patching TextSpitter.core to explicitly convert filename contents into File objects

Author: fsecada01

.idea/ryecharm-overrides.xml

@@ -1,6 +1,6 @@
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.6.0
+    rev: v5.0.0
     hooks:
       - id: check-added-large-files
         exclude: bin/
@@ -34,7 +34,7 @@ repos:
 
   - repo: https://github.com/charliermarsh/ruff-pre-commit
     # Ruff version.
-    rev: "v0.6.5"
+    rev: "v0.8.0"
     hooks:
       - id: ruff
 
@@ -45,7 +45,7 @@ repos:
         name: isort (python)
 
   - repo: https://github.com/psf/black
-    rev: 24.8.0
+    rev: 24.10.0
     hooks:
       - id: black
         exclude: tests/

.idea/ryecharm.xml

@@ -3,7 +3,7 @@
 """
 
 import mimetypes
-from io import BytesIO, FileIO
+from io import BytesIO
 from pathlib import Path
 from typing import IO
 
@@ -25,31 +25,36 @@ def __init__(
         object's file property; if a file-like object is provided instead of a
         name, then a file_ext arg will be required.
 
-        `file_name` is depreciated.
+        `filename` is depreciated.
 
         Args:
             file_obj: str | Path | None
             filename: : str | None
         """
 
         if filename:
-            self.file = FileIO(filename)
+            self.file = Path(filename)
             self.file_ext = filename.split(".")[-1]
-        elif isinstance(file_obj, str):
-            file_obj = Path(file_obj)
-
-        elif file_obj and any((isinstance(file_obj, x) for x in (Path, IO))):
-            pass
-
-        if hasattr(file_obj, "name"):
-            self.file = file_obj
-            self.file_ext = file_obj.name.split(".")[-1]
+            self.file_name = self.file.name
         else:
-            raise Exception(
-                "Your file object does not contain a name attribute. Please"
-                " add a name attribute with a file extension, and try "
-                "again. Need the file ext. data for mime-typing."
-            )
+            if isinstance(file_obj, str):
+                file_obj = Path(file_obj)
+
+            elif file_obj and any(
+                (isinstance(file_obj, x) for x in (Path, IO))
+            ):
+                pass
+
+            if hasattr(file_obj, "name"):
+                self.file = file_obj
+                self.file_ext = file_obj.name.split(".")[-1]
+                self.file_name = file_obj.name
+            else:
+                raise Exception(
+                    "Your file object does not contain a name attribute. Please"
+                    " add a name attribute with a file extension, and try "
+                    "again. Need the file ext. data for mime-typing."
+                )
 
     @staticmethod
     def get_file_type(file: str | Path):
@@ -74,7 +79,11 @@ def get_contents(self):
         Returns:
             str | int | bytes
         """
-        mime_type = self.get_file_type(self.file)
+        mime_type = (
+            self.get_file_type(self.file)
+            if isinstance(self.file, str)
+            else self.get_file_type(self.file_name)
+        )
         open_mode = "r" if "text" in mime_type else "rb+"
         with self.file.open(open_mode) as f:
             return f.read()

.pre-commit-config.yaml

@@ -1,19 +1,13 @@
-#
-# This file is autogenerated by pip-compile with Python 3.12
-# by the following command:
-#
-#    pip-compile core_requirements.in
-#
+# This file was autogenerated by uv via the following command:
+#    uv pip compile core_requirements.in -o core_requirements.txt
 colorama==0.4.6
     # via loguru
 loguru==0.7.2
     # via -r core_requirements.in
 lxml==5.3.0
     # via python-docx
-pymupdf==1.24.10
+pymupdf==1.24.14
     # via -r core_requirements.in
-pymupdfb==1.24.10
-    # via pymupdf
 pypdf2==3.0.1
     # via -r core_requirements.in
 python-docx==1.1.2

TextSpitter/core.py

@@ -1,10 +1,6 @@
-#
-# This file is autogenerated by pip-compile with Python 3.12
-# by the following command:
-#
-#    pip-compile dev_requirements.in
-#
-anyio==4.5.0
+# This file was autogenerated by uv via the following command:
+#    uv pip compile dev_requirements.in -o dev_requirements.txt
+anyio==4.6.2.post1
     # via
     #   httpx
     #   jupyter-server
@@ -26,9 +22,9 @@ babel==2.16.0
     # via jupyterlab-server
 beautifulsoup4==4.12.3
     # via nbconvert
-black==24.8.0
+black==24.10.0
     # via -r dev_requirements.in
-bleach==6.1.0
+bleach==6.2.0
     # via nbconvert
 certifi==2024.8.30
     # via
@@ -37,7 +33,7 @@ certifi==2024.8.30
     #   requests
 cffi==1.17.1
     # via argon2-cffi-bindings
-charset-normalizer==3.3.2
+charset-normalizer==3.4.0
     # via requests
 click==8.1.7
     # via black
@@ -48,7 +44,7 @@ colorama==0.4.6
     #   ipython
 comm==0.2.2
     # via ipykernel
-debugpy==1.8.5
+debugpy==1.8.9
     # via ipykernel
 decorator==5.1.1
     # via ipython
@@ -64,7 +60,7 @@ fqdn==1.5.1
     # via jsonschema
 h11==0.14.0
     # via httpcore
-httpcore==1.0.5
+httpcore==1.0.7
     # via httpx
 httpx==0.27.2
     # via jupyterlab
@@ -78,7 +74,7 @@ importlib-metadata==8.5.0
     # via twine
 ipykernel==6.29.5
     # via jupyterlab
-ipython==8.27.0
+ipython==8.29.0
     # via ipykernel
 isoduration==20.11.0
     # via jsonschema
@@ -88,26 +84,26 @@ jaraco-classes==3.4.0
     # via keyring
 jaraco-context==6.0.1
     # via keyring
-jaraco-functools==4.0.2
+jaraco-functools==4.1.0
     # via keyring
-jedi==0.19.1
+jedi==0.19.2
     # via ipython
 jinja2==3.1.4
     # via
     #   jupyter-server
     #   jupyterlab
     #   jupyterlab-server
     #   nbconvert
-json5==0.9.25
+json5==0.9.28
     # via jupyterlab-server
 jsonpointer==3.0.0
     # via jsonschema
-jsonschema[format-nongpl]==4.23.0
+jsonschema==4.23.0
     # via
     #   jupyter-events
     #   jupyterlab-server
     #   nbformat
-jsonschema-specifications==2023.12.1
+jsonschema-specifications==2024.10.1
     # via jsonschema
 jupyter-client==8.6.3
     # via
@@ -136,19 +132,19 @@ jupyter-server==2.14.2
     #   notebook-shim
 jupyter-server-terminals==0.5.3
     # via jupyter-server
-jupyterlab==4.2.5
+jupyterlab==4.3.1
     # via -r dev_requirements.in
 jupyterlab-code-formatter==3.0.2
     # via -r dev_requirements.in
 jupyterlab-pygments==0.3.0
     # via nbconvert
 jupyterlab-server==2.27.3
     # via jupyterlab
-keyring==25.4.1
+keyring==25.5.0
     # via twine
 markdown-it-py==3.0.0
     # via rich
-markupsafe==2.1.5
+markupsafe==3.0.2
     # via
     #   jinja2
     #   nbconvert
@@ -183,7 +179,7 @@ notebook-shim==0.2.4
     # via jupyterlab
 overrides==7.7.0
     # via jupyter-server
-packaging==24.1
+packaging==24.2
     # via
     #   black
     #   ipykernel
@@ -206,9 +202,9 @@ platformdirs==4.3.6
     #   jupyter-core
 prometheus-client==0.21.0
     # via jupyter-server
-prompt-toolkit==3.0.47
+prompt-toolkit==3.0.48
     # via ipython
-psutil==6.0.0
+psutil==6.1.0
     # via ipykernel
 pure-eval==0.2.3
     # via stack-data
@@ -226,11 +222,11 @@ python-dateutil==2.9.0.post0
     #   jupyter-client
 python-json-logger==2.0.7
     # via jupyter-events
-pywin32==306
+pywin32==308
     # via jupyter-core
 pywin32-ctypes==0.2.3
     # via keyring
-pywinpty==2.0.13
+pywinpty==2.0.14
     # via
     #   jupyter-server
     #   jupyter-server-terminals
@@ -266,20 +262,21 @@ rfc3986-validator==0.1.1
     # via
     #   jsonschema
     #   jupyter-events
-rich==13.8.1
+rich==13.9.4
     # via twine
-rpds-py==0.20.0
+rpds-py==0.21.0
     # via
     #   jsonschema
     #   referencing
-ruff==0.6.6
+ruff==0.8.0
     # via -r dev_requirements.in
 send2trash==1.8.3
     # via jupyter-server
+setuptools==75.6.0
+    # via jupyterlab
 six==1.16.0
     # via
     #   asttokens
-    #   bleach
     #   python-dateutil
     #   rfc3339-validator
 sniffio==1.3.1
@@ -294,9 +291,9 @@ terminado==0.18.1
     # via
     #   jupyter-server
     #   jupyter-server-terminals
-tinycss2==1.3.0
+tinycss2==1.4.0
     # via nbconvert
-tornado==6.4.1
+tornado==6.4.2
     # via
     #   ipykernel
     #   jupyter-client
@@ -319,7 +316,7 @@ traitlets==5.14.3
     #   nbformat
 twine==5.1.1
     # via -r dev_requirements.in
-types-python-dateutil==2.9.0.20240906
+types-python-dateutil==2.9.0.20241003
     # via arrow
 uri-template==1.3.0
     # via jsonschema
@@ -329,16 +326,13 @@ urllib3==2.2.3
     #   twine
 wcwidth==0.2.13
     # via prompt-toolkit
-webcolors==24.8.0
+webcolors==24.11.1
     # via jsonschema
 webencodings==0.5.1
     # via
     #   bleach
     #   tinycss2
 websocket-client==1.8.0
     # via jupyter-server
-zipp==3.20.2
+zipp==3.21.0
     # via importlib-metadata
-
-# The following packages are considered to be unsafe in a requirements file:
-# setuptools