抓word表格中的表格，按列排序到excel儲存格（包含文字顏色和底線）-- python

 Word2Excel/

│

├── input.docx

├── word_nested_table_to_excel.py

├── output.xlsx

└── README.txt

from docx import Document

from openpyxl import Workbook

from openpyxl.styles import Font

from openpyxl.utils import get_column_letter

WORD_FILE="input.docx"

OUTPUT_FILE="output.xlsx"

doc=Document(WORD_FILE)

wb=Workbook()

ws=wb.active

out_row=1

def first_run(cell):

for p in cell.paragraphs:

for r in p.runs:

if r.text:

return r

return None

def write_table(tbl, level):

global out_row

ws.cell(out_row,1).value=f"Level {level} Table"

ws.cell(out_row,1).font=Font(bold=True)

out_row+=1

for r,row in enumerate(tbl.rows):

for c,cell in enumerate(row.cells):

txt="".join(run.text for p in cell.paragraphs for run in p.runs)

ec=ws.cell(out_row+r,c+1)

ec.value=txt

run=first_run(cell)

if run:

color=None

try:

if run.font.color and run.font.color.rgb:

color=str(run.font.color.rgb)

except:

pass

ec.font=Font(

name=run.font.name,

size=run.font.size.pt if run.font.size else None,

bold=run.font.bold,

italic=run.font.italic,

underline="single" if run.font.underline else None,

color=color

)

out_row+=len(tbl.rows)+2

def walk(tbl, level=0):

write_table(tbl, level)

for row in tbl.rows:

for cell in row.cells:

for nt in cell.tables:

walk(nt, level+1)

for t in doc.tables:

walk(t)

for col in ws.columns:

m=0

for cell in col:

if cell.value:

m=max(m,len(str(cell.value)))

ws.column_dimensions[get_column_letter(col[0].column)].width=min(max(m+2,10),60)

wb.save(OUTPUT_FILE)

print("Done:",OUTPUT_FILE)