From 5f34b4561e43e9bf6362832cfde93d4e2d79573c Mon Sep 17 00:00:00 2001
From: xming521 <1223398803@qq.com>
Date: Wed, 9 Jul 2025 23:15:51 +0800
Subject: [PATCH] feat(dataset): adds option to include time

Adds a configuration option to include the current
datetime in the system prompt of the generated dataset.
This allows models to be aware of the temporal context
of the conversations.
---
 settings.template.jsonc        | 1 +
 weclone/data/qa_generator.py   | 6 +++++-
 weclone/utils/config_models.py | 1 +
 3 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/settings.template.jsonc b/settings.template.jsonc
index b465208..e6736f6 100644
--- a/settings.template.jsonc
+++ b/settings.template.jsonc
@@ -29,6 +29,7 @@
             "例如 密码",
             "//....."
         ],
+        "add_time": false,
         "single_combine_strategy": "time_window", // 单人组成单句策略
         "qa_match_strategy": "time_window", // 组成qa策略
         "single_combine_time_window": 2, // 单人组成单句时间窗口（分钟）,
diff --git a/weclone/data/qa_generator.py b/weclone/data/qa_generator.py
index a7c69d2..0a3c16c 100644
--- a/weclone/data/qa_generator.py
+++ b/weclone/data/qa_generator.py
@@ -278,13 +278,17 @@ class DataProcessor:
                     )
                     return qa_id
 
+                system_content = self.system_prompt
+                if self.c.add_time:
+                    system_content += f" Current datetime: {time_stamp.strftime('%Y-%m-%d %H:%M:%S')}"
+
                 qa_pair = self.QaPair(
                     id=qa_id,
                     time=time_stamp,
                     score=0,
                     messages=current_conversation_messages.copy(),
                     images=current_conversation_images.copy(),
-                    system=self.system_prompt,
+                    system=system_content,
                 )
                 qa_res.append(qa_pair)
                 return qa_id + 1
diff --git a/weclone/utils/config_models.py b/weclone/utils/config_models.py
index dfa8060..e4601c4 100644
--- a/weclone/utils/config_models.py
+++ b/weclone/utils/config_models.py
@@ -134,6 +134,7 @@ class MakeDatasetArgs(BaseConfigModel):
     include_type: List[DataModality] = Field([DataModality.TEXT], description="Types of data to include")
     max_image_num: int = Field(2, description="Maximum number of images per single data entry")
     blocked_words: List[str] = Field([], description="List of blocked words")
+    add_time: bool = Field(False, description="Whether to add time to the dataset")
     single_combine_strategy: CombineStrategy = Field(
         CombineStrategy.TIME_WINDOW,
         description="Strategy for combining single person's messages into a single sentence",