feat: http download file

api/core/workflow/nodes/http_request/http_executor.py

@@ -15,9 +15,9 @@ HTTP_REQUEST_DEFAULT_TIMEOUT = (10, 60)
 class HttpExecutorResponse:
     status_code: int
     headers: dict[str, str]
-    body: str
+    body: bytes
 
-    def __init__(self, status_code: int, headers: dict[str, str], body: str):
+    def __init__(self, status_code: int, headers: dict[str, str], body: bytes):
         """
         init
         """
@@ -25,6 +25,34 @@ class HttpExecutorResponse:
         self.headers = headers
         self.body = body
 
+    def get_content_type(self) -> str:
+        """
+        get content type
+        """
+        for key, val in self.headers.items():
+            if key.lower() == 'content-type':
+                return val
+        return ''
+
+    def extract_file(self) -> tuple[str, bytes]:
+        """
+        extract file from response if content type is file related
+        """
+        content_type = self.get_content_type()
+        file_content_types = ['image', 'audio', 'video']
+        for v in file_content_types:
+            if v in content_type:
+                return content_type, self.body
+            
+        return '', b''
+    
+    @property
+    def content(self) -> str:
+        """
+        get content
+        """
+        return self.body.decode('utf-8')
+
 class HttpExecutor:
     server_url: str
     method: str
@@ -192,14 +220,14 @@ class HttpExecutor:
             for k, v in response.headers.items():
                 headers[k] = v
 
-            return HttpExecutorResponse(response.status_code, headers, response.text)
+            return HttpExecutorResponse(response.status_code, headers, response.content)
         elif isinstance(response, requests.Response):
             # get key-value pairs headers
             headers = {}
             for k, v in response.headers.items():
                 headers[k] = v
 
-            return HttpExecutorResponse(response.status_code, headers, response.text)
+            return HttpExecutorResponse(response.status_code, headers, response.content)
         else:
             raise ValueError(f'Invalid response type {type(response)}')
         

api/core/workflow/nodes/http_request/http_file_transformer.py

@@ -0,0 +1,2 @@
+class HttpFileTransformer:
+    pass

api/core/workflow/nodes/http_request/http_request_node.py

@@ -1,10 +1,14 @@
+from mimetypes import guess_extension
+from os import path
 from typing import cast
 
+from core.file.file_obj import FileTransferMethod, FileType, FileVar
+from core.tools.tool_file_manager import ToolFileManager
 from core.workflow.entities.node_entities import NodeRunResult, NodeType
 from core.workflow.entities.variable_pool import VariablePool
 from core.workflow.nodes.base_node import BaseNode
 from core.workflow.nodes.http_request.entities import HttpRequestNodeData
-from core.workflow.nodes.http_request.http_executor import HttpExecutor
+from core.workflow.nodes.http_request.http_executor import HttpExecutor, HttpExecutorResponse
 from models.workflow import WorkflowNodeExecutionStatus
 
 
@@ -33,17 +37,20 @@ class HttpRequestNode(BaseNode):
                 inputs=variables,
                 error=str(e),
                 process_data={
-                    'request': http_executor.to_raw_request()
+                    'request': http_executor.to_raw_request(),
                 }
             )
+        
+        files = self.extract_files(http_executor.server_url, response)
 
         return NodeRunResult(
             status=WorkflowNodeExecutionStatus.SUCCEEDED,
             inputs=variables,
             outputs={
                 'status_code': response.status_code,
-                'body': response.body,
-                'headers': response.headers
+                'body': response.content if not files else '',
+                'headers': response.headers,
+                'files': files,
             },
             process_data={
                 'request': http_executor.to_raw_request(),
@@ -61,3 +68,39 @@ class HttpRequestNode(BaseNode):
         return {
             variable_selector.variable: variable_selector.value_selector for variable_selector in node_data.variables
         }
+
+    def extract_files(self, url: str, response: HttpExecutorResponse) -> list[FileVar]:
+        """
+        Extract files from response
+        """
+        files = []
+        mimetype, file_binary = response.extract_file()
+        # if not image, return directly
+        if 'image' not in mimetype:
+            return files
+        
+        if mimetype:
+            # extract filename from url
+            filename = path.basename(url)
+            # extract extension if possible
+            extension = guess_extension(mimetype) or '.bin'
+
+            tool_file = ToolFileManager.create_file_by_raw(
+                user_id=self.user_id, 
+                tenant_id=self.tenant_id, 
+                conversation_id=None, 
+                file_binary=file_binary, 
+                mimetype=mimetype,
+            )
+
+            files.append(FileVar(
+                tenant_id=self.tenant_id,
+                type=FileType.IMAGE,
+                transfer_method=FileTransferMethod.TOOL_FILE,
+                related_id=tool_file.id,
+                filename=filename,
+                extension=extension,
+                mime_type=mimetype,
+            ))
+
+        return files