kreuzberg 1.5.0__py3-none-any.whl → 1.6.0__py3-none-any.whl

kreuzberg/_extractors.py

@@ -4,6 +4,8 @@ import re
 from contextlib import suppress
 from html import escape
 from io import BytesIO
+from pathlib import Path
+from tempfile import NamedTemporaryFile
 from typing import TYPE_CHECKING
 
 import html_to_markdown
@@ -11,6 +13,7 @@ import pptx
 import pypdfium2
 from anyio import Path as AsyncPath
 from pptx.enum.shapes import MSO_SHAPE_TYPE
+from xlsx2csv import Xlsx2csv
 
 from kreuzberg._pandoc import process_content, process_file
 from kreuzberg._string import normalize_spaces, safe_decode
@@ -19,8 +22,6 @@ from kreuzberg._tesseract import batch_process_images
 from kreuzberg.exceptions import ParsingError
 
 if TYPE_CHECKING:  # pragma: no cover
-    from pathlib import Path
-
     from PIL.Image import Image
 
 
@@ -195,6 +196,40 @@ async def extract_pptx_file(file_path_or_contents: Path | bytes) -> str:
     return normalize_spaces(md_content)
 
 
+async def extract_xlsx_file(file_path_or_contents: Path | bytes) -> str:
+    """Extract text from an XLSX file by converting it to CSV and then to markdown.
+
+    Args:
+        file_path_or_contents: The path to the XLSX file or its contents as bytes.
+
+    Returns:
+        The extracted text content.
+
+    Raises:
+        ParsingError: If the XLSX file could not be parsed.
+    """
+    try:
+        with NamedTemporaryFile(suffix=".xlsx") as xlsx_file, NamedTemporaryFile(suffix=".csv") as csv_file:
+            if isinstance(file_path_or_contents, bytes):
+                xlsx_file.write(file_path_or_contents)
+                xlsx_file.flush()
+                xlsx_path = xlsx_file.name
+            else:
+                xlsx_path = str(file_path_or_contents)
+
+            await run_sync(Xlsx2csv(xlsx_path).convert, csv_file.name)
+            result = await process_file(csv_file.name, mime_type="text/csv")
+            return normalize_spaces(result.content)
+    except Exception as e:
+        raise ParsingError(
+            "Could not extract text from XLSX file",
+            context={
+                "error": str(e),
+                "file_path": str(file_path_or_contents) if isinstance(file_path_or_contents, Path) else None,
+            },
+        ) from e
+
+
 async def extract_html_string(file_path_or_contents: Path | bytes) -> str:
     """Extract text from an HTML string.
 

kreuzberg/_mime_types.py

@@ -10,7 +10,7 @@ MARKDOWN_MIME_TYPE: Final = "text/markdown"
 PDF_MIME_TYPE: Final = "application/pdf"
 PLAIN_TEXT_MIME_TYPE: Final = "text/plain"
 POWER_POINT_MIME_TYPE: Final = "application/vnd.openxmlformats-officedocument.presentationml.presentation"
-
+EXCEL_MIME_TYPE: Final = "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
 PLAIN_TEXT_MIME_TYPES: Final[set[str]] = {PLAIN_TEXT_MIME_TYPE, MARKDOWN_MIME_TYPE}
 
 IMAGE_MIME_TYPES: Final[set[str]] = {
@@ -89,5 +89,5 @@ SUPPORTED_MIME_TYPES: Final[set[str]] = (
     PLAIN_TEXT_MIME_TYPES
     | IMAGE_MIME_TYPES
     | PANDOC_SUPPORTED_MIME_TYPES
-    | {PDF_MIME_TYPE, POWER_POINT_MIME_TYPE, HTML_MIME_TYPE}
+    | {PDF_MIME_TYPE, POWER_POINT_MIME_TYPE, HTML_MIME_TYPE, EXCEL_MIME_TYPE}
 )

kreuzberg/extraction.py

@@ -22,8 +22,10 @@ from kreuzberg._extractors import (
     extract_html_string,
     extract_pdf_file,
     extract_pptx_file,
+    extract_xlsx_file,
 )
 from kreuzberg._mime_types import (
+    EXCEL_MIME_TYPE,
     HTML_MIME_TYPE,
     IMAGE_MIME_TYPE_EXT_MAP,
     IMAGE_MIME_TYPES,
@@ -75,6 +77,9 @@ async def extract_bytes(content: bytes, mime_type: str, force_ocr: bool = False)
                 content=await extract_pdf_file(Path(temp_file.name), force_ocr), mime_type=PLAIN_TEXT_MIME_TYPE
             )
 
+    if mime_type == EXCEL_MIME_TYPE or mime_type.startswith(EXCEL_MIME_TYPE):
+        return ExtractionResult(content=await extract_xlsx_file(content), mime_type=MARKDOWN_MIME_TYPE)
+
     if mime_type in IMAGE_MIME_TYPES or any(mime_type.startswith(value) for value in IMAGE_MIME_TYPES):
         with NamedTemporaryFile(suffix=IMAGE_MIME_TYPE_EXT_MAP[mime_type]) as temp_file:
             temp_file.write(content)
@@ -134,6 +139,9 @@ async def extract_file(
     if mime_type == PDF_MIME_TYPE or mime_type.startswith(PDF_MIME_TYPE):
         return ExtractionResult(content=await extract_pdf_file(file_path, force_ocr), mime_type=PLAIN_TEXT_MIME_TYPE)
 
+    if mime_type == EXCEL_MIME_TYPE or mime_type.startswith(EXCEL_MIME_TYPE):
+        return ExtractionResult(content=await extract_xlsx_file(file_path), mime_type=MARKDOWN_MIME_TYPE)
+
     if mime_type in IMAGE_MIME_TYPES or any(mime_type.startswith(value) for value in IMAGE_MIME_TYPES):
         return ExtractionResult(content=await process_image_with_tesseract(file_path), mime_type=PLAIN_TEXT_MIME_TYPE)
 

{kreuzberg-1.5.0.dist-info → kreuzberg-1.6.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: kreuzberg
-Version: 1.5.0
+Version: 1.6.0
 Summary: A text extraction library supporting PDFs, images, office documents and more
 Author-email: Na'aman Hirschfeld <nhirschfed@gmail.com>
 License: MIT
@@ -30,6 +30,7 @@ Requires-Dist: html-to-markdown>=1.2.0
 Requires-Dist: pypdfium2>=4.30.1
 Requires-Dist: python-pptx>=1.0.2
 Requires-Dist: typing-extensions>=4.12.2; python_version < "3.10"
+Requires-Dist: xlsx2csv>=0.8.4
 
 # Kreuzberg
 
@@ -68,16 +69,12 @@ pip install kreuzberg
 
 ### 2. Install System Dependencies
 
-Kreuzberg requires two open-source tools:
+Kreuzberg requires two system level dependencies:
 
 - [Pandoc](https://pandoc.org/installing.html) - For document format conversion
-
-  - GPL v2.0 licensed (used via CLI only)
-  - Handles office documents and markup formats
-
 - [Tesseract OCR](https://tesseract-ocr.github.io/) - For image and PDF OCR
-  - Apache License
-  - Required for scanned documents and images
+
+Please install these using their respective installation guides.
 
 ## Architecture
 
@@ -87,9 +84,10 @@ Kreuzberg is designed as a high-level async abstraction over established open-so
   - `pdfium2` for searchable PDFs
   - Tesseract OCR for scanned content
 - **Document Conversion**:
-  - Pandoc for office documents and markup
+  - Pandoc for many document and markup formats
   - `python-pptx` for PowerPoint files
   - `html-to-markdown` for HTML content
+  - `xlsx2csv` for Excel spreadsheets
 - **Text Processing**:
   - Smart encoding detection
   - Markdown and plain text handling
@@ -121,6 +119,7 @@ Kreuzberg is designed as a high-level async abstraction over established open-so
 
 #### Data and Research Formats
 
+- Excel spreadsheets (`.xlsx`)
 - CSV (`.csv`) and TSV (`.tsv`) files
 - Jupyter Notebooks (`.ipynb`)
 - BibTeX (`.bib`) and BibLaTeX (`.bib`)

{kreuzberg-1.5.0.dist-info → kreuzberg-1.6.0.dist-info}/RECORD

@@ -1,15 +1,15 @@
 kreuzberg/__init__.py,sha256=5IBPjPsZ7faK15gFB9ZEROHhkEX7KKQmrHPCZuGnhb0,285
-kreuzberg/_extractors.py,sha256=k6xO_2ItaftPmlqzfXyxTn8rdaWdwrJHGziBbo7gCio,6599
-kreuzberg/_mime_types.py,sha256=0ZYtRrMAaKpCMDkhpTbWAXHCsVob5MFRMGlbni8iYSA,2573
+kreuzberg/_extractors.py,sha256=cbDjitvqI35Gimh27iXvEE0Zczf9jZRJZS7Do8ugVNE,7934
+kreuzberg/_mime_types.py,sha256=nvRSWDUhtntO9-E9gv2l5BVYow61zim4llJ6n33k_BE,2682
 kreuzberg/_pandoc.py,sha256=DC6y_NN_CG9dF6fhAj3WumXqKIJLjYmnql2H53_KHnE,13766
 kreuzberg/_string.py,sha256=4txRDnkdR12oO6G8V-jXEMlA9ivgmw8E8EbjyhfL-W4,1106
 kreuzberg/_sync.py,sha256=ovsFHFdkcczz7gNEUJsbZzY8KHG0_GAOOYipQNE4hIY,874
 kreuzberg/_tesseract.py,sha256=nnhkjRIS0BSoovjMIqOlBEXlzngE0QJeFDe7BIqUik8,7872
 kreuzberg/exceptions.py,sha256=pxoEPS0T9e5QSgxsfXn1VmxsY_EGXvTwY0gETPiNn8E,945
-kreuzberg/extraction.py,sha256=gux3fkPIs8IbIKtRGuPFWJBLB5jO6Y9JsBfhHRcpQ0k,6160
+kreuzberg/extraction.py,sha256=G3_Uyzhe99qEib4WLE7_l1oC9JKlvoVdn3WEY56J_Wo,6572
 kreuzberg/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg-1.5.0.dist-info/LICENSE,sha256=-8caMvpCK8SgZ5LlRKhGCMtYDEXqTKH9X8pFEhl91_4,1066
-kreuzberg-1.5.0.dist-info/METADATA,sha256=O462ss7M6Cb8cO6fJXwqsOdzkzaZekqa1oGwb7Vrgx8,9641
-kreuzberg-1.5.0.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-kreuzberg-1.5.0.dist-info/top_level.txt,sha256=rbGkygffkZiyKhL8UN41ZOjLfem0jJPA1Whtndne0rE,10
-kreuzberg-1.5.0.dist-info/RECORD,,
+kreuzberg-1.6.0.dist-info/LICENSE,sha256=-8caMvpCK8SgZ5LlRKhGCMtYDEXqTKH9X8pFEhl91_4,1066
+kreuzberg-1.6.0.dist-info/METADATA,sha256=GQNbGnxmym5vAcXDivDUccdVBUGnYh-4M38xYEkKTJk,9663
+kreuzberg-1.6.0.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+kreuzberg-1.6.0.dist-info/top_level.txt,sha256=rbGkygffkZiyKhL8UN41ZOjLfem0jJPA1Whtndne0rE,10
+kreuzberg-1.6.0.dist-info/RECORD,,