从网页文本中提取电影标题失败，求技术解决方案

Let’s break down how to pull out those movie titles from your messy input string. First, let’s look at the pattern in your text: after the header content ("2018年3月23日-25日周末 标题 周末票房 上映周数"), each movie entry follows this consistent structure:

• Movie Title (can include spaces, colons, and other non-$/digit characters)
• Weekend gross (starts with $, like $28.1M)
• Secondary gross figure (also $-prefixed, like $17M)
• Release week number (a plain digit, like 1)

The key insight here is that movie titles are the only segments that don’t start with a $ or a digit, and they’re always immediately followed by a $-prefixed box office value. We can use regular expressions to capitalize on this pattern.

Example Code (Python) #

Here’s a practical implementation that works perfectly for your input:

import re

raw_text = "2018年3月23日-25日周末 标题 周末票房 上映周数 Pacific Rim: Uprising $28.1M $17M 1 Spider Man: Home Coming $37.8M $12M 3"

# Regex to target titles: sequences that don't start with $/digit, end right before a $ value
movie_titles = re.findall(r'(?<=\s)([^\$\d].*?)(?=\s\$)', raw_text)

print(movie_titles)
# Output: ['Pacific Rim: Uprising', 'Spider Man: Home Coming']

How the Regex Works #

Let’s unpack the pattern (?<=\s)([^\$\d].*?)(?=\s\$):

• (?<=\s): Positive lookbehind to ensure the title is preceded by a space (prevents matching header text)
• [^\$\d].*?: Matches any sequence that doesn’t start with $ or a digit, using non-greedy matching (*?) to stop as soon as the next box office value begins
• (?=\s\$): Positive lookahead to confirm the title is followed by a space and $ (the start of the box office number)

Edge Case Adjustment #

If you ever encounter titles that include digits (e.g., Fast & Furious 9), you can tweak the regex to allow digits inside the title while still stopping at the $ value:

# Allows digits in titles, still stops before the $ box office value
movie_titles = re.findall(r'(?<=\s)([^\$].*?)(?=\s\$)', raw_text)

内容的提问来源于stack exchange，提问作者Ahmad Egbaria