我在 txt 文件中有一个字典，格式如下：
house house$casa | casa, vivienda, hogar | edificio, casa | vivienda
$ 符号分隔翻译的术语。
我想通过带有文本编辑器的正则表达式来定位在同一行出现多次的字典单词，例如 Sublimetext、Notepad++、...删除那些重复的单词。在上面的例子中，正则表达式应该找到 house、casa 和 vivienda。我的目标是获得以下结果：
house$casa | vivienda, hogar | edificio
我尝试使用以下表达式，但它不能正常工作：
(\b\w+\b)\W+\1
FWIW，这是一个如何在 Python 中执行此操作的粗略示例：
import re
def distinct_words(block, seen, delim):
""" makes a list of words distinct, given a set of words seen earlier """
unique_words = []
for word in re.split(delim, block):
if not word in seen:
seen[word] = True
unique_words.append(word)
return unique_words
def process_line(line):
""" removes all duplicate words from a dictionary line """
# safeguard
if '$' not in line: return line
# split line at the '$'
original, translated = line.split('$')
# make original words distinct
distinct_original = distinct_words(original, {}, r' +')
# make translated words distinct, but keep block structure
# split the translated part at '|' into blocks
# split each block at ', ' into words
seen = {}
distinct_translated = [
distinct_list for distinct_list in (
distinct_words(block, seen, r', +') for block in (
re.split(r'\s*\|\s*', translated)
)
)
if len(distinct_list) > 0
]
# put everything back together again
part_original = ' '.join(distinct_original)
part_translated = [', '.join(block) for block in distinct_translated]
part_translated = ' | '.join(part_translated)
result = part_original + '$' + part_translated
return result
def process_dictionary(filename):
""" processes a dictionary text file, modifies the file in place """
lines = open(filename,'r').readlines()
lines_out = [process_line(line) for line in lines]
contents_out = '\n'.join(lines_out)
open(filename,'w').write(contents_out)
显然你会打电话process_dictionary()，像这样：
process_dictionary('dict_en_es.txt')
但是为了这个例子，假设你有一行：
line = "house house$casa | casa, vivienda, hogar | edificio, casa | vivienda"
line_out = process_line(line)
print line_out
打印想要的结果：
房子$casa | 维维安达，霍加尔| 高分辨率照片| CLIPARTO 大厦