siat 3.11.3__py3-none-any.whl → 3.11.5__py3-none-any.whl

siat/common.py

@@ -2889,7 +2889,7 @@ if __name__=='__main__':
     file='stooq.py'
     package='pandas_datareader'
 
-def fix_package(file='stooq.py',package='pandas_datareader'):
+def fix_package_x(file='stooq.py',package='pandas_datareader'):
     """
     功能：修复stooq.py，使用siat包中的stooq.py覆盖pandas_datareader中的同名文件
     注意：执行本程序需要系统管理员权限，可以系统管理员权限启动Jupyter或Spyder
@@ -2899,6 +2899,7 @@ def fix_package(file='stooq.py',package='pandas_datareader'):
     stooq.py            pandas_datareader
     bond_zh_sina.py     akshare
     
+    注意：在Python 3.13出错，暂时废弃！
     """
     #判断操作系统
     import sys; czxt=sys.platform
@@ -2923,7 +2924,7 @@ def fix_package(file='stooq.py',package='pandas_datareader'):
     
     #目标地址
     cmdstr1='import '+package
-    exec(cmdstr1)   #无返回值地执行字符串代码
+    exec(cmdstr1)   #无返回值地执行字符串代码，此句在Python 3.13后台不管用了！
     #import pandas_datareader
     #objpath=pandas_datareader.__path__[0]
     cmdstr2=package+'.__path__[0]'
@@ -2957,6 +2958,54 @@ def fix_package(file='stooq.py',package='pandas_datareader'):
     
     return    
 
+def fix_package(file='stooq.py', package='pandas_datareader'):
+    """
+    功能：修复指定包中的文件，例如将 siat 中的 stooq.py 覆盖 pandas_datareader 中的同名文件。
+    注意：执行本程序需要系统管理员权限，可以系统管理员权限启动 Jupyter 或 Spyder。
+    """
+
+    import sys
+    import importlib.util
+    import siat
+    from shutil import copyfile
+
+    # 判断操作系统
+    platform = sys.platform
+    if platform.startswith('win'):
+        os_type = 'windows'
+    elif platform == 'darwin':
+        os_type = 'mac'
+    elif platform.startswith('linux'):
+        os_type = 'linux'
+    else:
+        os_type = 'unknown'
+
+    # 获取源文件路径
+    src_path = siat.__path__[0].replace("\\", "/") if os_type == 'windows' else siat.__path__[0]
+    src_file = f"{src_path}/{file}"
+
+    # 获取目标包路径
+    spec = importlib.util.find_spec(package)
+    if spec is None or not spec.submodule_search_locations:
+        print(f"  #Error(fix_package): Package '{package}' not found.")
+        return
+
+    obj_path = spec.submodule_search_locations[0].replace("\\", "/") if os_type == 'windows' else spec.submodule_search_locations[0]
+    obj_file = f"{obj_path}/{file}"
+
+    # 执行文件复制
+    try:
+        copyfile(src_file, obj_file)
+    except IOError as e:
+        print(f"  #Error(fix_package): Unable to copy file. {e}")
+        print("  Solution: manually copy the file", src_file, "to the folder", obj_path)
+    except Exception as e:
+        print(f"  #Error(fix_package): Unexpected error: {e}")
+    else:
+        print(f"  Overrided '{file}' in '{package}'")
+        print("  Please RESTART Python kernel before using siat")
+
+    return
     
 
 
@@ -2964,11 +3013,12 @@ def fix_package(file='stooq.py',package='pandas_datareader'):
 if __name__=='__main__':
     file='stock_info.pickle'
     package='siat'
+    mode='read'
     developer=False
     
     file_position()
 
-def file_position(file='stock_info.pickle',package='siat',mode='read'):
+def file_position_x(file='stock_info.pickle',package='siat',mode='read'):
     """
     功能：给定文件名file，返回其路径
     注意：执行本程序可能需要系统管理员权限，可以系统管理员权限启动Jupyter或Spyder
@@ -2978,6 +3028,7 @@ def file_position(file='stock_info.pickle',package='siat',mode='read'):
     stooq.py            pandas_datareader
     bond_zh_sina.py     akshare
     
+    问题：在Python 3.13上后台运行出错，暂时废弃！
     """
     #判断操作系统
     import sys; czxt=sys.platform
@@ -3012,6 +3063,57 @@ def file_position(file='stock_info.pickle',package='siat',mode='read'):
     else:
         return objfile    
 #==============================================================================
+import sys
+import pickle
+import importlib
+
+def file_position(file='stock_info.pickle', package='siat', mode='read'):
+    """
+    功能：给定文件名file，返回其路径或读取其内容
+    参数：
+      file    - 目标文件名
+      package - 存放该文件的 Python 包名
+      mode    - 'read' 则加载并返回 pickle 中的对象，否则返回文件路径
+    注意：执行本程序可能需要系统管理员权限
+    """
+    # 1. 判断操作系统
+    czxt = sys.platform
+    if czxt in ('win32', 'win64'):
+        os_type = 'windows'
+    elif czxt == 'darwin':
+        os_type = 'mac'
+    elif czxt.startswith('linux'):
+        os_type = 'linux'
+    else:
+        os_type = 'windows'
+
+    # 2. 动态导入 package
+    try:
+        pkg = importlib.import_module(package)
+    except ImportError as e:
+        raise ImportError(f"无法导入包 '{package}': {e}")
+
+    # 3. 获取 package 的安装路径
+    try:
+        objpath = pkg.__path__[0]
+    except (AttributeError, IndexError):
+        # 如果是单文件模块，退而求其次取 __file__ 的目录
+        objpath = importlib.util.find_spec(package).origin
+        objpath = objpath.rsplit('/', 1)[0]
+
+    # 4. 拼接目标文件路径
+    if os_type == 'windows':
+        objpath_norm = objpath.replace('\\', '/')
+        objfile = objpath_norm + '/' + file
+    else:
+        objfile = objpath + '/' + file
+
+    # 5. 根据 mode 读取或返回路径
+    if mode == 'read':
+        with open(objfile, 'rb') as f:
+            return pickle.load(f)
+    else:
+        return objfile
 
 
 #==============================================================================
@@ -4848,8 +4950,28 @@ def ttest(sample1,sample2):
     return round(p_value,4)
 
 #==============================================================================
+import sys,os
+
+# —— 在程序最开始处，打印并校验 Python 版本 —— 
+def _check_python_version_jupyter2pdf():
+    ver = sys.version_info
+    version_str = f"{ver.major}.{ver.minor}.{ver.micro}"
+    #print(f"使用的 Python 版本：{version_str}")
+
+    # 只允许 3.7 <= 版本 < 3.13
+    usable=True
+    if not (ver.major == 3 and 7 <= ver.minor <= 12):
+        usable = False
+        print(f"不支持的 Python 版本：{version_str}，仅支持 Python 3.7–3.12")
+        print(f"解决方案：改用ipynb2pdf或ipynb2docx")
+        """
+        raise RuntimeError(
+            f"不支持的 Python 版本：{version_str}。"
+            " 请使用 Python 3.7–3.12。"
+        )
+        """
 
-
+    return usable
 
 async def jupyter2pdf(notebook_path, output_pdf_path, size="A3"):
     """
@@ -4866,6 +4988,9 @@ async def jupyter2pdf(notebook_path, output_pdf_path, size="A3"):
     await convert_notebook_to_pdf(notebook_path, output_pdf_path)
     注意3：notebook_path和output_pdf_path中可以带路径
     """
+    if not _check_python_version_jupyter2pdf():
+        return
+    
     size=size.upper()
     if not size in ['A4','A3','letter']:
         size='A3'
@@ -4945,6 +5070,8 @@ async def jupyter2pdf2(notebook_dir, notebook_file):
     注意2：调用本函数的格式是异步await开头，例如：
     await jupyter2pdf2(notebook_dir, notebook_file)
     """
+    if not _check_python_version_jupyter2pdf():
+        return
     
     # 路径分割符号
     if ('/' in notebook_dir) and not ('\\' in notebook_dir):
@@ -5074,6 +5201,9 @@ async def jupyter2pdf3(notebook_path):
     await jupyter2pdf3(notebook_path)
     """
     DEBUG=False
+
+    if not _check_python_version_jupyter2pdf():
+        return
     
     import os,sys
     # 分离目录和文件名

siat/save2docx.py

@@ -21,6 +21,8 @@ SIAT：Security Investment Analysis Tool
 """
 
 #==============================================================================
+#关闭所有警告
+import warnings; warnings.filterwarnings('ignore')
 
 import os
 import errno
@@ -39,6 +41,42 @@ from docx.enum.text import WD_ALIGN_PARAGRAPH
 from docx.enum.table import WD_TABLE_ALIGNMENT
 from docx.text.paragraph import Paragraph
 
+import contextlib
+import io
+
+import time
+from IPython.display import display, Javascript
+
+# —— 新增：Notebook 强制保存 —— 
+def _save_current_notebook():
+    """
+    在浏览器端触发一次保存：兼容 Classic Notebook、Lab 3.x/4.x。
+    """
+    js = """
+    (function() {
+      // Classic Notebook
+      if (window.Jupyter && Jupyter.notebook) {
+        Jupyter.notebook.save_checkpoint();
+      }
+      // JupyterLab >=3: 用 app.commands
+      else if (window.jupyterapp && jupyterapp.commands) {
+        jupyterapp.commands.execute('docmanager:save');
+      }
+      // JupyterLab <=2 或其他
+      else if (window.require) {
+        require(['@jupyterlab/docmanager'], function(docManager) {
+          docManager.save();
+        });
+      }
+    })();
+    """
+    try:
+        display(Javascript(js))
+        time.sleep(0.5)   # 给浏览器一点时间写盘
+    except Exception:
+        pass
+
+
 # 预设纸张尺寸（单位：毫米）
 PAGE_SIZES = {"A4": (210, 297), "A3": (297, 420)}
 
@@ -116,6 +154,10 @@ def convert_ipynb_to_docx(ipynb_path, docx_path=None, page_size="A3"):
       5. 表格等分列宽居中；图像放大至页宽并居中
       6. 若目标 docx 正被打开，抛出提示“请先关闭文件”
     """
+    # 0. 强制保存当前 Notebook
+    #print("Saving current ipynb ...")
+    _save_current_notebook()       
+    
     # ---- 1. 检查输入 & 输出路径 ----
     if not os.path.isfile(ipynb_path):
         raise FileNotFoundError(f"找不到输入文件：{ipynb_path}")
@@ -148,7 +190,11 @@ def convert_ipynb_to_docx(ipynb_path, docx_path=None, page_size="A3"):
     # ---- 4. Notebook → HTML（嵌入图像） ----
     exporter = HTMLExporter()
     exporter.embed_images = True
-    html_body, _ = exporter.from_notebook_node(nb)
+    
+    buf = io.StringIO()
+    # 屏蔽 stderr
+    with contextlib.redirect_stderr(buf):
+        html_body, _ = exporter.from_notebook_node(nb)
     html = f"<h1>{title}</h1>\n" + html_body
 
     # ---- 5. HTML → DOCX via Pandoc（或 subprocess fallback） ----
@@ -339,8 +385,7 @@ def ipynb2docx(ipynb_path, page_size="A3"):
     
     result = convert_ipynb_to_docx(ipynb_path, docx_path=None, page_size=page_size)
     print(f"{result} created with TOC in {page_size} size")
-    print(f"However, TOC needs update manually in Microsoft Word")
-    print(f"And, title and some other things may need fine tuned as well")
+    print(f"However, original title(s) and new TOC may need further adjustment")
     
     return
     

siat/save2pdf-playwright-20250712.py

@@ -0,0 +1,147 @@
+# -*- coding: utf-8 -*-
+"""
+本模块功能：转换ipynb文件为pdf，带有可跳转的目录（目前一级标题定位还不准确，二级以下目录定位较准确，但已可用）
+所属工具包：证券投资分析工具SIAT 
+SIAT：Security Investment Analysis Tool
+创建日期：2025年7月8日
+最新修订日期：2025年7月8日
+作者：王德宏 (WANG Dehong, Peter)
+作者单位：北京外国语大学国际商学院
+作者邮件：wdehong2000@163.com
+版权所有：王德宏
+用途限制：仅限研究与教学使用。
+特别声明：作者不对使用本工具进行证券投资导致的任何损益负责！
+"""
+
+#==============================================================================
+
+# 首次运行前，请安装依赖：
+# !pip install nbformat nbconvert playwright pymupdf nest_asyncio
+# !playwright install
+
+import os
+import re
+import tempfile
+import asyncio
+
+import nest_asyncio
+import nbformat
+from nbconvert import HTMLExporter
+from playwright.async_api import async_playwright
+import fitz  # PyMuPDF
+
+nest_asyncio.apply()  # 使 asyncio.run 在 Notebook 中可用
+
+def ipynb2pdf(ipynb_path: str) -> str:
+    """
+    将 .ipynb 转为带可跳转目录书签的 PDF。
+    返回生成的 PDF 文件路径。
+    """
+    if not os.path.isfile(ipynb_path):
+        raise FileNotFoundError(f"找不到文件：{ipynb_path}")
+    output_pdf = ipynb_path[:-6] + ".pdf"
+
+    print(f"Converting to PDF ...")
+
+    # 1. 读 notebook → 提取目录结构
+    nb = nbformat.read(ipynb_path, as_version=4)
+    toc = _extract_toc(nb)
+
+    # 2. nb → HTML
+    exporter = HTMLExporter()
+    html_body, _ = exporter.from_notebook_node(nb)
+
+    # 3. 临时写 HTML / PDF
+    with tempfile.NamedTemporaryFile("w", suffix=".html", encoding="utf-8", delete=False) as th:
+        th.write(html_body)
+        html_path = th.name
+    with tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) as tp:
+        tmp_pdf = tp.name
+
+    # 4. Playwright 渲染 HTML → PDF
+    asyncio.run(_html_to_pdf(html_path, tmp_pdf))
+
+    # 5. PyMuPDF 添加书签
+    _add_bookmarks(tmp_pdf, output_pdf, toc)
+
+    # 6. 清理
+    os.unlink(html_path)
+    os.unlink(tmp_pdf)
+
+    from pathlib import Path
+    full_path = Path(output_pdf)
+    # 提取文件名
+    filename = full_path.name  # 'report.pdf'
+    # 提取路径
+    directory = full_path.parent  # PosixPath('/Users/peter/Documents')
+    
+    print(f"✅ {filename} is created with TOC")
+    print(f"✅ It is in {directory}")
+    
+    #return output_pdf
+    return
+
+async def _html_to_pdf(html_path: str, pdf_path: str):
+    async with async_playwright() as p:
+        browser = await p.chromium.launch()
+        page = await browser.new_page()
+        await page.goto(f"file://{html_path}")
+        await page.pdf(
+            path=pdf_path,
+            #format="A4",
+            format="A3",
+            print_background=True,
+            margin={"top":"20mm","bottom":"20mm","left":"20mm","right":"20mm"},
+        )
+        await browser.close()
+
+def _extract_toc(nb_node) -> list[tuple[int,str]]:
+    """
+    从每个 markdown 单元首行提取 # 级别和标题文本，
+    返回 [(level, title), …]
+    """
+    toc = []
+    for cell in nb_node.cells:
+        if cell.cell_type != "markdown":
+            continue
+        first = cell.source.strip().splitlines()[0]
+        m = re.match(r"^(#{1,6})\s+(.*)", first)
+        if m:
+            toc.append((len(m.group(1)), m.group(2).strip()))
+    return toc
+
+def _add_bookmarks(input_pdf: str, output_pdf: str, toc: list[tuple[int,str]]):
+    """
+    用 PyMuPDF 打开临时 PDF，按 toc 列表查找页码，
+    然后用 set_toc() 批量写入书签。
+    """
+    doc = fitz.open(input_pdf)
+    outline = []
+    for level, title in toc:
+        page_num = 1
+        # 搜索标题出现在第几页（0-based → +1）
+        for i in range(doc.page_count):
+            if title in doc.load_page(i).get_text():
+                page_num = i + 1
+                break
+        outline.append([level, title, page_num])
+
+    # 批量设置目录书签
+    doc.set_toc(outline)
+    doc.save(output_pdf)
+
+# 使用示例（另起一个 cell 运行）：
+# ipynb = globals().get("__session__")
+# ipynb2pdf(ipynb)
+
+
+#==============================================================================
+
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================

siat/save2pdf-playwright-20250714.py

@@ -0,0 +1,202 @@
+# -*- coding: utf-8 -*-
+"""
+本模块功能：转换ipynb文件为pdf，带有可跳转的目录（目前一级标题定位还不准确，二级以下目录定位较准确，但已可用）
+所属工具包：证券投资分析工具SIAT 
+SIAT：Security Investment Analysis Tool
+创建日期：2025年7月8日
+最新修订日期：2025年7月8日
+作者：王德宏 (WANG Dehong, Peter)
+作者单位：北京外国语大学国际商学院
+作者邮件：wdehong2000@163.com
+版权所有：王德宏
+用途限制：仅限研究与教学使用。
+特别声明：作者不对使用本工具进行证券投资导致的任何损益负责！
+"""
+
+#==============================================================================
+
+# 首次运行前，请安装依赖：
+# !pip install nbformat nbconvert playwright pymupdf nest_asyncio
+# !playwright install
+
+# 针对Python 3.13在Windows下的修复
+# 在 Notebook 首格运行：
+import sys, asyncio
+
+if sys.platform.startswith("win"):
+    # SelectorEventLoop 无法启动 subprocess，改用 ProactorEventLoop
+    asyncio.set_event_loop_policy(
+        asyncio.WindowsProactorEventLoopPolicy()
+    )
+
+
+# 下面在Python < 3.13可正常运行
+import os
+import re
+import tempfile
+import asyncio
+
+import nest_asyncio
+import nbformat
+from nbconvert import HTMLExporter
+from playwright.async_api import async_playwright
+import fitz  # PyMuPDF
+
+nest_asyncio.apply()  # 使 asyncio.run 在 Notebook 中可用
+
+def ipynb2pdf(ipynb_path: str) -> str:
+    """
+    将 .ipynb 转为带可跳转目录书签的 PDF。
+    返回生成的 PDF 文件路径。
+    """
+    if not os.path.isfile(ipynb_path):
+        raise FileNotFoundError(f"找不到文件：{ipynb_path}")
+    output_pdf = ipynb_path[:-6] + ".pdf"
+
+    print(f"Converting to PDF ...")
+
+    # 1. 读 notebook → 提取目录结构
+    nb = nbformat.read(ipynb_path, as_version=4)
+    toc = _extract_toc(nb)
+
+    # 2. nb → HTML
+    exporter = HTMLExporter()
+    html_body, _ = exporter.from_notebook_node(nb)
+
+    # 3. 临时写 HTML / PDF
+    with tempfile.NamedTemporaryFile("w", suffix=".html", encoding="utf-8", delete=False) as th:
+        th.write(html_body)
+        html_path = th.name
+    with tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) as tp:
+        tmp_pdf = tp.name
+
+    # 4. Playwright 渲染 HTML → PDF
+    asyncio.run(_html_to_pdf(html_path, tmp_pdf))
+
+    # 5. PyMuPDF 添加书签
+    _add_bookmarks(tmp_pdf, output_pdf, toc)
+
+    # 6. 清理
+    os.unlink(html_path)
+    os.unlink(tmp_pdf)
+
+    from pathlib import Path
+    full_path = Path(output_pdf)
+    # 提取文件名
+    filename = full_path.name  # 'report.pdf'
+    # 提取路径
+    directory = full_path.parent  # PosixPath('/Users/peter/Documents')
+    
+    print(f"✅ {filename} is created with TOC")
+    print(f"✅ It is in {directory}")
+    
+    #return output_pdf
+    return
+
+#==============================================================================
+"""
+# 异步版本1
+async def _html_to_pdf(html_path: str, pdf_path: str):
+    async with async_playwright() as p:
+        browser = await p.chromium.launch()
+        page = await browser.new_page()
+        await page.goto(f"file://{html_path}")
+        await page.pdf(
+            path=pdf_path,
+            #format="A4",
+            format="A3",
+            print_background=True,
+            margin={"top":"20mm","bottom":"20mm","left":"20mm","right":"20mm"},
+        )
+        await browser.close()
+"""
+import nest_asyncio
+import asyncio
+from playwright.async_api import async_playwright
+
+nest_asyncio.apply()
+
+async def _html_to_pdf(html_path: str, pdf_path: str):
+    async with async_playwright() as p:
+        browser = await p.chromium.launch()
+        page = await browser.new_page()
+        await page.goto(f"file://{html_path}")
+        await page.pdf(
+            path=pdf_path,
+            format="A3",
+            print_background=True,
+            margin={"top": "20mm", "bottom": "20mm", "left": "20mm", "right": "20mm"},
+        )
+        await browser.close()
+
+
+
+"""
+# 同步版本：不能在Jupyter中使用
+from playwright.sync_api import sync_playwright
+
+def _html_to_pdf(html_path: str, pdf_path: str):
+    with sync_playwright() as p:
+        browser = p.chromium.launch()
+        page = browser.new_page()
+        page.goto(f"file://{html_path}")
+        page.pdf(
+            path=pdf_path,
+            format="A3",
+            print_background=True,
+            margin={"top": "20mm", "bottom": "20mm", "left": "20mm", "right": "20mm"},
+        )
+        browser.close()
+"""
+#==============================================================================
+
+def _extract_toc(nb_node) -> list[tuple[int,str]]:
+    """
+    从每个 markdown 单元首行提取 # 级别和标题文本，
+    返回 [(level, title), …]
+    """
+    toc = []
+    for cell in nb_node.cells:
+        if cell.cell_type != "markdown":
+            continue
+        first = cell.source.strip().splitlines()[0]
+        m = re.match(r"^(#{1,6})\s+(.*)", first)
+        if m:
+            toc.append((len(m.group(1)), m.group(2).strip()))
+    return toc
+
+def _add_bookmarks(input_pdf: str, output_pdf: str, toc: list[tuple[int,str]]):
+    """
+    用 PyMuPDF 打开临时 PDF，按 toc 列表查找页码，
+    然后用 set_toc() 批量写入书签。
+    """
+    doc = fitz.open(input_pdf)
+    outline = []
+    for level, title in toc:
+        page_num = 1
+        # 搜索标题出现在第几页（0-based → +1）
+        for i in range(doc.page_count):
+            if title in doc.load_page(i).get_text():
+                page_num = i + 1
+                break
+        outline.append([level, title, page_num])
+
+    # 批量设置目录书签
+    doc.set_toc(outline)
+    doc.save(output_pdf)
+
+# 使用示例（另起一个 cell 运行）：
+# ipynb = globals().get("__session__")
+# ipynb2pdf(ipynb)
+
+
+#==============================================================================
+
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================

siat/save2pdf-weasyprint-20250712.py

@@ -0,0 +1,133 @@
+# -*- coding: utf-8 -*-
+"""
+本模块功能：转换ipynb文件为pdf，带有可跳转的目录（目前一级标题定位还不准确，二级以下目录定位较准确，但已可用）
+所属工具包：证券投资分析工具SIAT 
+SIAT：Security Investment Analysis Tool
+创建日期：2025年7月8日
+最新修订日期：2025年7月8日
+作者：王德宏 (WANG Dehong, Peter)
+作者单位：北京外国语大学国际商学院
+作者邮件：wdehong2000@163.com
+版权所有：王德宏
+用途限制：仅限研究与教学使用。
+特别声明：作者不对使用本工具进行证券投资导致的任何损益负责！
+"""
+
+#==============================================================================
+
+# 首次运行前，请安装依赖：
+# !pip install nbformat nbconvert weasyprint pymupdf nest_asyncio
+# !playwright install
+
+import os
+import re
+import tempfile
+import nbformat
+from nbconvert import HTMLExporter
+from weasyprint import HTML, CSS
+import fitz  # PyMuPDF
+
+def ipynb2pdf(ipynb_path: str) -> str:
+    """
+    将 .ipynb 转为带可跳转目录书签的 PDF。
+    返回生成的 PDF 文件路径。
+    """
+    if not os.path.isfile(ipynb_path):
+        raise FileNotFoundError(f"找不到文件：{ipynb_path}")
+    output_pdf = ipynb_path[:-6] + ".pdf"
+
+    print(f"📄 正在转换为 PDF ...")
+
+    # 1. 读取 notebook → 提取目录结构
+    nb = nbformat.read(ipynb_path, as_version=4)
+    toc = _extract_toc(nb)
+
+    # 2. notebook → HTML
+    exporter = HTMLExporter()
+    html_body, _ = exporter.from_notebook_node(nb)
+
+    # 3. 写入临时 HTML 文件
+    with tempfile.NamedTemporaryFile("w", suffix=".html", encoding="utf-8", delete=False) as th:
+        th.write(html_body)
+        html_path = th.name
+
+    # 4. 使用 WeasyPrint 渲染 HTML → PDF
+    tmp_pdf = tempfile.NamedTemporaryFile(suffix=".pdf", delete=False).name
+    _html_to_pdf(html_path, tmp_pdf)
+
+    # 5. 使用 PyMuPDF 添加书签
+    _add_bookmarks(tmp_pdf, output_pdf, toc)
+
+    # 6. 清理临时文件
+    os.unlink(html_path)
+    os.unlink(tmp_pdf)
+
+    print(f"✅ PDF 已生成：{output_pdf}")
+    return output_pdf
+
+def _html_to_pdf(html_path: str, pdf_path: str):
+    """
+    使用 WeasyPrint 将 HTML 渲染为 PDF。
+    """
+    HTML(filename=html_path).write_pdf(
+        pdf_path,
+        stylesheets=[CSS(string="""
+            @page {
+                size: A4;
+                margin: 20mm;
+            }
+            body {
+                font-family: 'Arial', sans-serif;
+                line-height: 1.6;
+            }
+        """)]
+    )
+
+def _extract_toc(nb_node) -> list[tuple[int, str]]:
+    """
+    从每个 markdown 单元首行提取 # 级别和标题文本，
+    返回 [(level, title), …]
+    """
+    toc = []
+    for cell in nb_node.cells:
+        if cell.cell_type != "markdown":
+            continue
+        first = cell.source.strip().splitlines()[0]
+        m = re.match(r"^(#{1,6})\s+(.*)", first)
+        if m:
+            toc.append((len(m.group(1)), m.group(2).strip()))
+    return toc
+
+def _add_bookmarks(input_pdf: str, output_pdf: str, toc: list[tuple[int, str]]):
+    """
+    用 PyMuPDF 打开临时 PDF，按 toc 列表查找页码，
+    然后用 set_toc() 批量写入书签。
+    """
+    doc = fitz.open(input_pdf)
+    outline = []
+    for level, title in toc:
+        page_num = 1
+        for i in range(doc.page_count):
+            if title in doc.load_page(i).get_text():
+                page_num = i + 1
+                break
+        outline.append([level, title, page_num])
+    doc.set_toc(outline)
+    doc.save(output_pdf)
+
+
+# 使用示例（另起一个 cell 运行）：
+# ipynb = globals().get("__session__")
+# ipynb2pdf(ipynb)
+
+
+#==============================================================================
+
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================
+#==============================================================================

siat/save2pdf.py

@@ -19,37 +19,81 @@ SIAT：Security Investment Analysis Tool
 # !pip install nbformat nbconvert playwright pymupdf nest_asyncio
 # !playwright install
 
+#关闭所有警告
+import warnings; warnings.filterwarnings('ignore')
+
+# 能够在Python 3.13下运行了！
 import os
 import re
+import sys
 import tempfile
-import asyncio
-
-import nest_asyncio
+import subprocess
 import nbformat
 from nbconvert import HTMLExporter
-from playwright.async_api import async_playwright
-import fitz  # PyMuPDF
+import fitz           # PyMuPDF
+from pathlib import Path
+
+import contextlib
+import io
 
-nest_asyncio.apply()  # 使 asyncio.run 在 Notebook 中可用
+import time
+from IPython.display import display, Javascript
 
+# —— 新增：Notebook 强制保存 —— 
+def _save_current_notebook():
+    """
+    在浏览器端触发一次保存：兼容 Classic Notebook、Lab 3.x/4.x。
+    """
+    js = """
+    (function() {
+      // Classic Notebook
+      if (window.Jupyter && Jupyter.notebook) {
+        Jupyter.notebook.save_checkpoint();
+      }
+      // JupyterLab >=3: 用 app.commands
+      else if (window.jupyterapp && jupyterapp.commands) {
+        jupyterapp.commands.execute('docmanager:save');
+      }
+      // JupyterLab <=2 或其他
+      else if (window.require) {
+        require(['@jupyterlab/docmanager'], function(docManager) {
+          docManager.save();
+        });
+      }
+    })();
+    """
+    try:
+        display(Javascript(js))
+        time.sleep(0.5)   # 给浏览器一点时间写盘
+    except Exception:
+        pass
+    
+    
 def ipynb2pdf(ipynb_path: str) -> str:
     """
     将 .ipynb 转为带可跳转目录书签的 PDF。
     返回生成的 PDF 文件路径。
     """
+    # 0. 强制保存当前 Notebook
+    #print("Saving current ipynb ...")
+    _save_current_notebook()    
+    
     if not os.path.isfile(ipynb_path):
         raise FileNotFoundError(f"找不到文件：{ipynb_path}")
     output_pdf = ipynb_path[:-6] + ".pdf"
 
-    print(f"Converting to PDF ...")
+    print("Converting to PDF ...")
 
-    # 1. 读 notebook → 提取目录结构
+    # 1. 读 notebook → 提取目录
     nb = nbformat.read(ipynb_path, as_version=4)
     toc = _extract_toc(nb)
 
-    # 2. nb → HTML
+    # 2. nb → HTML（同时关闭图 alt 检查错误信息）
     exporter = HTMLExporter()
-    html_body, _ = exporter.from_notebook_node(nb)
+    buf = io.StringIO()
+    # 屏蔽 stderr
+    with contextlib.redirect_stderr(buf):
+        html_body, _ = exporter.from_notebook_node(nb)
 
     # 3. 临时写 HTML / PDF
     with tempfile.NamedTemporaryFile("w", suffix=".html", encoding="utf-8", delete=False) as th:
@@ -58,44 +102,43 @@ def ipynb2pdf(ipynb_path: str) -> str:
     with tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) as tp:
         tmp_pdf = tp.name
 
-    # 4. Playwright 渲染 HTML → PDF
-    asyncio.run(_html_to_pdf(html_path, tmp_pdf))
+    # 4. Playwright 渲染 HTML → PDF（在子进程中调用 sync API，避开 Jupyter 的 asyncio loop）
+    script = f"""
+import sys
+from playwright.sync_api import sync_playwright
+
+p = sync_playwright().start()
+browser = p.chromium.launch()
+page = browser.new_page()
+page.goto(r"file://{html_path}")
+page.pdf(
+    path=r"{tmp_pdf}",
+    format="A3",
+    print_background=True,
+    margin={{"top":"20mm","bottom":"20mm","left":"20mm","right":"20mm"}}
+)
+browser.close()
+p.stop()
+"""
+    subprocess.run([sys.executable, "-c", script], check=True)
 
     # 5. PyMuPDF 添加书签
     _add_bookmarks(tmp_pdf, output_pdf, toc)
 
-    # 6. 清理
+    # 6. 清理临时文件
     os.unlink(html_path)
     os.unlink(tmp_pdf)
 
-    from pathlib import Path
+    # 打印结果
     full_path = Path(output_pdf)
-    # 提取文件名
-    filename = full_path.name  # 'report.pdf'
-    # 提取路径
-    directory = full_path.parent  # PosixPath('/Users/peter/Documents')
-    
-    print(f"✅ {filename} is created with TOC")
-    print(f"✅ It is in {directory}")
-    
+    print(f"✅ {full_path.name} is created with TOC")
+    print(f"✅ It is in {full_path.parent}")
+
     #return output_pdf
     return
 
-async def _html_to_pdf(html_path: str, pdf_path: str):
-    async with async_playwright() as p:
-        browser = await p.chromium.launch()
-        page = await browser.new_page()
-        await page.goto(f"file://{html_path}")
-        await page.pdf(
-            path=pdf_path,
-            #format="A4",
-            format="A3",
-            print_background=True,
-            margin={"top":"20mm","bottom":"20mm","left":"20mm","right":"20mm"},
-        )
-        await browser.close()
 
-def _extract_toc(nb_node) -> list[tuple[int,str]]:
+def _extract_toc(nb_node) -> list[tuple[int, str]]:
     """
     从每个 markdown 单元首行提取 # 级别和标题文本，
     返回 [(level, title), …]
@@ -110,7 +153,8 @@ def _extract_toc(nb_node) -> list[tuple[int,str]]:
             toc.append((len(m.group(1)), m.group(2).strip()))
     return toc
 
-def _add_bookmarks(input_pdf: str, output_pdf: str, toc: list[tuple[int,str]]):
+
+def _add_bookmarks(input_pdf: str, output_pdf: str, toc: list[tuple[int, str]]):
     """
     用 PyMuPDF 打开临时 PDF，按 toc 列表查找页码，
     然后用 set_toc() 批量写入书签。
@@ -118,18 +162,18 @@ def _add_bookmarks(input_pdf: str, output_pdf: str, toc: list[tuple[int,str]]):
     doc = fitz.open(input_pdf)
     outline = []
     for level, title in toc:
-        page_num = 1
         # 搜索标题出现在第几页（0-based → +1）
-        for i in range(doc.page_count):
-            if title in doc.load_page(i).get_text():
-                page_num = i + 1
-                break
+        page_num = next(
+            (i+1 for i in range(doc.page_count)
+             if title in doc.load_page(i).get_text()),
+            1
+        )
         outline.append([level, title, page_num])
 
-    # 批量设置目录书签
     doc.set_toc(outline)
     doc.save(output_pdf)
 
+
 # 使用示例（另起一个 cell 运行）：
 # ipynb = globals().get("__session__")
 # ipynb2pdf(ipynb)

{siat-3.11.3.dist-info → siat-3.11.5.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: siat
-Version: 3.11.3
+Version: 3.11.5
 Summary: Securities Investment Analysis Tools (siat)
 Home-page: https://pypi.org/project/siat/
 Author: Prof. WANG Dehong, International Business School, Beijing Foreign Studies University
@@ -45,10 +45,12 @@ Requires-Dist: ipywidgets
 Requires-Dist: yahooquery
 Requires-Dist: alpha-vantage
 Requires-Dist: tiingo[pandas]
-Requires-Dist: numpy <2
+Requires-Dist: numpy<2
 Requires-Dist: playwright
 Requires-Dist: pymupdf
 Requires-Dist: pypandoc
+Requires-Dist: python-docx
+Requires-Dist: weasyprint
 
 
 # What is siat?

{siat-3.11.3.dist-info → siat-3.11.5.dist-info}/RECORD

@@ -10,7 +10,7 @@ siat/bond_china.py,sha256=WzUhjYYk8tsr3BDWLQcpuj9DqNxTzBSIi_wuAOZ48kY,3082
 siat/bond_zh_sina.py,sha256=26BohGcS120utwqg9dJvdGm5OkuNpNu5bco80uOuQpU,4423
 siat/capm_beta.py,sha256=t8-xr90II0JzbjsTOZNpRze_mKTvBRXjwN2o0N0tgD8,30521
 siat/capm_beta2.py,sha256=S2x6PrWp_1FyzVmG2MVzCf7LlpfHHEJxroJH2b26DvQ,35989
-siat/common.py,sha256=GLNRbXP7uDA_pibWXJQ-St0o9ylhvRut0k9KpCQ70bI,193909
+siat/common.py,sha256=L8ZAz0gkJjatm3CQ5mwJTPpATQtbagUPwI-7i3O5p18,198444
 siat/compare_cross.py,sha256=3iP9TH2h3w27F2ARZc7FjKcErYCzWRc-TPiymOyoVtw,24171
 siat/copyrights.py,sha256=YMLjZb328YpFMR-s_GUu0HBgeGce3pV7DgRut8S3I7w,690
 siat/cryptocurrency.py,sha256=QSc4jK9VFlqBWVu-0th1BIMt8wC-5R5sWky3EaNupy0,27940
@@ -50,8 +50,11 @@ siat/risk_adjusted_return.py,sha256=Q4ZRdTF57eNt4QCjeQ7uA8nG56Jls8f_QfJasZQEo0M,
 siat/risk_adjusted_return2.py,sha256=gCtHhfGNlV1wHqU9gfHJ_n17wRSyTMxc7lS8jgZ-GQk,87409
 siat/risk_evaluation.py,sha256=xfgLSKlIWYmRJrIL4kn2k2hp9fyOMAzYGIhi9ImvKOw,88917
 siat/risk_free_rate.py,sha256=IBuRqA2kppdZsW4D4fapW7vnM5HMEXOn95A5r9Pkwlo,12384
-siat/save2docx.py,sha256=c43X3IGgfli4gg2VJilRaLzd_KCZX4yF_lLO2S9FvQA,12302
-siat/save2pdf.py,sha256=cB1L5lH2n6RfgubCLFR7a617OGnrtT9IQhWXWGncoFs,5114
+siat/save2docx.py,sha256=cqdbPUEAPQD5PcByawMecXhz9ih_lBsWFvfCzGPFDuE,13494
+siat/save2pdf-playwright-20250712.py,sha256=cB1L5lH2n6RfgubCLFR7a617OGnrtT9IQhWXWGncoFs,5114
+siat/save2pdf-playwright-20250714.py,sha256=WPdjT4kjXiAoWri-nyvNlPhvjmQlAHX80qmIT5GGYxs,6785
+siat/save2pdf-weasyprint-20250712.py,sha256=ZVZq5yT-grcmdY3qq8XXZ7OCDCGqvh66o2WfszoK9ws,4570
+siat/save2pdf.py,sha256=YGxFCO6oOBpcyNnJPEc_skdohevUC5Hw5EedeL0HMvI,6136
 siat/sector_china.py,sha256=uLsDXdRBDVfgG6tnXWnQOTyDmyZfglVO9DRUYU2e3pk,157914
 siat/security_price2.py,sha256=DDiZ2dlv_TYPLhA8-gGb9i9xrl88r4rgSMEcxqQ6aU0,28065
 siat/security_prices.py,sha256=X3ip0q_m3OL3QRNRkr_lYQk-wsXLf6dWkFkyoZijhck,129368
@@ -73,8 +76,8 @@ siat/valuation.py,sha256=xGizcKJZ3ADLWWHm2TFQub18FxiDv2doQwBwbEqyqz0,51980
 siat/valuation_china.py,sha256=eSKIDckyjG8QkENlW_OKkqbQHno8pzDcomBO9iGNJVM,83079
 siat/var_model_validation.py,sha256=loqziBYO2p0xkeWm3Rb1rJsDhbcgAZ5aR9rBLRwLU5E,17624
 siat/yf_name.py,sha256=laNKMTZ9hdenGX3IZ7G0a2RLBKEWtUQJFY9CWuk_fp8,24058
-siat-3.11.3.dist-info/LICENSE,sha256=NTEMMROY9_4U1szoKC3N2BLHcDd_o5uTgqdVH8tbApw,1071
-siat-3.11.3.dist-info/METADATA,sha256=5NpbsFU3Bzzehbm1u5q8EFFX_61QA_ydPkJpXMGpHK8,8533
-siat-3.11.3.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-siat-3.11.3.dist-info/top_level.txt,sha256=X5R8wrVviq8agwJFVRVDsufkuOJuit-1qAT_kXeptrY,17
-siat-3.11.3.dist-info/RECORD,,
+siat-3.11.5.dist-info/LICENSE,sha256=NTEMMROY9_4U1szoKC3N2BLHcDd_o5uTgqdVH8tbApw,1071
+siat-3.11.5.dist-info/METADATA,sha256=4sdqgT0GBYVj46u0O_SgiUepBptHRxxTGE9EoFaawg8,8587
+siat-3.11.5.dist-info/WHEEL,sha256=R0nc6qTxuoLk7ShA2_Y-UWkN8ZdfDBG2B6Eqpz2WXbs,91
+siat-3.11.5.dist-info/top_level.txt,sha256=X5R8wrVviq8agwJFVRVDsufkuOJuit-1qAT_kXeptrY,17
+siat-3.11.5.dist-info/RECORD,,

{siat-3.11.3.dist-info → siat-3.11.5.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.43.0)
+Generator: setuptools (72.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any