Python文本生成器多字符统计时出现KeyError错误求助

问题描述 #

写代码统计文本字符序列并生成新文本，单字符统计功能正常，但换成3字符链统计时，出现KeyError错误，提示找不到键' -–'。

相关代码 #

import zipfile
from random import randint
from pprint import pprint

# file_zip = 'voyna-i-mir.txt.zip'

# zip = zipfile.ZipFile(file_zip, 'r')
# for file in zip.namelist():
#     zip.extract(file)

origin = 'voyna-i-mir.txt'

statistic = {}

chain = '   '
with open(origin, 'r', encoding='cp1251') as file:
    for sting in file:
        #print(sting)
        for symbol in sting:
            if chain in statistic:
                if symbol in statistic[chain]:
                    statistic[chain][symbol] += 1
                else:
                    statistic[chain][symbol] = 1
            else:
                statistic[chain] = {symbol: 1}
            chain = chain[1:] + symbol


dictionary = {}
stat_generator = {}
for chain, symbol_stat in statistic.items():
    dictionary[chain] = 0
    stat_generator[chain] = []
    for symbol, count in symbol_stat.items():
        dictionary[chain] += count
        stat_generator[chain].append([count, symbol])
    stat_generator[chain].sort(reverse=True)

gen = 1000
was_print = 0

chain = '   '
while was_print < gen:
    symbol_stat = stat_generator[chain]
    total = dictionary[chain]
    random = randint(1, total)
    position = 0
    for count, symbol in symbol_stat:
        position += count
        if random <= position:
            break
        print(symbol, end='')
        was_print += 1
        chain = chain[1:] + symbol

错误信息 #

Traceback (most recent call last):

  File "C:\Users\roman\Desktop\skillbox\[Skillbox] Профессия Python-
разработчик\9. Работа с файлами и форматированный
вывод-20210102T191733Z-001\9. Работа с файлами и форматированный
вывод\lesson_009\python_snippets\test.py", line 46, in <module>
    symbol_stat = stat_generator[chain]
              ~~~~~~~~~~~~~~^^^^^^^
KeyError: ' -–'

问题根源 #

1. 生成逻辑完全颠倒：代码里print和更新chain的操作写在了break的else分支，导致找到目标字符前就错误更新了链，生成原文本中从未出现过的3字符组合，自然在统计字典里找不到对应键。
2. 无容错处理：即使逻辑正确，生成过程中也可能出现原文本未覆盖的链组合，此时直接去字典取值就会触发KeyError。

修复方案 #

1. 修正生成循环的核心逻辑

把字符输出和链更新的代码移到找到目标字符之后，确保每次更新的链都是基于合法统计结果生成的：

2. 增加链不存在的容错处理

当生成的链不在统计字典中时，重置为初始链，避免报错。

修正后的生成循环代码：

gen = 1000
was_print = 0

chain = '   '
while was_print < gen:
    # 处理不存在的链，重置为初始状态
    if chain not in stat_generator:
        chain = '   '
        continue
        
    symbol_stat = stat_generator[chain]
    total = dictionary[chain]
    random = randint(1, total)
    position = 0
    target_symbol = None
    
    # 找到随机命中的字符
    for count, symbol in symbol_stat:
        position += count
        if random <= position:
            target_symbol = symbol
            break
    
    # 输出并更新链
    if target_symbol:
        print(target_symbol, end='')
        was_print += 1
        chain = chain[1:] + target_symbol

额外优化建议 #

• 修正变量名拼写：for sting in file:改为for string in file:，提升代码可读性。
• 用collections.defaultdict简化统计逻辑，省去多层if判断：

from collections import defaultdict

statistic = defaultdict(lambda: defaultdict(int))
chain = '   '
with open(origin, 'r', encoding='cp1251') as file:
    for string in file:
        for symbol in string:
            statistic[chain][symbol] += 1
            chain = chain[1:] + symbol

内容的提问来源于stack exchange，提问作者О. К.