假设有 11 个 PPT,每个 PPT 里面有若干张幻灯片包含表格。Python 自动遍历所有 PPT,找到表格,将表格数据逐一提取出来,按照原有顺序汇总到一个新的 Excel 文件。
PPT 原始内容 | Excel 中的处理 |
|---|---|
单元格文字 | 保留 |
字体、字号、粗体 | 尽量保留 |
字体颜色 | 保留 |
单元格背景颜色 | 保留 |
边框颜色和样式 | 尽量保留 |
水平、垂直对齐 | 保留 |
单元格宽度和高度 | 尽量映射 |
合并单元格 | 尽量保留 |
图片 | 可以提取并插入 |
公式 | PPT 普通表格通常没有原生 Excel 公式 |
import os
import re
import win32com.client
from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill, Border, Side, Alignment
from openpyxl.utils import get_column_letter
# ==============================
# 1. 配置
# ==============================
FOLDER = r"D:\PPT_Merge"
OUTPUT = os.path.join(FOLDER, "PPT表格汇总.xlsx")
# 是否保留每张表格自己的表头
KEEP_HEADERS = True
# 表格之间空几行
TABLE_GAP = 1
# ==============================
# 2. 辅助函数
# ==============================
def natural_sort_key(text):
"""按文件名中的数字自然排序"""
return [
int(x) if x.isdigit() else x.lower()
for x in re.split(r"(\d+)", text)
]
def ppt_rgb_to_hex(rgb):
"""将 PowerPoint RGB 整数转换为 Excel HEX"""
try:
rgb = int(rgb)
r = rgb & 255
g = (rgb >> 8) & 255
b = (rgb >> 16) & 255
return f"{r:02X}{g:02X}{b:02X}"
except Exception:
return None
def safe_get(func, default=None):
"""读取 COM 属性,避免个别特殊格式导致程序中断"""
try:
return func()
except Exception:
return default
def get_border(cell, border_id):
"""读取 PowerPoint 单元格边框"""
try:
line = cell.Borders(border_id)
visible = line.Visible
if not visible:
return Side(style=None)
color = ppt_rgb_to_hex(line.ForeColor.RGB)
return Side(
style="thin",
color=color or "000000"
)
except Exception:
return Side(style=None)
def get_cell_text(cell):
"""获取 PPT 单元格文字"""
try:
return cell.Shape.TextFrame.TextRange.Text.rstrip("\r\x0b")
except Exception:
return ""
# ==============================
# 3. 读取单元格格式
# ==============================
def copy_cell_format(ppt_cell, excel_cell):
shape = ppt_cell.Shape
# 字体格式
text_range = safe_get(
lambda: shape.TextFrame.TextRange
)
if text_range is not None:
font = text_range.Font
font_name = safe_get(lambda: font.Name)
font_size = safe_get(lambda: font.Size)
bold = safe_get(lambda: font.Bold, False)
italic = safe_get(lambda: font.Italic, False)
underline = safe_get(lambda: font.Underline, False)
font_color = safe_get(lambda: font.Color.RGB)
excel_cell.font = Font(
name=font_name,
size=font_size,
bold=bool(bold),
italic=bool(italic),
underline="single" if underline else None,
color=ppt_rgb_to_hex(font_color)
)
# 水平对齐
paragraph_alignment = safe_get(
lambda: text_range.ParagraphFormat.Alignment
)
alignment_map = {
1: "left",
2: "center",
3: "right",
4: "justify",
}
horizontal = alignment_map.get(
paragraph_alignment, "left"
)
# 垂直对齐
vertical_anchor = safe_get(
lambda: shape.TextFrame.VerticalAnchor
)
vertical_map = {
1: "top",
3: "center",
4: "bottom",
}
vertical = vertical_map.get(
vertical_anchor, "center"
)
excel_cell.alignment = Alignment(
horizontal=horizontal,
vertical=vertical,
wrap_text=True
)
# 背景颜色
try:
fill = shape.Fill
if fill.Visible:
fill_type = fill.Type
# 仅对纯色填充直接复制背景颜色
if fill_type == 1:
color = ppt_rgb_to_hex(fill.ForeColor.RGB)
if color:
excel_cell.fill = PatternFill(
fill_type="solid",
fgColor=color
)
except Exception:
pass
# 单元格边框
# PowerPoint 边框常量:
# 1 = Top, 2 = Left, 3 = Bottom, 4 = Right
excel_cell.border = Border(
top=get_border(ppt_cell, 1),
left=get_border(ppt_cell, 2),
bottom=get_border(ppt_cell, 3),
right=get_border(ppt_cell, 4)
)
# ==============================
# 4. 提取 PPT 表格
# ==============================
def extract_tables(presentation, filename, worksheet):
table_count = 0
for slide_index in range(1, presentation.Slides.Count + 1):
slide = presentation.Slides(slide_index)
for shape_index in range(1, slide.Shapes.Count + 1):
shape = slide.Shapes(shape_index)
# 判断是否为真正的 PowerPoint 表格
if not shape.HasTable:
continue
table = shape.Table
rows = table.Rows.Count
cols = table.Columns.Count
print(
f"发现表格:{filename} "
f"第{slide_index}页 "
f"{rows}行×{cols}列"
)
table_count += 1
# 来源标记
source_row = worksheet.max_row + 1
worksheet.merge_cells(
start_row=source_row,
start_column=1,
end_row=source_row,
end_column=cols
)
source_cell = worksheet.cell(
row=source_row,
column=1,
value=f"来源:{filename} | 第{slide_index}页 | 表格{table_count}"
)
source_cell.font = Font(
bold=True,
color="FFFFFF",
size=10
)
source_cell.fill = PatternFill(
fill_type="solid",
fgColor="4472C4"
)
source_cell.alignment = Alignment(
horizontal="left",
vertical="center"
)
worksheet.row_dimensions[source_row].height = 22
# 表格数据开始位置
start_row = source_row + 1
# 逐行提取
for r in range(1, rows + 1):
for c in range(1, cols + 1):
ppt_cell = table.Cell(r, c)
value = get_cell_text(ppt_cell)
excel_cell = worksheet.cell(
row=start_row + r - 1,
column=c,
value=value
)
# 复制格式
copy_cell_format(
ppt_cell,
excel_cell
)
# 尽量保留行高
try:
height = table.Rows(r).Height
worksheet.row_dimensions[
start_row + r - 1
].height = height
except Exception:
pass
# 尽量保留列宽
for c in range(1, cols + 1):
try:
width = table.Columns(c).Width
# PowerPoint 宽度单位为 point
# Excel 列宽为近似字符宽度
excel_width = width / 5.25
col_letter = get_column_letter(c)
current_width = worksheet.column_dimensions[
col_letter
].width
if current_width is None or excel_width > current_width:
worksheet.column_dimensions[
col_letter
].width = excel_width
except Exception:
pass
# 下一张表格之前留空行
next_row = start_row + rows + TABLE_GAP
worksheet.cell(
row=next_row - 1,
column=1,
value=None
)
return table_count
# ==============================
# 5. 主程序
# ==============================
def main():
files = sorted(
[
os.path.join(FOLDER, f)
for f in os.listdir(FOLDER)
if f.lower().endswith(".pptx")
and f.lower() != os.path.basename(OUTPUT).lower()
and not f.startswith("~$")
],
key=lambda x: natural_sort_key(os.path.basename(x))
)
if not files:
raise RuntimeError("没有找到 PPTX 文件")
# 创建 Excel
wb = Workbook()
ws = wb.active
ws.title = "汇总"
# 设置工作表
ws.freeze_panes = "A1"
app = win32com.client.DispatchEx(
"PowerPoint.Application"
)
app.Visible = True
total_tables = 0
total_files = 0
try:
for index, ppt_path in enumerate(files, 1):
filename = os.path.basename(ppt_path)
print("\n" + "=" * 50)
print(f"[{index}/{len(files)}] 正在处理:{filename}")
presentation = None
try:
presentation = app.Presentations.Open(
ppt_path,
ReadOnly=True,
Untitled=False,
WithWindow=False
)
count = extract_tables(
presentation,
filename,
ws
)
total_tables += count
total_files += 1
print(f"提取完成:{count} 个表格")
except Exception as e:
print(f"处理失败:{filename}")
print("错误信息:", e)
finally:
if presentation is not None:
presentation.Close()
# 保存 Excel
wb.save(OUTPUT)
print("\n" + "=" * 50)
print("全部处理结束!")
print("成功处理 PPT:", total_files)
print("提取表格数量:", total_tables)
print("Excel 保存位置:", OUTPUT)
finally:
app.Quit()
if __name__ == "__main__":
main()
以上只是简单的处理办法,下面还有可以处理更多的方案:
方案一
# -*- coding: utf-8 -*-
import sys
sys.stdout.reconfigure(encoding="utf-8")
sys.stderr.reconfigure(encoding="utf-8")
import os
import re
import win32com.client
from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill, Border, Side, Alignment
from openpyxl.utils import get_column_letter
# ==============================
# 1. 配置
# ==============================
FOLDER = r"D:\试用合并"
OUTPUT = os.path.join(FOLDER, "PPT表格汇总.xlsx")
TABLE_GAP = 1
# ==============================
# 2. 辅助函数
# ==============================
def natural_sort_key(text):
return [
int(x) if x.isdigit() else x.lower()
for x in re.split(r"(\d+)", text)
]
def safe_get(func, default=None):
try:
return func()
except Exception:
return default
def ppt_rgb_to_hex(rgb):
try:
rgb = int(rgb)
r = rgb & 255
g = (rgb >> 8) & 255
b = (rgb >> 16) & 255
return f"{r:02X}{g:02X}{b:02X}"
except Exception:
return None
def get_border(cell, border_id):
try:
line = cell.Borders(border_id)
if not line.Visible:
return Side(style=None)
color = ppt_rgb_to_hex(line.ForeColor.RGB)
return Side(
style="thin",
color=color or "000000"
)
except Exception:
return Side(style=None)
def get_cell_text(cell):
try:
return cell.Shape.TextFrame.TextRange.Text.rstrip("\r\x0b")
except Exception:
return ""
def make_sheet_name(filename, used_names):
"""生成合法且不重复的 Excel Sheet 名称"""
name = os.path.splitext(filename)[0]
# Excel 不允许的字符
name = re.sub(r'[\\/*?:\[\]]', "_", name)
name = name[:31] or "Sheet"
original = name
index = 1
while name.lower() in used_names:
suffix = f"_{index}"
name = original[:31 - len(suffix)] + suffix
index += 1
used_names.add(name.lower())
return name
# ==============================
# 3. 复制单元格格式
# ==============================
def copy_cell_format(ppt_cell, excel_cell):
shape = ppt_cell.Shape
text_range = safe_get(
lambda: shape.TextFrame.TextRange
)
if text_range is not None:
font = text_range.Font
font_name = safe_get(lambda: font.Name)
font_size = safe_get(lambda: font.Size)
bold = safe_get(lambda: font.Bold, False)
italic = safe_get(lambda: font.Italic, False)
underline = safe_get(lambda: font.Underline, False)
font_color = safe_get(lambda: font.Color.RGB)
excel_cell.font = Font(
name=font_name,
size=font_size,
bold=bool(bold),
italic=bool(italic),
underline="single" if underline else None,
color=ppt_rgb_to_hex(font_color)
)
paragraph_alignment = safe_get(
lambda: text_range.ParagraphFormat.Alignment
)
alignment_map = {
1: "left",
2: "center",
3: "right",
4: "justify",
}
horizontal = alignment_map.get(
paragraph_alignment, "left"
)
vertical_anchor = safe_get(
lambda: shape.TextFrame.VerticalAnchor
)
vertical_map = {
1: "top",
3: "center",
4: "bottom",
}
vertical = vertical_map.get(
vertical_anchor, "center"
)
excel_cell.alignment = Alignment(
horizontal=horizontal,
vertical=vertical,
wrap_text=True
)
# 单元格背景
try:
fill = shape.Fill
if fill.Visible and fill.Type == 1:
color = ppt_rgb_to_hex(fill.ForeColor.RGB)
if color:
excel_cell.fill = PatternFill(
fill_type="solid",
fgColor=color
)
except Exception:
pass
# 单元格边框
excel_cell.border = Border(
top=get_border(ppt_cell, 1),
left=get_border(ppt_cell, 2),
bottom=get_border(ppt_cell, 3),
right=get_border(ppt_cell, 4)
)
# ==============================
# 4. 提取一个 PPT 的全部表格
# ==============================
def extract_ppt_tables(presentation, worksheet, filename):
table_count = 0
for slide_index in range(1, presentation.Slides.Count + 1):
slide = presentation.Slides(slide_index)
for shape_index in range(1, slide.Shapes.Count + 1):
shape = slide.Shapes(shape_index)
if not shape.HasTable:
continue
table = shape.Table
rows = table.Rows.Count
cols = table.Columns.Count
table_count += 1
print(
f"{filename} | 第{slide_index}页 | "
f"表格{table_count} | {rows}行×{cols}列"
)
# 来源标题
source_row = worksheet.max_row + 1
worksheet.merge_cells(
start_row=source_row,
start_column=1,
end_row=source_row,
end_column=cols
)
cell = worksheet.cell(
row=source_row,
column=1,
value=f"来源:{filename} | 第{slide_index}页 | 表格{table_count}"
)
cell.font = Font(
bold=True,
color="FFFFFF",
size=10
)
cell.fill = PatternFill(
fill_type="solid",
fgColor="4472C4"
)
cell.alignment = Alignment(
horizontal="left",
vertical="center"
)
worksheet.row_dimensions[source_row].height = 22
start_row = source_row + 1
# 逐单元格提取
for r in range(1, rows + 1):
for c in range(1, cols + 1):
ppt_cell = table.Cell(r, c)
value = get_cell_text(ppt_cell)
excel_cell = worksheet.cell(
row=start_row + r - 1,
column=c,
value=value
)
copy_cell_format(
ppt_cell,
excel_cell
)
# 行高
try:
height = table.Rows(r).Height
worksheet.row_dimensions[
start_row + r - 1
].height = height
except Exception:
pass
# 列宽
for c in range(1, cols + 1):
try:
width = table.Columns(c).Width
excel_width = width / 5.25
letter = get_column_letter(c)
current = worksheet.column_dimensions[
letter
].width
if current is None or excel_width > current:
worksheet.column_dimensions[
letter
].width = excel_width
except Exception:
pass
# 表格间隔
worksheet.cell(
row=start_row + rows + TABLE_GAP - 1,
column=1
)
return table_count
# ==============================
# 5. 主程序
# ==============================
def main():
files = sorted(
[
os.path.join(FOLDER, f)
for f in os.listdir(FOLDER)
if f.lower().endswith(".pptx")
and f.lower() != os.path.basename(OUTPUT).lower()
and not f.startswith("~$")
],
key=lambda x: natural_sort_key(os.path.basename(x))
)
if not files:
raise RuntimeError("没有找到 PPTX 文件")
wb = Workbook()
# 删除默认工作表
default_sheet = wb.active
wb.remove(default_sheet)
app = win32com.client.DispatchEx(
"PowerPoint.Application"
)
app.Visible = True
used_names = set()
total_tables = 0
success_files = 0
try:
for index, ppt_path in enumerate(files, 1):
filename = os.path.basename(ppt_path)
print("\n" + "=" * 50)
print(f"[{index}/{len(files)}] 正在处理:{filename}")
# 每个 PPT 创建一个 Sheet
sheet_name = make_sheet_name(
filename,
used_names
)
ws = wb.create_sheet(sheet_name)
ws.freeze_panes = "A1"
presentation = None
try:
presentation = app.Presentations.Open(
ppt_path,
ReadOnly=True,
Untitled=False,
WithWindow=False
)
count = extract_ppt_tables(
presentation,
ws,
filename
)
total_tables += count
success_files += 1
if count == 0:
ws["A1"] = "该 PPT 未发现原生表格对象"
print(f"完成:提取 {count} 个表格")
except Exception as e:
print(f"处理失败:{filename}")
print("错误:", e)
ws["A1"] = f"提取失败:{e}"
finally:
if presentation is not None:
presentation.Close()
# 保存 Excel
wb.save(OUTPUT)
print("\n" + "=" * 50)
print("全部处理完成!")
print("PPT 文件数量:", len(files))
print("成功处理数量:", success_files)
print("提取表格数量:", total_tables)
print("Excel Sheet 数量:", len(wb.sheetnames))
print("输出位置:", OUTPUT)
finally:
app.Quit()
if __name__ == "__main__":
main()
方案二
# -*- coding: utf-8 -*-
import os
import re
import sys
import json
import time
import hashlib
import logging
import traceback
from pathlib import Path
import pythoncom
import win32com.client
# ==================================================
# 1. 配置
# ==================================================
ROOT = Path(r"D:\PPT_Extractor")
INPUT_DIR = ROOT / "input"
OUTPUT_DIR = ROOT / "output"
LOG_DIR = ROOT / "logs"
CHECKPOINT_DIR = ROOT / "checkpoint"
TEMP_DIR = ROOT / "temp"
OUTPUT_FILE = OUTPUT_DIR / "PPT表格汇总.xlsx"
CHECKPOINT_FILE = CHECKPOINT_DIR / "progress.json"
# 每个 PPT 处理完成后保存
SAVE_AFTER_EACH_PPT = True
# 是否导出原始表格图片
EXPORT_IMAGES = True
# Excel 工作表名称最大长度
MAX_SHEET_NAME = 31
# PowerPoint 常量
PP_SHAPE_FORMAT_PNG = 2
# Excel 常量
XL_UP = -4162
XL_TO_LEFT = -4159
# ==================================================
# 2. 初始化目录和日志
# ==================================================
for folder in [
INPUT_DIR,
OUTPUT_DIR,
LOG_DIR,
CHECKPOINT_DIR,
TEMP_DIR
]:
folder.mkdir(parents=True, exist_ok=True)
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s | %(levelname)s | %(message)s",
handlers=[
logging.FileHandler(
LOG_DIR / "process.log",
encoding="utf-8"
),
logging.StreamHandler(sys.stdout)
]
)
logger = logging.getLogger("PPTExtractor")
# ==================================================
# 3. 通用函数
# ==================================================
def safe_get(func, default=None):
try:
return func()
except Exception:
return default
def natural_sort_key(text):
return [
int(x) if x.isdigit() else x.lower()
for x in re.split(r"(\d+)", text)
]
def sha256_file(path):
sha = hashlib.sha256()
with open(path, "rb") as f:
for block in iter(lambda: f.read(1024 * 1024), b""):
sha.update(block)
return sha.hexdigest()
def atomic_json_save(path, data):
temp = path.with_suffix(".tmp")
with open(temp, "w", encoding="utf-8") as f:
json.dump(
data,
f,
ensure_ascii=False,
indent=2
)
os.replace(temp, path)
def load_checkpoint():
if not CHECKPOINT_FILE.exists():
return {"completed": {}}
try:
with open(CHECKPOINT_FILE, "r", encoding="utf-8") as f:
return json.load(f)
except Exception:
logger.exception("断点文件读取失败,将创建新的任务记录")
return {"completed": {}}
def make_sheet_name(filename, used_names):
name = Path(filename).stem
name = re.sub(r'[\\/*?:\[\]]', "_", name)
name = name[:MAX_SHEET_NAME] or "Sheet"
original = name
index = 1
while name.lower() in used_names:
suffix = f"_{index}"
name = original[:MAX_SHEET_NAME - len(suffix)] + suffix
index += 1
used_names.add(name.lower())
return name
def ppt_rgb_to_hex(rgb):
try:
rgb = int(rgb)
r = rgb & 255
g = (rgb >> 8) & 255
b = (rgb >> 16) & 255
return f"{r:02X}{g:02X}{b:02X}"
except Exception:
return None
def log_error(message):
with open(
LOG_DIR / "errors.log",
"a",
encoding="utf-8"
) as f:
f.write(
time.strftime("%Y-%m-%d %H:%M:%S")
+ " | "
+ str(message)
+ "\n"
)
# ==================================================
# 4. Excel 工作簿初始化
# ==================================================
def open_or_create_excel(excel_app):
if OUTPUT_FILE.exists():
logger.info("检测到已有结果文件,尝试恢复")
wb = excel_app.Workbooks.Open(
str(OUTPUT_FILE)
)
else:
wb = excel_app.Workbooks.Add()
# 删除默认多余工作表,先保留一个
while wb.Worksheets.Count > 1:
wb.Worksheets(wb.Worksheets.Count).Delete()
wb.Worksheets(1).Name = "处理说明"
ws = wb.Worksheets("处理说明")
ws.Cells(1, 1).Value = "PPT 批量表格提取结果"
ws.Cells(2, 1).Value = "每个源 PPT 对应一个工作表"
ws.Cells(3, 1).Value = "原始图片位于 output\\原始表格图片"
wb.SaveAs(str(OUTPUT_FILE), FileFormat=51)
return wb
def get_or_create_sheet(wb, name):
for i in range(1, wb.Worksheets.Count + 1):
ws = wb.Worksheets(i)
if ws.Name.lower() == name.lower():
return ws
ws = wb.Worksheets.Add(
After=wb.Worksheets(wb.Worksheets.Count)
)
ws.Name = name
return ws
# ==================================================
# 5. Excel 单元格格式映射
# ==================================================
def copy_native_cell(ppt_cell, excel_cell):
shape = ppt_cell.Shape
text_range = safe_get(
lambda: shape.TextFrame.TextRange
)
if text_range is not None:
excel_cell.Value = safe_get(
lambda: text_range.Text,
""
)
font = text_range.Font
font_name = safe_get(lambda: font.Name)
font_size = safe_get(lambda: font.Size)
bold = safe_get(lambda: font.Bold)
italic = safe_get(lambda: font.Italic)
underline = safe_get(lambda: font.Underline)
font_color = safe_get(lambda: font.Color.RGB)
try:
excel_cell.Font.Name = font_name
excel_cell.Font.Size = font_size
excel_cell.Font.Bold = bool(bold)
excel_cell.Font.Italic = bool(italic)
if underline:
excel_cell.Font.Underline = True
color = ppt_rgb_to_hex(font_color)
if color:
excel_cell.Font.Color = int(color, 16)
except Exception:
pass
# 水平对齐
alignment = safe_get(
lambda: text_range.ParagraphFormat.Alignment
)
# PowerPoint: 1左、2中、3右
alignment_map = {
1: -4131,
2: -4108,
3: -4152
}
if alignment in alignment_map:
excel_cell.HorizontalAlignment = alignment_map[alignment]
# 垂直对齐
vertical = safe_get(
lambda: shape.TextFrame.VerticalAnchor
)
vertical_map = {
1: -4160,
3: -4108,
4: -4107
}
if vertical in vertical_map:
excel_cell.VerticalAlignment = vertical_map[vertical]
excel_cell.WrapText = True
# 单元格背景
try:
fill = shape.Fill
if fill.Visible and fill.Type == 1:
color = ppt_rgb_to_hex(fill.ForeColor.RGB)
if color:
excel_cell.Interior.Color = int(color, 16)
except Exception:
pass
# 单元格边框
# 尝试映射四条边
border_map = {
1: 7, # Top
2: 10, # Left
3: 8, # Bottom
4: 9 # Right
}
for ppt_border, excel_border in border_map.items():
try:
line = ppt_cell.Borders(ppt_border)
if line.Visible:
border = excel_cell.Borders(excel_border)
border.LineStyle = 1
border.Weight = 2
color = ppt_rgb_to_hex(
line.ForeColor.RGB
)
if color:
border.Color = int(color, 16)
except Exception:
pass
# ==================================================
# 6. 原生表格提取
# ==================================================
def extract_native_table(
shape,
worksheet,
start_row,
filename,
slide_index,
object_index,
image_dir
):
table = shape.Table
rows = table.Rows.Count
cols = table.Columns.Count
title = worksheet.Cells(start_row, 1)
title.Value = (
f"来源:{filename} | 幻灯片:{slide_index} "
f"| 对象:{object_index} | 原生表格"
)
title.Font.Bold = True
title.Font.Color = 16777215
title.Interior.Color = 4472 * 256 + 68
title.RowHeight = 24
data_start = start_row + 1
# 逐个单元格提取
for r in range(1, rows + 1):
for c in range(1, cols + 1):
ppt_cell = table.Cell(r, c)
excel_cell = worksheet.Cells(
data_start + r - 1,
c
)
copy_native_cell(
ppt_cell,
excel_cell
)
# 行高
height = safe_get(
lambda: table.Rows(r).Height
)
if height:
worksheet.Rows(data_start + r - 1).RowHeight = height
# 列宽
for c in range(1, cols + 1):
width = safe_get(
lambda: table.Columns(c).Width
)
if width:
worksheet.Columns(c).ColumnWidth = width / 5.25
# 导出视觉备份
if EXPORT_IMAGES:
image_path = image_dir / (
f"{Path(filename).stem}"
f"_slide{slide_index}"
f"_object{object_index}.png"
)
try:
shape.Export(
str(image_path),
PP_SHAPE_FORMAT_PNG
)
except Exception as e:
log_error(f"表格图片导出失败:{image_path} | {e}")
return rows, cols, data_start + rows + 2
# ==================================================
# 7. 嵌入式 Excel 提取
# ==================================================
def extract_embedded_excel(
shape,
worksheet,
start_row,
filename,
slide_index,
object_index
):
ole = None
embedded_wb = None
try:
ole = shape.OLEFormat
# 尝试获取嵌入式 Excel 对象
embedded_wb = ole.Object
if embedded_wb is None:
raise RuntimeError("无法获取嵌入式 Excel 工作簿")
worksheet.Cells(start_row, 1).Value = (
f"来源:{filename} | 幻灯片:{slide_index} "
f"| 对象:{object_index} | 嵌入式 Excel"
)
worksheet.Cells(start_row, 1).Font.Bold = True
current_row = start_row + 1
# 遍历内部工作簿的所有工作表
for sheet_index in range(
1,
embedded_wb.Worksheets.Count + 1
):
source_ws = embedded_wb.Worksheets(sheet_index)
worksheet.Cells(
current_row,
1
).Value = (
f"内部工作表:{source_ws.Name}"
)
current_row += 1
used = source_ws.UsedRange
# 跨工作簿复制数据及 Excel 原生格式
used.Copy(
Destination=worksheet.Cells(
current_row,
1
)
)
current_row += used.Rows.Count + 2
return current_row
except Exception as e:
log_error(
f"嵌入式 Excel 提取失败:"
f"{filename} 第{slide_index}页 对象{object_index} | {e}"
)
worksheet.Cells(start_row, 1).Value = (
f"嵌入式 Excel 提取失败:{e}"
)
return start_row + 2
finally:
try:
if embedded_wb is not None:
# 不主动关闭 OLE 对象,避免误关闭宿主 PowerPoint
pass
except Exception:
pass
# ==================================================
# 8. 递归处理组合对象
# ==================================================
def process_shape(
shape,
worksheet,
row,
filename,
slide_index,
object_index,
image_dir,
visited=None
):
if visited is None:
visited = set()
shape_id = safe_get(lambda: shape.Id)
if shape_id in visited:
return row
if shape_id is not None:
visited.add(shape_id)
# 组合对象:递归处理内部形状
shape_type = safe_get(lambda: shape.Type)
# msoGroup = 6
if shape_type == 6:
try:
group = shape.GroupItems
for i in range(1, group.Count + 1):
row = process_shape(
group.Item(i),
worksheet,
row,
filename,
slide_index,
object_index * 100 + i,
image_dir,
visited
)
except Exception as e:
log_error(f"组合对象解析失败:{filename} | {e}")
return row
# 原生 PowerPoint 表格
has_table = safe_get(
lambda: bool(shape.HasTable),
False
)
if has_table:
try:
_, _, next_row = extract_native_table(
shape,
worksheet,
row,
filename,
slide_index,
object_index,
image_dir
)
return next_row
except Exception as e:
log_error(
f"原生表格提取失败:{filename} "
f"第{slide_index}页 | {e}"
)
worksheet.Cells(row, 1).Value = (
f"表格提取失败:{e}"
)
return row + 2
# 嵌入式 Excel / OLE
has_ole = safe_get(
lambda: shape.Type == 10,
False
)
if has_ole:
return extract_embedded_excel(
shape,
worksheet,
row,
filename,
slide_index,
object_index
)
return row
# ==================================================
# 9. PPT 文件提取
# ==================================================
def process_ppt(
powerpoint,
excel_wb,
ppt_path,
checkpoint,
used_names,
image_dir
):
filename = ppt_path.name
file_hash = sha256_file(ppt_path)
old_record = checkpoint["completed"].get(filename)
if old_record and old_record.get("sha256") == file_hash:
logger.info(f"已完成且文件未变化,跳过:{filename}")
return
sheet_name = make_sheet_name(
filename,
used_names
)
worksheet = get_or_create_sheet(
excel_wb,
sheet_name
)
# 文件发生变化时,先清理旧 Sheet 内容
worksheet.Cells.Clear()
presentation = None
stats = {
"native_tables": 0,
"embedded_objects": 0,
"slides": 0
}
try:
presentation = powerpoint.Presentations.Open(
str(ppt_path),
ReadOnly=True,
Untitled=False,
WithWindow=False
)
total_slides = presentation.Slides.Count
stats["slides"] = total_slides
row = 1
worksheet.Cells(1, 1).Value = (
f"源文件:{filename}"
)
worksheet.Cells(2, 1).Value = (
f"SHA256:{file_hash}"
)
row = 4
for slide_index in range(1, total_slides + 1):
slide = presentation.Slides(slide_index)
logger.info(
f"{filename} | 幻灯片 {slide_index}/{total_slides}"
)
for object_index in range(
1,
slide.Shapes.Count + 1
):
shape = slide.Shapes(object_index)
before_row = row
row = process_shape(
shape,
worksheet,
row,
filename,
slide_index,
object_index,
image_dir
)
if row > before_row:
stats["native_tables"] += int(
safe_get(
lambda: bool(shape.HasTable),
False
)
)
stats["embedded_objects"] += int(
safe_get(
lambda: shape.Type == 10,
False
)
)
worksheet.Columns.AutoFit()
# 记录完成状态
checkpoint["completed"][filename] = {
"sha256": file_hash,
"sheet": sheet_name,
"completed_at": time.strftime(
"%Y-%m-%d %H:%M:%S"
),
"stats": stats,
"status": "completed"
}
excel_wb.Save()
atomic_json_save(CHECKPOINT_FILE, checkpoint)
logger.info(
f"完成:{filename} | "
f"幻灯片:{stats['slides']} | "
f"原生表格:{stats['native_tables']} | "
f"嵌入对象:{stats['embedded_objects']}"
)
except Exception as e:
logger.exception(f"PPT 处理失败:{filename}")
log_error(
f"{filename} | {traceback.format_exc()}"
)
checkpoint["completed"][filename] = {
"sha256": file_hash,
"sheet": sheet_name,
"status": "failed",
"error": str(e)
}
# 保存已经写入的结果
try:
excel_wb.Save()
except Exception:
logger.exception("异常状态下 Excel 保存失败")
atomic_json_save(CHECKPOINT_FILE, checkpoint)
finally:
if presentation is not None:
try:
presentation.Close()
except Exception:
logger.exception("关闭 PPT 失败")
# ==================================================
# 10. 主程序
# ==================================================
def main():
ppt_files = sorted(
[
p for p in INPUT_DIR.iterdir()
if p.is_file()
and p.suffix.lower() == ".pptx"
and not p.name.startswith("~$")
],
key=lambda p: natural_sort_key(p.name)
)
if not ppt_files:
raise RuntimeError(
f"输入目录没有找到 PPTX:{INPUT_DIR}"
)
image_dir = OUTPUT_DIR / "原始表格图片"
image_dir.mkdir(exist_ok=True)
checkpoint = load_checkpoint()
pythoncom.CoInitialize()
powerpoint = None
excel = None
wb = None
try:
logger.info("启动 PowerPoint")
powerpoint = win32com.client.DispatchEx(
"PowerPoint.Application"
)
powerpoint.Visible = True
logger.info("启动 Excel")
excel = win32com.client.DispatchEx(
"Excel.Application"
)
excel.Visible = False
excel.DisplayAlerts = False
excel.AskToUpdateLinks = False
wb = open_or_create_excel(excel)
used_names = {
wb.Worksheets(i).Name.lower()
for i in range(1, wb.Worksheets.Count + 1)
}
for index, ppt_path in enumerate(ppt_files, 1):
logger.info(
f"========== [{index}/{len(ppt_files)}] "
f"{ppt_path.name} =========="
)
process_ppt(
powerpoint,
wb,
ppt_path,
checkpoint,
used_names,
image_dir
)
wb.Save()
logger.info("=" * 60)
logger.info("批量处理结束")
logger.info(f"输出文件:{OUTPUT_FILE}")
finally:
if wb is not None:
try:
wb.Save()
wb.Close()
except Exception:
logger.exception("Excel 工作簿关闭失败")
if excel is not None:
try:
excel.Quit()
except Exception:
logger.exception("Excel 退出失败")
if powerpoint is not None:
try:
powerpoint.Quit()
except Exception:
logger.exception("PowerPoint 退出失败")
pythoncom.CoUninitialize()
if __name__ == "__main__":
main()
gridSpan、rowSpan、hMerge、vMerge 等信息,再映射成 Excel 的合并区域。ppt/embeddings 文件,提取原始嵌入工作簿;同时通过 PowerPoint 导出当前显示效果,两者交叉核对。原创声明:本文系作者授权腾讯云开发者社区发表,未经许可,不得转载。
如有侵权,请联系 cloudcommunity@tencent.com 删除。