diff --git a/requirements/base.in b/requirements/base.in index ff38c57bb6..c9c328d9ea 100644 --- a/requirements/base.in +++ b/requirements/base.in @@ -23,3 +23,4 @@ psutil python-oxmsg html5lib ndjson +langchain-text-splitters \ No newline at end of file diff --git a/requirements/base.txt b/requirements/base.txt index 6fecb30c04..1dfc435585 100644 --- a/requirements/base.txt +++ b/requirements/base.txt @@ -4,7 +4,9 @@ # # pip-compile ./base.in # -anyio==4.7.0 +annotated-types==0.7.0 + # via pydantic +anyio==4.8.0 # via httpx backoff==2.2.1 # via -r ./base.in @@ -49,7 +51,9 @@ html5lib==1.1 httpcore==1.0.7 # via httpx httpx==0.28.1 - # via unstructured-client + # via + # langsmith + # unstructured-client idna==3.10 # via # anyio @@ -58,13 +62,23 @@ idna==3.10 # unstructured-client joblib==1.4.2 # via nltk +jsonpatch==1.33 + # via langchain-core jsonpath-python==1.0.6 # via unstructured-client +jsonpointer==3.0.0 + # via jsonpatch +langchain-core==0.3.29 + # via langchain-text-splitters +langchain-text-splitters==0.3.5 + # via -r ./base.in langdetect==1.0.9 # via -r ./base.in +langsmith==0.2.10 + # via langchain-core lxml==5.3.0 # via -r ./base.in -marshmallow==3.23.2 +marshmallow==3.25.1 # via # dataclasses-json # unstructured-client @@ -84,14 +98,23 @@ olefile==0.47 # via python-oxmsg orderly-set==5.2.3 # via deepdiff +orjson==3.10.14 + # via langsmith packaging==24.2 # via + # langchain-core # marshmallow # unstructured-client psutil==6.1.1 # via -r ./base.in pycparser==2.22 # via cffi +pydantic==2.10.5 + # via + # langchain-core + # langsmith +pydantic-core==2.27.2 + # via pydantic pypdf==5.1.0 # via unstructured-client python-dateutil==2.9.0.post0 @@ -102,6 +125,8 @@ python-magic==0.4.27 # via -r ./base.in python-oxmsg==0.0.1 # via -r ./base.in +pyyaml==6.0.2 + # via langchain-core rapidfuzz==3.11.0 # via -r ./base.in regex==2024.11.6 @@ -109,10 +134,13 @@ regex==2024.11.6 requests==2.32.3 # via # -r ./base.in + # langsmith # requests-toolbelt # unstructured-client requests-toolbelt==1.0.0 - # via unstructured-client + # via + # langsmith + # unstructured-client six==1.17.0 # via # html5lib @@ -123,6 +151,8 @@ sniffio==1.3.1 # via anyio soupsieve==2.6 # via beautifulsoup4 +tenacity==9.0.0 + # via langchain-core tqdm==4.67.1 # via # -r ./base.in @@ -131,6 +161,9 @@ typing-extensions==4.12.2 # via # -r ./base.in # anyio + # langchain-core + # pydantic + # pydantic-core # pypdf # python-oxmsg # typing-inspect @@ -150,5 +183,5 @@ urllib3==1.26.20 # unstructured-client webencodings==0.5.1 # via html5lib -wrapt==1.17.0 +wrapt==1.17.2 # via -r ./base.in diff --git a/requirements/dev.txt b/requirements/dev.txt index 30e42eb0ac..eced667c12 100644 --- a/requirements/dev.txt +++ b/requirements/dev.txt @@ -17,7 +17,7 @@ distlib==0.3.9 # via virtualenv filelock==3.16.1 # via virtualenv -identify==2.6.4 +identify==2.6.5 # via pre-commit importlib-metadata==8.5.0 # via @@ -44,6 +44,7 @@ pyproject-hooks==1.2.0 # pip-tools pyyaml==6.0.2 # via + # -c ./base.txt # -c ./test.txt # pre-commit tomli==2.2.1 diff --git a/requirements/extra-epub.txt b/requirements/extra-epub.txt index a9533059da..460408c418 100644 --- a/requirements/extra-epub.txt +++ b/requirements/extra-epub.txt @@ -4,5 +4,5 @@ # # pip-compile ./extra-epub.in # -pypandoc==1.14 +pypandoc==1.15 # via -r ./extra-epub.in diff --git a/requirements/extra-odt.txt b/requirements/extra-odt.txt index 28ebf301a6..362c53ed74 100644 --- a/requirements/extra-odt.txt +++ b/requirements/extra-odt.txt @@ -8,7 +8,7 @@ lxml==5.3.0 # via # -c ./base.txt # python-docx -pypandoc==1.14 +pypandoc==1.15 # via -r ./extra-odt.in python-docx==1.1.2 # via -r ./extra-odt.in diff --git a/requirements/extra-paddleocr.txt b/requirements/extra-paddleocr.txt index d7c0fe7226..3a47b09058 100644 --- a/requirements/extra-paddleocr.txt +++ b/requirements/extra-paddleocr.txt @@ -4,7 +4,7 @@ # # pip-compile ./extra-paddleocr.in # -anyio==4.7.0 +anyio==4.8.0 # via # -c ./base.txt # httpx @@ -58,7 +58,7 @@ imageio==2.36.1 # scikit-image imgaug==0.4.0 # via unstructured-paddleocr -importlib-resources==6.5.1 +importlib-resources==6.5.2 # via matplotlib kiwisolver==1.4.7 # via matplotlib @@ -126,7 +126,9 @@ python-dateutil==2.9.0.post0 # -c ./base.txt # matplotlib pyyaml==6.0.2 - # via unstructured-paddleocr + # via + # -c ./base.txt + # unstructured-paddleocr rapidfuzz==3.11.0 # via # -c ./base.txt diff --git a/requirements/extra-pandoc.txt b/requirements/extra-pandoc.txt index 4125059733..dd397c3845 100644 --- a/requirements/extra-pandoc.txt +++ b/requirements/extra-pandoc.txt @@ -4,5 +4,5 @@ # # pip-compile ./extra-pandoc.in # -pypandoc==1.14 +pypandoc==1.15 # via -r ./extra-pandoc.in diff --git a/requirements/extra-pdf-image.txt b/requirements/extra-pdf-image.txt index 81b61276ef..822992953a 100644 --- a/requirements/extra-pdf-image.txt +++ b/requirements/extra-pdf-image.txt @@ -60,14 +60,14 @@ googleapis-common-protos==1.66.0 # via # google-api-core # grpcio-status -grpcio==1.68.1 +grpcio==1.69.0 # via # -c ././deps/constraints.txt # google-api-core # grpcio-status grpcio-status==1.62.3 # via google-api-core -huggingface-hub==0.27.0 +huggingface-hub==0.27.1 # via # timm # tokenizers @@ -79,7 +79,7 @@ idna==3.10 # via # -c ./base.txt # requests -importlib-resources==6.5.1 +importlib-resources==6.5.2 # via matplotlib iopath==0.1.10 # via layoutparser @@ -153,7 +153,7 @@ pdfplumber==0.11.5 # via layoutparser pi-heif==0.21.0 # via -r ./extra-pdf-image.in -pikepdf==9.5.0 +pikepdf==9.5.1 # via -r ./extra-pdf-image.in pillow==11.1.0 # via @@ -212,6 +212,7 @@ pytz==2024.2 # via pandas pyyaml==6.0.2 # via + # -c ./base.txt # huggingface-hub # layoutparser # omegaconf @@ -233,7 +234,7 @@ requests==2.32.3 # transformers rsa==4.9 # via google-auth -safetensors==0.5.0 +safetensors==0.5.2 # via # timm # transformers @@ -247,7 +248,7 @@ sympy==1.13.1 # via # onnxruntime # torch -timm==1.0.12 +timm==1.0.13 # via # effdet # unstructured-inference @@ -291,7 +292,7 @@ urllib3==1.26.20 # -c ././deps/constraints.txt # -c ./base.txt # requests -wrapt==1.17.0 +wrapt==1.17.2 # via # -c ./base.txt # deprecated diff --git a/requirements/huggingface.txt b/requirements/huggingface.txt index 7051a2233b..698140cda0 100644 --- a/requirements/huggingface.txt +++ b/requirements/huggingface.txt @@ -25,7 +25,7 @@ fsspec==2024.12.0 # via # huggingface-hub # torch -huggingface-hub==0.27.0 +huggingface-hub==0.27.1 # via # tokenizers # transformers @@ -60,6 +60,7 @@ packaging==24.2 # transformers pyyaml==6.0.2 # via + # -c ./base.txt # huggingface-hub # transformers regex==2024.11.6 @@ -74,7 +75,7 @@ requests==2.32.3 # transformers sacremoses==0.1.1 # via -r ./huggingface.in -safetensors==0.5.0 +safetensors==0.5.2 # via transformers sentencepiece==0.2.0 # via -r ./huggingface.in diff --git a/requirements/test.txt b/requirements/test.txt index a7e1d2cfa2..29a5a5105f 100644 --- a/requirements/test.txt +++ b/requirements/test.txt @@ -5,8 +5,10 @@ # pip-compile ./test.in # annotated-types==0.7.0 - # via pydantic -anyio==4.7.0 + # via + # -c ./base.txt + # pydantic +anyio==4.8.0 # via # -c ./base.txt # httpx @@ -54,7 +56,7 @@ exceptiongroup==1.2.2 # -c ./base.txt # anyio # pytest -faker==33.1.0 +faker==33.3.1 # via jsf flake8==7.1.1 # via @@ -66,7 +68,7 @@ freezegun==1.5.1 # via -r ./test.in genson==1.3.0 # via datamodel-code-generator -grpcio==1.68.1 +grpcio==1.69.0 # via # -c ././deps/constraints.txt # -r ./test.in @@ -164,14 +166,17 @@ pycodestyle==2.12.1 # via # flake8 # flake8-print -pydantic[email]==2.10.4 +pydantic[email]==2.10.5 # via + # -c ./base.txt # -r ./test.in # datamodel-code-generator # jsf # label-studio-sdk pydantic-core==2.27.2 - # via pydantic + # via + # -c ./base.txt + # pydantic pyflakes==3.2.0 # via # autoflake @@ -194,6 +199,7 @@ pytz==2024.2 # via pandas pyyaml==6.0.2 # via + # -c ./base.txt # datamodel-code-generator # vcrpy referencing==0.35.1 @@ -218,7 +224,7 @@ rpds-py==0.22.3 # referencing rstr==3.2.2 # via jsf -ruff==0.8.5 +ruff==0.9.1 # via -r ./test.in semantic-version==2.10.0 # via liccheck @@ -281,7 +287,7 @@ urllib3==1.26.20 # vcrpy vcrpy==7.0.0 # via -r ./test.in -wrapt==1.17.0 +wrapt==1.17.2 # via # -c ./base.txt # smart-open