Skip to content

Slack

SlackReader #

Bases: BasePydanticReader

Slack阅读器。

从频道中读取对话。如果提供了最早日期,还可以提供可选的最晚日期。如果没有提供最晚日期,我们假设最晚日期是当前时间戳。

Source code in llama_index/readers/slack/base.py
 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
class SlackReader(BasePydanticReader):
    """Slack阅读器。

从频道中读取对话。如果提供了最早日期,还可以提供可选的最晚日期。如果没有提供最晚日期,我们假设最晚日期是当前时间戳。

Args:
    slack_token(可选[str]):Slack令牌。如果未提供,则假定环境变量`SLACK_BOT_TOKEN`已设置。
    ssl(可选[str]):自定义SSL上下文。如果未提供,假定已经有一个SSL上下文可用。
    earliest_date(可选[datetime]):从中读取对话的最早日期。如果未提供,我们读取所有消息。
    latest_date(可选[datetime]):从中读取对话的最晚日期。如果未提供,默认为当前时间戳与最早日期的组合。"""

    is_remote: bool = True
    slack_token: str
    earliest_date_timestamp: Optional[float]
    latest_date_timestamp: float

    _client: Any = PrivateAttr()

    def __init__(
        self,
        slack_token: Optional[str] = None,
        ssl: Optional[SSLContext] = None,
        earliest_date: Optional[datetime] = None,
        latest_date: Optional[datetime] = None,
        earliest_date_timestamp: Optional[float] = None,
        latest_date_timestamp: Optional[float] = None,
    ) -> None:
        """使用参数进行初始化。"""
        from slack_sdk import WebClient

        if slack_token is None:
            slack_token = os.environ["SLACK_BOT_TOKEN"]
        if slack_token is None:
            raise ValueError(
                "Must specify `slack_token` or set environment "
                "variable `SLACK_BOT_TOKEN`."
            )
        if ssl is None:
            self._client = WebClient(token=slack_token)
        else:
            self._client = WebClient(token=slack_token, ssl=ssl)
        if latest_date is not None and earliest_date is None:
            raise ValueError(
                "Must specify `earliest_date` if `latest_date` is specified."
            )
        if earliest_date is not None:
            earliest_date_timestamp = earliest_date.timestamp()
        else:
            earliest_date_timestamp = None or earliest_date_timestamp
        if latest_date is not None:
            latest_date_timestamp = latest_date.timestamp()
        else:
            latest_date_timestamp = datetime.now().timestamp() or latest_date_timestamp
        res = self._client.api_test()
        if not res["ok"]:
            raise ValueError(f"Error initializing Slack API: {res['error']}")

        super().__init__(
            slack_token=slack_token,
            earliest_date_timestamp=earliest_date_timestamp,
            latest_date_timestamp=latest_date_timestamp,
        )

    @classmethod
    def class_name(cls) -> str:
        return "SlackReader"

    def _read_message(self, channel_id: str, message_ts: str) -> str:
        from slack_sdk.errors import SlackApiError

        """Read a message."""

        messages_text: List[str] = []
        next_cursor = None
        while True:
            try:
                # https://slack.com/api/conversations.replies
                # List all replies to a message, including the message itself.
                if self.earliest_date_timestamp is None:
                    result = self._client.conversations_replies(
                        channel=channel_id, ts=message_ts, cursor=next_cursor
                    )
                else:
                    conversations_replies_kwargs = {
                        "channel": channel_id,
                        "ts": message_ts,
                        "cursor": next_cursor,
                        "latest": str(self.latest_date_timestamp),
                    }
                    if self.earliest_date_timestamp is not None:
                        conversations_replies_kwargs["oldest"] = str(
                            self.earliest_date_timestamp
                        )
                    result = self._client.conversations_replies(
                        **conversations_replies_kwargs  # type: ignore
                    )
                messages = result["messages"]
                messages_text.extend(message["text"] for message in messages)
                if not result["has_more"]:
                    break

                next_cursor = result["response_metadata"]["next_cursor"]
            except SlackApiError as e:
                if e.response["error"] == "ratelimited":
                    logger.error(
                        "Rate limit error reached, sleeping for: {} seconds".format(
                            e.response.headers["retry-after"]
                        )
                    )
                    time.sleep(int(e.response.headers["retry-after"]))
                else:
                    logger.error(f"Error parsing conversation replies: {e}")
                    break

        return "\n\n".join(messages_text)

    def _read_channel(self, channel_id: str, reverse_chronological: bool) -> str:
        from slack_sdk.errors import SlackApiError

        """Read a channel."""

        result_messages: List[str] = []
        next_cursor = None
        while True:
            try:
                # Call the conversations.history method using the WebClient
                # conversations.history returns the first 100 messages by default
                # These results are paginated,
                # see: https://api.slack.com/methods/conversations.history$pagination
                conversations_history_kwargs = {
                    "channel": channel_id,
                    "cursor": next_cursor,
                    "latest": str(self.latest_date_timestamp),
                }
                if self.earliest_date_timestamp is not None:
                    conversations_history_kwargs["oldest"] = str(
                        self.earliest_date_timestamp
                    )
                result = self._client.conversations_history(
                    **conversations_history_kwargs  # type: ignore
                )
                conversation_history = result["messages"]
                # Print results
                logger.info(
                    f"{len(conversation_history)} messages found in {channel_id}"
                )
                result_messages.extend(
                    self._read_message(channel_id, message["ts"])
                    for message in conversation_history
                )
                if not result["has_more"]:
                    break
                next_cursor = result["response_metadata"]["next_cursor"]

            except SlackApiError as e:
                if e.response["error"] == "ratelimited":
                    logger.error(
                        "Rate limit error reached, sleeping for: {} seconds".format(
                            e.response.headers["retry-after"]
                        )
                    )
                    time.sleep(int(e.response.headers["retry-after"]))
                else:
                    logger.error(f"Error parsing conversation replies: {e}")

        return (
            "\n\n".join(result_messages)
            if reverse_chronological
            else "\n\n".join(result_messages[::-1])
        )

    def load_data(
        self, channel_ids: List[str], reverse_chronological: bool = True
    ) -> List[Document]:
        """从输入目录加载数据。

Args:
    channel_ids(List[str]):要读取的频道id列表。

Returns:
    List[Document]:文档列表。
"""
        results = []
        for channel_id in channel_ids:
            channel_content = self._read_channel(
                channel_id, reverse_chronological=reverse_chronological
            )
            results.append(
                Document(
                    id_=channel_id,
                    text=channel_content,
                    metadata={"channel": channel_id},
                )
            )
        return results

load_data #

load_data(
    channel_ids: List[str],
    reverse_chronological: bool = True,
) -> List[Document]

从输入目录加载数据。

Returns:

Type Description
List[Document]

List[Document]:文档列表。

Source code in llama_index/readers/slack/base.py
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
    def load_data(
        self, channel_ids: List[str], reverse_chronological: bool = True
    ) -> List[Document]:
        """从输入目录加载数据。

Args:
    channel_ids(List[str]):要读取的频道id列表。

Returns:
    List[Document]:文档列表。
"""
        results = []
        for channel_id in channel_ids:
            channel_content = self._read_channel(
                channel_id, reverse_chronological=reverse_chronological
            )
            results.append(
                Document(
                    id_=channel_id,
                    text=channel_content,
                    metadata={"channel": channel_id},
                )
            )
        return results