Merge pull request #66 from alkem-io/fix-ingest-in-docker

Fix ingest in Docker and bump version to 0.8.2

Dockerfile

@@ -9,38 +9,32 @@ ARG GO_VERSION=1.21.6
 ARG HUGO_VERSION=0.121.2
 ARG TARGETARCH
 
-# install git, go and hugo
-RUN apt update && apt upgrade -y && apt install -y git wget
+# install wget and git
+RUN apt-get update -y \
+  && apt-get upgrade -y \
+  && apt-get install -y git wget \
+  && apt-get clean \
+  && rm -rf /var/lib/apt/lists/*
+
 RUN wget https://go.dev/dl/go${GO_VERSION}.linux-${TARGETARCH}.tar.gz && tar -C /usr/local -xzf go${GO_VERSION}.linux-${TARGETARCH}.tar.gz 
-RUN export PATH=$PATH:/usr/local/go/bin:/usr/local && go version
-RUN wget https://github.com/gohugoio/hugo/releases/download/v${HUGO_VERSION}/hugo_extended_${HUGO_VERSION}_linux-${TARGETARCH}.tar.gz && tar -C /usr/local -xzf hugo_extended_${HUGO_VERSION}_linux-${TARGETARCH}.tar.gz && ls -al /usr/local
-RUN /usr/local/hugo version
+
+ENV PATH="/usr/local/go/bin:/usr/local:${PATH}"
+RUN go version
+
+RUN wget https://github.com/gohugoio/hugo/releases/download/v${HUGO_VERSION}/hugo_extended_${HUGO_VERSION}_linux-${TARGETARCH}.tar.gz \
+  && tar -C /bin -xzf hugo_extended_${HUGO_VERSION}_linux-${TARGETARCH}.tar.gz \
+  && rm hugo_extended_${HUGO_VERSION}_linux-${TARGETARCH}.tar.gz
+
+RUN hugo version
 
 # Install Poetry
 RUN pip install poetry
 
 # Copy the current directory contents into the container at /app
 COPY . /app
 
-# Use Poetry to install dependencies
+# # Use Poetry to install dependencies
 RUN poetry config virtualenvs.create true && poetry install --no-interaction --no-ansi
 
-# Start a new, final stage
-FROM python:${PYTHON_VERSION}-slim-bullseye
-
-WORKDIR /app
-
-# Install git and Poetry in the final stage
-RUN apt update && apt install -y git && pip install poetry
-
-# Copy the compiled app, Hugo executable, Go, and the virtual environment from the previous stage
-COPY --from=builder /app /app
-COPY --from=builder /usr/local/hugo /usr/local/hugo
-COPY --from=builder /usr/local/go /usr/local/go
-COPY --from=builder /root/.cache/pypoetry/virtualenvs /root/.cache/pypoetry/virtualenvs
-
-# Add Hugo and Go to the PATH
-ENV PATH="/usr/local/hugo:/usr/local/go/bin:${PATH}"
-
-# Run virtual_contributor_engine_guidance.py when the container launches
+# Run main.py when the container launches
 CMD ["poetry", "run", "python", "main.py"]

config.py

@@ -46,8 +46,8 @@ def __init__(self):
         self.site_url = os.getenv("AI_SOURCE_WEBSITE", "")
         self.welcome_site_url = os.getenv("AI_SOURCE_WEBSITE2", "")
 
-        self.site_repo = os.getenv("AI_WEBSITE_REPO", "")
-        self.welcome_site_repo = os.getenv("AI_WEBSITE_REPO2", "")
+        self.site_repo = f"https://{os.getenv('AI_GITHUB_PAT', '')}@{os.getenv('AI_WEBSITE_REPO', '')}"
+        self.welcome_site_repo = f"https://{os.getenv('AI_GITHUB_PAT', '')}@{os.getenv('AI_WEBSITE_REPO2', '')}"
 
         self.github_user = os.getenv("AI_GITHUB_USER", "")
         self.github_pat = os.getenv("AI_GITHUB_PAT", "")

ingest.py

@@ -45,10 +45,10 @@ def embed_text(documents: list[Document]):
 
 async def ensure_cloned(repo_name: str, path: str):
     if os.path.exists(path):
-        logger.info(f"Repository {repo_name} already cloned.")
+        logger.info(f"Repository {path} already cloned.")
         repo = Repo.init(path)
     else:
-        logger.info(f"Closning repository: {repo_name}")
+        logger.info(f"Cloning repository: {path}")
         repo = Repo.clone_from(repo_name, path, branch="main")
 
     repo.heads.main.checkout()

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "virtual-contributor-engine-guidance"
-version = "0.8.1"
+version = "0.8.2"
 description = "Alkemio Generative AI Guidance Engine"
 authors = ["Alkemio BV <[email protected]>"]
 license = "EUPL-1.2"