nestharus

Here's a redesign that fully treats:

object IDs as entities,
field names as part of the fact,
and object membership (ancestor chain) as the core context.

I'll start with the shared data model and then show how extract_ids_and_text, resolution_tracker.py, and candidate_extraction.py change, plus how this plugs into your fact pipeline.

	You will be given a very unrefined prompt containing many interleaved requirements. Do not create an implementation plan. Do not add assumptions about my repo/system. Your job is only to unmangle and restructure what I said.

	Produce three sections:

	1. Topics (Unmangled)

	* Extract and list all distinct topics/concerns present in my text.
	* Phrase them as short topic titles.

	2. Organized Requirements (Same content, reordered)

	# Plan

	## Overview

	## Current State (Problems)

	## Target State

	## Additional Info

	Download as zip
	Create empty auths folder in directory
	Create empty logs folder in directory

	Within directory

	docker-compose up -d

	OpenAI: docker exec -it cli-proxy-api ./CLIProxyAPI --codex-login
	Google: docker exec -it cli-proxy-api ./CLIProxyAPI --login

	from fastapi import FastAPI, HTTPException
	from pydantic import BaseModel
	from typing import List, Optional
	from vllm import LLM, SamplingParams
	from vllm.inputs.data import TokensPrompt
	from transformers import AutoTokenizer
	import math

	class RerankRequest(BaseModel):
	query: str

	public static Map<String, Set<String>> idsAndFields(final Collection<Match> matches) {
	return matches.stream()
	.collect(Collectors.groupingBy(
	match -> match.id,
	Collectors.flatMapping(
	match -> match.positions().stream()
	.map(position -> position.fieldName),
	Collectors.toSet()
	)
	));

	# 2 [ 1]
	# 3 [ 3, 1]
	# 5 [5, 3, 1, 3, 1, 3, 5, 1]
	# (19) 7 [9, 1, 3, 1, 3, 5, 1, 5, 3, 1, 3, 5, 5, 1, 5, 3, 1, 5, 3, 5, 7, 3, 1, 3, 1, 3********************************************************************, 7, 5, 3, 5, 1, 3, 5, 1, 5, 5, 3, 1, 3, 5, 1, 5, 3, 1, 3, 1, 9, 1]
	# ( 9) 11 [11, 3, 1, 3, 5, 1, 5, 3, 1, 3, 5, 5, 1, 5, 3, 1, 5, 3, 5, 7, 3, 1, 3, 1, 3, 13, 3, 5, 1, 9, 1, 5, 5, 3********************************************************************, 1, 3, 5, 1, 9, 1, 3, 1, 11, 9, 1, 3, 1, 3, 5, 1, 5, 3, 5, 5, 5, 1, 5, 3, 1, 5, 3, 5, 7, 3, 1, 3, 5, 7, 5, 9, 1, 3, 5, 1, 5, 5, 3, 1, 3, 5, 1, 5, 3, 1, 5, 9, 1, 9, 1, 3, 1, 3, 5, 7, 3, 1, 3, 11, 1, 5, 3, 1, 5, 3, 5, 11, 1, 3, 1, 3, 7, 5, 3, 5, 1, 3, 5, 1, 5, 9, 1, 3,5, 1, 5, 3, 1, 3, 1, 9, 1, 9, 1, 3, 5, 5, 1, 5, 5, 3, 5, 5, 1, 5, 3, 1, 5, 3, 5, 7, 3, 1, 5,3, 7, 5, 3, 5, 1, 3, 5, 7, 5, 3, 1, 9, 1, 5, 3, 1, 3, 1, 9, 1, 9, 1, 3, 1, 3, 7, 5, 3, 1, 3,5, 5, 1, 5, 3, 7, 3, 5, 7, 3, 1, 3, 1, 3, 7, 5, 3, 5, 5, 5, 1, 5, 5, 3, 1, 3, 5, 1, 5

	class ClassTest {
	public static class A_MODEL {
	public B_MODEL b;
	public C_MODEL c;
	}

	public static class A_DTO {
	public B_DTO b;
	}

	export function once(func) {
	let invoked = false;

	return function() {
	if (invoked) {
	return;
	}

	invoked = true;