chore: 更改本地文件参考路径

deepsearcher/agent/deep_search.py

@@ -6,6 +6,9 @@ from deepsearcher.utils.message_stream import send_info, send_answer
 from deepsearcher.vector_db import RetrievalResult
 from deepsearcher.vector_db.base import BaseVectorDB, deduplicate
 from collections import defaultdict
+import os
+import urllib.parse
+
 
 
 COLLECTION_ROUTE_PROMPT = """
@@ -22,7 +25,7 @@ COLLECTION_ROUTE_PROMPT = """
 
 
 SUB_QUERY_PROMPT = """
-为了能够全面的回答这个问题，请你尝试把原本的问题拆分或扩展为几个子问题
+为了能够全面的回答这个问题，请你尝试把原本的问题拆分或扩展为几个子问题（三五个即可）
 请你使用自顶向下和自底向上两种方向来思考如何拆分问题
 子问题的数量不可以太多，但是也不可以太少，应当保证问题的回答全面性，请根据问题复杂程度来决定子问题的数量
 如果原问题本身非常简单，没有必要进行拆分，则保留输出原问题本身
@@ -95,6 +98,7 @@ SUMMARY_PROMPT = """
 来自<chunk><reference>的引用序号从[^index]从index=1开始，来源需要与前文<reference>中的"id"一致。
 不需要对每个<chunk>分配一个引用，而是相同<reference>的<chunk>共用引用，并确保每一个<reference>都被引用。
 另外，如果回答的内容文内引用需要引用多个<reference>，请添加多个[^index]到句尾。
+注意，不需要你生成文末脚注，即不需要参考列表，只需要在回答中添加文内引用即可。
 
 <EXAMPLE>
 
@@ -431,7 +435,7 @@ class DeepSearch(BaseAgent):
             print(formated_chunk)
             formated_chunks.append(formated_chunk)
             chunk_count += len(ref_dict[reference])
-            formated_refs.append(f"[{i + 1}]: " + str(reference) + "\n")
+            formated_refs.append(f"[^{i + 1}]: "  + str(urllib.parse.quote_plus('file://' + str(os.path.abspath(reference)))) + "\n")
         formated_chunks = "".join(formated_chunks)
         formated_refs = "".join(formated_refs)
         return formated_chunks, formated_refs

pyproject.toml

@@ -11,6 +11,7 @@ dependencies = [
     "langchain-text-splitters>=0.3.8",
     "numpy>=1.26.4",
     "openai>=1.77.0",
+    "pathlib>=1.0.1",
     "pdfplumber>=0.11.6",
     "pymilvus>=2.5.8",
     "requests>=2.32.3",

uv.lock

@@ -880,6 +880,7 @@ dependencies = [
     { name = "langchain-text-splitters" },
     { name = "numpy" },
     { name = "openai" },
+    { name = "pathlib" },
     { name = "pdfplumber" },
     { name = "pymilvus" },
     { name = "requests" },
@@ -1001,6 +1002,7 @@ requires-dist = [
     { name = "openai", specifier = ">=1.77.0" },
     { name = "oracledb", marker = "extra == 'all'", specifier = ">=3.1.0" },
     { name = "oracledb", marker = "extra == 'oracledb'", specifier = ">=3.1.0" },
+    { name = "pathlib", specifier = ">=1.0.1" },
     { name = "pdfplumber", specifier = ">=0.11.6" },
     { name = "pymilvus", specifier = ">=2.5.8" },
     { name = "qdrant-client", marker = "extra == 'all'", specifier = ">=1.14.2" },
@@ -4045,6 +4047,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/c6/ac/dac4a63f978e4dcb3c6d3a78c4d8e0192a113d288502a1216950c41b1027/parso-0.8.4-py2.py3-none-any.whl", hash = "sha256:a418670a20291dacd2dddc80c377c5c3791378ee1e8d12bffc35420643d43f18", size = 103650, upload-time = "2024-04-05T09:43:53.299Z" },
 ]
 
+[[package]]
+name = "pathlib"
+version = "1.0.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ac/aa/9b065a76b9af472437a0059f77e8f962fe350438b927cb80184c32f075eb/pathlib-1.0.1.tar.gz", hash = "sha256:6940718dfc3eff4258203ad5021090933e5c04707d5ca8cc9e73c94a7894ea9f", size = 49298, upload-time = "2014-09-03T15:41:57.18Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/78/f9/690a8600b93c332de3ab4a344a4ac34f00c8f104917061f779db6a918ed6/pathlib-1.0.1-py3-none-any.whl", hash = "sha256:f35f95ab8b0f59e6d354090350b44a80a80635d22efdedfa84c7ad1cf0a74147", size = 14363, upload-time = "2022-05-04T13:37:20.585Z" },
+]
+
 [[package]]
 name = "pathspec"
 version = "0.12.1"