Skip to content

RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.

docs: document the tenant_id parameter in resume parsing (#18169)

iridescentWena0cc2be2dc4d
diff --git a/rag/app/resume.py b/rag/app/resume.py
index 8475e172c5..e8ad271a6c 100644
--- a/rag/app/resume.py
+++ b/rag/app/resume.py
@@ -1045,6 +1045,7 @@ def _call_llm(prompt: str, tenant_id, lang: str) -> Optional[dict]:
Args:
prompt: User prompt
+ tenant_id: Tenant ID used to build the LLMBundle that serves the request
lang: Language
Returns:
Parsed dictionary, or None on failure
@@ -1295,6 +1296,7 @@ def parse_with_llm(indexed_text: str, lines: list[str], tenant_id, lang: str) ->
Args:
indexed_text: Line-indexed resume text
lines: List of original line texts (for index-based extraction)
+ tenant_id: Tenant ID used to resolve the chat model for each subtask
lang: Language
Returns:
Merged structured resume dictionary, or None on failure
@@ -2072,6 +2074,7 @@ def parse_resume(filename: str, binary: bytes, tenant_id, lang: str = "Chinese")
Args:
filename: File name
binary: File binary content
+ tenant_id: Tenant ID forwarded to the LLM extraction stage
lang: Language, default "Chinese"
Returns:
(resume, lines, line_positions) tuple: