Bump packages, clean up uv commands (#564)

<!-- CURSOR_SUMMARY -->
> [!NOTE]
> **Medium Risk**
> Moderate risk because this updates the runtime dependency set and
changes build/CI/Docker provisioning (uv lock enforcement and spaCy
model preloading), which can cause install or runtime regressions if the
new `unstructured`/model behavior differs.
> 
> **Overview**
> Bumps the release to `0.1.2` and refreshes dependencies (including
adding `python-multipart`), aligning with an `unstructured` update that
replaces NLTK usage with spaCy.
> 
> Updates Dockerfile, Makefile, and GitHub workflows to **pre-download
spaCy models** (replacing `download_nltk_packages`) and standardizes
dependency installs by switching `uv sync` from `--frozen` to `--locked`
across CI, Docker, and local install targets.
> 
> <sup>Written by [Cursor
Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit
d9d6362. This will update automatically
on new commits. Configure
[here](https://cursor.com/dashboard?tab=bugbot).</sup>
<!-- /CURSOR_SUMMARY -->

Author: PastelStorm

.github/workflows/ci.yml

@@ -24,7 +24,7 @@ jobs:
     - name: Set up Python ${{ env.PYTHON_VERSION }}
       run: uv python install ${{ env.PYTHON_VERSION }}
     - name: Install lint dependencies
-      run: uv sync --only-group lint --frozen
+      run: uv sync --only-group lint --locked
     - name: Lint
       run: make check
 
@@ -52,9 +52,9 @@ jobs:
     - name: Install dependencies and run core tests
       run: |
         sudo apt-get update && sudo apt-get install --yes poppler-utils libreoffice
-        uv sync --group test --frozen
+        uv sync --group test --locked
         make install-pandoc
-        make install-nltk-models
+        make install-nlp-models
         sudo add-apt-repository -y ppa:alex-p/tesseract-ocr5
         sudo apt-get install -y tesseract-ocr tesseract-ocr-kor
         tesseract --version
@@ -114,6 +114,6 @@ jobs:
         df -h
     - name: Test Dockerfile
       run: |
-        uv sync --group test --frozen
+        uv sync --group test --locked
         make docker-build
         make docker-test

.github/workflows/docker-publish.yml

@@ -83,7 +83,7 @@ jobs:
     - name: Set up Python ${{ env.PYTHON_VERSION }}
       run: uv python install ${{ env.PYTHON_VERSION }}
     - name: Install test dependencies
-      run: uv sync --group test --frozen
+      run: uv sync --group test --locked
     - name: Test image
       run: |
         export DOCKER_IMAGE="$DOCKER_BUILD_REPOSITORY:${{ matrix.arch }}-$SHORT_SHA"

CHANGELOG.md

@@ -1,3 +1,8 @@
+## 0.1.2
+* Bump all packages (refresh uv.lock), pulling `unstructured==0.22.12` which replaces NLTK with spaCy
+* Replace `download_nltk_packages` calls with spaCy model pre-download in Makefile, Dockerfile, and CI
+* Switch `uv sync --frozen` to `uv sync --locked` across Dockerfile, Makefile, and CI workflows
+
 ## 0.1.1
 * Switch arm64 Docker build runner from custom `opensource-linux-arm64-4core` to GitHub-hosted `ubuntu-24.04-arm`
 * Consolidate multiarch Docker manifest creation into a single `docker buildx imagetools create` call

Dockerfile

@@ -61,7 +61,7 @@ ENV UV_PROJECT_ENVIRONMENT="${HOME}/.local"
 
 COPY --chown=${NB_USER}:${NB_USER} pyproject.toml pyproject.toml
 COPY --chown=${NB_USER}:${NB_USER} uv.lock uv.lock
-RUN uv sync --no-dev --no-install-project --frozen
+RUN uv sync --no-dev --no-install-project --locked
 
 ARG PANDOC_VERSION="3.9"
 RUN ARCH=$(uname -m) && \
@@ -71,7 +71,7 @@ RUN ARCH=$(uname -m) && \
     cp /tmp/pandoc-${PANDOC_VERSION}/bin/pandoc /home/${USER}/.local/bin/ && \
     rm -rf /tmp/pandoc*
 
-RUN ${PYTHON} -c "from unstructured.nlp.tokenize import download_nltk_packages; download_nltk_packages()" && \
+RUN ${PYTHON} -c "from unstructured.nlp.tokenize import _load_spacy_model; _load_spacy_model()" && \
     ${PYTHON} -c "from unstructured.partition.model_init import initialize; initialize()" && \
     ${PYTHON} -c "from unstructured_inference.models.tables import UnstructuredTableTransformerModel; model = UnstructuredTableTransformerModel(); model.initialize('microsoft/table-transformer-structure-recognition')"
 

Makefile

@@ -13,23 +13,23 @@ help: Makefile
 
 ## install-base:                installs minimum requirements to run the API
 .PHONY: install-base
-install-base: install-base-packages install-nltk-models
+install-base: install-base-packages install-nlp-models
 
 ## install:                     installs all test and dev requirements
 .PHONY: install
 install: install-base install-test
 
 .PHONY: install-base-packages
 install-base-packages:
-	uv sync --no-dev --frozen
+	uv sync --no-dev --locked
 
 .PHONY: install-test
 install-test:
-	uv sync --group test --frozen
+	uv sync --group test --locked
 
-.PHONY: install-nltk-models
-install-nltk-models:
-	uv run python -c "from unstructured.nlp.tokenize import download_nltk_packages; download_nltk_packages()"
+.PHONY: install-nlp-models
+install-nlp-models:
+	uv run python -c "from unstructured.nlp.tokenize import _load_spacy_model; _load_spacy_model()"
 
 ## lock:                        regenerates uv.lock
 .PHONY: lock

prepline_general/api/__version__.py

@@ -1 +1 @@
-__version__ = "0.1.1"  # pragma: no cover
+__version__ = "0.1.2"  # pragma: no cover

prepline_general/api/general.py

@@ -145,7 +145,7 @@ def partition_file_via_api(
     if not request_url:
         raise HTTPException(status_code=500, detail="Parallel mode enabled but no url set!")
 
-    api_key = request.headers.get("unstructured-api-key", default="")
+    api_key = request.headers.get("unstructured-api-key", "")
     partition_kwargs["starting_page_number"] = (
         partition_kwargs.get("starting_page_number", 1) + page_offset
     )

pyproject.toml

@@ -6,6 +6,7 @@ requires-python = ">=3.12"
 dependencies = [
     "unstructured[all-docs] >=0.18.31, <1.0.0",
     "fastapi >=0.128.4, <1.0.0",
+    "python-multipart >=0.0.18",
     "uvicorn >=0.40.0, <1.0.0",
     "backoff >=2.2.1, <3.0.0",
     "pandas >=3.0.0, <4.0.0",