pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-23 15:06:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix: Merge main branch (#10377)

Browse Source 
### What problem does this PR solve?


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

---------

Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: jinhai <haijin.chn@gmail.com>
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
Co-authored-by: Lynn <lynn_inf@hotmail.com>
Co-authored-by: chanx <1243304602@qq.com>
Co-authored-by: balibabu <cike8899@users.noreply.github.com>
Co-authored-by: 纷繁下的无奈 <zhileihuang@126.com>
Co-authored-by: huangzl <huangzl@shinemo.com>
Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com>
Co-authored-by: Wilmer <33392318@qq.com>
Co-authored-by: Adrian Weidig <adrianweidig@gmx.net>
Co-authored-by: Zhichang Yu <yuzhichang@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Yongteng Lei <yongtengrey@outlook.com>
Co-authored-by: Liu An <asiro@qq.com>
Co-authored-by: buua436 <66937541+buua436@users.noreply.github.com>
Co-authored-by: BadwomanCraZY <511528396@qq.com>
Co-authored-by: cucusenok <31804608+cucusenok@users.noreply.github.com>
Co-authored-by: Russell Valentine <russ@coldstonelabs.org>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Billy Bao <newyorkupperbay@gmail.com>
Co-authored-by: Zhedong Cen <cenzhedong2@126.com>
Co-authored-by: TensorNull <129579691+TensorNull@users.noreply.github.com>
Co-authored-by: TensorNull <tensor.null@gmail.com>
Co-authored-by: Ajay <160579663+aybanda@users.noreply.github.com>
Co-authored-by: AB <aj@Ajays-MacBook-Air.local>
Co-authored-by: 天海蒼灆 <huangaoqin@tecpie.com>
Co-authored-by: He Wang <wanghechn@qq.com>
Co-authored-by: Atsushi Hatakeyama <atu729@icloud.com>
Co-authored-by: Jin Hai <haijin.chn@gmail.com>
Co-authored-by: Mohamed Mathari <155896313+melmathari@users.noreply.github.com>
Co-authored-by: Mohamed Mathari <nocodeventure@Mac-mini-van-Mohamed.fritz.box>
Co-authored-by: Stephen Hu <stephenhu@seismic.com>
Co-authored-by: Shaun Zhang <zhangwfjh@users.noreply.github.com>
Co-authored-by: zhimeng123 <60221886+zhimeng123@users.noreply.github.com>
Co-authored-by: mxc <mxc@example.com>
Co-authored-by: Dominik Novotný <50611433+SgtMarmite@users.noreply.github.com>
Co-authored-by: EVGENY M <168018528+rjohny55@users.noreply.github.com>
Co-authored-by: mcoder6425 <mcoder64@gmail.com>
Co-authored-by: TeslaZY <TeslaZY@outlook.com>
Co-authored-by: lemsn <lemsn@msn.com>
Co-authored-by: lemsn <lemsn@126.com>
Co-authored-by: Adrian Gora <47756404+adagora@users.noreply.github.com>
Co-authored-by: Womsxd <45663319+Womsxd@users.noreply.github.com>
Co-authored-by: FatMii <39074672+FatMii@users.noreply.github.com>
This commit is contained in:
Kevin Hu
2025-09-30 13:13:15 +08:00
committed by GitHub
parent 4d6ff672eb
commit 20b577a72c
201 changed files with 7929 additions and 1110 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
agent/tools/arxiv.py
View File

@ -61,7 +61,7 @@ class ArXivParam(ToolParamBase):
class ArXiv(ToolBase, ABC):
component_name = "ArXiv"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -97,6 +97,6 @@ class ArXiv(ToolBase, ABC):
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Looking for the most relevant articles.
""".format(self.get_input().get("query", "-_-!"))
""".format(self.get_input().get("query", "-_-!"))

2
agent/tools/base.py
View File

@ -22,7 +22,7 @@ from typing import TypedDict, List, Any
from agent.component.base import ComponentParamBase, ComponentBase
from api.utils import hash_str2int
from rag.llm.chat_model import ToolCallSession
from rag.prompts.prompts import kb_prompt
from rag.prompts.generator import kb_prompt
from rag.utils.mcp_tool_call_conn import MCPToolCallSession
from timeit import default_timer as timer

4
agent/tools/code_exec.py
View File

@ -129,7 +129,7 @@ module.exports = { main };
class CodeExec(ToolBase, ABC):
component_name = "CodeExec"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60)))
def _invoke(self, **kwargs):
lang = kwargs.get("lang", self._param.lang)
script = kwargs.get("script", self._param.script)
@ -157,7 +157,7 @@ class CodeExec(ToolBase, ABC):
try:
resp = requests.post(url=f"http://{settings.SANDBOX_HOST}:9385/run", json=code_req, timeout=os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60))
logging.info(f"http://{settings.SANDBOX_HOST}:9385/run, code_req: {code_req}, resp.status_code {resp.status_code}:")
logging.info(f"http://{settings.SANDBOX_HOST}:9385/run", code_req, resp.status_code)
if resp.status_code != 200:
resp.raise_for_status()
body = resp.json()

6
agent/tools/duckduckgo.py
View File

@ -73,7 +73,7 @@ class DuckDuckGoParam(ToolParamBase):
class DuckDuckGo(ToolBase, ABC):
component_name = "DuckDuckGo"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -115,6 +115,6 @@ class DuckDuckGo(ToolBase, ABC):
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Looking for the most relevant articles.
""".format(self.get_input().get("query", "-_-!"))
""".format(self.get_input().get("query", "-_-!"))

6
agent/tools/email.py
View File

@ -98,8 +98,8 @@ class EmailParam(ToolParamBase):
class Email(ToolBase, ABC):
component_name = "Email"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60)))
def _invoke(self, **kwargs):
if not kwargs.get("to_email"):
self.set_output("success", False)
@ -212,4 +212,4 @@ class Email(ToolBase, ABC):
To: {}
Subject: {}
Your email is on its way—sit tight!
""".format(inputs.get("to_email", "-_-!"), inputs.get("subject", "-_-!"))
""".format(inputs.get("to_email", "-_-!"), inputs.get("subject", "-_-!"))

55
agent/tools/exesql.py
View File

@ -53,7 +53,7 @@ class ExeSQLParam(ToolParamBase):
self.max_records = 1024
def check(self):
self.check_valid_value(self.db_type, "Choose DB type", ['mysql', 'postgres', 'mariadb', 'mssql'])
self.check_valid_value(self.db_type, "Choose DB type", ['mysql', 'postgres', 'mariadb', 'mssql', 'IBM DB2'])
self.check_empty(self.database, "Database name")
self.check_empty(self.username, "database username")
self.check_empty(self.host, "IP Address")
@ -78,7 +78,7 @@ class ExeSQLParam(ToolParamBase):
class ExeSQL(ToolBase, ABC):
component_name = "ExeSQL"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60)))
def _invoke(self, **kwargs):
def convert_decimals(obj):
@ -123,6 +123,55 @@ class ExeSQL(ToolBase, ABC):
r'PWD=' + self._param.password
)
db = pyodbc.connect(conn_str)
elif self._param.db_type == 'IBM DB2':
import ibm_db
conn_str = (
f"DATABASE={self._param.database};"
f"HOSTNAME={self._param.host};"
f"PORT={self._param.port};"
f"PROTOCOL=TCPIP;"
f"UID={self._param.username};"
f"PWD={self._param.password};"
)
try:
conn = ibm_db.connect(conn_str, "", "")
except Exception as e:
raise Exception("Database Connection Failed! \n" + str(e))
sql_res = []
formalized_content = []
for single_sql in sqls:
single_sql = single_sql.replace("```", "").strip()
if not single_sql:
continue
single_sql = re.sub(r"\[ID:[0-9]+\]", "", single_sql)
stmt = ibm_db.exec_immediate(conn, single_sql)
rows = []
row = ibm_db.fetch_assoc(stmt)
while row and len(rows) < self._param.max_records:
rows.append(row)
row = ibm_db.fetch_assoc(stmt)
if not rows:
sql_res.append({"content": "No record in the database!"})
continue
df = pd.DataFrame(rows)
for col in df.columns:
if pd.api.types.is_datetime64_any_dtype(df[col]):
df[col] = df[col].dt.strftime("%Y-%m-%d")
df = df.where(pd.notnull(df), None)
sql_res.append(convert_decimals(df.to_dict(orient="records")))
formalized_content.append(df.to_markdown(index=False, floatfmt=".6f"))
ibm_db.close(conn)
self.set_output("json", sql_res)
self.set_output("formalized_content", "\n\n".join(formalized_content))
return self.output("formalized_content")
try:
cursor = db.cursor()
except Exception as e:
@ -150,6 +199,8 @@ class ExeSQL(ToolBase, ABC):
if pd.api.types.is_datetime64_any_dtype(single_res[col]):
single_res[col] = single_res[col].dt.strftime('%Y-%m-%d')
single_res = single_res.where(pd.notnull(single_res), None)
sql_res.append(convert_decimals(single_res.to_dict(orient='records')))
formalized_content.append(single_res.to_markdown(index=False, floatfmt=".6f"))

4
agent/tools/github.py
View File

@ -57,7 +57,7 @@ class GitHubParam(ToolParamBase):
class GitHub(ToolBase, ABC):
component_name = "GitHub"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -88,4 +88,4 @@ class GitHub(ToolBase, ABC):
assert False, self.output()
def thoughts(self) -> str:
return "Scanning GitHub repos related to `{}`.".format(self.get_input().get("query", "-_-!"))
return "Scanning GitHub repos related to `{}`.".format(self.get_input().get("query", "-_-!"))

6
agent/tools/google.py
View File

@ -116,7 +116,7 @@ class GoogleParam(ToolParamBase):
class Google(ToolBase, ABC):
component_name = "Google"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("q"):
self.set_output("formalized_content", "")
@ -154,6 +154,6 @@ class Google(ToolBase, ABC):
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Looking for the most relevant articles.
""".format(self.get_input().get("query", "-_-!"))
""".format(self.get_input().get("query", "-_-!"))

4
agent/tools/googlescholar.py
View File

@ -63,7 +63,7 @@ class GoogleScholarParam(ToolParamBase):
class GoogleScholar(ToolBase, ABC):
component_name = "GoogleScholar"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -93,4 +93,4 @@ class GoogleScholar(ToolBase, ABC):
assert False, self.output()
def thoughts(self) -> str:
return "Looking for scholarly papers on `{}`,” prioritising reputable sources.".format(self.get_input().get("query", "-_-!"))
return "Looking for scholarly papers on `{}`,” prioritising reputable sources.".format(self.get_input().get("query", "-_-!"))

6
agent/tools/pubmed.py
View File

@ -33,7 +33,7 @@ class PubMedParam(ToolParamBase):
self.meta:ToolMeta = {
"name": "pubmed_search",
"description": """
PubMed is an openly accessible, free database which includes primarily the MEDLINE database of references and abstracts on life sciences and biomedical topics.
PubMed is an openly accessible, free database which includes primarily the MEDLINE database of references and abstracts on life sciences and biomedical topics.
In addition to MEDLINE, PubMed provides access to:
- older references from the print version of Index Medicus, back to 1951 and earlier
- references to some journals before they were indexed in Index Medicus and MEDLINE, for instance Science, BMJ, and Annals of Surgery
@ -69,7 +69,7 @@ In addition to MEDLINE, PubMed provides access to:
class PubMed(ToolBase, ABC):
component_name = "PubMed"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -105,4 +105,4 @@ class PubMed(ToolBase, ABC):
assert False, self.output()
def thoughts(self) -> str:
return "Looking for scholarly papers on `{}`,” prioritising reputable sources.".format(self.get_input().get("query", "-_-!"))
return "Looking for scholarly papers on `{}`,” prioritising reputable sources.".format(self.get_input().get("query", "-_-!"))

16
agent/tools/retrieval.py
View File

@ -23,8 +23,7 @@ from api.db.services.llm_service import LLMBundle
from api import settings
from api.utils.api_utils import timeout
from rag.app.tag import label_question
from rag.prompts import kb_prompt
from rag.prompts.prompts import cross_languages
from rag.prompts.generator import cross_languages, kb_prompt
class RetrievalParam(ToolParamBase):
@ -75,7 +74,7 @@ class RetrievalParam(ToolParamBase):
class Retrieval(ToolBase, ABC):
component_name = "Retrieval"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", self._param.empty_response)
@ -163,13 +162,20 @@ class Retrieval(ToolBase, ABC):
self.set_output("formalized_content", self._param.empty_response)
return
# Format the chunks for JSON output (similar to how other tools do it)
json_output = kbinfos["chunks"].copy()
self._canvas.add_reference(kbinfos["chunks"], kbinfos["doc_aggs"])
form_cnt = "\n".join(kb_prompt(kbinfos, 200000, True))
# Set both formalized content and JSON output
self.set_output("formalized_content", form_cnt)
self.set_output("json", json_output)
return form_cnt
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Looking for the most relevant articles.
""".format(self.get_input().get("query", "-_-!"))
""".format(self.get_input().get("query", "-_-!"))

21
agent/tools/searxng.py
View File

@ -77,7 +77,7 @@ class SearXNGParam(ToolParamBase):
class SearXNG(ToolBase, ABC):
component_name = "SearXNG"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
# Gracefully handle try-run without inputs
query = kwargs.get("query")
@ -94,7 +94,6 @@ class SearXNG(ToolBase, ABC):
last_e = ""
for _ in range(self._param.max_retries+1):
try:
# 构建搜索参数
search_params = {
'q': query,
'format': 'json',
@ -104,33 +103,29 @@ class SearXNG(ToolBase, ABC):
'pageno': 1
}
# 发送搜索请求
response = requests.get(
f"{searxng_url}/search",
params=search_params,
timeout=10
)
response.raise_for_status()
data = response.json()
# 验证响应数据
if not data or not isinstance(data, dict):
raise ValueError("Invalid response from SearXNG")
results = data.get("results", [])
if not isinstance(results, list):
raise ValueError("Invalid results format from SearXNG")
# 限制结果数量
results = results[:self._param.top_n]
# 处理搜索结果
self._retrieve_chunks(results,
get_title=lambda r: r.get("title", ""),
get_url=lambda r: r.get("url", ""),
get_content=lambda r: r.get("content", ""))
self.set_output("json", results)
return self.output("formalized_content")
@ -151,6 +146,6 @@ class SearXNG(ToolBase, ABC):
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Searching with SearXNG for relevant results...
""".format(self.get_input().get("query", "-_-!"))

10
agent/tools/tavily.py
View File

@ -31,7 +31,7 @@ class TavilySearchParam(ToolParamBase):
self.meta:ToolMeta = {
"name": "tavily_search",
"description": """
Tavily is a search engine optimized for LLMs, aimed at efficient, quick and persistent search results.
Tavily is a search engine optimized for LLMs, aimed at efficient, quick and persistent search results.
When searching:
- Start with specific query which should focus on just a single aspect.
- Number of keywords in query should be less than 5.
@ -101,7 +101,7 @@ When searching:
class TavilySearch(ToolBase, ABC):
component_name = "TavilySearch"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -136,7 +136,7 @@ class TavilySearch(ToolBase, ABC):
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Looking for the most relevant articles.
""".format(self.get_input().get("query", "-_-!"))
@ -199,7 +199,7 @@ class TavilyExtractParam(ToolParamBase):
class TavilyExtract(ToolBase, ABC):
component_name = "TavilyExtract"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 10*60)))
def _invoke(self, **kwargs):
self.tavily_client = TavilyClient(api_key=self._param.api_key)
last_e = None
@ -224,4 +224,4 @@ class TavilyExtract(ToolBase, ABC):
assert False, self.output()
def thoughts(self) -> str:
return "Opened {}—pulling out the main text…".format(self.get_input().get("urls", "-_-!"))
return "Opened {}—pulling out the main text…".format(self.get_input().get("urls", "-_-!"))

4
agent/tools/wencai.py
View File

@ -68,7 +68,7 @@ fund selection platform: through AI technology, is committed to providing excell
class WenCai(ToolBase, ABC):
component_name = "WenCai"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 12)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("report", "")
@ -111,4 +111,4 @@ class WenCai(ToolBase, ABC):
assert False, self.output()
def thoughts(self) -> str:
return "Pulling live financial data for `{}`.".format(self.get_input().get("query", "-_-!"))
return "Pulling live financial data for `{}`.".format(self.get_input().get("query", "-_-!"))

6
agent/tools/wikipedia.py
View File

@ -64,7 +64,7 @@ class WikipediaParam(ToolParamBase):
class Wikipedia(ToolBase, ABC):
component_name = "Wikipedia"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60)))
def _invoke(self, **kwargs):
if not kwargs.get("query"):
self.set_output("formalized_content", "")
@ -99,6 +99,6 @@ class Wikipedia(ToolBase, ABC):
def thoughts(self) -> str:
return """
Keywords: {}
Keywords: {}
Looking for the most relevant articles.
""".format(self.get_input().get("query", "-_-!"))
""".format(self.get_input().get("query", "-_-!"))

4
agent/tools/yahoofinance.py
View File

@ -72,7 +72,7 @@ class YahooFinanceParam(ToolParamBase):
class YahooFinance(ToolBase, ABC):
component_name = "YahooFinance"
@timeout(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60))
@timeout(int(os.environ.get("COMPONENT_EXEC_TIMEOUT", 60)))
def _invoke(self, **kwargs):
if not kwargs.get("stock_code"):
self.set_output("report", "")
@ -111,4 +111,4 @@ class YahooFinance(ToolBase, ABC):
assert False, self.output()
def thoughts(self) -> str:
return "Pulling live financial data for `{}`.".format(self.get_input().get("stock_code", "-_-!"))
return "Pulling live financial data for `{}`.".format(self.get_input().get("stock_code", "-_-!"))