ChuanhuChatGPT / overwrites.py
JohnSmith9982's picture
GitHub 94adb4f
a51e754
raw
history blame
1.55 kB
from __future__ import annotations
import logging
from llama_index import Prompt
from typing import List, Tuple
import mdtex2html
from presets import *
from llama_func import *
def compact_text_chunks(self, prompt: Prompt, text_chunks: List[str]) -> List[str]:
logging.debug("Compacting text chunks...πŸš€πŸš€πŸš€")
combined_str = [c.strip() for c in text_chunks if c.strip()]
combined_str = [f"[{index+1}] {c}" for index, c in enumerate(combined_str)]
combined_str = "\n\n".join(combined_str)
# resplit based on self.max_chunk_overlap
text_splitter = self.get_text_splitter_given_prompt(prompt, 1, padding=1)
return text_splitter.split_text(combined_str)
def postprocess(
self, y: List[Tuple[str | None, str | None]]
) -> List[Tuple[str | None, str | None]]:
"""
Parameters:
y: List of tuples representing the message and response pairs. Each message and response should be a string, which may be in Markdown format.
Returns:
List of tuples representing the message and response. Each message and response will be a string of HTML.
"""
if y is None:
return []
for i, (message, response) in enumerate(y):
y[i] = (
# None if message is None else markdown.markdown(message),
# None if response is None else markdown.markdown(response),
None if message is None else message,
None if response is None else mdtex2html.convert(response, extensions=['fenced_code','codehilite','tables']),
)
return y