Update langchain version to 0.1.5 and fix _split_sources() method in …

…BaseQAWithSourcesChain
skypointcloud · Jan 4, 2024 · 8a3f835 · 8a3f835
1 parent 02efd94
commit 8a3f835
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 7 deletions.
diff --git a/libs/langchain/langchain/chains/qa_with_sources/base.py b/libs/langchain/langchain/chains/qa_with_sources/base.py
@@ -119,17 +119,27 @@ def validate_naming(cls, values: Dict) -> Dict:
             values["combine_documents_chain"] = values.pop("combine_document_chain")
         return values
 
-    def _split_sources(self, answer: str) -> Tuple[str, str]:
+    def _split_sources(self, raw_answer: str) -> Tuple[str, str]:
         """Split sources from answer."""
-        if re.search(r"SOURCES?:", answer, re.IGNORECASE):
-            answer, sources = re.split(
-                r"SOURCES?:|QUESTION:\s", answer, flags=re.IGNORECASE
-            )[:2]
-            sources = re.split(r"\n", sources)[0].strip()
+        if re.search(r"SOURCES?:", raw_answer, re.IGNORECASE):
+            try:
+                answer, raw_sources = re.split(
+                    r"SOURCES?:|QUESTION:\s", raw_answer, flags=re.IGNORECASE
+                )[:2]
+                sources = re.split(r"\n", raw_sources)[0].strip()
+                if sources == "":
+                    regex = r"- \s*(.+\.pdf)"
+                    sources_list = re.findall(regex, raw_sources)
+                    sources = ', '.join(sources_list)
+            except Exception as e:
+                print(f"An Exception has occured for answer : {raw_answer}",e)
+                sources = ""
         else:
             sources = ""
+
         return answer, sources
 
+
     @abstractmethod
     def _get_docs(
         self,

diff --git a/libs/langchain/pyproject.toml b/libs/langchain/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "langchain"
-version = "0.1.3"
+version = "0.1.5"
 description = "Building applications with LLMs through composability"
 authors = []
 license = "MIT"