Skip to content

Instantly share code, notes, and snippets.

@yukiarimo

yukiarimo/character-ai-wrapper.py

Last active December 24, 2023 08:07
Show Gist options
  • Save yukiarimo/f3eea376e96f79e1c537bd2fad93328b to your computer and use it in GitHub Desktop.
Save yukiarimo/f3eea376e96f79e1c537bd2fad93328b to your computer and use it in GitHub Desktop.
Download ZIP
Character AI Dialog Extractor
Raw
character-ai-wrapper.py
import json
from bs4 import BeautifulSoup
with open("Main - Rushia Uruha.html") as fp:
soup = BeautifulSoup(fp, "html.parser")
items = []
main_text = None
target_text = None
for tag in soup.find_all("span", class_="s1"):
text = tag.get_text().strip()
if text == "Yuki Arimo":
if target_text:
items.append({"main_text": main_text, "target_text": target_text})
main_text = None
target_text = None
main_text = ""
elif text == "Rushia Uruha":
if main_text:
items.append({"main_text": main_text, "target_text": target_text})
main_text = None
target_text = None
target_text = ""
elif tag.name == "img":
continue
else:
if main_text is not None:
main_text += " " + text
elif target_text is not None:
target_text += " " + text
if main_text or target_text:
items.append({"main_text": main_text, "target_text": target_text})
new_list = []
for i in range(len(items)-1):
if items[i].get("main_text") and items[i+1].get("target_text"):
new_dict = {"main_text": items[i].get("main_text"), "target_text": items[i+1].get("target_text")}
new_list.append(new_dict)
json.dump(new_list, open("output.json", "w"), ensure_ascii=False, indent=4)
@yukiarimo
Copy link
Author

yukiarimo commented Apr 7, 2023 •
edited

How to use it:

  1. Share your dialog with AI on beta.character.ai (you can use the "link only" option)
  2. Save the webpage as .webarchive file
  3. Convert to HTML using: textutil -convert html 'Main - Rushia Uruha.webarchive'
  4. Use my script

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment