Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
Mirrored from https://github.com/Camo67/composio.git
Synced Nov 22, 13:15 UTC
fix: make MIME type lookup case-insensitive and decode URL-encoded filenames
Addresses review feedback from cursorbot: - MIME type lookup now uses .lower() to be case-insensitive per RFC 2045 - URL paths are now decoded using urllib.parse.unquote before extracting filenames This fixes issues where: - Servers returning Content-Type headers like 'Image/JPEG' would fail lookup - URLs like 'My%20Document.pdf' would result in percent-encoded filenames
jkomyno committed
Jan 27, 2026 at 05:00 UTC
f5e17626dc544c7f6eeb74fff6c561bb5cd4b55e
2 files changed
+69
-4
python/composio/core/models/_files.py
+4
-4
@@ -4,7 +4,7 @@ import hashlib
4
import os
5
import typing as t
6
from pathlib import Path
7
-from urllib.parse import urlparse
7
+from urllib.parse import unquote, urlparse
8
import uuid
9
import datetime
10
@@ -123,7 +123,7 @@ def _get_extension_from_mimetype(mimetype: str) -> str:
123
"video/mp4": ".mp4",
124
"video/webm": ".webm",
125
}
126
- return mime_to_ext.get(mimetype, "")
126
+ return mime_to_ext.get(mimetype.lower(), "")
127
128
129
def _generate_timestamped_filename(extension: str) -> str:
@@ -150,9 +150,9 @@ def _fetch_file_from_url(url: str) -> t.Tuple[str, bytes, str]:
150
# Handle mimetypes with charset or other parameters (e.g., "text/html; charset=utf-8")
151
mimetype = mimetype.split(";")[0].strip()
152
153
- # Extract filename from URL
153
+ # Extract filename from URL (decode percent-encoded characters)
154
parsed_url = urlparse(url)
155
- pathname = parsed_url.path
155
+ pathname = unquote(parsed_url.path)
156
filename = os.path.basename(pathname) if pathname else ""
157
158
# If no filename from URL or no extension, generate one
python/tests/test_files.py
+65
@@ -1010,6 +1010,19 @@ class TestUrlHelperFunctions:
1010
assert _get_extension_from_mimetype("application/unknown") == ""
1011
assert _get_extension_from_mimetype("custom/type") == ""
1012
1013
+ def test_get_extension_from_mimetype_case_insensitive(self):
1014
+ """Test _get_extension_from_mimetype handles case-insensitive mimetypes (RFC 2045)."""
1015
+ # Uppercase variations
1016
+ assert _get_extension_from_mimetype("IMAGE/JPEG") == ".jpg"
1017
+ assert _get_extension_from_mimetype("APPLICATION/PDF") == ".pdf"
1018
+ # Mixed case variations
1019
+ assert _get_extension_from_mimetype("Image/Jpeg") == ".jpg"
1020
+ assert _get_extension_from_mimetype("Application/Pdf") == ".pdf"
1021
+ assert _get_extension_from_mimetype("Text/Plain") == ".txt"
1022
+ # Edge cases
1023
+ assert _get_extension_from_mimetype("IMAGE/png") == ".png"
1024
+ assert _get_extension_from_mimetype("video/MP4") == ".mp4"
1025
+
1026
def test_generate_timestamped_filename(self):
1027
"""Test _generate_timestamped_filename generates valid filenames."""
1028
filename = _generate_timestamped_filename(".jpg")
@@ -1104,6 +1117,58 @@ class TestFetchFileFromUrl:
1117
assert "Failed to fetch file from URL" in str(exc_info.value)
1118
assert "404" in str(exc_info.value)
1119
1120
+ @patch("composio.core.models._files.requests.get")
1121
+ def test_fetch_file_from_url_decodes_percent_encoded_filename(self, mock_get):
1122
+ """Test that percent-encoded characters in URL filenames are decoded."""
1123
+ mock_response = MagicMock()
1124
+ mock_response.ok = True
1125
+ mock_response.content = b"document content"
1126
+ mock_response.headers = {"content-type": "application/pdf"}
1127
+ mock_get.return_value = mock_response
1128
+
1129
+ # URL with percent-encoded spaces and special characters
1130
+ filename, content, mimetype = _fetch_file_from_url(
1131
+ "https://example.com/My%20Document%20%282024%29.pdf"
1132
+ )
1133
+
1134
+ # Filename should be decoded
1135
+ assert filename == "My Document (2024).pdf"
1136
+ assert content == b"document content"
1137
+ assert mimetype == "application/pdf"
1138
+
1139
+ @patch("composio.core.models._files.requests.get")
1140
+ def test_fetch_file_from_url_decodes_unicode_filename(self, mock_get):
1141
+ """Test that percent-encoded unicode characters in URL filenames are decoded."""
1142
+ mock_response = MagicMock()
1143
+ mock_response.ok = True
1144
+ mock_response.content = b"image data"
1145
+ mock_response.headers = {"content-type": "image/jpeg"}
1146
+ mock_get.return_value = mock_response
1147
+
1148
+ # URL with percent-encoded unicode (e.g., Japanese characters)
1149
+ filename, content, mimetype = _fetch_file_from_url(
1150
+ "https://example.com/%E3%83%95%E3%82%A1%E3%82%A4%E3%83%AB.jpg"
1151
+ )
1152
+
1153
+ # Filename should be decoded to unicode
1154
+ assert filename == "ファイル.jpg"
1155
+
1156
+ @patch("composio.core.models._files.requests.get")
1157
+ def test_fetch_file_from_url_handles_plus_sign_in_filename(self, mock_get):
1158
+ """Test that plus signs in URL paths are preserved (not converted to spaces)."""
1159
+ mock_response = MagicMock()
1160
+ mock_response.ok = True
1161
+ mock_response.content = b"file content"
1162
+ mock_response.headers = {"content-type": "text/plain"}
1163
+ mock_get.return_value = mock_response
1164
+
1165
+ # Plus signs in path should remain as plus signs (unquote doesn't convert + to space)
1166
+ filename, content, mimetype = _fetch_file_from_url(
1167
+ "https://example.com/file+name.txt"
1168
+ )
1169
+
1170
+ assert filename == "file+name.txt"
1171
+
1172
1173
class TestUploadBytesToS3:
1174
"""Test cases for _upload_bytes_to_s3 function."""