如何将文档扫描代码的for循环转为while循环并解决死循环

我尝试将文档扫描功能中的for循环转换为while循环，但每次转换后程序都会陷入无限循环。原for循环可正常识别表单页眉、结束文本并记录表单起止页码，我认为通过索引迭代的while循环应能实现相同效果。以下是原for循环代码及存在问题的while循环代码，求解决思路。

原for循环代码 #

from fuzzywuzzy import fuzz
import time
import fitz
from date_check import locate_date

## Each header is a list containing the header text and the form name ##
headers = ["header1", "Header1"]

## cast to lowercase ##
for header in headers:
    header[0] = header[0].lower()

## One of the following is expected to be on the last page of the form ##
end_texts = ["Signature", "Signed"]
## cast to lowercase ##
for i in range(len(end_texts)):
    end_texts[i] = end_texts[i].lower()


## set variables ##
forms = []
first_page = 0
header = ""

## Scan entire document for headers ##
def scan_document(document):
    document = fitz.open(document)
    first_page = False
    last_page = False
    index = 0
    ## This is the loop in question ##
    for i in range(len(document)):
        page = document[i]
        text = page.get_text("text")
        text = text.lower()
        if first_page == False:
            for header in headers:
                if fuzz.partial_ratio(header[0], text) > 90:
                    first_page = i
                    ## Find the date on the page ##
                    date = locate_date(text)
                    forms.append([date, header[1], first_page])
                    break

        elif  first_page != False and last_page == False:
            for end_text in end_texts:
                if end_text in text:
                    last_page = i
                    forms[index].append(last_page)
                    first_page = False
                    last_page = False
                    index += 1
                    break


    ## Return forms list containing first and last page of each form as well as the header ##
    return(forms)

存在问题的while循环代码（运行后程序挂起） #

## set variables ##
forms = []
first_page = 0
header = ""

## Scan entire document for headers ##
def scan_document(document):
    document = fitz.open(document)
    first_page = False
    last_page = False
    page_num = 0
    index = 0
    
    while page_num <= len(document):
        page = document[page_num]
        text = page.get_text("text")
        text = text.lower()
        if first_page == False:
            for header in headers:
                if fuzz.partial_ratio(header[0], text) > 90:
                    first_page = page_num
                    ## Find the date on the page ##
                    date = locate_date(text)
                    forms.append([date, header[1], first_page])
                    page_num += 1
                    break

        elif  first_page != False and last_page == False:
            for end_text in end_texts:
                if end_text in text:
                    last_page = page_num
                    forms[index].append(last_page)
                    first_page = False
                    last_page = False
                    index += 1
                    page_num += 1
                    break
        else:
            page_num += 1

    ## Return forms list containing first and last page of each form as well as the header ##
    return(forms)

问题原因分析 #

1. 循环条件错误：文档页码索引从0开始，总页数为len(document)时，最大有效索引是len(document)-1。原while循环条件page_num <= len(document)会导致最后一次循环访问document[len(document)]触发索引越界，同时也会让循环多执行一次，加重卡住概率。
2. 页码递增逻辑漏洞：
   • 在first_page == False分支中，仅当找到匹配页眉时才执行page_num +=1；若当前页无匹配页眉，page_num不会递增，循环永远停在当前页，触发无限循环。
   • 同理，在first_page != False and last_page == False分支中，仅找到匹配结束文本时才递增页码；若当前页无匹配结束文本，页码不变，循环卡住。

修复方案 #

• 修正循环条件为page_num < len(document)，确保只访问有效页码索引。
• 将page_num +=1移到循环末尾，无论当前页是否匹配到页眉或结束文本，都强制递增页码，避免循环停滞。

修复后的while循环代码 #

## set variables ##
forms = []
first_page = 0
header = ""

## Scan entire document for headers ##
def scan_document(document):
    document = fitz.open(document)
    first_page = False
    last_page = False
    page_num = 0
    index = 0
    
    # 修正循环条件：只遍历有效页码索引
    while page_num < len(document):
        page = document[page_num]
        text = page.get_text("text")
        text = text.lower()
        
        if first_page == False:
            for header in headers:
                if fuzz.partial_ratio(header[0], text) > 90:
                    first_page = page_num
                    date = locate_date(text)
                    forms.append([date, header[1], first_page])
                    break

        elif first_page != False and last_page == False:
            for end_text in end_texts:
                if end_text in text:
                    last_page = page_num
                    forms[index].append(last_page)
                    first_page = False
                    last_page = False
                    index += 1
                    break
        
        # 无论是否匹配成功，都递增页码
        page_num += 1

    return forms

内容的提问来源于stack exchange，提问作者Jcwscience