From d71fce7a6aff2ae4e94ec29a02ba396ac939ff32 Mon Sep 17 00:00:00 2001
From: Matteo Rosati <rosati.matteo@gmail.com>
Date: Fri, 24 Apr 2026 18:40:51 +0200
Subject: [PATCH] cannot import non-text files!

---
 chromy/cli.py                  |  5 ++++-
 chromy/handlers/import_data.py |  8 +++++++
 chromy/utilities.py            | 39 ++++++++++++++++++++++++++++++++++
 3 files changed, 51 insertions(+), 1 deletion(-)

diff --git a/chromy/cli.py b/chromy/cli.py
index d142254..2f2f1b9 100644
--- a/chromy/cli.py
+++ b/chromy/cli.py
@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from plistlib import InvalidFileException
 from typing import Annotated, Callable
 
 import typer
@@ -122,7 +123,9 @@ def import_data(
     except NotFoundError:
         _fail(f"Collection '{collection}' does not exist.")
     except FileNotFoundError:
-        _fail(f"The file {file} was not found.")
+        _fail(f"The file '{file}' was not found.")
+    except InvalidFileException:
+        _fail(f"The file '{file}' is not a text file.")
 
 
 # ------------------------------------------------------------------------------
diff --git a/chromy/handlers/import_data.py b/chromy/handlers/import_data.py
index 9bee641..365736c 100644
--- a/chromy/handlers/import_data.py
+++ b/chromy/handlers/import_data.py
@@ -2,11 +2,14 @@ from __future__ import annotations
 
 import os
 from pathlib import Path
+from plistlib import InvalidFileException
 
 from rich import print
 
 from chromy.utilities import ingest_file
 
+from ..utilities import is_probably_text_file
+
 
 def _get_absolute_path(file: str) -> str:
     """
@@ -27,6 +30,11 @@ def _get_absolute_path(file: str) -> str:
 
 
 def handle_import(collection: str, file: str) -> int:
+    absolute_path = _get_absolute_path(file)
+
+    if not is_probably_text_file(absolute_path):
+        raise InvalidFileException()
+
     records_added = ingest_file(collection, _get_absolute_path(file))
     print(f"[bold green]Added[/] {records_added} records to collection '{collection}'.")
     return 0
diff --git a/chromy/utilities.py b/chromy/utilities.py
index fdb086f..3ddf44c 100644
--- a/chromy/utilities.py
+++ b/chromy/utilities.py
@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 from collections.abc import Mapping, Sequence
+from pathlib import Path
 
 from chromadb import QueryResult
 from rich.console import Console
@@ -76,3 +77,41 @@ def format_query_result(result: QueryResult) -> list[Rule | Text]:
         lines.append(Rule())
 
     return lines
+
+
+def is_probably_text_file(path: str | Path, sample_size: int = 8192) -> bool:
+    """
+    Return whether a file appears to contain text.
+
+    Args:
+        path (str | Path): The path to the file to inspect.
+        sample_size (int): The maximum number of bytes to read from the file.
+
+    Returns:
+        bool: ``True`` if the sampled bytes decode as UTF-8, UTF-8 with BOM,
+        UTF-16, or UTF-32, or if the file is empty. Otherwise, ``False``.
+    """
+
+    path = Path(path)
+
+    with path.open("rb") as f:
+        sample = f.read(sample_size)
+
+    if not sample:
+        return True
+
+    encodings = (
+        "utf-8",
+        "utf-8-sig",
+        "utf-16",
+        "utf-32",
+    )
+
+    for encoding in encodings:
+        try:
+            sample.decode(encoding)
+            return True
+        except UnicodeDecodeError:
+            pass
+
+    return False