chore(build): fix broken build logic

Standardizes project structure so the dockerbuild works

Dockerfile

@@ -19,10 +19,10 @@ ENV PATH="$POETRY_HOME/bin:$PATH"
 
 RUN curl -sSL https://install.python-poetry.org | python3 -
 
-COPY ./data ./data
+COPY ./src ./src
-COPY ./cv_analysis ./cv_analysis
+COPY ./config ./config
-COPY ./scripts ./scripts
+COPY pyproject.toml poetry.lock ./
-COPY pyproject.toml poetry.lock ./src ./
+
 
 RUN ls -hal
 RUN echo "${USERNAME} ${TOKEN} ${PYPI_REGISTRY_RED} ${POETRY_SOURCE_REF_RED} "
@@ -33,9 +33,9 @@ RUN poetry config virtualenvs.create false && \
     poetry config http-basic.${POETRY_SOURCE_REF_RESEARCH} ${USERNAME} ${TOKEN} && \
     poetry config repositories.${POETRY_SOURCE_REF_RED} ${PYPI_REGISTRY_RED} && \
     poetry config http-basic.${POETRY_SOURCE_REF_RED} ${USERNAME} ${TOKEN} && \
-    poetry install --without=test -vv --no-interaction --no-root
+    poetry install --without=test -vv --no-interaction
 
 EXPOSE 5000
 EXPOSE 8080
 
-CMD [ "python", "serve.py"]
+CMD [ "python", "src/serve.py"]

pyproject.toml

@@ -4,7 +4,7 @@ version = "2.0.0"
 description = ""
 authors = []
 readme = "README.md"
-packages = [{ include = "cv_analysis" }]
+packages = [{ include = "cv_analysis", from = "src" }]
 
 [tool.poetry.dependencies]
 python = ">=3.10,<3.11"

scripts/annotate_pdf.py

@@ -1,16 +1,13 @@
 import argparse
 import timeit
-from time import process_time
-from itertools import starmap
-from pathlib import Path
-
 from PIL import Image
-from funcy import lmap
-
 from cv_analysis.figure_detection.figure_detection import detect_figures
 from cv_analysis.layout_parsing import parse_layout
 from cv_analysis.table_parsing import parse_tables
 from cv_analysis.utils.draw import draw_rectangles
+from funcy import lmap
+from itertools import starmap
+from pathlib import Path
 from pdf2img.conversion import convert_pages_to_images
 
 

src/cv_analysis/config.py

@@ -4,6 +4,6 @@ from pyinfra.config.loader import load_settings
 
 
 def get_config():
-    local_root_path = Path(__file__).parents[1]
+    local_root_path = Path(__file__).parents[2]
     settings = load_settings(root_path=local_root_path, settings_path="config")
     return settings

src/serve.py

@@ -1,6 +1,5 @@
 from sys import stdout
 
-import IPython
 from kn_utils.logging import logger
 from pyinfra.examples import start_standard_queue_consumer
 from pyinfra.queue.callback import make_download_process_upload_callback