else:
+                # If all files can be processed by docling, do so
+                allow_advanced = all(not file_path.endswith((".csv", ".xlsx", ".parquet")) for file_path in paths)
+                build_config["advanced_mode"]["show"] = allow_advanced
+                if not allow_advanced:
+                    build_config["advanced_mode"]["value"] = False
+                    docling_fields = (
+                        "pipeline",
+                        "ocr_engine",
+                        "doc_key",
+                        "md_image_placeholder",
+                        "md_page_break_placeholder",
+                    )
+                    for field in docling_fields:
+                        if field in build_config:
+                            build_config[field]["show"] = False
 
         # Docling Processing
         elif field_name == "advanced_mode":
-            for f in ("pipeline", "ocr_engine", "doc_key", "md_image_placeholder", "md_page_break_placeholder"):
-                if f in build_config:
-                    build_config[f]["show"] = bool(field_value)
-                    if f == "pipeline":
-                        build_config[f]["advanced"] = not bool(field_value)
+            # Disable in cloud environments - don't show Docling fields even if advanced_mode is toggled
+            if is_astra_cloud_environment():
+                self._disable_docling_fields_in_cloud(build_config)
+            else:
+                docling_fields = (
+                    "pipeline",
+                    "ocr_engine",
+                    "doc_key",
+                    "md_image_placeholder",
+                    "md_page_break_placeholder",
+                )
+                for field in docling_fields:
+                    if field in build_config:
+                        build_config[field]["show"] = bool(field_value)
+                        if field == "pipeline":
+                            build_config[field]["advanced"] = not bool(field_value)
 
         elif field_name == "pipeline":
-            if field_value == "standard":
+            # Disable in cloud environments - don't show OCR engine even if pipeline is changed
+            if is_astra_cloud_environment():
+                self._disable_docling_fields_in_cloud(build_config)
+            elif field_value == "standard":
                 build_config["ocr_engine"]["show"] = True
                 build_config["ocr_engine"]["value"] = "easyocr"
             else:
@@ -1043,10 +1083,16 @@ class FileComponent(BaseFileComponent):
             for file in file_list:
                 extension = file.path.suffix[1:].lower()
                 if extension in self.DOCLING_ONLY_EXTENSIONS:
-                    msg = (
-                        f"File '{file.path.name}' has extension '.{extension}' which requires "
-                        f"Advanced Parser mode. Please enable 'Advanced Parser' to process this file."
-                    )
+                    if is_astra_cloud_environment():
+                        msg = (
+                            f"File '{file.path.name}' has extension '.{extension}' which requires "
+                            f"Advanced Parser mode. Advanced Parser is not available in cloud environments."
+                        )
+                    else:
+                        msg = (
+                            f"File '{file.path.name}' has extension '.{extension}' which requires "
+                            f"Advanced Parser mode. Please enable 'Advanced Parser' to process this file."
+                        )
                     self.log(msg)
                     raise ValueError(msg)