github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-26 16:43:35 +00:00
Code Issues Packages Projects Releases Wiki Activity

Allow passing in encoding to csv_loader (#1836)

Browse Source
This commit is contained in:
Daniel Dror (Dubovski) 2023-03-21 01:03:00 -04:00 committed by GitHub
parent 9555bbd5bb
commit c592b12043
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
langchain/document_loaders/csv_loader.py
View File

@ -31,9 +31,11 @@ class CSVLoader(BaseLoader):
file_path: str, file_path: str,
source_column: Optional[str] = None, source_column: Optional[str] = None,
csv_args: Optional[Dict] = None, csv_args: Optional[Dict] = None,
encoding: Optional[str] = None,
): ):
self.file_path = file_path self.file_path = file_path
self.source_column = source_column self.source_column = source_column
self.encoding = encoding
if csv_args is None: if csv_args is None:
self.csv_args = { self.csv_args = {
"delimiter": ",", "delimiter": ",",
@ -45,7 +47,7 @@ class CSVLoader(BaseLoader):
def load(self) -> List[Document]: def load(self) -> List[Document]:
docs = [] docs = []
with open(self.file_path, newline="") as csvfile: with open(self.file_path, newline="", encoding=self.encoding) as csvfile:
csv = DictReader(csvfile, **self.csv_args) # type: ignore csv = DictReader(csvfile, **self.csv_args) # type: ignore
for i, row in enumerate(csv): for i, row in enumerate(csv):
content = "\n".join(f"{k.strip()}: {v.strip()}" for k, v in row.items()) content = "\n".join(f"{k.strip()}: {v.strip()}" for k, v in row.items())