build: drop remaining Python 3.9 refs (#4049)

Dropped variables that said we support Python 3.9 in `setup.py`, as well
as any remaining references to Python 3.9.

I also checked the pins and removed several that don't seem necessary
any more.

Author: qued

CHANGELOG.md

@@ -1,7 +1,9 @@
-## 0.18.5-dev0
+## 0.18.5-dev1
 
 ### Enhancements
 
+- **Bump dependencies and remove lingering Python 3.9 artifacts** Cleaned up some references to 3.9 that were left When we dropped Python 3.9 support.
+
 ### Features
 
 ### Fixes

pyproject.toml

@@ -22,7 +22,7 @@ unstructured/nlp/patterns\.py
 
 [tool.pyright]
 pythonPlatform = "Linux"
-pythonVersion = "3.9"
+pythonVersion = "3.10"
 reportUnnecessaryCast = true
 reportUnnecessaryTypeIgnoreComment = true
 stubPath = "./typings"
@@ -31,7 +31,7 @@ verboseOutput = true
 
 [tool.ruff]
 line-length = 100
-target-version = "py39"
+target-version = "py310"
 
 [tool.ruff.lint]
 ignore = [

requirements/base.txt

@@ -10,7 +10,7 @@ backoff==2.2.1
     # via -r ./base.in
 beautifulsoup4==4.13.4
     # via -r ./base.in
-certifi==2025.6.15
+certifi==2025.7.9
     # via
     #   httpcore
     #   httpx
@@ -28,7 +28,7 @@ click==8.2.1
     # via
     #   nltk
     #   python-oxmsg
-cryptography==45.0.4
+cryptography==45.0.5
     # via unstructured-client
 dataclasses-json==0.6.7
     # via
@@ -62,7 +62,7 @@ jsonpath-python==1.0.6
     # via unstructured-client
 langdetect==1.0.9
     # via -r ./base.in
-lxml==5.4.0
+lxml==6.0.0
     # via -r ./base.in
 marshmallow==3.26.1
     # via
@@ -90,7 +90,7 @@ psutil==7.0.0
     # via -r ./base.in
 pycparser==2.22
     # via cffi
-pypdf==5.6.0
+pypdf==5.7.0
     # via unstructured-client
 python-dateutil==2.9.0.post0
     # via unstructured-client
@@ -125,7 +125,7 @@ tqdm==4.67.1
     # via
     #   -r ./base.in
     #   nltk
-typing-extensions==4.14.0
+typing-extensions==4.14.1
     # via
     #   -r ./base.in
     #   anyio
@@ -143,7 +143,7 @@ unstructured-client==0.25.9
     # via
     #   -c requirements/deps/constraints.txt
     #   -r ./base.in
-urllib3==2.4.0
+urllib3==2.5.0
     # via
     #   -c requirements/deps/constraints.txt
     #   requests

requirements/deps/constraints.txt

@@ -5,19 +5,10 @@
 ####################################################################################################
 # we are using v3 client https://weaviate.io/developers/weaviate/client-libraries/python/python_v3
 weaviate-client>=3.26.7,<4.0.0
-# TODO: Constriant due to multiple versions being installed during pip-compile
-protobuf>=6.30.0
-# TODO: Constriant due to multiple versions being installed during pip-compile
-grpcio>=1.65.5
 # TODO: Pinned in transformers package, remove when that gets updated (https://github.com/huggingface/transformers/blob/main/setup.py)
 tokenizers>=0.21,<0.22
-# TODO: Constaint due to boto, with python before 3.10 not requiring openssl 1.1.1, remove when that gets
-# updated or we drop support for 3.9
+# NOTE(alan): Okay to drop pin once this version exists and we verify compatibility.
 urllib3<3.0.0
-# TODO: Constriant due to aiobotocore, remove when that gets updates:
-botocore<1.34.132
-# TODO: Constriant due to both 8.5.0 and 8.4.0 being installed during pip-compile
-importlib-metadata>=8.5.0
 # (austin): Versions below this have a different interface for passing parameters
 unstructured-client>=0.23.0,<0.26.0
 # paddle constrains protobuf; maybe we should put paddle here since its version is pinned in .in file

requirements/extra-csv.txt

@@ -8,7 +8,7 @@ numpy==2.2.6
     # via
     #   -c requirements/base.txt
     #   pandas
-pandas==2.3.0
+pandas==2.3.1
     # via -r ./extra-csv.in
 python-dateutil==2.9.0.post0
     # via

requirements/extra-docx.txt

@@ -4,13 +4,13 @@
 #
 #    pip-compile ./extra-docx.in
 #
-lxml==5.4.0
+lxml==6.0.0
     # via
     #   -c requirements/base.txt
     #   python-docx
 python-docx==1.2.0
     # via -r ./extra-docx.in
-typing-extensions==4.14.0
+typing-extensions==4.14.1
     # via
     #   -c requirements/base.txt
     #   python-docx

requirements/extra-markdown.txt

@@ -4,5 +4,5 @@
 #
 #    pip-compile ./extra-markdown.in
 #
-markdown==3.8
+markdown==3.8.2
     # via -r ./extra-markdown.in

requirements/extra-odt.txt

@@ -4,15 +4,15 @@
 #
 #    pip-compile ./extra-odt.in
 #
-lxml==5.4.0
+lxml==6.0.0
     # via
     #   -c requirements/base.txt
     #   python-docx
 pypandoc==1.15
     # via -r ./extra-odt.in
 python-docx==1.2.0
     # via -r ./extra-odt.in
-typing-extensions==4.14.0
+typing-extensions==4.14.1
     # via
     #   -c requirements/base.txt
     #   python-docx

requirements/extra-paddleocr.txt

@@ -16,13 +16,11 @@ anyio==4.9.0
     # via
     #   -c requirements/base.txt
     #   httpx
-astor==0.8.1
-    # via paddlepaddle
 beautifulsoup4==4.13.4
     # via
     #   -c requirements/base.txt
     #   unstructured-paddleocr
-certifi==2025.6.15
+certifi==2025.7.9
     # via
     #   -c requirements/base.txt
     #   httpcore
@@ -42,7 +40,7 @@ exceptiongroup==1.3.0
     #   anyio
 fire==0.7.0
     # via unstructured-paddleocr
-fonttools==4.58.4
+fonttools==4.58.5
     # via unstructured-paddleocr
 h11==0.16.0
     # via
@@ -66,7 +64,7 @@ imageio==2.37.0
     # via scikit-image
 lazy-loader==0.4
     # via scikit-image
-lxml==5.4.0
+lxml==6.0.0
     # via
     #   -c requirements/base.txt
     #   python-docx
@@ -90,11 +88,11 @@ numpy==2.2.6
     #   shapely
     #   tifffile
     #   unstructured-paddleocr
-opencv-contrib-python==4.11.0.86
+opencv-contrib-python==4.12.0.88
     # via unstructured-paddleocr
-opencv-python==4.11.0.86
+opencv-python==4.12.0.88
     # via unstructured-paddleocr
-opencv-python-headless==4.11.0.86
+opencv-python-headless==4.12.0.88
     # via
     #   albucore
     #   albumentations
@@ -105,7 +103,7 @@ packaging==25.0
     #   -c requirements/base.txt
     #   lazy-loader
     #   scikit-image
-paddlepaddle==3.0.0
+paddlepaddle==3.1.0
     # via -r ./extra-paddleocr.in
 pillow==11.3.0
     # via
@@ -145,7 +143,7 @@ scipy==1.15.3
     #   scikit-image
 shapely==2.1.1
     # via unstructured-paddleocr
-simsimd==6.4.9
+simsimd==6.5.0
     # via albucore
 sniffio==1.3.1
     # via
@@ -165,7 +163,7 @@ tqdm==4.67.1
     # via
     #   -c requirements/base.txt
     #   unstructured-paddleocr
-typing-extensions==4.14.0
+typing-extensions==4.14.1
     # via
     #   -c requirements/base.txt
     #   anyio
@@ -180,7 +178,7 @@ typing-inspection==0.4.1
     # via pydantic
 unstructured-paddleocr==2.10.0
     # via -r ./extra-paddleocr.in
-urllib3==2.4.0
+urllib3==2.5.0
     # via
     #   -c requirements/base.txt
     #   -c requirements/deps/constraints.txt

requirements/extra-pdf-image.txt

@@ -4,13 +4,13 @@
 #
 #    pip-compile ./extra-pdf-image.in
 #
-accelerate==1.7.0
+accelerate==1.8.1
     # via unstructured-inference
 antlr4-python3-runtime==4.9.3
     # via omegaconf
 cachetools==5.5.2
     # via google-auth
-certifi==2025.6.15
+certifi==2025.7.9
     # via
     #   -c requirements/base.txt
     #   requests
@@ -27,7 +27,7 @@ coloredlogs==15.0.1
     # via onnxruntime
 contourpy==1.3.2
     # via matplotlib
-cryptography==45.0.4
+cryptography==45.0.5
     # via
     #   -c requirements/base.txt
     #   pdfminer-six
@@ -44,7 +44,7 @@ filelock==3.18.0
     #   transformers
 flatbuffers==25.2.10
     # via onnxruntime
-fonttools==4.58.4
+fonttools==4.58.5
     # via matplotlib
 fsspec==2025.5.1
     # via
@@ -62,16 +62,16 @@ googleapis-common-protos==1.70.0
     # via
     #   google-api-core
     #   grpcio-status
-grpcio==1.73.0
+grpcio==1.73.1
     # via
     #   -c requirements/deps/constraints.txt
     #   google-api-core
     #   grpcio-status
-grpcio-status==1.73.0
+grpcio-status==1.73.1
     # via google-api-core
-hf-xet==1.1.4
+hf-xet==1.1.5
     # via huggingface-hub
-huggingface-hub==0.33.0
+huggingface-hub==0.33.2
     # via
     #   accelerate
     #   timm
@@ -88,7 +88,7 @@ jinja2==3.1.6
     # via torch
 kiwisolver==1.4.8
     # via matplotlib
-lxml==5.4.0
+lxml==6.0.0
     # via
     #   -c requirements/base.txt
     #   pikepdf
@@ -125,7 +125,7 @@ onnxruntime==1.22.0
     # via
     #   -r ./extra-pdf-image.in
     #   unstructured-inference
-opencv-python==4.11.0.86
+opencv-python==4.12.0.88
     # via unstructured-inference
 packaging==25.0
     # via
@@ -137,7 +137,7 @@ packaging==25.0
     #   pikepdf
     #   transformers
     #   unstructured-pytesseract
-pandas==2.3.0
+pandas==2.3.1
     # via unstructured-inference
 pdf2image==1.17.0
     # via -r ./extra-pdf-image.in
@@ -146,9 +146,9 @@ pdfminer-six==20250327
     #   -c requirements/deps/constraints.txt
     #   -r ./extra-pdf-image.in
     #   unstructured-inference
-pi-heif==0.22.0
+pi-heif==1.0.0
     # via -r ./extra-pdf-image.in
-pikepdf==9.8.1
+pikepdf==9.9.0
     # via -r ./extra-pdf-image.in
 pillow==11.3.0
     # via
@@ -190,7 +190,7 @@ pycparser==2.22
     #   cffi
 pyparsing==3.2.3
     # via matplotlib
-pypdf==5.6.0
+pypdf==5.7.0
     # via
     #   -c requirements/base.txt
     #   -r ./extra-pdf-image.in
@@ -243,11 +243,11 @@ sympy==1.14.0
     # via
     #   onnxruntime
     #   torch
-timm==1.0.15
+timm==1.0.16
     # via
     #   effdet
     #   unstructured-inference
-tokenizers==0.21.1
+tokenizers==0.21.2
     # via
     #   -c requirements/deps/constraints.txt
     #   transformers
@@ -267,9 +267,9 @@ tqdm==4.67.1
     #   -c requirements/base.txt
     #   huggingface-hub
     #   transformers
-transformers==4.52.4
+transformers==4.53.1
     # via unstructured-inference
-typing-extensions==4.14.0
+typing-extensions==4.14.1
     # via
     #   -c requirements/base.txt
     #   huggingface-hub
@@ -282,7 +282,7 @@ unstructured-inference==1.0.5
     # via -r ./extra-pdf-image.in
 unstructured-pytesseract==0.3.15
     # via -r ./extra-pdf-image.in
-urllib3==2.4.0
+urllib3==2.5.0
     # via
     #   -c requirements/base.txt
     #   -c requirements/deps/constraints.txt