Skip to content

Python tool for converting files and office documents to Markdown.

Add CSV to Markdown table conversion - fixes #1144 (#1176)

Turdıbek8576f1d9153d
diff --git a/packages/markitdown/src/markitdown/_markitdown.py b/packages/markitdown/src/markitdown/_markitdown.py
index 54a0dc80..682902b4 100644
--- a/packages/markitdown/src/markitdown/_markitdown.py
+++ b/packages/markitdown/src/markitdown/_markitdown.py
@@ -41,6 +41,7 @@ from .converters import (
ZipConverter,
EpubConverter,
DocumentIntelligenceConverter,
+ CsvConverter,
)
from ._base_converter import DocumentConverter, DocumentConverterResult
@@ -194,6 +195,7 @@ class MarkItDown:
self.register_converter(PdfConverter())
self.register_converter(OutlookMsgConverter())
self.register_converter(EpubConverter())
+ self.register_converter(CsvConverter())
# Register Document Intelligence converter at the top of the stack if endpoint is provided
docintel_endpoint = kwargs.get("docintel_endpoint")
diff --git a/packages/markitdown/src/markitdown/converters/__init__.py b/packages/markitdown/src/markitdown/converters/__init__.py
index c68d0c34..e4437a58 100644
--- a/packages/markitdown/src/markitdown/converters/__init__.py
+++ b/packages/markitdown/src/markitdown/converters/__init__.py
@@ -22,6 +22,7 @@ from ._doc_intel_converter import (
DocumentIntelligenceFileType,
)
from ._epub_converter import EpubConverter
+from ._csv_converter import CsvConverter
__all__ = [
"PlainTextConverter",
@@ -43,4 +44,5 @@ __all__ = [
"DocumentIntelligenceConverter",
"DocumentIntelligenceFileType",
"EpubConverter",
+ "CsvConverter",
]
diff --git a/packages/markitdown/src/markitdown/converters/_csv_converter.py b/packages/markitdown/src/markitdown/converters/_csv_converter.py
new file mode 100644
index 00000000..7162889d
--- /dev/null
+++ b/packages/markitdown/src/markitdown/converters/_csv_converter.py
@@ -0,0 +1,79 @@
+import sys
+import csv
+import io
+from typing import BinaryIO, Any
+from charset_normalizer import from_bytes
+from ._html_converter import HtmlConverter
+from .._base_converter import DocumentConverter, DocumentConverterResult
+from .._stream_info import StreamInfo
+
+ACCEPTED_MIME_TYPE_PREFIXES = [
+ "text/csv",
+ "application/csv",
+]
+ACCEPTED_FILE_EXTENSIONS = [".csv"]
+
+
+class CsvConverter(DocumentConverter):
+ """
+ Converts CSV files to Markdown tables.
+ """
+
+ def __init__(self):
+ super().__init__()
+
+ def accepts(
+ self,
+ file_stream: BinaryIO,
+ stream_info: StreamInfo,
+ **kwargs: Any, # Options to pass to the converter
+ ) -> bool:
+ mimetype = (stream_info.mimetype or "").lower()
+ extension = (stream_info.extension or "").lower()
+ if extension in ACCEPTED_FILE_EXTENSIONS:
+ return True
+ for prefix in ACCEPTED_MIME_TYPE_PREFIXES:
+ if mimetype.startswith(prefix):
+ return True
+ return False
+
+ def convert(
+ self,
+ file_stream: BinaryIO,
+ stream_info: StreamInfo,
+ **kwargs: Any, # Options to pass to the converter
+ ) -> DocumentConverterResult:
+ # Read the file content
+ if stream_info.charset:
+ content = file_stream.read().decode(stream_info.charset)
+ else:
+ content = str(from_bytes(file_stream.read()).best())
+
+ # Parse CSV content
+ reader = csv.reader(io.StringIO(content))
+ rows = list(reader)
+
+ if not rows:
+ return DocumentConverterResult(markdown="")
+
+ # Create markdown table
+ markdown_table = []
+
+ # Add header row
+ markdown_table.append("| " + " | ".join(rows[0]) + " |")
+
+ # Add separator row
+ markdown_table.append("| " + " | ".join(["---"] * len(rows[0])) + " |")
+
+ # Add data rows
+ for row in rows[1:]:
+ # Make sure row has the same number of columns as header
+ while len(row) < len(rows[0]):
+ row.append("")
+ # Truncate if row has more columns than header
+ row = row[: len(rows[0])]
+ markdown_table.append("| " + " | ".join(row) + " |")
+
+ result = "\n".join(markdown_table)
+
+ return DocumentConverterResult(markdown=result)
diff --git a/packages/markitdown/src/markitdown/converters/_doc_intel_converter.py b/packages/markitdown/src/markitdown/converters/_doc_intel_converter.py
index 5f4069b0..d2dce917 100644
--- a/packages/markitdown/src/markitdown/converters/_doc_intel_converter.py
+++ b/packages/markitdown/src/markitdown/converters/_doc_intel_converter.py
@@ -1,8 +1,7 @@
import sys
import re
import os
-
-from typing import BinaryIO, Any, List
+from typing import BinaryIO, Any, List, Optional, Union
from enum import Enum
from ._html_converter import HtmlConverter
@@ -26,6 +25,28 @@ except ImportError:
# Preserve the error and stack trace for later
_dependency_exc_info = sys.exc_info()
+ # Define these types for type hinting when the package is not available
+ class AzureKeyCredential:
+ pass
+
+ class TokenCredential:
+ pass
+
+ class DocumentIntelligenceClient:
+ pass
+
+ class AnalyzeDocumentRequest:
+ pass
+
+ class AnalyzeResult:
+ pass
+
+ class DocumentAnalysisFeature:
+ pass
+
+ class DefaultAzureCredential:
+ pass
+
# TODO: currently, there is a bug in the document intelligence SDK with importing the "ContentFormat" enum.
# This constant is a temporary fix until the bug is resolved.
diff --git a/packages/markitdown/tests/_test_vectors.py b/packages/markitdown/tests/_test_vectors.py
index 4a7b54a0..74fa9bd0 100644
--- a/packages/markitdown/tests/_test_vectors.py
+++ b/packages/markitdown/tests/_test_vectors.py
@@ -144,10 +144,11 @@ GENERAL_TEST_VECTORS = [
charset="cp932",
url=None,
must_include=[
- "名前,年齢,住所",
- "佐藤太郎,30,東京",
- "三木英子,25,大阪",
- "髙橋淳,35,名古屋",
+ "| 名前 | 年齢 | 住所 |",
+ "| --- | --- | --- |",
+ "| 佐藤太郎 | 30 | 東京 |",
+ "| 三木英子 | 25 | 大阪 |",
+ "| 髙橋淳 | 35 | 名古屋 |",
],
must_not_include=[],
),