Dockerfile

# Dockerfile for development purposes.
# Read docs/development.md for more information.
# vi: ft=dockerfile

###############################################################################
# Base build image
FROM golang:1.21-alpine AS build_base
RUN apk add bash ca-certificates git gcc g++ libc-dev
WORKDIR /go/src/github.com/weaviate/weaviate
ENV GO111MODULE=on
# Populate the module cache based on the go.{mod,sum} files.
COPY go.mod .
COPY go.sum .
RUN go mod download

###############################################################################
# This image builds the weaviate server
FROM build_base AS server_builder
ARG TARGETARCH
ARG GITHASH="unknown"
ARG EXTRA_BUILD_ARGS=""
COPY . .
RUN CGO_ENABLED=0 GOARCH=$TARGETARCH go build $EXTRA_BUILD_ARGS \
      -a -ldflags '-w -extldflags "-static" -X github.com/weaviate/weaviate/usecases/config.GitHash='"$GITHASH"'' \
      -o /weaviate-server ./cmd/weaviate-server

###############################################################################
# This creates an image that can be used to fake an api for telemetry acceptance test purposes
FROM build_base AS telemetry_mock_api
COPY . .
ENTRYPOINT ["./tools/dev/telemetry_mock_api.sh"]

###############################################################################
# This image gets grpc health check probe
FROM build_base AS grpc_health_probe_builder
ARG TARGETARCH
RUN GRPC_HEALTH_PROBE_VERSION=v0.4.22 && \
      wget -qO/bin/grpc_health_probe https://github.com/grpc-ecosystem/grpc-health-probe/releases/download/${GRPC_HEALTH_PROBE_VERSION}/grpc_health_probe-linux-${TARGETARCH} && \
      chmod +x /bin/grpc_health_probe

###############################################################################
# Weaviate (no differentiation between dev/test/prod - 12 factor!)
FROM alpine AS weaviate
#ENTRYPOINT ["/bin/weaviate"]
COPY --from=grpc_health_probe_builder /bin/grpc_health_probe /bin/
COPY --from=server_builder /weaviate-server /bin/weaviate
RUN chmod 755 /bin/weaviate
RUN apk add --no-cache --upgrade ca-certificates openssl
RUN mkdir ./modules
RUN mkdir -p /var/lib/weaviate/data
RUN chmod -R 777 /var
ENV AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED true
ENV PERSISTENCE_DATA_PATH /var/lib/weaviate/data
#ENV ENABLE_MODULES text2vec-transformers
#CMD [ "--host", "0.0.0.0", "--port", "8080", "--scheme", "http", ""]

###############################################################################
# python environment and app.
FROM python:3.11.5
ENTRYPOINT ["/app/startup.sh"]
RUN apt update
WORKDIR /app

COPY ./requirements.txt /app/requirements.txt
COPY ./semsearch.py     /app/semsearch.py
COPY ./startup.sh       /app/startup.sh
RUN  chmod 755          /app/startup.sh

COPY --from=weaviate /bin/weaviate /app/weaviate
COPY --from=weaviate ./modules ./
COPY --from=server_builder /lib/libc.musl-x86_64.so.1 /lib
RUN  mkdir -p /var/lib/weaviate/data y
RUN  chmod -R 777 /var
COPY --from=weaviate /var/lib/weaviate/data /var/lib/weaviate/data
ENV AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED true
ENV AUTHENTICATION_OIDC_ENABLED false
ENV AUTHENTICATION_APIKEY_ENABLED false
ENV PERSISTENCE_DATA_PATH /var/lib/weaviate/data
#ENV ENABLE_MODULES text2vec-transformers
#ENV TRANSFORMERS_INFERENCE_API "http://127.0.0.1:999"

RUN mkdir -p /app/inputDocs
COPY ./inputDocs/*      /app/inputDocs

RUN pip install --no-cache-dir --upgrade -r /app/requirements.txt
RUN pip install https://files.pythonhosted.org/packages/13/87/e0cb08c2d4bd7d38ab63816b306c8b1e7cfdc0e59bd54462e8b0df069078/semantic_text_splitter-0.6.3-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
RUN pip show semantic-text-splitter
RUN useradd -m -u 1000 user

##############################################################################
# Start the weaviate vector database and the semantic search  app.
#RUN /app/startup.sh
CMD []