ref: Add aiofile to do async file operations (#4705)

Add aiofile to do async file operations

pyproject.toml

@@ -112,6 +112,7 @@ dependencies = [
     "sqlalchemy[aiosqlite,postgresql_psycopg2binary,postgresql_psycopgbinary]>=2.0.36,<3.0.0",
     "atlassian-python-api>=3.41.16,<4.0.0",
     "mem0ai>=0.1.26,<1.0.0",
+    "aiofile>=3.9.0,<4.0.0",
 ]
 
 [project.urls]

src/backend/base/langflow/custom/directory_reader/directory_reader.py

@@ -3,6 +3,7 @@ import asyncio
 import zlib
 from pathlib import Path
 
+from aiofile import async_open
 from loguru import logger
 
 from langflow.custom import Component
@@ -104,6 +105,22 @@ class DirectoryReader:
             with _file_path.open("rb") as f:
                 return f.read().decode("utf-8")
 
+    async def aread_file_content(self, file_path):
+        """Read and return the content of a file."""
+        _file_path = Path(file_path)
+        if not _file_path.is_file():
+            return None
+        try:
+            async with async_open(_file_path, encoding="utf-8") as file:
+                # UnicodeDecodeError: 'charmap' codec can't decode byte 0x9d in position 3069:
+                # character maps to <undefined>
+                return await file.read()
+        except UnicodeDecodeError:
+            # This is happening in Windows, so we need to open the file in binary mode
+            # The file is always just a python file, so we can safely read it as utf-8
+            with async_open(_file_path, "rb") as f:
+                return (await f.read()).decode("utf-8")
+
     def get_files(self):
         """Walk through the directory path and return a list of all .py files."""
         if not (safe_path := self.get_safe_path()):
@@ -257,7 +274,7 @@ class DirectoryReader:
 
     async def process_file_async(self, file_path):
         try:
-            file_content = await asyncio.to_thread(self.read_file_content, file_path)
+            file_content = await self.aread_file_content(file_path)
         except Exception:  # noqa: BLE001
             logger.exception(f"Error while reading file {file_path}")
             return False, f"Could not read {file_path}"

uv.lock

@@ -13,6 +13,18 @@ members = [
     "langflow-base",
 ]
 
+[[package]]
+name = "aiofile"
+version = "3.9.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "caio" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/67/e2/d7cb819de8df6b5c1968a2756c3cb4122d4fa2b8fc768b53b7c9e5edb646/aiofile-3.9.0.tar.gz", hash = "sha256:e5ad718bb148b265b6df1b3752c4d1d83024b93da9bd599df74b9d9ffcf7919b", size = 17943 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/50/25/da1f0b4dd970e52bf5a36c204c107e11a0c6d3ed195eba0bfbc664c312b2/aiofile-3.9.0-py3-none-any.whl", hash = "sha256:ce2f6c1571538cbdfa0143b04e16b208ecb0e9cb4148e528af8a640ed51cc8aa", size = 19539 },
+]
+
 [[package]]
 name = "aiofiles"
 version = "24.1.0"
@@ -512,6 +524,21 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a4/07/14f8ad37f2d12a5ce41206c21820d8cb6561b728e51fad4530dff0552a67/cachetools-5.5.0-py3-none-any.whl", hash = "sha256:02134e8439cdc2ffb62023ce1debca2944c3f289d66bb17ead3ab3dede74b292", size = 9524 },
 ]
 
+[[package]]
+name = "caio"
+version = "0.9.17"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/88/cf/59b868909a85ad9eca985ad5bbeb3c0a8cd435e50ae12def770d16911753/caio-0.9.17.tar.gz", hash = "sha256:8f30511526814d961aeef389ea6885273abe6c655f1e08abbadb95d12fdd9b4f", size = 25001 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/db/c3/17bc41b7c795d91d58ee7a70ad98e23f1ba0d50bdeadd82173bb02cddc8e/caio-0.9.17-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3f69395fdd45c115b2ef59732e3c8664722a2b51de2d6eedb3d354b2f5f3be3c", size = 37849 },
+    { url = "https://files.pythonhosted.org/packages/bf/3f/0ae9f69deb3dc96b20bf084cc262b438a154d5de08a064628272713ff239/caio-0.9.17-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3028b746e9ec7f6d6ebb386a7fd8caf0eebed5d6e6b4f18c8ef25861934b1673", size = 79417 },
+    { url = "https://files.pythonhosted.org/packages/52/44/a79c7004a9562a176d78437816c736c34ab9fb6233a4b8164eb25628a09c/caio-0.9.17-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:079730a353bbde03796fab681e969472eace09ffbe5000e584868a7fe389ba6f", size = 37842 },
+    { url = "https://files.pythonhosted.org/packages/57/e2/1d04e506a5fd735856f0bb95b4d03b800947bd43c98193ee57d37070a51e/caio-0.9.17-cp311-cp311-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:549caa51b475877fe32856a26fe937366ae7a1c23a9727005b441db9abb12bcc", size = 79985 },
+    { url = "https://files.pythonhosted.org/packages/c3/b8/37dcee4bc4fae1701a86373a297bbca797f6b7bfe5f85993a11049649c63/caio-0.9.17-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:0ddb253b145a53ecca76381677ce465bc5efeaecb6aaf493fac43ae79659f0fb", size = 37909 },
+    { url = "https://files.pythonhosted.org/packages/80/f5/5e993120daeb4ec084f5f84c118bbd48b65379f63ed56919bf224e0eab42/caio-0.9.17-cp312-cp312-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b3e320b0ea371c810359934f8e8fe81777c493cc5fb4d41de44277cbe7336e74", size = 81709 },
+    { url = "https://files.pythonhosted.org/packages/58/72/3f4895adb1d23b0ac1d8afc748405a2ad3c77d8d0f23b05a64ff583c11e5/caio-0.9.17-py3-none-any.whl", hash = "sha256:c55d4dc6b3a36f93237ecd6360e1c131c3808bc47d4191a130148a99b80bb311", size = 19062 },
+]
+
 [[package]]
 name = "cassandra-driver"
 version = "3.29.2"
@@ -3494,6 +3521,7 @@ name = "langflow"
 version = "1.1.1"
 source = { editable = "." }
 dependencies = [
+    { name = "aiofile" },
     { name = "assemblyai" },
     { name = "astra-assistants", extra = ["tools"] },
     { name = "atlassian-python-api" },
@@ -3635,6 +3663,7 @@ dev = [
 
 [package.metadata]
 requires-dist = [
+    { name = "aiofile", specifier = ">=3.9.0,<4.0.0" },
     { name = "assemblyai", specifier = ">=0.33.0,<1.0.0" },
     { name = "astra-assistants", extras = ["tools"], specifier = "~=2.2.6" },
     { name = "atlassian-python-api", specifier = ">=3.41.16,<4.0.0" },