Windows下Python3.5实现歌德戏剧场次台词至嵌套列表的追加问题

Alright, let's break down how to build that nested list structure for Goethe's plays using Python 3.5 on Windows. Since the plays follow a consistent structure, we can automate parsing lines into (character, line) tuples and grouping them by scene. Here's a practical, step-by-step solution:

First, let's clarify the end goal: we want a top-level list where each element is a sublist representing one scene. Each sublist will contain tuples formatted as (角色名, 台词). For example:

[
    [("FAUST", "Was ziehst du mich so hastig hin?"), ("MEPHISTOPHELES", "Ich will dich führen, wo du froh bist.")],  # 场次1
    [("GRETCHEN", "Mein lieber Herr, was macht ihr hier?"), ...]  # 场次2
]

Goethe's plays follow predictable formatting cues we can leverage:

• Scenes are usually marked with clear headers (like SCENE 1 or ACT I, SCENE 2)
• Character names are often in all caps, followed by a colon (e.g., FAUST:)
• Lines belonging to a character follow immediately after their name

We'll use these cues to split the text into scenes and map lines to their respective characters.

Here's a code snippet tailored to your Windows environment and Python 3.5 constraints (no 3.6+ features like f-strings):

import os

def parse_goethe_play(file_path):
    # Initialize our nested structure: top list = scenes, sublists = character-line tuples
    play_scenes = []
    current_scene = []
    current_character = None

    # Handle Windows file encoding (adjust to 'gbk' if your files use Chinese encoding)
    with open(file_path, 'r', encoding='utf-8') as play_file:
        for line in play_file:
            stripped_line = line.strip()
            # Skip empty lines
            if not stripped_line:
                continue

            # Detect scene changes (adjust this condition to match your play's scene headers)
            if stripped_line.startswith("SCENE") or stripped_line.startswith("ACT"):
                # Save the current scene if it has content, then reset for the next scene
                if current_scene:
                    play_scenes.append(current_scene)
                    current_scene = []
                continue

            # Detect character names (assuming format: "CHARACTER:")
            if ':' in stripped_line and stripped_line.split(':')[0].isupper():
                char_name = stripped_line.split(':')[0].strip()
                current_character = char_name
                # Check if the line has dialogue after the colon
                line_content = stripped_line.split(':', 1)[1].strip()
                if line_content:
                    current_scene.append((current_character, line_content))
            # If we have an active character, this line is their dialogue
            elif current_character:
                current_scene.append((current_character, stripped_line))

    # Don't forget to add the last scene to the top-level list
    if current_scene:
        play_scenes.append(current_scene)

    return play_scenes

# Example usage (Windows path uses raw string to avoid escape issues)
faust_data = parse_goethe_play(r"C:\GoethePlays\Faust_Part1.txt")

# Test: print the first 3 lines of the first scene
print("First 3 lines of Scene 1:")
for line_tuple in faust_data[0][:3]:
    print(f"{line_tuple[0]}: {line_tuple[1]}")

If your text files use slightly different formatting, adjust these parts:

• Scene detection: Modify the startswith condition to match your play's scene headers (e.g., if "场次" in stripped_line for Chinese-labeled scenes)
• Character detection: If character names aren't all caps, use other cues like checking for a period instead of a colon, or that the line is short and doesn't look like dialogue
• Stage directions: Add a check to skip or tag stage directions (e.g., if stripped_line.startswith('['): continue to ignore bracketed directions)

• File paths: Use raw strings (like r"C:\...") or forward slashes ("C:/...") to avoid escape character errors
• Encoding: If you get a UnicodeDecodeError, try encoding='gbk' or encoding='utf-8-sig' (for files with a UTF-8 BOM)
• Python 3.5 compatibility: Avoid using features introduced in 3.6+, like f-strings or the walrus operator (:=)—stick to .format() for string formatting

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