如何从python docx中删除粗体字

Question

I have a docx that I need to preprocess using spaCy. I need to remove all words that appear in bold in the document.

I tried the following:

def delete_paragraph(paragraph):
p = paragraph._element
p.getparent().remove(p)
p._p = p._element = None


length = len(document.paragraphs)
for i in range(0,length):
  for j in range(0,len(document.paragraphs[i].runs)):
     if document.paragraphs[i].runs[j].bold == True:
        delete_paragraph(document.paragraphs[i])
        length = length-1
        continue
document.save("/home/nikita/Desktop/Internship/new topic_mod/AXIS new.docx")

But I get the following error:

IndexError: Traceback (most recent call last)
<ipython-input-12-d144bd42e95e> in <module>()
  3     #print(document.paragraphs[i].text)
  4     for j in range(0,len(document.paragraphs[i].runs)):
----> 5         if document.paragraphs[i].runs[j].bold == True:
  6             delete_paragraph(document.paragraphs[i])
  7             length = length-1

IndexError: list index out of range

I cannot figure out why its out of range. How can I remove bold words from a python-docx ?

Please help!

Answer 1

There are a couple probable reasons:

1. You need to break (not continue ) out of your inner loop once you've deleted the paragraph, otherwise you try to delete the same paragraph multiple times if it has more than one bold run.

    for j in range(0,len(document.paragraphs[i].runs)): if document.paragraphs[i].runs[j].bold == True: delete_paragraph(document.paragraphs[i]) length = length-1 break 

2. Your list of paragraphs is getting shorter each time you delete one, which changes the index of each paragraph that follows. If you traverse the paragraphs from bottom to top, that won't be a problem. Also, you can ditch all that (i, j) index management; Python rarely needs that.

    for paragraph in reversed(list(document.paragraphs)): for run in paragraph.runs: if run.bold: delete_paragraph(paragraph) break