fix: update image URL paths to use OSS endpoints

Author: liuruibin

apps/common/handle/impl/qa/xlsx_parse_qa_handle.py

@@ -37,7 +37,7 @@ def handle_sheet(file_name, sheet, image_dict):
         content = str(content.value)
         image = image_dict.get(content, None)
         if image is not None:
-            content = f'![](/api/image/{image.id})'
+            content = f'![](/oss/file/{image.id})'
         paragraph_list.append({'title': title[0:255],
                                'content': content[0:102400],
                                'problem_list': problem_list})

apps/common/handle/impl/qa/zip_parse_qa_handle.py

@@ -90,20 +90,20 @@ def get_image_list(result_list: list, zip_files: List[str]):
                         '/') else source_image_path)
                     if not zip_files.__contains__(image_path):
                         continue
-                    if image_path.startswith('api/file/') or image_path.startswith('api/image/'):
-                        image_id = image_path.replace('api/file/', '').replace('api/image/', '')
+                    if image_path.startswith('oss/file/') or image_path.startswith('oss/image/'):
+                        image_id = image_path.replace('oss/file/', '')
                         if is_valid_uuid(image_id):
                             image_file_list.append({'source_file': image_path,
                                                     'image_id': image_id})
                         else:
                             image_file_list.append({'source_file': image_path,
                                                     'image_id': new_image_id})
-                            content = content.replace(source_image_path, f'/api/image/{new_image_id}')
+                            content = content.replace(source_image_path, f'/oss/file/{new_image_id}')
                             p['content'] = content
                     else:
                         image_file_list.append({'source_file': image_path,
                                                 'image_id': new_image_id})
-                        content = content.replace(source_image_path, f'/api/image/{new_image_id}')
+                        content = content.replace(source_image_path, f'/oss/file/{new_image_id}')
                         p['content'] = content
 
     return image_file_list

apps/common/handle/impl/table/xlsx_parse_table_handle.py

@@ -43,7 +43,7 @@ def fill_merged_cells(self, sheet, image_dict):
 
                 image = image_dict.get(cell_value, None)
                 if image is not None:
-                    cell_value = f'![](/api/image/{image.id})'
+                    cell_value = f'![](/oss/file/{image.id})'
 
                 # 使用标题作为键，单元格的值作为值存入字典
                 row_data[headers[col_idx]] = cell_value
@@ -110,7 +110,6 @@ def get_content(self, file, save_image):
 
                 md_tables += md_table + '\n\n'
 
-                md_tables = md_tables.replace('/api/image/', '/oss/file/')
             return md_tables
         except Exception as e:
             max_kb.error(f'excel split handle error: {e}')

apps/common/handle/impl/text/doc_split_handle.py

@@ -43,7 +43,7 @@ def image_to_mode(image, doc: Document, images_list, get_image_id):
             if len([i for i in images_list if i.id == image_uuid]) == 0:
                 image = File(id=image_uuid, file_name=part.filename, meta={'debug': False, 'content': part.blob})
                 images_list.append(image)
-            return f'![](/api/image/{image_uuid})'
+            return f'![](/oss/file/{image_uuid})'
         return None
     return None
 
@@ -226,7 +226,6 @@ def get_content(self, file, save_image):
             doc = Document(io.BytesIO(buffer))
             content = self.to_md(doc, image_list, get_image_id_func())
             if len(image_list) > 0:
-                content = content.replace('/api/image/', '/oss/file/')
                 save_image(image_list)
             return content
         except BaseException as e:

apps/common/handle/impl/text/xlsx_split_handle.py

@@ -18,7 +18,7 @@
 def post_cell(image_dict, cell_value):
     image = image_dict.get(cell_value, None)
     if image is not None:
-        return f'![](/api/image/{image.id})'
+        return f'![](/oss/file/{image.id})'
     return cell_value.replace('\n', '<br>').replace('|', '&#124;')
 
 

apps/knowledge/serializers/common.py

@@ -181,20 +181,6 @@ def write_image(zip_path: str, image_list: List[str]):
                     os.makedirs(os.path.dirname(file_path))
                 with open(os.path.join(zip_path, file_path), 'wb') as f:
                     f.write(file.get_bytes())
-            # else:
-            #     r = text.replace('(/api/image/', '').replace(')', '')
-            #     r = r.strip().split(" ")[0]
-            #     if not is_valid_uuid(r):
-            #         break
-            #     image_model = QuerySet(Image).filter(id=r).first()
-            #     if image_model is None:
-            #         break
-            #     zip_inner_path = os.path.join('api', 'image', r)
-            #     file_path = os.path.join(zip_path, zip_inner_path)
-            #     if not os.path.exists(os.path.dirname(file_path)):
-            #         os.makedirs(os.path.dirname(file_path))
-            #     with open(file_path, 'wb') as f:
-            #         f.write(image_model.image)
 
 
 def update_document_char_length(document_id: str):
@@ -223,7 +209,6 @@ def or_get(exists_problem_list, content, knowledge_id, document_id, paragraph_id
         return problem, document_id, paragraph_id
 
 
-
 def get_knowledge_operation_object(knowledge_id: str):
     knowledge_model = QuerySet(model=Knowledge).filter(id=knowledge_id).first()
     if knowledge_model is not None: