|
|
@@ -20,7 +20,7 @@ import requests
|
|
|
import yaml
|
|
|
|
|
|
|
|
|
-VERSION = "3.4.0"
|
|
|
+VERSION = "3.4.1"
|
|
|
|
|
|
|
|
|
# === SMTP邮件配置 ===
|
|
|
@@ -1071,6 +1071,9 @@ def format_rank_display(ranks: List[int], rank_threshold: int, format_type: str)
|
|
|
elif format_type == "telegram":
|
|
|
highlight_start = "<b>"
|
|
|
highlight_end = "</b>"
|
|
|
+ elif format_type == "slack":
|
|
|
+ highlight_start = "*"
|
|
|
+ highlight_end = "*"
|
|
|
else:
|
|
|
highlight_start = "**"
|
|
|
highlight_end = "**"
|
|
|
@@ -1576,7 +1579,8 @@ def format_title_for_platform(
|
|
|
|
|
|
return result
|
|
|
|
|
|
- elif platform == "wework":
|
|
|
+ elif platform in ("wework", "bark"):
|
|
|
+ # WeWork 和 Bark 使用 markdown 格式
|
|
|
if link_url:
|
|
|
formatted_title = f"[{cleaned_title}]({link_url})"
|
|
|
else:
|
|
|
@@ -1642,6 +1646,34 @@ def format_title_for_platform(
|
|
|
|
|
|
return result
|
|
|
|
|
|
+ elif platform == "slack":
|
|
|
+ # Slack 使用 mrkdwn 格式
|
|
|
+ if link_url:
|
|
|
+ # Slack 链接格式: <url|text>
|
|
|
+ formatted_title = f"<{link_url}|{cleaned_title}>"
|
|
|
+ else:
|
|
|
+ formatted_title = cleaned_title
|
|
|
+
|
|
|
+ title_prefix = "🆕 " if title_data.get("is_new") else ""
|
|
|
+
|
|
|
+ if show_source:
|
|
|
+ result = f"[{title_data['source_name']}] {title_prefix}{formatted_title}"
|
|
|
+ else:
|
|
|
+ result = f"{title_prefix}{formatted_title}"
|
|
|
+
|
|
|
+ # 排名（使用 * 加粗）
|
|
|
+ rank_display = format_rank_display(
|
|
|
+ title_data["ranks"], title_data["rank_threshold"], "slack"
|
|
|
+ )
|
|
|
+ if rank_display:
|
|
|
+ result += f" {rank_display}"
|
|
|
+ if title_data["time_display"]:
|
|
|
+ result += f" `- {title_data['time_display']}`"
|
|
|
+ if title_data["count"] > 1:
|
|
|
+ result += f" `({title_data['count']}次)`"
|
|
|
+
|
|
|
+ return result
|
|
|
+
|
|
|
elif platform == "html":
|
|
|
rank_display = format_rank_display(
|
|
|
title_data["ranks"], title_data["rank_threshold"], "html"
|
|
|
@@ -2906,6 +2938,90 @@ def render_dingtalk_content(
|
|
|
return text_content
|
|
|
|
|
|
|
|
|
+def _get_batch_header(format_type: str, batch_num: int, total_batches: int) -> str:
|
|
|
+ """根据 format_type 生成对应格式的批次头部"""
|
|
|
+ if format_type == "telegram":
|
|
|
+ return f"<b>[第 {batch_num}/{total_batches} 批次]</b>\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ return f"*[第 {batch_num}/{total_batches} 批次]*\n\n"
|
|
|
+ elif format_type in ("wework_text", "bark"):
|
|
|
+ # 企业微信文本模式和 Bark 使用纯文本格式
|
|
|
+ return f"[第 {batch_num}/{total_batches} 批次]\n\n"
|
|
|
+ else:
|
|
|
+ # 飞书、钉钉、ntfy、企业微信 markdown 模式
|
|
|
+ return f"**[第 {batch_num}/{total_batches} 批次]**\n\n"
|
|
|
+
|
|
|
+
|
|
|
+def _get_max_batch_header_size(format_type: str) -> int:
|
|
|
+ """估算批次头部的最大字节数（假设最多 99 批次）
|
|
|
+
|
|
|
+ 用于在分批时预留空间，避免事后截断破坏内容完整性。
|
|
|
+ """
|
|
|
+ # 生成最坏情况的头部（99/99 批次）
|
|
|
+ max_header = _get_batch_header(format_type, 99, 99)
|
|
|
+ return len(max_header.encode("utf-8"))
|
|
|
+
|
|
|
+
|
|
|
+def _truncate_to_bytes(text: str, max_bytes: int) -> str:
|
|
|
+ """安全截断字符串到指定字节数，避免截断多字节字符"""
|
|
|
+ text_bytes = text.encode("utf-8")
|
|
|
+ if len(text_bytes) <= max_bytes:
|
|
|
+ return text
|
|
|
+
|
|
|
+ # 截断到指定字节数
|
|
|
+ truncated = text_bytes[:max_bytes]
|
|
|
+
|
|
|
+ # 处理可能的不完整 UTF-8 字符
|
|
|
+ for i in range(min(4, len(truncated))):
|
|
|
+ try:
|
|
|
+ return truncated[: len(truncated) - i].decode("utf-8")
|
|
|
+ except UnicodeDecodeError:
|
|
|
+ continue
|
|
|
+
|
|
|
+ # 极端情况：返回空字符串
|
|
|
+ return ""
|
|
|
+
|
|
|
+
|
|
|
+def add_batch_headers(
|
|
|
+ batches: List[str], format_type: str, max_bytes: int
|
|
|
+) -> List[str]:
|
|
|
+ """为批次添加头部，动态计算确保总大小不超过限制
|
|
|
+
|
|
|
+ Args:
|
|
|
+ batches: 原始批次列表
|
|
|
+ format_type: 推送类型（bark, telegram, feishu 等）
|
|
|
+ max_bytes: 该推送类型的最大字节限制
|
|
|
+
|
|
|
+ Returns:
|
|
|
+ 添加头部后的批次列表
|
|
|
+ """
|
|
|
+ if len(batches) <= 1:
|
|
|
+ return batches
|
|
|
+
|
|
|
+ total = len(batches)
|
|
|
+ result = []
|
|
|
+
|
|
|
+ for i, content in enumerate(batches, 1):
|
|
|
+ # 生成批次头部
|
|
|
+ header = _get_batch_header(format_type, i, total)
|
|
|
+ header_size = len(header.encode("utf-8"))
|
|
|
+
|
|
|
+ # 动态计算允许的最大内容大小
|
|
|
+ max_content_size = max_bytes - header_size
|
|
|
+ content_size = len(content.encode("utf-8"))
|
|
|
+
|
|
|
+ # 如果超出，截断到安全大小
|
|
|
+ if content_size > max_content_size:
|
|
|
+ print(
|
|
|
+ f"警告：{format_type} 第 {i}/{total} 批次内容({content_size}字节) + 头部({header_size}字节) 超出限制({max_bytes}字节)，截断到 {max_content_size} 字节"
|
|
|
+ )
|
|
|
+ content = _truncate_to_bytes(content, max_content_size)
|
|
|
+
|
|
|
+ result.append(header + content)
|
|
|
+
|
|
|
+ return result
|
|
|
+
|
|
|
+
|
|
|
def split_content_into_batches(
|
|
|
report_data: Dict,
|
|
|
format_type: str,
|
|
|
@@ -2932,7 +3048,7 @@ def split_content_into_batches(
|
|
|
now = get_beijing_time()
|
|
|
|
|
|
base_header = ""
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
base_header = f"**总新闻数：** {total_titles}\n\n\n\n"
|
|
|
elif format_type == "telegram":
|
|
|
base_header = f"总新闻数： {total_titles}\n\n"
|
|
|
@@ -2945,9 +3061,11 @@ def split_content_into_batches(
|
|
|
base_header += f"**时间：** {now.strftime('%Y-%m-%d %H:%M:%S')}\n\n"
|
|
|
base_header += f"**类型：** 热点分析报告\n\n"
|
|
|
base_header += "---\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ base_header = f"*总新闻数：* {total_titles}\n\n"
|
|
|
|
|
|
base_footer = ""
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
base_footer = f"\n\n\n> 更新时间：{now.strftime('%Y-%m-%d %H:%M:%S')}"
|
|
|
if update_info:
|
|
|
base_footer += f"\n> TrendRadar 发现新版本 **{update_info['remote_version']}**，当前 **{update_info['current_version']}**"
|
|
|
@@ -2967,10 +3085,14 @@ def split_content_into_batches(
|
|
|
base_footer = f"\n\n> 更新时间：{now.strftime('%Y-%m-%d %H:%M:%S')}"
|
|
|
if update_info:
|
|
|
base_footer += f"\n> TrendRadar 发现新版本 **{update_info['remote_version']}**，当前 **{update_info['current_version']}**"
|
|
|
+ elif format_type == "slack":
|
|
|
+ base_footer = f"\n\n_更新时间：{now.strftime('%Y-%m-%d %H:%M:%S')}_"
|
|
|
+ if update_info:
|
|
|
+ base_footer += f"\n_TrendRadar 发现新版本 *{update_info['remote_version']}*，当前 *{update_info['current_version']}_"
|
|
|
|
|
|
stats_header = ""
|
|
|
if report_data["stats"]:
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
stats_header = f"📊 **热点词汇统计**\n\n"
|
|
|
elif format_type == "telegram":
|
|
|
stats_header = f"📊 热点词汇统计\n\n"
|
|
|
@@ -2980,6 +3102,8 @@ def split_content_into_batches(
|
|
|
stats_header = f"📊 **热点词汇统计**\n\n"
|
|
|
elif format_type == "dingtalk":
|
|
|
stats_header = f"📊 **热点词汇统计**\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ stats_header = f"📊 *热点词汇统计*\n\n"
|
|
|
|
|
|
current_batch = base_header
|
|
|
current_batch_has_content = False
|
|
|
@@ -3026,7 +3150,7 @@ def split_content_into_batches(
|
|
|
|
|
|
# 构建词组标题
|
|
|
word_header = ""
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
if count >= 10:
|
|
|
word_header = (
|
|
|
f"🔥 {sequence_display} **{word}** : **{count}** 条\n\n"
|
|
|
@@ -3073,12 +3197,23 @@ def split_content_into_batches(
|
|
|
)
|
|
|
else:
|
|
|
word_header = f"📌 {sequence_display} **{word}** : {count} 条\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ if count >= 10:
|
|
|
+ word_header = (
|
|
|
+ f"🔥 {sequence_display} *{word}* : *{count}* 条\n\n"
|
|
|
+ )
|
|
|
+ elif count >= 5:
|
|
|
+ word_header = (
|
|
|
+ f"📈 {sequence_display} *{word}* : *{count}* 条\n\n"
|
|
|
+ )
|
|
|
+ else:
|
|
|
+ word_header = f"📌 {sequence_display} *{word}* : {count} 条\n\n"
|
|
|
|
|
|
# 构建第一条新闻
|
|
|
first_news_line = ""
|
|
|
if stat["titles"]:
|
|
|
first_title_data = stat["titles"][0]
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"wework", first_title_data, show_source=True
|
|
|
)
|
|
|
@@ -3098,6 +3233,10 @@ def split_content_into_batches(
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"dingtalk", first_title_data, show_source=True
|
|
|
)
|
|
|
+ elif format_type == "slack":
|
|
|
+ formatted_title = format_title_for_platform(
|
|
|
+ "slack", first_title_data, show_source=True
|
|
|
+ )
|
|
|
else:
|
|
|
formatted_title = f"{first_title_data['title']}"
|
|
|
|
|
|
@@ -3127,7 +3266,7 @@ def split_content_into_batches(
|
|
|
# 处理剩余新闻条目
|
|
|
for j in range(start_index, len(stat["titles"])):
|
|
|
title_data = stat["titles"][j]
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"wework", title_data, show_source=True
|
|
|
)
|
|
|
@@ -3147,6 +3286,10 @@ def split_content_into_batches(
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"dingtalk", title_data, show_source=True
|
|
|
)
|
|
|
+ elif format_type == "slack":
|
|
|
+ formatted_title = format_title_for_platform(
|
|
|
+ "slack", title_data, show_source=True
|
|
|
+ )
|
|
|
else:
|
|
|
formatted_title = f"{title_data['title']}"
|
|
|
|
|
|
@@ -3170,7 +3313,7 @@ def split_content_into_batches(
|
|
|
# 词组间分隔符
|
|
|
if i < len(report_data["stats"]) - 1:
|
|
|
separator = ""
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
separator = f"\n\n\n\n"
|
|
|
elif format_type == "telegram":
|
|
|
separator = f"\n\n"
|
|
|
@@ -3180,6 +3323,8 @@ def split_content_into_batches(
|
|
|
separator = f"\n{CONFIG['FEISHU_MESSAGE_SEPARATOR']}\n\n"
|
|
|
elif format_type == "dingtalk":
|
|
|
separator = f"\n---\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ separator = f"\n\n"
|
|
|
|
|
|
test_content = current_batch + separator
|
|
|
if (
|
|
|
@@ -3191,7 +3336,7 @@ def split_content_into_batches(
|
|
|
# 处理新增新闻（同样确保来源标题+第一条新闻的原子性）
|
|
|
if report_data["new_titles"]:
|
|
|
new_header = ""
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
new_header = f"\n\n\n\n🆕 **本次新增热点新闻** (共 {report_data['total_new_count']} 条)\n\n"
|
|
|
elif format_type == "telegram":
|
|
|
new_header = (
|
|
|
@@ -3203,6 +3348,8 @@ def split_content_into_batches(
|
|
|
new_header = f"\n{CONFIG['FEISHU_MESSAGE_SEPARATOR']}\n\n🆕 **本次新增热点新闻** (共 {report_data['total_new_count']} 条)\n\n"
|
|
|
elif format_type == "dingtalk":
|
|
|
new_header = f"\n---\n\n🆕 **本次新增热点新闻** (共 {report_data['total_new_count']} 条)\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ new_header = f"\n\n🆕 *本次新增热点新闻* (共 {report_data['total_new_count']} 条)\n\n"
|
|
|
|
|
|
test_content = current_batch + new_header
|
|
|
if (
|
|
|
@@ -3220,7 +3367,7 @@ def split_content_into_batches(
|
|
|
# 逐个处理新增新闻来源
|
|
|
for source_data in report_data["new_titles"]:
|
|
|
source_header = ""
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
source_header = f"**{source_data['source_name']}** ({len(source_data['titles'])} 条):\n\n"
|
|
|
elif format_type == "telegram":
|
|
|
source_header = f"{source_data['source_name']} ({len(source_data['titles'])} 条):\n\n"
|
|
|
@@ -3230,6 +3377,8 @@ def split_content_into_batches(
|
|
|
source_header = f"**{source_data['source_name']}** ({len(source_data['titles'])} 条):\n\n"
|
|
|
elif format_type == "dingtalk":
|
|
|
source_header = f"**{source_data['source_name']}** ({len(source_data['titles'])} 条):\n\n"
|
|
|
+ elif format_type == "slack":
|
|
|
+ source_header = f"*{source_data['source_name']}* ({len(source_data['titles'])} 条):\n\n"
|
|
|
|
|
|
# 构建第一条新增新闻
|
|
|
first_news_line = ""
|
|
|
@@ -3238,7 +3387,7 @@ def split_content_into_batches(
|
|
|
title_data_copy = first_title_data.copy()
|
|
|
title_data_copy["is_new"] = False
|
|
|
|
|
|
- if format_type == "wework":
|
|
|
+ if format_type in ("wework", "bark"):
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"wework", title_data_copy, show_source=False
|
|
|
)
|
|
|
@@ -3254,6 +3403,10 @@ def split_content_into_batches(
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"dingtalk", title_data_copy, show_source=False
|
|
|
)
|
|
|
+ elif format_type == "slack":
|
|
|
+ formatted_title = format_title_for_platform(
|
|
|
+ "slack", title_data_copy, show_source=False
|
|
|
+ )
|
|
|
else:
|
|
|
formatted_title = f"{title_data_copy['title']}"
|
|
|
|
|
|
@@ -3299,6 +3452,10 @@ def split_content_into_batches(
|
|
|
formatted_title = format_title_for_platform(
|
|
|
"dingtalk", title_data_copy, show_source=False
|
|
|
)
|
|
|
+ elif format_type == "slack":
|
|
|
+ formatted_title = format_title_for_platform(
|
|
|
+ "slack", title_data_copy, show_source=False
|
|
|
+ )
|
|
|
else:
|
|
|
formatted_title = f"{title_data_copy['title']}"
|
|
|
|
|
|
@@ -3533,14 +3690,20 @@ def send_to_feishu(
|
|
|
proxies = {"http": proxy_url, "https": proxy_url}
|
|
|
|
|
|
# 获取分批内容，使用飞书专用的批次大小
|
|
|
+ feishu_batch_size = CONFIG.get("FEISHU_BATCH_SIZE", 29000)
|
|
|
+ # 预留批次头部空间，避免添加头部后超限
|
|
|
+ header_reserve = _get_max_batch_header_size("feishu")
|
|
|
batches = split_content_into_batches(
|
|
|
report_data,
|
|
|
"feishu",
|
|
|
update_info,
|
|
|
- max_bytes=CONFIG.get("FEISHU_BATCH_SIZE", 29000),
|
|
|
+ max_bytes=feishu_batch_size - header_reserve,
|
|
|
mode=mode,
|
|
|
)
|
|
|
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, "feishu", feishu_batch_size)
|
|
|
+
|
|
|
print(f"飞书消息分为 {len(batches)} 批次发送 [{report_type}]")
|
|
|
|
|
|
# 逐批发送
|
|
|
@@ -3550,18 +3713,6 @@ def send_to_feishu(
|
|
|
f"发送飞书第 {i}/{len(batches)} 批次，大小：{batch_size} 字节 [{report_type}]"
|
|
|
)
|
|
|
|
|
|
- # 添加批次标识
|
|
|
- if len(batches) > 1:
|
|
|
- batch_header = f"**[第 {i}/{len(batches)} 批次]**\n\n"
|
|
|
- # 将批次标识插入到适当位置（在统计标题之后）
|
|
|
- if "📊 **热点词汇统计**" in batch_content:
|
|
|
- batch_content = batch_content.replace(
|
|
|
- "📊 **热点词汇统计**\n\n", f"📊 **热点词汇统计** {batch_header}"
|
|
|
- )
|
|
|
- else:
|
|
|
- # 如果没有统计标题，直接在开头添加
|
|
|
- batch_content = batch_header + batch_content
|
|
|
-
|
|
|
total_titles = sum(
|
|
|
len(stat["titles"]) for stat in report_data["stats"] if stat["count"] > 0
|
|
|
)
|
|
|
@@ -3623,14 +3774,20 @@ def send_to_dingtalk(
|
|
|
proxies = {"http": proxy_url, "https": proxy_url}
|
|
|
|
|
|
# 获取分批内容，使用钉钉专用的批次大小
|
|
|
+ dingtalk_batch_size = CONFIG.get("DINGTALK_BATCH_SIZE", 20000)
|
|
|
+ # 预留批次头部空间，避免添加头部后超限
|
|
|
+ header_reserve = _get_max_batch_header_size("dingtalk")
|
|
|
batches = split_content_into_batches(
|
|
|
report_data,
|
|
|
"dingtalk",
|
|
|
update_info,
|
|
|
- max_bytes=CONFIG.get("DINGTALK_BATCH_SIZE", 20000),
|
|
|
+ max_bytes=dingtalk_batch_size - header_reserve,
|
|
|
mode=mode,
|
|
|
)
|
|
|
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, "dingtalk", dingtalk_batch_size)
|
|
|
+
|
|
|
print(f"钉钉消息分为 {len(batches)} 批次发送 [{report_type}]")
|
|
|
|
|
|
# 逐批发送
|
|
|
@@ -3640,18 +3797,6 @@ def send_to_dingtalk(
|
|
|
f"发送钉钉第 {i}/{len(batches)} 批次，大小：{batch_size} 字节 [{report_type}]"
|
|
|
)
|
|
|
|
|
|
- # 添加批次标识
|
|
|
- if len(batches) > 1:
|
|
|
- batch_header = f"**[第 {i}/{len(batches)} 批次]**\n\n"
|
|
|
- # 将批次标识插入到适当位置（在标题之后）
|
|
|
- if "📊 **热点词汇统计**" in batch_content:
|
|
|
- batch_content = batch_content.replace(
|
|
|
- "📊 **热点词汇统计**\n\n", f"📊 **热点词汇统计** {batch_header}\n\n"
|
|
|
- )
|
|
|
- else:
|
|
|
- # 如果没有统计标题，直接在开头添加
|
|
|
- batch_content = batch_header + batch_content
|
|
|
-
|
|
|
payload = {
|
|
|
"msgtype": "markdown",
|
|
|
"markdown": {
|
|
|
@@ -3756,21 +3901,23 @@ def send_to_wework(
|
|
|
else:
|
|
|
print(f"企业微信使用 markdown 格式（群机器人模式）[{report_type}]")
|
|
|
|
|
|
- # 获取分批内容
|
|
|
- batches = split_content_into_batches(report_data, "wework", update_info, mode=mode)
|
|
|
+ # text 模式使用 wework_text，markdown 模式使用 wework
|
|
|
+ header_format_type = "wework_text" if is_text_mode else "wework"
|
|
|
+
|
|
|
+ # 获取分批内容，预留批次头部空间
|
|
|
+ wework_batch_size = CONFIG.get("MESSAGE_BATCH_SIZE", 4000)
|
|
|
+ header_reserve = _get_max_batch_header_size(header_format_type)
|
|
|
+ batches = split_content_into_batches(
|
|
|
+ report_data, "wework", update_info, max_bytes=wework_batch_size - header_reserve, mode=mode
|
|
|
+ )
|
|
|
+
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, header_format_type, wework_batch_size)
|
|
|
|
|
|
print(f"企业微信消息分为 {len(batches)} 批次发送 [{report_type}]")
|
|
|
|
|
|
# 逐批发送
|
|
|
for i, batch_content in enumerate(batches, 1):
|
|
|
- # 添加批次标识
|
|
|
- if len(batches) > 1:
|
|
|
- if is_text_mode:
|
|
|
- batch_header = f"[第 {i}/{len(batches)} 批次]\n\n"
|
|
|
- else:
|
|
|
- batch_header = f"**[第 {i}/{len(batches)} 批次]**\n\n"
|
|
|
- batch_content = batch_header + batch_content
|
|
|
-
|
|
|
# 根据消息类型构建 payload
|
|
|
if is_text_mode:
|
|
|
# text 格式：去除 markdown 语法
|
|
|
@@ -3832,11 +3979,16 @@ def send_to_telegram(
|
|
|
if proxy_url:
|
|
|
proxies = {"http": proxy_url, "https": proxy_url}
|
|
|
|
|
|
- # 获取分批内容
|
|
|
+ # 获取分批内容，预留批次头部空间
|
|
|
+ telegram_batch_size = CONFIG.get("MESSAGE_BATCH_SIZE", 4000)
|
|
|
+ header_reserve = _get_max_batch_header_size("telegram")
|
|
|
batches = split_content_into_batches(
|
|
|
- report_data, "telegram", update_info, mode=mode
|
|
|
+ report_data, "telegram", update_info, max_bytes=telegram_batch_size - header_reserve, mode=mode
|
|
|
)
|
|
|
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, "telegram", telegram_batch_size)
|
|
|
+
|
|
|
print(f"Telegram消息分为 {len(batches)} 批次发送 [{report_type}]")
|
|
|
|
|
|
# 逐批发送
|
|
|
@@ -3846,11 +3998,6 @@ def send_to_telegram(
|
|
|
f"发送Telegram第 {i}/{len(batches)} 批次，大小：{batch_size} 字节 [{report_type}]"
|
|
|
)
|
|
|
|
|
|
- # 添加批次标识
|
|
|
- if len(batches) > 1:
|
|
|
- batch_header = f"<b>[第 {i}/{len(batches)} 批次]</b>\n\n"
|
|
|
- batch_content = batch_header + batch_content
|
|
|
-
|
|
|
payload = {
|
|
|
"chat_id": chat_id,
|
|
|
"text": batch_content,
|
|
|
@@ -4058,7 +4205,7 @@ def send_to_ntfy(
|
|
|
|
|
|
if token:
|
|
|
headers["Authorization"] = f"Bearer {token}"
|
|
|
-
|
|
|
+
|
|
|
# 构建完整URL，确保格式正确
|
|
|
base_url = server_url.rstrip("/")
|
|
|
if not base_url.startswith(("http://", "https://")):
|
|
|
@@ -4069,18 +4216,23 @@ def send_to_ntfy(
|
|
|
if proxy_url:
|
|
|
proxies = {"http": proxy_url, "https": proxy_url}
|
|
|
|
|
|
- # 获取分批内容，使用ntfy专用的4KB限制
|
|
|
+ # 获取分批内容，使用ntfy专用的4KB限制，预留批次头部空间
|
|
|
+ ntfy_batch_size = 3800
|
|
|
+ header_reserve = _get_max_batch_header_size("ntfy")
|
|
|
batches = split_content_into_batches(
|
|
|
- report_data, "ntfy", update_info, max_bytes=3800, mode=mode
|
|
|
+ report_data, "ntfy", update_info, max_bytes=ntfy_batch_size - header_reserve, mode=mode
|
|
|
)
|
|
|
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, "ntfy", ntfy_batch_size)
|
|
|
+
|
|
|
total_batches = len(batches)
|
|
|
print(f"ntfy消息分为 {total_batches} 批次发送 [{report_type}]")
|
|
|
|
|
|
# 反转批次顺序，使得在ntfy客户端显示时顺序正确
|
|
|
# ntfy显示最新消息在上面，所以我们从最后一批开始推送
|
|
|
reversed_batches = list(reversed(batches))
|
|
|
-
|
|
|
+
|
|
|
print(f"ntfy将按反向顺序推送（最后批次先推送），确保客户端显示顺序正确")
|
|
|
|
|
|
# 逐批发送（反向顺序）
|
|
|
@@ -4088,7 +4240,7 @@ def send_to_ntfy(
|
|
|
for idx, batch_content in enumerate(reversed_batches, 1):
|
|
|
# 计算正确的批次编号（用户视角的编号）
|
|
|
actual_batch_num = total_batches - idx + 1
|
|
|
-
|
|
|
+
|
|
|
batch_size = len(batch_content.encode("utf-8"))
|
|
|
print(
|
|
|
f"发送ntfy第 {actual_batch_num}/{total_batches} 批次（推送顺序: {idx}/{total_batches}），大小：{batch_size} 字节 [{report_type}]"
|
|
|
@@ -4098,11 +4250,9 @@ def send_to_ntfy(
|
|
|
if batch_size > 4096:
|
|
|
print(f"警告：ntfy第 {actual_batch_num} 批次消息过大（{batch_size} 字节），可能被拒绝")
|
|
|
|
|
|
- # 添加批次标识（使用正确的批次编号）
|
|
|
+ # 更新 headers 的批次标识
|
|
|
current_headers = headers.copy()
|
|
|
if total_batches > 1:
|
|
|
- batch_header = f"**[第 {actual_batch_num}/{total_batches} 批次]**\n\n"
|
|
|
- batch_content = batch_header + batch_content
|
|
|
current_headers["Title"] = (
|
|
|
f"{report_type_en} ({actual_batch_num}/{total_batches})"
|
|
|
)
|
|
|
@@ -4185,16 +4335,35 @@ def send_to_bark(
|
|
|
proxy_url: Optional[str] = None,
|
|
|
mode: str = "daily",
|
|
|
) -> bool:
|
|
|
- """发送到Bark（支持分批发送，使用纯文本格式）"""
|
|
|
+ """发送到Bark（支持分批发送，使用 markdown 格式）"""
|
|
|
proxies = None
|
|
|
if proxy_url:
|
|
|
proxies = {"http": proxy_url, "https": proxy_url}
|
|
|
|
|
|
- # 获取分批内容（Bark 限制为 3600 字节以避免 413 错误）
|
|
|
+ # 解析 Bark URL，提取 device_key 和 API 端点
|
|
|
+ # Bark URL 格式: https://api.day.app/device_key 或 https://bark.day.app/device_key
|
|
|
+ from urllib.parse import urlparse
|
|
|
+
|
|
|
+ parsed_url = urlparse(bark_url)
|
|
|
+ device_key = parsed_url.path.strip('/').split('/')[0] if parsed_url.path else None
|
|
|
+
|
|
|
+ if not device_key:
|
|
|
+ print(f"Bark URL 格式错误，无法提取 device_key: {bark_url}")
|
|
|
+ return False
|
|
|
+
|
|
|
+ # 构建正确的 API 端点
|
|
|
+ api_endpoint = f"{parsed_url.scheme}://{parsed_url.netloc}/push"
|
|
|
+
|
|
|
+ # 获取分批内容（Bark 限制为 3600 字节以避免 413 错误），预留批次头部空间
|
|
|
+ bark_batch_size = CONFIG["BARK_BATCH_SIZE"]
|
|
|
+ header_reserve = _get_max_batch_header_size("bark")
|
|
|
batches = split_content_into_batches(
|
|
|
- report_data, "wework", update_info, max_bytes=CONFIG["BARK_BATCH_SIZE"], mode=mode
|
|
|
+ report_data, "bark", update_info, max_bytes=bark_batch_size - header_reserve, mode=mode
|
|
|
)
|
|
|
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, "bark", bark_batch_size)
|
|
|
+
|
|
|
total_batches = len(batches)
|
|
|
print(f"Bark消息分为 {total_batches} 批次发送 [{report_type}]")
|
|
|
|
|
|
@@ -4210,15 +4379,7 @@ def send_to_bark(
|
|
|
# 计算正确的批次编号（用户视角的编号）
|
|
|
actual_batch_num = total_batches - idx + 1
|
|
|
|
|
|
- # 添加批次标识（使用正确的批次编号）
|
|
|
- if total_batches > 1:
|
|
|
- batch_header = f"[第 {actual_batch_num}/{total_batches} 批次]\n\n"
|
|
|
- batch_content = batch_header + batch_content
|
|
|
-
|
|
|
- # 清理 markdown 语法（Bark 不支持 markdown）
|
|
|
- plain_content = strip_markdown(batch_content)
|
|
|
-
|
|
|
- batch_size = len(plain_content.encode("utf-8"))
|
|
|
+ batch_size = len(batch_content.encode("utf-8"))
|
|
|
print(
|
|
|
f"发送Bark第 {actual_batch_num}/{total_batches} 批次（推送顺序: {idx}/{total_batches}），大小：{batch_size} 字节 [{report_type}]"
|
|
|
)
|
|
|
@@ -4232,14 +4393,16 @@ def send_to_bark(
|
|
|
# 构建JSON payload
|
|
|
payload = {
|
|
|
"title": report_type,
|
|
|
- "body": plain_content,
|
|
|
+ "markdown": batch_content,
|
|
|
+ "device_key": device_key,
|
|
|
"sound": "default",
|
|
|
"group": "TrendRadar",
|
|
|
+ "action": "none", # 点击推送跳到 APP 不弹出弹框,方便阅读
|
|
|
}
|
|
|
|
|
|
try:
|
|
|
response = requests.post(
|
|
|
- bark_url,
|
|
|
+ api_endpoint,
|
|
|
json=payload,
|
|
|
proxies=proxies,
|
|
|
timeout=30,
|
|
|
@@ -4319,20 +4482,20 @@ def send_to_slack(
|
|
|
if proxy_url:
|
|
|
proxies = {"http": proxy_url, "https": proxy_url}
|
|
|
|
|
|
- # 获取分批内容（使用 Slack 批次大小）
|
|
|
+ # 获取分批内容（使用 Slack 批次大小），预留批次头部空间
|
|
|
+ slack_batch_size = CONFIG["SLACK_BATCH_SIZE"]
|
|
|
+ header_reserve = _get_max_batch_header_size("slack")
|
|
|
batches = split_content_into_batches(
|
|
|
- report_data, "wework", update_info, max_bytes=CONFIG["SLACK_BATCH_SIZE"], mode=mode
|
|
|
+ report_data, "slack", update_info, max_bytes=slack_batch_size - header_reserve, mode=mode
|
|
|
)
|
|
|
|
|
|
+ # 统一添加批次头部（已预留空间，不会超限）
|
|
|
+ batches = add_batch_headers(batches, "slack", slack_batch_size)
|
|
|
+
|
|
|
print(f"Slack消息分为 {len(batches)} 批次发送 [{report_type}]")
|
|
|
|
|
|
# 逐批发送
|
|
|
for i, batch_content in enumerate(batches, 1):
|
|
|
- # 添加批次标识
|
|
|
- if len(batches) > 1:
|
|
|
- batch_header = f"*[第 {i}/{len(batches)} 批次]*\n\n"
|
|
|
- batch_content = batch_header + batch_content
|
|
|
-
|
|
|
# 转换 Markdown 到 mrkdwn 格式
|
|
|
mrkdwn_content = convert_markdown_to_mrkdwn(batch_content)
|
|
|
|
sansan