add basic docker build

.dockerignore

@@ -0,0 +1,2 @@
+/test_data
+/target

.gitignore

@@ -2,3 +2,4 @@
 ff_fasttext/*.so
 __pycache__
 *.egg-info
+/taxonomy.json

.gitlab-ci.yml

@@ -1,13 +1,35 @@
-# You can override the included template(s) by including variable overrides
-# SAST customization: https://docs.gitlab.com/ee/user/application_security/sast/#customizing-the-sast-settings
-# Secret Detection customization: https://docs.gitlab.com/ee/user/application_security/secret_detection/#customizing-settings
-# Dependency Scanning customization: https://docs.gitlab.com/ee/user/application_security/dependency_scanning/#customizing-the-dependency-scanning-settings
-# Container Scanning customization: https://docs.gitlab.com/ee/user/application_security/container_scanning/#customizing-the-container-scanning-settings
-# Note that environment variables can be set in several places
-# See https://docs.gitlab.com/ee/ci/variables/#cicd-variable-precedence
+# https://docs.openfaas.com/reference/cicd/gitlab/
+
+variables:
+  CONTAINER_TAG_IMAGE: $CI_REGISTRY_IMAGE:build-$CI_BUILD_REF_NAME
+  CONTAINER_IMAGE: $CI_REGISTRY_IMAGE:build-$CI_PIPELINE_ID
+  CONTAINER_RELEASE_IMAGE: $CI_REGISTRY_IMAGE:latest
+  DOCKER_DRIVER: overlay
+  DOCKER_HOST: tcp://localhost:2375/
+
+
+services:
+- docker:18.09.7-dind
+
+before_script:
+- docker login -u gitlab-ci-token -p $CI_BUILD_TOKEN registry.gitlab.com
+
 stages:
-- test
-sast:
-  stage: test
-include:
-- template: Security/SAST.gitlab-ci.yml
+  - build
+  - deploy
+
+# Cache the templates and build-context to speed things up
+cache:
+  key: ${CI_COMMIT_REF_SLUG} # i.e. master
+  paths:
+  - ./faas-cli
+  - ./template
+
+# Build the whole stack using only the faas-cli
+build:
+  script:
+  - docker build -f Dockerfile -t $CONTAINER_IMAGE .
+  - docker push $CONTAINER_IMAGE
+  - docker tag $CONTAINER_IMAGE $CONTAINER_TAG_IMAGE && docker push $CONTAINER_TAG_IMAGE
+  only:
+    - main

Dockerfile

@@ -0,0 +1,52 @@
+FROM rust as rust_build
+
+RUN mkdir -p /app/build/ff_fasttext
+RUN mkdir -p /app/build/test_data
+WORKDIR /app/build/test_data
+RUN curl -L -O "http://www.sfs.uni-tuebingen.de/a3-public-data/finalfusion-fasttext/wiki/wiki.en.fifu"
+WORKDIR /app/build
+
+COPY Cargo.lock /app/build
+COPY Cargo.toml /app/build
+
+RUN apt-get update && apt-get install -y python3-pip liblapack-dev libatlas-base-dev
+
+RUN RUSTFLAGS="-C link-args=-lcblas -llapack" cargo install finalfusion-utils --features=opq
+
+COPY setup.py /app/build
+COPY setup.cfg /app/build
+COPY pyproject.toml /app/build
+COPY poetry.lock /app/build
+COPY requirements-dev.txt /app/build
+COPY src /app/build/src
+COPY ff_fasttext/__init__.py /app/build/ff_fasttext
+
+WORKDIR /app/build
+
+RUN pip install --no-cache-dir --upgrade -r requirements-dev.txt
+
+RUN python3 setup.py develop
+
+FROM python:3-bullseye
+
+RUN mkdir -p /app/ff_fasttext
+
+COPY --from=rust_build /app/build/ff_fasttext/_ff_fasttext.abi3.so /app/ff_fasttext
+COPY --from=rust_build /app/build/test_data /app/test_data
+
+RUN pip install poetry
+WORKDIR /app
+
+COPY setup.py /app
+COPY setup.cfg /app
+COPY pyproject.toml /app
+COPY poetry.lock /app
+COPY requirements-dev.txt /app
+COPY ff_fasttext /app/ff_fasttext
+# COPY taxonomy.json /app
+
+WORKDIR /app
+
+RUN poetry install
+
+CMD ["poetry", "run", "uvicorn", "ff_fasttext.server:app", "--host", "0.0.0.0", "--port", "80"]

ff_fasttext/server.py

@@ -1,16 +1,17 @@
-from fastapi import FastApi
+import logging
+from fastapi import FastAPI
 from .extract import load
 
 THRESHOLD = 0.4
 
 def make_app(category_manager):
-    app = FastApi()
+    app = FastAPI()
 
     @app.get('/categories')
     def get_categories(query: str):
         categories = category_manager.test(query.strip(), 'onyxcats')
         return [
-            c for c in categories if c[1] > THRESHOLD
+            [float(c[0]), list(c[1])] for c in categories if c[0] > THRESHOLD
         ]
 
     return app

ff_fasttext/taxonomy.py

@@ -8,8 +8,6 @@
 
 import json
 
-from .extract import append_discovered_terms_from_elasticsearch
-
 def get_taxonomy():
     with open('taxonomy.json', 'r') as f:
         taxonomy = json.load(f)

pyproject.toml

@@ -19,9 +19,10 @@ sortedcontainers = "^2.4.0"
 elasticsearch2 = "^2.5.1"
 elasticsearch-dsl = "2"
 click = "^8.0.3"
-fastapi = "^0.73.0"
+fastapi = "0.52"
 pydantic = "^1.9.0"
 uvicorn = "^0.17.4"
+numpy = "^1.22.2"
 
 [tool.poetry.scripts]
 ff_fasttext_cli = "ff_fasttext.scripts.ff_fasttext:main"

src/lib.rs

@@ -43,16 +43,6 @@ impl FfModel {
         success
     }
 
-    fn embedding_similarities(
-        self_: PyRef<Self>,
-        x: PyReadonlyArrayDyn<f32>,
-        y: PyReadonlyArrayDyn<f32>
-    ) -> f32 {
-        let py = self_.py();
-        let X = x.as_array();
-        X.sum() / (X.len() as f32)
-    }
-
     fn eval(self_: PyRef<Self>, haystack: &str) -> PyResult<()> {
         if let Some(embedding) = self_.embeddings.embedding(haystack) {
              println!("{:#?}", embedding);