v0.58.2 - improved formatting

README.md

@@ -52,6 +52,7 @@ timezonefinder>=6.4.0
 - Use the `configmerger.py` to update old configuration files into a newer version's `config.ini`. You can do this by creating a copy of your existing config to i.e. a file named `myconfig.txt` and including in it the lines you want to keep for the newer version. Then, just run `python configmerger.py config.ini myconfig.txt` and all your existing config lines will be migrated to the new one. Works in most cases, but remember to be careful and double-check any migration issues with i.e. `diff`!
 
 # Changelog
+- v0.58.2 - improved formatting in pplx API calls
 - v0.58.1 - improved markdown parsing in translated Perplexity API calls
 - v0.58 - chunking, parsing and other small fixes
 - v0.57.5 - changes made to Perplexity API handling; new sonar-online models

api_perplexity_search.py

@@ -21,7 +21,7 @@
 
 # Global variable for chunk size
 # Set this value as needed
-CHUNK_SIZE = 350
+CHUNK_SIZE = 500
 
 # Assuming you've set PERPLEXITY_API_KEY in your environment variables
 PERPLEXITY_API_KEY = os.getenv("PERPLEXITY_API_KEY")
@@ -269,8 +269,13 @@ async def translate_response_chunked(bot, user_message, perplexity_response, con
  # After building the full translated_response, apply the paragraph breaks adjustment
  adjusted_response = add_paragraph_breaks_to_headers(translated_response)
 
- # Then convert the markdown to HTML
- html_response = markdown_to_html(adjusted_response)
+ # Apply the header formatting for Telegram before converting to HTML
+ telegram_formatted_response = format_headers_for_telegram(adjusted_response)
+
+ # Then convert the Telegram-formatted response to HTML
+ html_response = markdown_to_html(telegram_formatted_response)
+
+ logging.info(f"Parsed translated response: {html_response}")
 
  return html_response
 
@@ -357,6 +362,41 @@ def add_paragraph_breaks_to_headers(translated_response):
  adjusted_response = '\n'.join(adjusted_lines)
  return adjusted_response
 
+# formatting headers to look neater for TG and ensure proper breaks
+def format_headers_for_telegram(translated_response):
+ # Split the translated response into lines
+ lines = translated_response.split('\n')
+
+ # Initialize a list to hold the formatted lines
+ formatted_lines = []
+
+ # Iterate over the lines, adding symbols before headers and ensuring proper line breaks
+ for i, line in enumerate(lines):
+ if line.startswith('###'):
+ # Add a newline before the header if it's not the first line and the previous line is not empty
+ if i > 0 and lines[i - 1].strip() != '':
+ formatted_lines.append('')
+ # Format the header and add a newline after
+ formatted_line = '• <b>' + line[3:].strip() + '</b>'
+ formatted_lines.append(formatted_line)
+ if i < len(lines) - 1 and lines[i + 1].strip() != '':
+ formatted_lines.append('')
+ elif line.startswith('##'):
+ # Add a newline before the main header if it's not the first line and the previous line is not empty
+ if i > 0 and lines[i - 1].strip() != '':
+ formatted_lines.append('')
+ # Format the main header and add a newline after
+ formatted_line = '➤ <b>' + line[2:].strip() + '</b>'
+ formatted_lines.append(formatted_line)
+ if i < len(lines) - 1 and lines[i + 1].strip() != '':
+ formatted_lines.append('')
+ else:
+ formatted_lines.append(line)
+
+ # Join the adjusted lines back into a single string
+ formatted_response = '\n'.join(formatted_lines)
+ return formatted_response
+
 # smart chunking (v1.09)
 """ def smart_chunk(text, chunk_size=CHUNK_SIZE):
  chunks = []

main.py

@@ -5,7 +5,7 @@
 # https://github.com/FlyingFathead/TelegramBot-OpenAI-API
 #
 # version of this program
-version_number = "0.58.1"
+version_number = "0.58.2"
 
 # experimental modules
 import requests