add: readme & fix: remove dep

.github/workflows/backend-core-tests.yml

@@ -39,6 +39,5 @@ jobs:
           OPENAI_API_KEY: this-is-a-test-key
         run: |
           sudo apt-get update
-          sudo apt-get install -y libmagic-dev poppler-utils libreoffice tesseract-ocr  pandoc
           cd core
           rye test -p quivr-core

core/Dockerfile.test

@@ -4,25 +4,15 @@ FROM python:3.11.6-slim-bullseye
 # Install GEOS library, Rust, and other dependencies, then clean up
 RUN apt-get clean && apt-get update && apt-get install -y \
     curl \
-    gcc \
-    autoconf \
-    automake \
     build-essential \
-    # Additional dependencies for document handling
-    libmagic-dev \
-    tesseract-ocr \
-    poppler-utils \
-    libreoffice \
-    pandoc && \
-    rm -rf /var/lib/apt/lists/*
-
+    libgeos-dev \
+    && apt-get clean && rm -rf /var/lib/apt/lists/*
 # Set the working directory
 WORKDIR /code
 
-# Install Poetry
-RUN curl -sSL https://install.python-poetry.org | POETRY_HOME=/opt/poetry python && \
+RUN curl -sSL https://install.python-poetry.org | python3 - && \
     cd /usr/local/bin && \
-    ln -s /opt/poetry/bin/poetry && \
+    ln -s /root/.local/bin/poetry && \
     poetry config virtualenvs.create false
 
 # Add Poetry to PATH

core/README.md

@@ -12,6 +12,18 @@ This project is licensed under the Apache 2.0 License
 pip install quivr-core
 ```
 
+### Use with [MegaParse](https://github.com/QuivrHQ/MegaParse)
+
+By **default** megaparse-core uses MegaParse to parse files, in order to use it we need to either :
+
+**Use Quivr hosted Megaparse API**
+* Send email to **[email protected]** with object  : "[Megaparse] API Key Request"
+* Add ```MEGAPARSE_API_KEY = md-...``` in your *.env* file
+
+**Use self hosted Megaparse API**
+* Clone https://github.com/QuivrHQ/MegaParse
+* Run the api with ```make dev``` in ```cd MegaParse```
+* Back in quivr-core add ```MEGAPARSE_URL = http://localhost:8000``` in your env variables
 
 
 

core/pyproject.toml

@@ -60,7 +60,6 @@ markers = [
     "slow: marks tests as slow (deselect with '-m \"not slow\"')",
     "base: these tests require quivr-core with extra `base` to be installed",
     "tika: these tests require a tika server to be running",
-    "unstructured: these tests require `unstructured` dependency",
 ]
 
 [[tool.mypy.overrides]]