[nix] flake config added and working

.dockerignore.old

-*
-!requirements.txt
-!src/main.py
-!src/s2v_old/
\ No newline at end of file

.gitlab-ci.yml

@@ -44,7 +44,7 @@ docker-hub:
   stage: deploy
   image: docker:dind
   script: sh scripts/deploy.sh
-  timeout: 2h
+  #timeout: 2h
   only: [master]
 
 pages:
@@ -54,4 +54,4 @@ pages:
     - cp redoc-static.html public/index.html
   artifacts:
     paths: [public]
-  only: [master]
+  only: [master]
\ No newline at end of file

Dockerfile

-FROM python:3.9-slim-bullseye
-
-ENV VIRTUAL_ENV=/opt/venv
-RUN python3 -m venv $VIRTUAL_ENV
-ENV PATH="$VIRTUAL_ENV/bin:$PATH"
-
-WORKDIR /tmp
-
-# Install dependencies:
-COPY ./requirements.txt .
-RUN pip install -r requirements.txt -f https://download.pytorch.org/whl/torch_stable.html
-
-# Build args
-# https://docs.docker.com/build/guide/build-args/
-#ARG PORT 8001
-# https://spacy.io/usage/models/
-ARG MODEL en_core_web_trf
-
-RUN python -m spacy download ${MODEL}
-
-#RUN python -m spacy download en_core_web_trf
-#RUN python -m spacy download fr_dep_news_trf    # 8001
-#RUN python -m spacy download de_dep_news_trf   # 8002
-#RUN python -m spacy download uk_core_news_trf   # 8003
-#RUN python -m spacy download ru_core_news_trf   # 8004
-#RUN python -m spacy download pl_core_news_trf   # 8004
-#RUN python -m spacy download es_core_news_trf   # 8005
-#RUN python -m spacy download pt_core_news_trf   # 8006
-#RUN python -m spacy download it_core_news_trf   # 8007
-#RUN python -m spacy download zh_core_news_trf   # 8008
-#RUN python -m spacy download el_core_news_trf   # 8009 (greek)
-#RUN python -m spacy download xx_ent_wiki_trf    # 8010
-
-# nl_core_news_sm
-
-WORKDIR /code
-
-COPY src/main.py .
-
-ENV SPACY_MODEL=${MODEL}
-CMD uvicorn main:app --host 0.0.0.0 --port 8001

README.md

@@ -25,6 +25,14 @@ README!
 2. Optionally, generate a wrapper for the HTTP API using [OpenAPI Generator](https://openapi-generator.tech/) on the
    file [`docs/spec/openapi.yaml`](https://github.com/neelkamath/spacy-server/raw/master/docs/spec/openapi.yaml).
 
+### Nix
+
+One can run spacy server directly via:
+```shell
+nix -L develop .#en
+```
+(we need to use `develop`, see https://blog.ysndr.de/posts/guides/2021-12-01-nix-shells/#shell-hooks).
+
 ## [Usage](https://hub.docker.com/r/neelkamath/spacy-server)
 
 ## [Contributing](docs/CONTRIBUTING.md)

default.nix

-with (import <nixpkgs> {});
-stdenv.mkDerivation rec {
-  name = "env";
-  env = buildEnv {
-    name = name;
-    paths = buildInputs;
-  };
-  buildInputs = [
-    docker-compose
-  ];
-}

docker-compose.yaml

-services:
-  spacy-server:
-    build:
-      context: ./
-      args:
-        MODEL: en_core_web_trf
-    ports:
-      - 8003:8001
-      
-  spacy-server-fr:
-    build:
-      context: ./
-      args:
-        MODEL: fr_dep_news_trf
-    ports:
-      - 8002:8001
-
-  # spacy-server-de:
-  #   build:
-  #     context: ./
-  #     args:
-  #       MODEL: de_dep_news_trf
-  #   ports:
-  #     - 8002:8000
-
-  # spacy-server-uk:
-  #   build:
-  #     context: ./
-  #     args:
-  #       MODEL: uk_core_news_trf
-  #   ports:
-  #     - 8003:8000
-
-  # spacy-server-ru:
-  #   build:
-  #     context: ./
-  #     args:
-  #       MODEL: ru_core_news_trf
-  #   ports:
-  #     - 8004:8000
-
-  # spacy-server-pl:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8004
-  #       MODEL: pl_core_news_trf
-  #   ports:
-  #     - 8004:8000
-
-  # spacy-server-es:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8005
-  #       MODEL: es_core_news_trf
-  #   ports:
-  #     - 8005:8000
-
-  # spacy-server-pt:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8006
-  #       MODEL: es_core_news_trf
-  #   ports:
-  #     - 8006:8000
-
-  # spacy-server-it:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8007
-  #       MODEL: it_core_news_trf
-  #   ports:
-  #     - 8007:8000
-
-  # spacy-server-zh:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8008
-  #       MODEL: it_core_news_trf
-  #   ports:
-  #     - 8008:8000
-
-  # spacy-server-el:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8009
-  #       MODEL: el_core_news_trf
-  #   ports:
-  #     - 8009:8000
-
-  # spacy-server-xx:
-  #   build:
-  #     context: ./
-  #     args:
-  #       PORT: 8010
-  #       MODEL: xx_ent_wiki_trf
-  #   ports:
-  #     - 8010:8000
-

docker/Dockerfile

-FROM python:3.8 AS base
-WORKDIR /app
-COPY requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt
-ARG SPACY_MODEL
-ENV PYTHONBUFFERED=1 SENSE2VEC=0 SPACY_MODEL=$SPACY_MODEL
-RUN python -m spacy download $SPACY_MODEL
-COPY src/main.py src/main.py
-EXPOSE 8000
-HEALTHCHECK --timeout=2s --start-period=2s --retries=1 \
-    CMD curl -f http://localhost:8000/health_check
-RUN useradd user
-USER user
-CMD ["uvicorn", "src.main:app", "--host", "0.0.0.0"]
-
-FROM base
-ENV SENSE2VEC 1
-COPY src/s2v_old/ src/s2v_old/
\ No newline at end of file

docker/docker-compose.override.yml

-version: '3.7'
-services:
-  app:
-    command: sh -c '. scripts/setup.sh && uvicorn src.main:app --host 0.0.0.0 --reload'
-    ports: ['8000:8000']
-    environment:
-      SPACY_MODEL:
-      SENSE2VEC:
\ No newline at end of file

docker/docker-compose.test.yml

-version: '3.7'
-services:
-  app:
-    environment:
-      SENSE2VEC: 1
-      # Since any model will do, tests have been written only for the en_core_web_sm model because it has a small size
-      # and has every necessary pipeline component.
-      SPACY_MODEL: en_core_web_sm
\ No newline at end of file

docker/docker-compose.yml

-# A virtual environment caches dependencies instead of a Docker volume because the volume randomly gets corrupted.
-version: '3.7'
-services:
-  app:
-    image: python:3.8
-    working_dir: /app
-    volumes:
-      - type: bind
-        source: .
-        target: /app
\ No newline at end of file

flake.lock

+{
+  "nodes": {
+    "flake-utils": {
+      "inputs": {
+        "systems": "systems"
+      },
+      "locked": {
+        "lastModified": 1731533236,
+        "narHash": "sha256-l0KFg5HjrsfsO/JpG+r7fRrqm12kzFHyUHqHCVpMMbI=",
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "rev": "11707dc2f618dd54ca8739b309ec4fc024de578b",
+        "type": "github"
+      },
+      "original": {
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "type": "github"
+      }
+    },
+    "nixpkgs": {
+      "locked": {
+        "lastModified": 1741246872,
+        "narHash": "sha256-Q6pMP4a9ed636qilcYX8XUguvKl/0/LGXhHcRI91p0U=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "10069ef4cf863633f57238f179a0297de84bd8d3",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixos-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "root": {
+      "inputs": {
+        "flake-utils": "flake-utils",
+        "nixpkgs": "nixpkgs"
+      }
+    },
+    "systems": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
+    }
+  },
+  "root": "root",
+  "version": 7
+}

flake.nix

+{
+  description = "Spacy-server flake";
+
+  inputs.flake-utils.url = "github:numtide/flake-utils";
+  inputs.nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
+
+  outputs = { self, nixpkgs, flake-utils }:
+    flake-utils.lib.eachDefaultSystem (system:
+      let pkgs = nixpkgs.legacyPackages.${system};
+          python = (pkgs.python3.override {
+            packageOverrides = self: super: {
+              sense2vec = self.buildPythonPackage rec {
+                pname = "sense2vec" ;
+                version = "2.0.2" ;
+                src = pkgs.fetchFromGitHub {
+                  owner = "explosion" ;
+                  repo = "sense2vec";
+                  rev = "v${version}";
+                  sha256 = "sha256-o862emhbYEaIhzI3H4XDc7SjpDAQ05gL8BhSSW/KNE0=";
+                };
+              };
+            };
+          });
+          ppkgs = model: pp: [ pp.fastapi
+                               pp.ipython
+                               pp.sense2vec
+                               pp.spacy
+                               pp.spacy-models."${model}"
+                               pp.uvicorn ];
+          pythonWithModel = model:
+            python.withPackages (ppkgs model);
+
+          server = model: (
+            let pwm = pythonWithModel model;
+            in
+              pkgs.stdenv.mkDerivation rec {
+                name = "server-${model}";
+                src = ./.;
+                buildInputs = [
+                  pkgs.openblas
+                  pwm
+                ];
+                nativeBuildInputs = [ pkgs.makeWrapper ];
+                installPhase = with pwm.pkgs; ''
+                  mkdir -p $out/bin
+                  makeWrapper ${src}/startServer.sh $out/bin/startServer.sh \
+                    --set PYTHONPATH "${pwm}/${pwm.sitePackages}" 
+                '';
+              }
+          );
+
+          langMap = {
+            de = "de_dep_news_trf";
+            el = "el_core_web_trf";
+            en = "en_core_web_trf";
+            es = "es_core_news_trf";
+            fr = "fr_dep_news_trf";
+            it = "it_core_news_trf";
+            pl = "pl_core_news_trf";
+            pt = "pt_core_news_trf";
+            ru = "ru_core_news_trf";
+            uk = "uk_core_news_trf";
+            zh = "zh_core_news_trf";
+          };
+
+          mkDevShell = lang: pkgs.mkShell {
+            buildInputs = [
+              (pythonWithModel (langMap."${lang}"))
+              (server (langMap."${lang}"))
+            ];
+            shellHook = ''
+              ${(server (langMap."${lang}"))}/bin/startServer.sh
+            '';
+          };
+      in
+        {
+          devShells.en = mkDevShell "en";
+          devShells.fr = mkDevShell "fr";
+        }
+    );
+}

src/main.py

@@ -133,7 +133,7 @@ class PhraseInSentence(BaseModel):
 
     @model_validator(mode='after')
     def phrase_must_be_in_sentence(cls, values):
-        if values.get('phrase') not in values.get('sentence'):
+        if values.phrase not in values.sentence:
             raise ValueError('phrase must be in sentence')
         return values
 

startServer.sh

+#!/bin/bash
+
+uvicorn src.main:app --host 0.0.0.0 "$@"