File size: 4,746 Bytes
79d4fd5
 
 
 
 
 
 
 
 
0ff62f0
79d4fd5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
92925b6
 
 
 
 
 
 
 
 
 
 
 
 
79d4fd5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0ff62f0
79d4fd5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import hashlib
import logging
import time
from collections import OrderedDict
from typing import Any, Dict, List, Optional

from groq import Groq

from config import GROQ_API_KEY
from prompts import SUMMARIZATION_PROMPT

logger = logging.getLogger(__name__)


class ConversationSummarizer:

    MAX_HISTORY_TURNS = 10
    SUMMARY_MAX_TOKENS = 300
    RECENT_MESSAGES_TO_KEEP = 8
    MAX_INPUT_TEXT_LENGTH = 2000
    MAX_CACHE_SIZE = 100

    def __init__(
        self,
        api_key: Optional[str] = None,
        model: str = "llama-3.1-8b-instant",
        max_history_turns: int = 10,
    ):
        self._api_key = api_key or GROQ_API_KEY
        self._model = model
        self._max_turns = max_history_turns
        self._client: Optional[Groq] = None
        self._summary_cache: OrderedDict[str, str] = OrderedDict()

    @property
    def client(self) -> Groq:
        if self._client is None:
            if not self._api_key:
                raise ValueError("GROQ_API_KEY not configured")
            self._client = Groq(api_key=self._api_key, timeout=5.0)
        return self._client

    def _generate_cache_key(self, messages: List[Dict[str, str]]) -> str:
        parts = []
        for msg in messages:
            content = str(msg.get("content", ""))[:50]
            parts.append(content)
        raw = "||".join(parts)
        return hashlib.md5(raw.encode("utf-8")).hexdigest()

    def summarize_if_needed(self, chat_history: List[Dict[str, str]]) -> List[Dict[str, str]]:
        if len(chat_history) <= self._max_turns * 2:
            return chat_history

        try:
            recent_messages = chat_history[-self.RECENT_MESSAGES_TO_KEEP:]
            older_messages = chat_history[:-self.RECENT_MESSAGES_TO_KEEP]

            db_messages = [m for m in older_messages if m.get("is_database")]
            non_db_older = [m for m in older_messages if not m.get("is_database")]

            summary = self._summarize_conversation(non_db_older) if non_db_older else ""
            result = []
            if summary:
                result.append({
                    "role": "system",
                    "content": f"Önceki konuşma özeti: {summary}",
                })
            result.extend(db_messages)
            result.extend(recent_messages)
            return result
        except Exception as e:
            logger.warning(f"Summarization failed, returning original history: {e}")
            return chat_history

    def _summarize_conversation(self, messages: List[Dict[str, str]]) -> str:
        cache_key = self._generate_cache_key(messages)

        if cache_key in self._summary_cache:
            self._summary_cache.move_to_end(cache_key)
            logger.debug("Conversation summary cache hit")
            return self._summary_cache[cache_key]

        conversation_parts = []
        for msg in messages:
            role = msg.get("role", "unknown")
            content = str(msg.get("content", ""))
            conversation_parts.append(f"{role}: {content}")

        conversation_text = "\n".join(conversation_parts)
        if len(conversation_text) > self.MAX_INPUT_TEXT_LENGTH:
            conversation_text = conversation_text[-self.MAX_INPUT_TEXT_LENGTH:]

        prompt = SUMMARIZATION_PROMPT.format(conversation_text=conversation_text)

        try:
            response = self.client.chat.completions.create(
                model=self._model,
                messages=[{"role": "user", "content": prompt}],
                temperature=0.3,
                max_tokens=self.SUMMARY_MAX_TOKENS,
            )
            summary = response.choices[0].message.content.strip()
            if not summary:
                summary = "Konuşma geçmişi mevcut ama özeti oluşturulamadı."
        except Exception as e:
            logger.warning(f"Summary generation failed: {e}")
            summary = "Konuşma geçmişi mevcut ama özeti oluşturulamadı."

        while len(self._summary_cache) >= self.MAX_CACHE_SIZE:
            self._summary_cache.popitem(last=False)
        self._summary_cache[cache_key] = summary

        return summary

    def clear_cache(self) -> None:
        self._summary_cache.clear()


_summarizer_instance: Optional[ConversationSummarizer] = None


def get_conversation_summarizer(
    api_key: Optional[str] = None,
    model: Optional[str] = None,
) -> ConversationSummarizer:
    global _summarizer_instance
    if _summarizer_instance is None:
        _summarizer_instance = ConversationSummarizer(
            api_key=api_key,
            model=model or "llama-3.1-8b-instant",
        )
    return _summarizer_instance


def reset_conversation_summarizer() -> None:
    global _summarizer_instance
    _summarizer_instance = None