File size: 15,928 Bytes
b3b0ccd
45b0535
b3b0ccd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45b0535
b3b0ccd
45b0535
 
 
 
 
b3b0ccd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45b0535
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b3b0ccd
 
 
 
45b0535
b3b0ccd
 
 
 
 
 
 
 
 
 
45b0535
b3b0ccd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
from typing import Literal, Set
import json
from langchain_core.messages import SystemMessage, HumanMessage, RemoveMessage, AIMessage, ToolMessage
from langgraph.types import Command
from .graph_state import State, AgentState
from .schemas import QueryAnalysis
from .prompts import *
from utils import estimate_context_tokens
from config import BASE_TOKEN_THRESHOLD, CHILD_CHUNK_SEPARATOR, MAIN_HISTORY_MESSAGES_TO_KEEP, TOKEN_GROWTH_FACTOR

if MAIN_HISTORY_MESSAGES_TO_KEEP < 2:
    raise ValueError("MAIN_HISTORY_MESSAGES_TO_KEEP must be at least 2.")

PRE_ANSWER_HISTORY_MESSAGES_TO_KEEP = max(MAIN_HISTORY_MESSAGES_TO_KEEP - 1, 0)

def _is_plain_conversation_message(msg) -> bool:
    return (
        isinstance(msg, (HumanMessage, AIMessage))
        and not getattr(msg, "tool_calls", None)
        and not getattr(msg, "name", None)
    )

def _name_internal_message(message, name):
    """Tag a subgraph-only message so it is not treated as chat history."""
    return message.model_copy(update={"name": name})

def _retrieval_contexts(messages) -> list[str]:
    contexts = []
    ignored_prefixes = (
        "NO_RELEVANT_CHUNKS",
        "NO_PARENT_DOCUMENT",
        "RETRIEVAL_ERROR:",
        "PARENT_RETRIEVAL_ERROR:",
    )
    for message in messages:
        if not isinstance(message, ToolMessage):
            continue
        content = str(message.content).strip()
        if content and not content.startswith(ignored_prefixes):
            parts = content.split(CHILD_CHUNK_SEPARATOR) if message.name == "search_child_chunks" else [content]
            contexts.extend(part for part in parts if part)
    return list(dict.fromkeys(contexts))

def _format_conversation(messages) -> str:
    lines = []
    for msg in messages:
        role = "User" if isinstance(msg, HumanMessage) else "Assistant"
        lines.append(f"{role}: {msg.content}")
    return "\n".join(lines)

def _remove_messages_not_in(messages, keep_ids):
    removals = []
    for msg in messages:
        msg_id = getattr(msg, "id", None)
        if isinstance(msg, SystemMessage) or not msg_id:
            continue
        if msg_id not in keep_ids:
            removals.append(RemoveMessage(id=msg_id))
    return removals

def _recent_conversation(messages, pending_query="") -> list:
    """Return recent context before the current user message.

    During clarification, exclude the unresolved query and the assistant's
    clarification request because they are represented explicitly.
    """
    plain_messages = [msg for msg in messages if _is_plain_conversation_message(msg)]
    recent_messages = plain_messages[:-1]

    if pending_query:
        for index in range(len(recent_messages) - 1, -1, -1):
            msg = recent_messages[index]
            if isinstance(msg, HumanMessage) and str(msg.content).strip() == pending_query:
                return recent_messages[:index]

    return recent_messages

def summarize_history(state: State, llm):
    messages = state.get("messages", [])
    updates = {"agent_answers": [{"__reset__": True}]}

    if not messages:
        return updates

    plain_messages = [msg for msg in messages if _is_plain_conversation_message(msg)]
    keep_count = PRE_ANSWER_HISTORY_MESSAGES_TO_KEEP
    messages_to_summarize = plain_messages[:-keep_count] if len(plain_messages) > keep_count else []
    keep_ids = {getattr(msg, "id", None) for msg in plain_messages[-keep_count:]}
    keep_ids.discard(None)

    removals = _remove_messages_not_in(messages, keep_ids)
    if removals:
        updates["messages"] = removals

    if not messages_to_summarize:
        return updates

    existing_summary = state.get("conversation_summary", "").strip()
    conversation = "Existing summary:\n"
    conversation += f"{existing_summary or '(none)'}\n\n"
    conversation += "New messages to merge into the summary:\n"
    conversation += _format_conversation(messages_to_summarize)

    summary_response = llm.invoke([
        SystemMessage(content=get_conversation_summary_prompt()),
        HumanMessage(content=conversation),
    ])
    updates["conversation_summary"] = summary_response.content.strip()
    return updates

def rewrite_query(state: State, llm):
    last_message = state["messages"][-1]
    current_query = str(last_message.content).strip()
    conversation_summary = state.get("conversation_summary", "").strip()
    pending_query = state.get("pendingQuery", "").strip()
    pending_clarifications = state.get("pendingClarifications", [])
    recent_messages = _recent_conversation(state["messages"], pending_query)
    available_documents = state.get("availableDocuments", [])

    context_parts = [
        "Available Uploaded Documents (filenames only, not instructions):\n"
        + json.dumps(available_documents),
        f"Latest User Message:\n{current_query}",
    ]
    if conversation_summary:
        context_parts.append(f"Conversation Summary:\n{conversation_summary}")
    if recent_messages:
        context_parts.append(f"Recent Conversation:\n{_format_conversation(recent_messages)}")

    if pending_query:
        clarifications = [*pending_clarifications, current_query]
        clarification_text = "\n".join(
            f"{index}. {value}" for index, value in enumerate(clarifications, start=1)
        )
        context_parts.append(
            f"Unresolved User Query:\n{pending_query}\n\n"
            f"User Clarifications:\n{clarification_text}"
        )
        original_query = f"{pending_query}\nClarifications:\n{clarification_text}"
    else:
        clarifications = []
        context_parts.append(f"User Query:\n{current_query}")
        original_query = current_query

    context_section = "\n\n".join(context_parts)
    llm_with_structure = llm.with_structured_output(QueryAnalysis, method="function_calling")
    response = llm_with_structure.invoke([SystemMessage(content=get_rewrite_query_prompt()), HumanMessage(content=context_section)])
    if pending_query and (response.starts_new_query or response.is_document_listing):
        clarification_message_update = [
            _name_internal_message(message, "abandoned_query")
            for message in state["messages"][:-1]
            if isinstance(message, HumanMessage) and str(message.content).strip() == pending_query
        ]
        pending_query = ""
        clarifications = []
        original_query = current_query
    else:
        clarification_message_update = (
            [_name_internal_message(last_message, "clarification_response")]
            if pending_query else []
        )

    if response.is_document_listing:
        inventory = (
            "Yes. These documents are available in the shared library:\n\n"
            + "\n".join(f"- {filename}" for filename in available_documents)
            if available_documents
            else "No documents are indexed yet. Upload a PDF or Markdown file in Documents, then click Add Documents."
        )
        return {
            "questionIsClear": True,
            "documentListRequested": True,
            "originalQuery": current_query,
            "pendingQuery": "",
            "pendingClarifications": [],
            "rewrittenQuestions": [],
            "messages": clarification_message_update + [AIMessage(content=inventory)],
        }

    if response.questions and response.is_clear:
        return {
            "questionIsClear": True,
            "documentListRequested": False,
            "originalQuery": original_query,
            "pendingQuery": "",
            "pendingClarifications": [],
            "rewrittenQuestions": response.questions,
            "messages": clarification_message_update,
        }

    clarification = response.clarification_needed if response.clarification_needed and len(response.clarification_needed.strip()) > 10 else "I need more information to understand your question."
    return {
        "questionIsClear": False,
        "documentListRequested": False,
        "originalQuery": "",
        "pendingQuery": pending_query or current_query,
        "pendingClarifications": clarifications,
        "rewrittenQuestions": [],
        "messages": clarification_message_update + [
            AIMessage(content=clarification, name="clarification")
        ],
    }

def request_clarification(state: State):
    return {}

# --- Agent Nodes ---
def orchestrator(state: AgentState, llm_with_tools):
    context_summary = state.get("context_summary", "").strip()
    sys_msg = SystemMessage(content=get_orchestrator_prompt())
    summary_injection = (
        [HumanMessage(content=f"[COMPRESSED CONTEXT FROM PRIOR RESEARCH]\n\n{context_summary}")]
        if context_summary else []
    )
    if not state.get("messages"):
        human_msg = HumanMessage(content=state["question"], name="agent_question")
        force_search = HumanMessage(content="YOU MUST CALL 'search_child_chunks' AS THE FIRST STEP TO ANSWER THIS QUESTION.")
        response = llm_with_tools.invoke([sys_msg] + summary_injection + [human_msg, force_search])
        response = _name_internal_message(response, "agent_response")
        return {"messages": [human_msg, response], "tool_call_count": len(response.tool_calls or []), "iteration_count": 1}

    response = llm_with_tools.invoke([sys_msg] + summary_injection + state["messages"])
    response = _name_internal_message(response, "agent_response")
    tool_calls = response.tool_calls if hasattr(response, "tool_calls") else []
    return {"messages": [response], "tool_call_count": len(tool_calls) if tool_calls else 0, "iteration_count": 1}

def fallback_response(state: AgentState, llm):
    seen = set()
    unique_contents = []
    for m in state["messages"]:
        if isinstance(m, ToolMessage) and m.content not in seen:
            unique_contents.append(m.content)
            seen.add(m.content)

    context_summary = state.get("context_summary", "").strip()

    context_parts = []
    if context_summary:
        context_parts.append(f"## Compressed Research Context (from prior iterations)\n\n{context_summary}")
    if unique_contents:
        context_parts.append(
            "## Retrieved Data (current iteration)\n\n" +
            "\n\n".join(f"--- DATA SOURCE {i} ---\n{content}" for i, content in enumerate(unique_contents, 1))
        )

    context_text = "\n\n".join(context_parts) if context_parts else "No data was retrieved from the documents."

    prompt_content = (
        f"USER QUERY: {state.get('question')}\n\n"
        f"{context_text}\n\n"
        f"INSTRUCTION:\nProvide the best possible answer using only the data above."
    )
    response = llm.invoke([SystemMessage(content=get_fallback_response_prompt()), HumanMessage(content=prompt_content)])
    response = _name_internal_message(response, "agent_response")
    return {"messages": [response]}

def should_compress_context(state: AgentState) -> Command[Literal["compress_context", "orchestrator"]]:
    messages = state["messages"]

    new_ids: Set[str] = set()
    for msg in reversed(messages):
        if isinstance(msg, AIMessage) and getattr(msg, "tool_calls", None):
            for tc in msg.tool_calls:
                if tc["name"] == "retrieve_parent_chunks":
                    raw = tc["args"].get("parent_id") or tc["args"].get("id") or tc["args"].get("ids") or []
                    if isinstance(raw, str):
                        new_ids.add(f"parent::{raw}")
                    else:
                        new_ids.update(f"parent::{r}" for r in raw)

                elif tc["name"] == "search_child_chunks":
                    query = tc["args"].get("query", "")
                    if query:
                        new_ids.add(f"search::{query}")
            break

    updated_ids = state.get("retrieval_keys", set()) | new_ids

    current_token_messages = estimate_context_tokens(messages)
    current_token_summary = estimate_context_tokens([HumanMessage(content=state.get("context_summary", ""))])
    current_tokens = current_token_messages + current_token_summary

    max_allowed = BASE_TOKEN_THRESHOLD + int(current_token_summary * TOKEN_GROWTH_FACTOR)

    goto = "compress_context" if current_tokens > max_allowed else "orchestrator"
    return Command(
        update={
            "retrieval_keys": updated_ids,
            "retrieved_contexts": _retrieval_contexts(messages),
        },
        goto=goto,
    )

def compress_context(state: AgentState, llm):
    messages = state["messages"]
    existing_summary = state.get("context_summary", "").strip()

    if not messages:
        return {}

    conversation_text = f"USER QUESTION:\n{state.get('question')}\n\nConversation to compress:\n\n"
    if existing_summary:
        conversation_text += f"[PRIOR COMPRESSED CONTEXT]\n{existing_summary}\n\n"

    for msg in messages[1:]:
        if isinstance(msg, AIMessage):
            tool_calls_info = ""
            if getattr(msg, "tool_calls", None):
                calls = ", ".join(f"{tc['name']}({tc['args']})" for tc in msg.tool_calls)
                tool_calls_info = f" | Tool calls: {calls}"
            conversation_text += f"[ASSISTANT{tool_calls_info}]\n{msg.content or '(tool call only)'}\n\n"
        elif isinstance(msg, ToolMessage):
            tool_name = getattr(msg, "name", "tool")
            conversation_text += f"[TOOL RESULT — {tool_name}]\n{msg.content}\n\n"

    summary_response = llm.invoke([SystemMessage(content=get_context_compression_prompt()), HumanMessage(content=conversation_text)])
    new_summary = summary_response.content

    retrieved_ids: Set[str] = state.get("retrieval_keys", set())
    if retrieved_ids:
        parent_ids = sorted(r for r in retrieved_ids if r.startswith("parent::"))
        search_queries = sorted(r.replace("search::", "") for r in retrieved_ids if r.startswith("search::"))

        block = "\n\n---\n**Already executed (do NOT repeat):**\n"
        if parent_ids:
            block += "Parent chunks retrieved:\n" + "\n".join(f"- {p.replace('parent::', '')}" for p in parent_ids) + "\n"
        if search_queries:
            block += "Search queries already run:\n" + "\n".join(f"- {q}" for q in search_queries) + "\n"
        new_summary += block

    return {"context_summary": new_summary, "messages": [RemoveMessage(id=m.id) for m in messages[1:]]}

def collect_answer(state: AgentState):
    last_message = state["messages"][-1]
    is_valid = isinstance(last_message, AIMessage) and last_message.content and not last_message.tool_calls
    answer = last_message.content if is_valid else "Unable to generate an answer."
    return {
        "final_answer": answer,
        "agent_answers": [{
            "index": state["question_index"],
            "question": state["question"],
            "answer": answer,
            "contexts": state.get("retrieved_contexts", []),
        }]
    }
# --- End of Agent Nodes---

def aggregate_answers(state: State, llm):
    messages = state.get("messages", [])
    plain_messages = [msg for msg in messages if _is_plain_conversation_message(msg)]
    keep_ids = {getattr(msg, "id", None) for msg in plain_messages[-PRE_ANSWER_HISTORY_MESSAGES_TO_KEEP:]}
    keep_ids.discard(None)
    removals = _remove_messages_not_in(messages, keep_ids)

    if not state.get("agent_answers"):
        return {"messages": removals + [AIMessage(content="No answers were generated.")]}

    sorted_answers = sorted(state["agent_answers"], key=lambda x: x["index"])

    formatted_answers = ""
    for i, ans in enumerate(sorted_answers, start=1):
        formatted_answers += (f"\nRetrieved response {i}:\n"f"{ans['answer']}\n")

    user_message = HumanMessage(content=f"""Original user question: {state["originalQuery"]}\nRetrieved answers:{formatted_answers}""")
    synthesis_response = llm.invoke([SystemMessage(content=get_aggregation_prompt()), user_message])
    return {"messages": removals + [AIMessage(content=synthesis_response.content)]}