fix: correct Zhipu API thinking parameter and add non-interactive archive mode #37

Merged
starmen1265 merged 1 commits from fix/crawler-proxy-config into main 2026-03-20 00:51:28 +08:00
4 changed files with 34 additions and 7 deletions
+3
View File
@@ -12,3 +12,6 @@ export KB_BASE_PATH="~/work/knowledge-base"
# 可选: Git 配置 # 可选: Git 配置
# export KB_GIT_USER="Your Name" # export KB_GIT_USER="Your Name"
# export KB_GIT_EMAIL="your@email.com" # export KB_GIT_EMAIL="your@email.com"
# 可选: 工作流行为配置
# export KB_OVERWRITE_ARCHIVES="false" # 非交互模式下是否覆盖已存在的归档文件(true/false)
+9 -5
View File
@@ -90,15 +90,19 @@ class SimpleAIClient:
def _call_zhipu(self, prompt: str, max_tokens: int, api_key: str, model: str, base_url: str) -> str: def _call_zhipu(self, prompt: str, max_tokens: int, api_key: str, model: str, base_url: str) -> str:
import requests import requests
import sys
base_url = base_url or "https://open.bigmodel.cn/api/coding/paas/v4" base_url = base_url or "https://open.bigmodel.cn/api/coding/paas/v4"
url = f"{base_url}/chat/completions" url = f"{base_url}/chat/completions"
headers = {"Authorization": f"Bearer {api_key}"} headers = {"Authorization": f"Bearer {api_key}"}
# GLM-4.5/4.6/4.7/5 默认启用thinking mode,添加 disable_reasoning 参数获取正常响应 # GLM-4.5/4.6/4.7/5 默认启用thinking mode,使用 thinking.type 参数禁用
payload = { payload = {
"model": model, "model": model,
"messages": [{"role": "user", "content": prompt}], "messages": [{"role": "user", "content": prompt}],
"max_tokens": max_tokens, "max_tokens": max_tokens,
"disable_reasoning": True # 禁用思考模式,获取直接响应 "thinking": {
"type": "disabled" # 禁用思考模式,获取直接响应
}
} }
response = requests.post(url, headers=headers, json=payload, timeout=600) response = requests.post(url, headers=headers, json=payload, timeout=600)
response.raise_for_status() response.raise_for_status()
@@ -110,9 +114,9 @@ class SimpleAIClient:
# 如果仍然为空,检查是否有reasoning_content(用于调试) # 如果仍然为空,检查是否有reasoning_content(用于调试)
reasoning = message.get('reasoning_content', '') reasoning = message.get('reasoning_content', '')
if reasoning: if reasoning:
print(f"[DEBUG] Zhipu API返回空的content,但有reasoning_content。") print(f"[DEBUG] Zhipu API返回空的content,但有reasoning_content。", file=sys.stderr)
print(f"[DEBUG] 这可能表示disable_reasoning参数未生效或模型不支持。") print(f"[DEBUG] 这可能表示thinking.type参数未生效或模型不支持。", file=sys.stderr)
print(f"[DEBUG] reasoning_content前200字符: {reasoning[:200]}") print(f"[DEBUG] reasoning_content前200字符: {reasoning[:200]}", file=sys.stderr)
raise ValueError(f"Zhipu API返回空内容。请检查API配置和模型参数。响应数据: {data}") raise ValueError(f"Zhipu API返回空内容。请检查API配置和模型参数。响应数据: {data}")
return content return content
+5
View File
@@ -58,6 +58,11 @@ class KBSettings(BaseSettings):
default=SecretStr(""), description="B站 Cookie (for video download)" default=SecretStr(""), description="B站 Cookie (for video download)"
) )
# 工作流行为配置
overwrite_archives: bool = Field(
default=False, description="覆盖已存在的归档文件(非交互模式)"
)
@lru_cache @lru_cache
def get_settings() -> KBSettings: def get_settings() -> KBSettings:
+17 -2
View File
@@ -46,6 +46,8 @@ class ArchiveProcessor(BaseProcessor):
Archive 处理结果 Archive 处理结果
""" """
try: try:
import sys
# Step 1: 检查是否已完成(幂等执行) # Step 1: 检查是否已完成(幂等执行)
if self.state_manager: if self.state_manager:
state = self.state_manager.load() state = self.state_manager.load()
@@ -73,7 +75,9 @@ class ArchiveProcessor(BaseProcessor):
platform_dir, input_data.content_id platform_dir, input_data.content_id
) )
if existing_file: if existing_file:
should_overwrite = self._ask_overwrite(existing_file) # 检测是否为非交互式环境
is_non_interactive = not sys.stdin.isatty()
should_overwrite = self._ask_overwrite(existing_file, is_non_interactive)
if not should_overwrite: if not should_overwrite:
logger.info(f"User chose not to overwrite, skipping archive") logger.info(f"User chose not to overwrite, skipping archive")
return ArchiveOutput( return ArchiveOutput(
@@ -143,17 +147,28 @@ class ArchiveProcessor(BaseProcessor):
matches = list(platform_dir.glob(pattern)) matches = list(platform_dir.glob(pattern))
return matches[0] if matches else None return matches[0] if matches else None
def _ask_overwrite(self, existing_file: Path) -> bool: def _ask_overwrite(self, existing_file: Path, is_non_interactive: bool = False) -> bool:
"""询问用户是否覆盖已存在的文件 """询问用户是否覆盖已存在的文件
使用 rich.prompt.Confirm 模式(与 Pipeline._ask_rerun_step 相同) 使用 rich.prompt.Confirm 模式(与 Pipeline._ask_rerun_step 相同)
Args: Args:
existing_file: 已存在的文件路径 existing_file: 已存在的文件路径
is_non_interactive: 是否为非交互式环境(stdin不是tty)
Returns: Returns:
True 如果用户选择覆盖 True 如果用户选择覆盖
""" """
# 非交互式模式:使用配置决定
if is_non_interactive:
if self.settings.overwrite_archives:
logger.info(f"Non-interactive mode: Overwriting existing archive (KB_OVERWRITE_ARCHIVES=true): {existing_file.name}")
return True
else:
logger.info(f"Non-interactive mode: Skipping existing archive (KB_OVERWRITE_ARCHIVES=false): {existing_file.name}")
return False
# 交互式模式:询问用户
try: try:
from rich.console import Console from rich.console import Console
from rich.prompt import Confirm from rich.prompt import Confirm