feat: support xlsx file parsing (#304)

Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>

api/core/index/readers/xlsx_parser.py

@@ -0,0 +1,31 @@
+from pathlib import Path
+import json
+from typing import Dict
+from openpyxl import load_workbook
+
+from llama_index.readers.file.base_parser import BaseParser
+from flask import current_app
+
+
+class XLSXParser(BaseParser):
+    """XLSX parser."""
+
+    def _init_parser(self) -> Dict:
+        """Init parser"""
+        return {}
+
+    def parse_file(self, file: Path, errors: str = "ignore") -> str:
+        data = []
+        keys = []
+        with open(file, "r") as fp:
+            wb = load_workbook(filename=file, read_only=True)
+            # loop over all sheets
+            for sheet in wb:
+                for row in sheet.iter_rows(values_only=True):
+                    if all(v is None for v in row):
+                        continue
+                    if keys == []:
+                        keys = row
+                    else:
+                        data.append(json.dumps(dict(zip(keys, row)), ensure_ascii=False))
+        return data

api/core/indexing_runner.py

@@ -12,6 +12,8 @@ from llama_index.data_structs import Node
 from llama_index.data_structs.node_v2 import DocumentRelationship
 from llama_index.node_parser import SimpleNodeParser, NodeParser
 from llama_index.readers.file.base import DEFAULT_FILE_EXTRACTOR
+from llama_index.readers.file.markdown_parser import MarkdownParser
+from core.index.readers.xlsx_parser import XLSXParser
 from core.docstore.dataset_docstore import DatesetDocumentStore
 from core.index.keyword_table_index import KeywordTableIndex
 from core.index.readers.html_parser import HTMLParser
@@ -250,6 +252,7 @@ class IndexingRunner:
             file_extractor[".html"] = HTMLParser()
             file_extractor[".htm"] = HTMLParser()
             file_extractor[".pdf"] = PDFParser({'upload_file': upload_file})
+            file_extractor[".xlsx"] = XLSXParser()
 
             loader = SimpleDirectoryReader(input_files=[filepath], file_extractor=file_extractor)
             text_docs = loader.load_data()