bella-openapi 1.0.3.1__tar.gz → 1.0.3.2__tar.gz

{bella_openapi-1.0.3.1/src/bella_openapi.egg-info → bella_openapi-1.0.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bella-openapi
-Version: 1.0.3.1
+Version: 1.0.3.2
 Summary: client for openapi service.
 Home-page: 
 Author: ['tangxiaolong', 'fanqiangwei', 'zhangxiaojia', 'liumin', 'wangyukun']
@@ -12,7 +12,8 @@ License-File: LICENSE
 Requires-Dist: httpx<=0.26.0,>=0.10.0
 Requires-Dist: Werkzeug==3.0.1
 Requires-Dist: tiktoken>=0.5.0
-Requires-Dist: pydantic==2.11.7
+Requires-Dist: pydantic==1.10.14
+Requires-Dist: fastapi==0.110.0
 Dynamic: author
 Dynamic: classifier
 Dynamic: description
@@ -260,3 +261,5 @@ get接口地址：/v1/openapi/log/{requestId}
     * rename 包名
 * 1.0.2
     * 修复bug
+* 1.0.3 
+    * 新增standard domtree

{bella_openapi-1.0.3.1 → bella_openapi-1.0.3.2}/README.md

@@ -235,4 +235,6 @@ get接口地址：/v1/openapi/log/{requestId}
 * 1.0.1
     * rename 包名
 * 1.0.2
-    * 修复bug
+    * 修复bug
+* 1.0.3 
+    * 新增standard domtree

{bella_openapi-1.0.3.1 → bella_openapi-1.0.3.2}/setup.py

@@ -3,7 +3,7 @@ from setuptools import setup, find_packages
 
 SHORT = "client for openapi service."
 
-__version__ = "1.0.3.1"
+__version__ = "1.0.3.2"
 __author__ = ["tangxiaolong", "fanqiangwei", "zhangxiaojia", 'liumin', 'wangyukun']
 __email__ = ''
 readme_path = 'README.md'
@@ -17,7 +17,8 @@ setup(
         'httpx>=0.10.0,<=0.26.0',
         'Werkzeug==3.0.1',
         'tiktoken>=0.5.0',
-        'pydantic==2.11.7'
+        'pydantic==1.10.14',
+        'fastapi==0.110.0'
     ],
     url='',
     author=__author__,

{bella_openapi-1.0.3.1 → bella_openapi-1.0.3.2}/src/bella_openapi/entity/standard_domtree.py

@@ -9,7 +9,7 @@ StandardDomTree - 新的DOM树协议定义
 """
 
 from typing import List, Optional, Literal, Any, Union
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, Field, root_validator
 import tiktoken
 
 # 布局类型映射表：将旧解析器的布局类型映射到新的标准类型
@@ -92,6 +92,23 @@ class StandardNode(BaseModel):
     path: Optional[List[int]] = Field(default_factory=list)  # 编号的层级信息，例如：1.2.1
     element: Optional[Union[StandardElement, StandardImageElement, StandardTableElement]] = None  # 元素信息，当前节点的元素详情
     children: Optional[List["StandardNode"]] = Field(default_factory=list)  # 子节点信息，当前节点的所有子节点
+    
+    @root_validator(pre=True)
+    def parse_element(cls, values):
+        """根据element的type字段正确实例化element"""
+        if 'element' in values and values['element'] is not None:
+            element_data = values['element']
+            # 只有当element是字典时才需要转换，如果已经是正确的实例则保持不变
+            if isinstance(element_data, dict):
+                element_type = element_data.get('type', '')
+                if element_type == 'Figure':
+                    values['element'] = StandardImageElement(**element_data)
+                elif element_type == 'Table':
+                    values['element'] = StandardTableElement(**element_data)
+                else:
+                    values['element'] = StandardElement(**element_data)
+            # 如果element已经是正确的实例类型，则保持不变
+        return values
 
 
 class StandardDomTree(BaseModel):
@@ -129,7 +146,7 @@ class StandardDomTree(BaseModel):
                     # 添加表格名称
                     if node.element.name:
                         markdown_res += f"**{node.element.name}**\n\n"
-                    table_md = self._list_to_markdown_table(node.element.rows)
+                    table_md = self._list_to_html_table(node.element.rows)
                     markdown_res += f"{table_md}\n\n"
                     # 添加表格描述
                     if node.element.description:
@@ -168,25 +185,28 @@ class StandardDomTree(BaseModel):
         quoted_lines = ['> ' + line for line in lines]
         return '\n'.join(quoted_lines)
 
-    def _list_to_markdown_table(self, rows: List[StandardRow]) -> str:
-        """将表格行转换为 markdown 表格格式"""
+    def _list_to_html_table(self, rows: List[StandardRow]) -> str:
+        """将表格行转换为HTML表格"""
         if not rows:
             return ""
 
-        markdown_table = ""
-
-        # 添加表头
-        if rows:
-            header_cells = [cell.text or "" for cell in rows[0].cells]
-            markdown_table += "| " + " | ".join(header_cells) + " |\n"
-            markdown_table += "| " + " | ".join(["---"] * len(header_cells)) + " |\n"
-
-        # 添加数据行
+        html_text = "<table>"
         for row in rows:
-            cells = [cell.text or "" for cell in row.cells]
-            markdown_table += "| " + " | ".join(cells) + " |\n"
+            html_text += "<tr>"
+            for cell in row.cells:
+                # 从path中提取rowspan和colspan信息
+                if len(cell.path) >= 4:
+                    start_row, end_row, start_col, end_col = cell.path[:4]
+                    rowspan = end_row - start_row + 1
+                    colspan = end_col - start_col + 1
+                else:
+                    rowspan = colspan = 1
 
-        return markdown_table
+                cell_text = cell.text or ""
+                html_text += f"<td rowspan='{rowspan}' colspan='{colspan}'>{cell_text}</td>"
+            html_text += "</tr>"
+        html_text += "</table>"
+        return html_text
 
     @classmethod
     def from_domtree_dict(cls, domtree: dict, file_info):
@@ -397,19 +417,22 @@ class StandardDomTree(BaseModel):
                     url=element['image_link']
                 )
 
-            # 创建 StandardNode
-            standard_node = StandardNode(
+            # 创建StandardImageElement实例
+            image_element = StandardImageElement(
+                type=element_type,
+                positions=positions,
+                name="",
+                description="",
+                text=text,
+                image=image,
+            )
+            
+            # 使用construct方法跳过validator，保持正确的element类型
+            standard_node = StandardNode.construct(
                 summary="",
                 tokens=0,  # 先设置为 0，后面再计算
                 path=[],  # 初始化为空列表，后续再计算
-                element=StandardImageElement(
-                    type=element_type,
-                    positions=positions,
-                    name="",
-                    description="",
-                    text=text,
-                    image=image,
-                ),
+                element=image_element,
                 children=[]
             )
         elif element_type == "Table":
@@ -444,17 +467,21 @@ class StandardDomTree(BaseModel):
             # 将所有单元格的文本合并，用于计算 token 数量
             text = " ".join(cell_texts)
 
-            standard_node = StandardNode(
+            # 创建StandardTableElement实例
+            table_element = StandardTableElement(
+                type=element_type,
+                positions=positions,
+                name="",
+                description="",
+                rows=rows
+            )
+            
+            # 使用construct方法跳过validator，保持正确的element类型
+            standard_node = StandardNode.construct(
                 summary="",
                 tokens=0,  # 先设置为 0，后面再计算
                 path=[],  # 初始化为空列表，后续再计算
-                element=StandardTableElement(
-                    type=element_type,
-                    positions=positions,
-                    name="",
-                    description="",
-                    rows=rows
-                ),
+                element=table_element,
                 children=[]
             )
         else:
@@ -507,4 +534,9 @@ class StandardDomTree(BaseModel):
         # 计算标记列表的长度，即标记的数量
         token_count = len(tokens)
         # 返回标记的数量
-        return token_count
+        return token_count
+
+
+# 更新forward references
+StandardNode.update_forward_refs()
+Cell.update_forward_refs()

{bella_openapi-1.0.3.1 → bella_openapi-1.0.3.2/src/bella_openapi.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bella-openapi
-Version: 1.0.3.1
+Version: 1.0.3.2
 Summary: client for openapi service.
 Home-page: 
 Author: ['tangxiaolong', 'fanqiangwei', 'zhangxiaojia', 'liumin', 'wangyukun']
@@ -12,7 +12,8 @@ License-File: LICENSE
 Requires-Dist: httpx<=0.26.0,>=0.10.0
 Requires-Dist: Werkzeug==3.0.1
 Requires-Dist: tiktoken>=0.5.0
-Requires-Dist: pydantic==2.11.7
+Requires-Dist: pydantic==1.10.14
+Requires-Dist: fastapi==0.110.0
 Dynamic: author
 Dynamic: classifier
 Dynamic: description
@@ -260,3 +261,5 @@ get接口地址：/v1/openapi/log/{requestId}
     * rename 包名
 * 1.0.2
     * 修复bug
+* 1.0.3 
+    * 新增standard domtree

{bella_openapi-1.0.3.1 → bella_openapi-1.0.3.2}/src/bella_openapi.egg-info/requires.txt

@@ -1,4 +1,5 @@
 httpx<=0.26.0,>=0.10.0
 Werkzeug==3.0.1
 tiktoken>=0.5.0
-pydantic==2.11.7
+pydantic==1.10.14
+fastapi==0.110.0