Skip to content

Conversation

@EvenSol
Copy link
Owner

@EvenSol EvenSol commented Oct 25, 2025

Summary

  • persist AI helper chat history in session state and include recent turns when forming prompts and offline fallbacks
  • upgrade the what-if sidebar chat to retain structured history, feed it into result summaries, and reset inputs between turns
  • document that both assistants remember conversation context across a session

Testing

  • python -m compileall components services docs

https://chatgpt.com/codex/tasks/task_e_68fc4aff8724832db2c12d97d0f72f4c

@EvenSol EvenSol marked this pull request as ready for review October 26, 2025 11:57
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment on lines +14 to +16
_COMPONENT_PATTERN = re.compile(
r"(?P<amount>[-+]?\d+(?:\.\d+)?)\s*(?:%|percent|mole\s*%|mol\s*%)\s*(?:of\s+)?(?P<name>[A-Za-z0-9\-\s]+)",
flags=re.IGNORECASE,

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Parse multiple components in natural language scenario prompts

The component matcher uses a greedy name capture ([A-Za-z0-9-\s]+), so the first match absorbs the rest of the prompt and no additional components are detected. For example, plan_scenario("90% methane and 10% ethane at 40 °C") only produces one match and normalizes to 100 % methane, ignoring the ethane specification. This makes the new natural-language setup silently produce the wrong fluid when users separate components with common phrases like “and”, which is a primary use case of the feature.

Useful? React with 👍 / 👎.

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

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants