From 9761fed94faae63e8f77eefc329d613188c2dcbe Mon Sep 17 00:00:00 2001
From: daere1 <tiimmii12345@gmail.com>
Date: Sun, 13 Sep 2026 13:25:45 +0200
Subject: [PATCH] Add tests for PlainTextConverter

PlainTextConverter (packages/markitdown/src/markitdown/converters/_plain_text_converter.py)
had no dedicated unit tests. Add coverage for:
- accepted file extensions (.txt, .text, .md, .markdown, .json, .jsonl)
- accepted mimetype prefixes (text/, application/json, application/markdown)
- accepts() short-circuiting to True whenever stream_info.charset is set
- decoding via a declared charset vs. charset_normalizer auto-detection

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
---
 .../tests/test_plain_text_converter.py        | 62 +++++++++++++++++++
 1 file changed, 62 insertions(+)
 create mode 100644 packages/markitdown/tests/test_plain_text_converter.py

diff --git a/packages/markitdown/tests/test_plain_text_converter.py b/packages/markitdown/tests/test_plain_text_converter.py
new file mode 100644
index 000000000..729f7a945
--- /dev/null
+++ b/packages/markitdown/tests/test_plain_text_converter.py
@@ -0,0 +1,62 @@
+"""Tests for PlainTextConverter (src/markitdown/converters/_plain_text_converter.py)."""
+
+import io
+
+import pytest
+
+from markitdown import MarkItDown, StreamInfo
+
+
+@pytest.fixture(scope="module")
+def converter() -> MarkItDown:
+    return MarkItDown(enable_plugins=False)
+
+
+@pytest.mark.parametrize(
+    "extension", [".txt", ".text", ".md", ".markdown", ".json", ".jsonl"]
+)
+def test_accepted_extensions_pass_content_through(
+    converter: MarkItDown, extension: str
+) -> None:
+    result = converter.convert_stream(
+        io.BytesIO(b"hello, world"),
+        stream_info=StreamInfo(extension=extension, charset="utf-8"),
+    )
+    assert result.markdown == "hello, world"
+
+
+@pytest.mark.parametrize(
+    "mimetype", ["text/plain", "text/x-log", "application/json", "application/markdown"]
+)
+def test_accepted_mimetype_prefixes_pass_content_through(
+    converter: MarkItDown, mimetype: str
+) -> None:
+    result = converter.convert_stream(
+        io.BytesIO(b"hello, world"),
+        stream_info=StreamInfo(mimetype=mimetype, charset="utf-8"),
+    )
+    assert result.markdown == "hello, world"
+
+
+def test_declared_charset_is_honored_even_with_an_unrecognized_extension(
+    converter: MarkItDown,
+) -> None:
+    # accepts() short-circuits to True whenever stream_info.charset is set,
+    # regardless of extension/mimetype -- exercise that branch directly.
+    body = "grüße".encode("utf-8")
+    result = converter.convert_stream(
+        io.BytesIO(body),
+        stream_info=StreamInfo(extension=".some-unknown-ext", charset="utf-8"),
+    )
+    assert result.markdown == "grüße"
+
+
+def test_charset_is_auto_detected_when_not_declared(converter: MarkItDown) -> None:
+    # UTF-8 encodes non-ASCII text distinctively enough for charset_normalizer
+    # to detect reliably without a declared charset.
+    body = "grüße".encode("utf-8")
+    result = converter.convert_stream(
+        io.BytesIO(body),
+        stream_info=StreamInfo(extension=".txt"),
+    )
+    assert result.markdown == "grüße"