Merge branch 'access-pipeline' of gitlab.ub.uni-bielefeld.de:sfb1288inf/nopaque into access-pipeline

.dockerignore

@@ -8,5 +8,6 @@
 !.flaskenv
 !boot.sh
 !config.py
+!docker-nopaque-entrypoint.sh
 !nopaque.py
 !requirements.txt

.env.tpl

@@ -1,204 +1,32 @@
-################################################################################
-# Docker                                                                       #
-################################################################################
-# DEFAULT: ./data
-# NOTE: Use `.` as <project-basedir>
-# HOST_DATA_DIR=
-
-# Example: 1000
+##############################################################################
+# Variables for use in Docker Compose YAML files                             #
+##############################################################################
 # HINT: Use this bash command `id -u`
+# NOTE: 0 (= root user) is not allowed
 HOST_UID=
 
-# Example: 1000
 # HINT: Use this bash command `id -g`
 HOST_GID=
 
-# Example: 999
 # HINT: Use this bash command `getent group docker | cut -d: -f3`
 HOST_DOCKER_GID=
 
-# DEFAULT: ./logs
-# NOTES: Use `.` as <project-basedir>
-# HOST_LOG_DIR=
+# DEFAULT: nopaque
+# DOCKER_DEFAULT_NETWORK_NAME=
 
-# DEFAULT: nopaque_default
-# DOCKER_NETWORK_NAME=
+# DEFAULT: ./volumes/db/data
+# NOTE: Use `.` as <project-basedir>
+# DOCKER_DB_SERVICE_DATA_VOLUME_SOURCE_PATH=
 
-################################################################################
-# Flask                                                                        #
-# https://flask.palletsprojects.com/en/1.1.x/config/                           #
-################################################################################
-# CHOOSE ONE: http, https
-# DEFAULT: http
-# PREFERRED_URL_SCHEME=
+# DEFAULT: ./volumes/mq/data
+# NOTE: Use `.` as <project-basedir>
+# DOCKER_MQ_SERVICE_DATA_VOLUME_SOURCE_PATH=
 
-# DEFAULT: hard to guess string
-# HINT: Use this bash command `python -c "import uuid; print(uuid.uuid4().hex)"`
-# SECRET_KEY=
+# NOTE: This must be a network share and it must be available on all
+#       Docker Swarm nodes, mounted to the same path with the same
+#       user and group ownership.
+DOCKER_NOPAQUE_SERVICE_DATA_VOLUME_SOURCE_PATH=
 
-# DEFAULT: localhost:5000
-# Example: nopaque.example.com/nopaque.example.com:5000
-# HINT: If your instance is publicly available on a different Port then 80/443,
-#       you will have to add this to the server name
-# SERVER_NAME=
-
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# HINT: Set to true if you redirect http to https
-# SESSION_COOKIE_SECURE=
-
-
-################################################################################
-# Flask-Assets                                                                 #
-# https://webassets.readthedocs.io/en/latest/                                  #
-################################################################################
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# ASSETS_DEBUG=
-
-
-################################################################################
-# Flask-Hashids                                                                #
-# https://github.com/Pevtrick/Flask-Hashids                                    #
-################################################################################
-# DEFAULT: 16
-# HASHIDS_MIN_LENGTH=
-
-# NOTE: Use this bash command `python -c "import uuid; print(uuid.uuid4().hex)"`
-#       It is strongly recommended that this is NEVER the same as the SECRET_KEY
-HASHIDS_SALT=
-
-
-################################################################################
-# Flask-Login                                                                  #
-# https://flask-login.readthedocs.io/en/latest/                                #
-################################################################################
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# HINT: Set to true if you redirect http to https
-# REMEMBER_COOKIE_SECURE=
-
-
-################################################################################
-# Flask-Mail                                                                   #
-# https://pythonhosted.org/Flask-Mail/                                         #
-################################################################################
-# EXAMPLE: nopaque Admin <nopaque@example.com>
-MAIL_DEFAULT_SENDER=
-
-MAIL_PASSWORD=
-
-# EXAMPLE: smtp.example.com
-MAIL_SERVER=
-
-# EXAMPLE: 587
-MAIL_PORT=
-
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# MAIL_USE_SSL=
-
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# MAIL_USE_TLS=
-
-# EXAMPLE: nopaque@example.com
-MAIL_USERNAME=
-
-
-################################################################################
-# Flask-SQLAlchemy                                                             #
-# https://flask-sqlalchemy.palletsprojects.com/en/2.x/config/                  #
-################################################################################
-# DEFAULT: 'sqlite:///<nopaque-basedir>/data.sqlite'
-# NOTE: Use `.` as <nopaque-basedir>,
-#       Don't use a SQLite database when using Docker
-# SQLALCHEMY_DATABASE_URI=
-
-
-################################################################################
-# nopaque                                                                      #
-################################################################################
-# An account is registered with this email adress gets automatically assigned
-# the administrator role.
-# EXAMPLE: admin.nopaque@example.com
-NOPAQUE_ADMIN=
-
-# DEFAULT: /mnt/nopaque
-# NOTE: This must be a network share and it must be available on all Docker
-#       Swarm nodes
-# NOPAQUE_DATA_DIR=
-
-# CHOOSE ONE: False, True
-# DEFAULT: True
-# NOPAQUE_IS_PRIMARY_INSTANCE=
-
-# transport://[userid:password]@hostname[:port]/[virtual_host]
-NOPAQUE_SOCKETIO_MESSAGE_QUEUE_URI=
-
-# NOTE: Get these from the nopaque development team
-NOPAQUE_DOCKER_REGISTRY_USERNAME=
-NOPAQUE_DOCKER_REGISTRY_PASSWORD=
-
-# DEFAULT: %Y-%m-%d %H:%M:%S
-# NOPAQUE_LOG_DATE_FORMAT=
-
-# DEFAULT: [%(asctime)s] %(levelname)s in %(pathname)s (function: %(funcName)s, line: %(lineno)d): %(message)s
-# NOPAQUE_LOG_FORMAT=
-
-# DEFAULT: INFO
-# CHOOSE ONE: CRITICAL, ERROR, WARNING, INFO, DEBUG
-# NOPAQUE_LOG_LEVEL=
-
-# CHOOSE ONE: False, True
-# DEFAULT: True
-# NOPAQUE_LOG_FILE_ENABLED=
-
-# DEFAULT: <nopaque-basedir>/logs
-# NOTE: Use `.` as <nopaque-basedir>
-# NOPAQUE_LOG_FILE_DIR=
-
-# DEFAULT: NOPAQUE_LOG_LEVEL
-# CHOOSE ONE: CRITICAL, ERROR, WARNING, INFO, DEBUG
-# NOPAQUE_LOG_FILE_LEVEL=
-
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# NOPAQUE_LOG_STDERR_ENABLED=
-
-# CHOOSE ONE: CRITICAL, ERROR, WARNING, INFO, DEBUG
-# DEFAULT: NOPAQUE_LOG_LEVEL
-# NOPAQUE_LOG_STDERR_LEVEL=
-
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# HINT: Set this to True only if you are using a proxy in front of nopaque
-# NOPAQUE_PROXY_FIX_ENABLED=
-
-# DEFAULT: 0
-# Number of values to trust for X-Forwarded-For
-# NOPAQUE_PROXY_FIX_X_FOR=
-
-# DEFAULT: 0
-# Number of values to trust for X-Forwarded-Host
-# NOPAQUE_PROXY_FIX_X_HOST=
-
-# DEFAULT: 0
-# Number of values to trust for X-Forwarded-Port
-# NOPAQUE_PROXY_FIX_X_PORT=
-
-# DEFAULT: 0
-# Number of values to trust for X-Forwarded-Prefix
-# NOPAQUE_PROXY_FIX_X_PREFIX=
-
-# DEFAULT: 0
-# Number of values to trust for X-Forwarded-Proto
-# NOPAQUE_PROXY_FIX_X_PROTO=
-
-# CHOOSE ONE: False, True
-# DEFAULT: False
-# NOPAQUE_TRANSKRIBUS_ENABLED=
-
-# READ-COOP account data: https://readcoop.eu/
-# NOPAQUE_READCOOP_USERNAME=
-# NOPAQUE_READCOOP_PASSWORD=
+# DEFAULT: ./volumes/nopaque/logs
+# NOTE: Use `.` as <project-basedir>
+# DOCKER_NOPAQUE_SERVICE_LOGS_VOLUME_SOURCE_PATH=.

.gitignore

@@ -1,6 +1,6 @@
 # nopaque specifics
 app/static/gen/
-data/
+volumes/
 docker-compose.override.yml
 logs/
 !logs/dummy

.gitlab-ci.yml

@@ -0,0 +1,84 @@
+include:
+  - template: Security/Container-Scanning.gitlab-ci.yml
+
+##############################################################################
+# Pipeline stages in order of execution                                      #
+##############################################################################
+stages:
+  - build
+  - publish
+  - sca
+
+##############################################################################
+# Pipeline behavior                                                          #
+##############################################################################
+workflow:
+  rules:
+    # Run the pipeline on commits to the default branch
+    - if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH
+      variables:
+        # Set the Docker image tag to `latest`
+        DOCKER_IMAGE: $CI_REGISTRY_IMAGE:latest
+      when: always
+    # Run the pipeline on tag creation
+    - if: $CI_COMMIT_TAG
+      variables:
+        # Set the Docker image tag to the Git tag name
+        DOCKER_IMAGE: $CI_REGISTRY_IMAGE:$CI_COMMIT_REF_NAME
+      when: always
+    # Don't run the pipeline on all other occasions
+    - when: never
+
+##############################################################################
+# Default values for pipeline jobs                                           #
+##############################################################################
+default:
+  image: docker:24.0.6
+  services:
+    - docker:24.0.6-dind
+  tags:
+    - docker
+
+##############################################################################
+# CI/CD variables for all jobs in the pipeline                               #
+##############################################################################
+variables:
+  DOCKER_TLS_CERTDIR: /certs
+  DOCKER_BUILD_PATH: .
+  DOCKERFILE: Dockerfile
+
+##############################################################################
+# Pipeline jobs                                                              #
+##############################################################################
+build:
+  stage: build
+  script:
+    - docker build --tag $DOCKER_IMAGE --file $DOCKERFILE $DOCKER_BUILD_PATH
+    - docker save $DOCKER_IMAGE > docker_image.tar
+  artifacts:
+    paths:
+      - docker_image.tar
+
+publish:
+  stage: publish
+  before_script:
+    - docker login --username gitlab-ci-token --password $CI_JOB_TOKEN $CI_REGISTRY
+  script:
+    - docker load --input docker_image.tar
+    - docker push $DOCKER_IMAGE
+  after_script:
+    - docker logout $CI_REGISTRY
+
+container_scanning:
+  stage: sca
+  rules:
+    # Run the job on commits to the default branch
+    - if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH
+      when: always
+    # Run the job on tag creation
+    - if: $CI_COMMIT_TAG
+      when: always
+    # Don't run the job on all other occasions
+    - when: never
+  variables:
+    CS_IMAGE: $DOCKER_IMAGE

.vscode/extensions.json

@@ -1,7 +1,8 @@
 {
     "recommendations": [
-        "samuelcolvin.jinjahtml",
+        "irongeek.vscode-env",
         "ms-azuretools.vscode-docker",
-        "ms-python.python"
+        "ms-python.python",
+        "samuelcolvin.jinjahtml"
     ]
 }

.vscode/settings.json

@@ -1,13 +1,9 @@
 {
     "editor.rulers": [79],
     "files.insertFinalNewline": true,
-    "python.terminal.activateEnvironment": false,
     "[css]": {
         "editor.tabSize": 2
     },
-    "[scss]": {
-        "editor.tabSize": 2
-    },
     "[html]": {
         "editor.tabSize": 2
     },
@@ -17,7 +13,7 @@
     "[jinja-html]": {
         "editor.tabSize": 2
     },
-    "[jinja-js]": {
+    "[scss]": {
         "editor.tabSize": 2
     }
 }

Dockerfile

@@ -1,50 +1,60 @@
-FROM python:3.8.10-slim-buster
+FROM python:3.10.13-slim-bookworm
 
 
 LABEL authors="Patrick Jentsch <p.jentsch@uni-bielefeld.de>"
 
 
-ARG DOCKER_GID
-ARG UID
-ARG GID
-
-
+# Set environment variables
 ENV LANG="C.UTF-8"
 ENV PYTHONDONTWRITEBYTECODE="1"
 ENV PYTHONUNBUFFERED="1"
 
 
+# Install system dependencies
 RUN apt-get update \
  && apt-get install --no-install-recommends --yes \
       build-essential \
+      gosu \
       libpq-dev \
  && rm --recursive /var/lib/apt/lists/*
 
 
-RUN groupadd --gid "${DOCKER_GID}" docker \
- && groupadd --gid "${GID}" nopaque \
- && useradd --create-home --gid nopaque --groups "${DOCKER_GID}" --no-log-init --uid "${UID}" nopaque
+# Create a non-root user
+RUN useradd --create-home --no-log-init nopaque \
+ && groupadd docker \
+ && usermod --append --groups docker nopaque
+
 USER nopaque
 WORKDIR /home/nopaque
 
 
-ENV PYTHON3_VENV_PATH="/home/nopaque/venv"
-RUN python3 -m venv "${PYTHON3_VENV_PATH}"
-ENV PATH="${PYTHON3_VENV_PATH}/bin:${PATH}"
+# Create a Python virtual environment
+ENV NOPAQUE_PYTHON3_VENV_PATH="/home/nopaque/.venv"
+RUN python3 -m venv "${NOPAQUE_PYTHON3_VENV_PATH}"
+ENV PATH="${NOPAQUE_PYTHON3_VENV_PATH}/bin:${PATH}"
 
 
-COPY --chown=nopaque:nopaque requirements.txt .
+# Install Python dependencies
+COPY --chown=nopaque:nopaque requirements.txt requirements.txt
 RUN python3 -m pip install --requirement requirements.txt \
  && rm requirements.txt
 
 
+# Install the application
+COPY docker-nopaque-entrypoint.sh /usr/local/bin/
+
 COPY --chown=nopaque:nopaque app app
 COPY --chown=nopaque:nopaque migrations migrations
 COPY --chown=nopaque:nopaque tests tests
-COPY --chown=nopaque:nopaque .flaskenv boot.sh config.py nopaque.py ./
+COPY --chown=nopaque:nopaque .flaskenv boot.sh config.py nopaque.py requirements.txt ./
+
+RUN mkdir logs
 
 
 EXPOSE 5000
 
 
-ENTRYPOINT ["./boot.sh"]
+USER root
+
+
+ENTRYPOINT ["docker-nopaque-entrypoint.sh"]

README.md

@@ -1,5 +1,8 @@
 # nopaque
 
+![release badge](https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nopaque/-/badges/release.svg)
+![pipeline badge](https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nopaque/badges/master/pipeline.svg?ignore_skipped=true)
+
 nopaque bundles various tools and services that provide humanities scholars with DH methods and thus can support their various individual research processes. Using nopaque, researchers can subject digitized sources to Optical Character Recognition (OCR). The resulting text files can then be used as a data basis for Natural Language Processing (NLP). The texts are automatically subjected to various linguistic annotations. The data processed via NLP can then be summarized in the web application as corpora and analyzed by means of an information retrieval system through complex search queries. The range of functions of the web application will be successively extended according to the needs of the researchers.
 
 ## Prerequisites and requirements

app/__init__.py

@@ -13,7 +13,6 @@ from flask_paranoid import Paranoid
 from flask_socketio import SocketIO
 from flask_sqlalchemy import SQLAlchemy
 from flask_hashids import Hashids
-from werkzeug.exceptions import HTTPException
 
 
 apifairy = APIFairy()
@@ -58,6 +57,9 @@ def create_app(config: Config = Config) -> Flask:
     scheduler.init_app(app)
     socketio.init_app(app, message_queue=app.config['NOPAQUE_SOCKETIO_MESSAGE_QUEUE_URI'])  # noqa
 
+    from .models.event_listeners import register_event_listeners
+    register_event_listeners()
+
     from .admin import bp as admin_blueprint
     default_breadcrumb_root(admin_blueprint, '.admin')
     app.register_blueprint(admin_blueprint, url_prefix='/admin')
@@ -100,7 +102,7 @@ def create_app(config: Config = Config) -> Flask:
 
     from .users import bp as users_blueprint
     default_breadcrumb_root(users_blueprint, '.users')
-    app.register_blueprint(users_blueprint, url_prefix='/users')
+    app.register_blueprint(users_blueprint, cli_group='user', url_prefix='/users')
 
     from .workshops import bp as workshops_blueprint
     app.register_blueprint(workshops_blueprint, url_prefix='/workshops')

app/contributions/spacy_nlp_pipeline_models/forms.py

@@ -16,8 +16,8 @@ class CreateSpaCyNLPPipelineModelForm(ContributionBaseForm):
     )
 
     def validate_spacy_model_file(self, field):
-        if not field.data.filename.lower().endswith('.tar.gz'):
-            raise ValidationError('.tar.gz files only!')
+        if not field.data.filename.lower().endswith(('.tar.gz', ('.whl'))):
+            raise ValidationError('.tar.gz or .whl files only!')
 
     def __init__(self, *args, **kwargs):
         if 'prefix' not in kwargs:

app/converters/sandpaper.py

@@ -2,80 +2,69 @@ from flask import current_app
 from app import db
 from app.models import User, Corpus, CorpusFile
 from datetime import datetime
+from pathlib import Path
+from typing import Dict, List
 import json
-import os
 import shutil
 
 
 class SandpaperConverter:
-    def __init__(self, json_db_file, data_dir):
+    def __init__(self, json_db_file: Path, data_dir: Path):
         self.json_db_file = json_db_file
         self.data_dir = data_dir
 
     def run(self):
-        with open(self.json_db_file, 'r') as f:
-            json_db = json.loads(f.read())
+        with self.json_db_file.open('r') as f:
+            json_db: List[Dict] = json.load(f)
 
         for json_user in json_db:
             if not json_user['confirmed']:
                 current_app.logger.info(f'Skip unconfirmed user {json_user["username"]}')
                 continue
-            user_dir = os.path.join(self.data_dir, str(json_user['id']))
+            user_dir = self.data_dir / f'{json_user["id"]}'
             self.convert_user(json_user, user_dir)
             db.session.commit()
 
 
-    def convert_user(self, json_user, user_dir):
+    def convert_user(self, json_user: Dict, user_dir: Path):
         current_app.logger.info(f'Create User {json_user["username"]}...')
-        user = User(
-            confirmed=json_user['confirmed'],
-            email=json_user['email'],
-            last_seen=datetime.fromtimestamp(json_user['last_seen']),
-            member_since=datetime.fromtimestamp(json_user['member_since']),
-            password_hash=json_user['password_hash'],  # TODO: Needs to be added manually
-            username=json_user['username']
-        )
-        db.session.add(user)
-        db.session.flush(objects=[user])
-        db.session.refresh(user)
         try:
-            user.makedirs()
-        except OSError as e:
-            current_app.logger.error(e)
-            db.session.rollback()
+            user = User.create(
+                confirmed=json_user['confirmed'],
+                email=json_user['email'],
+                last_seen=datetime.fromtimestamp(json_user['last_seen']),
+                member_since=datetime.fromtimestamp(json_user['member_since']),
+                password_hash=json_user['password_hash'],  # TODO: Needs to be added manually
+                username=json_user['username']
+            )
+        except OSError:
             raise Exception('Internal Server Error')
         for json_corpus in json_user['corpora'].values():
             if not json_corpus['files'].values():
                 current_app.logger.info(f'Skip empty corpus {json_corpus["title"]}')
                 continue
-            corpus_dir = os.path.join(user_dir, 'corpora', str(json_corpus['id']))
+            corpus_dir = user_dir / 'corpora' / f'{json_corpus["id"]}'
             self.convert_corpus(json_corpus, user, corpus_dir)
         current_app.logger.info('Done')
 
 
-    def convert_corpus(self, json_corpus, user, corpus_dir):
+    def convert_corpus(self, json_corpus: Dict, user: User, corpus_dir: Path):
         current_app.logger.info(f'Create Corpus {json_corpus["title"]}...')
-        corpus = Corpus(
-            user=user,
-            creation_date=datetime.fromtimestamp(json_corpus['creation_date']),
-            description=json_corpus['description'],
-            title=json_corpus['title']
-        )
-        db.session.add(corpus)
-        db.session.flush(objects=[corpus])
-        db.session.refresh(corpus)
         try:
-            corpus.makedirs()
-        except OSError as e:
-            current_app.logger.error(e)
-            db.session.rollback()
+            corpus = Corpus.create(
+                user=user,
+                creation_date=datetime.fromtimestamp(json_corpus['creation_date']),
+                description=json_corpus['description'],
+                title=json_corpus['title']
+            )
+        except OSError:
             raise Exception('Internal Server Error')
         for json_corpus_file in json_corpus['files'].values():
             self.convert_corpus_file(json_corpus_file, corpus, corpus_dir)
         current_app.logger.info('Done')
 
 
-    def convert_corpus_file(self, json_corpus_file, corpus, corpus_dir):
+    def convert_corpus_file(self, json_corpus_file: Dict, corpus: Corpus, corpus_dir: Path):
         current_app.logger.info(f'Create CorpusFile {json_corpus_file["title"]}...')
         corpus_file = CorpusFile(
             corpus=corpus,
@@ -99,13 +88,13 @@ class SandpaperConverter:
         db.session.refresh(corpus_file)
         try:
             shutil.copy2(
-                os.path.join(corpus_dir, json_corpus_file['filename']),
+                corpus_dir / json_corpus_file['filename'],
                 corpus_file.path
             )
         except:
             current_app.logger.warning(
                 'Can not convert corpus file: '
-                f'{os.path.join(corpus_dir, json_corpus_file["filename"])}'
+                f'{corpus_dir / json_corpus_file["filename"]}'
                 ' -> '
                 f'{corpus_file.path}'
             )

app/corpora/cli.py

@@ -1,7 +1,7 @@
-from app.models import Corpus, CorpusStatus
-import os
+from flask import current_app
 import shutil
 from app import db
+from app.models import Corpus, CorpusStatus
 from . import bp
 
 
@@ -18,10 +18,17 @@ def reset():
     ]
     for corpus in [x for x in Corpus.query.all() if x.status in status]:
         print(f'Resetting corpus {corpus}')
-        shutil.rmtree(os.path.join(corpus.path, 'cwb'), ignore_errors=True)
-        os.mkdir(os.path.join(corpus.path, 'cwb'))
-        os.mkdir(os.path.join(corpus.path, 'cwb', 'data'))
-        os.mkdir(os.path.join(corpus.path, 'cwb', 'registry'))
+        corpus_cwb_dir = corpus.path / 'cwb'
+        corpus_cwb_data_dir = corpus_cwb_dir / 'data'
+        corpus_cwb_registry_dir = corpus_cwb_dir / 'registry'
+        try:
+            shutil.rmtree(corpus.path / 'cwb', ignore_errors=True)
+            corpus_cwb_dir.mkdir()
+            corpus_cwb_data_dir.mkdir()
+            corpus_cwb_registry_dir.mkdir()
+        except OSError as e:
+            current_app.logger.error(e)
+            raise
         corpus.status = CorpusStatus.UNPREPARED
         corpus.num_analysis_sessions = 0
     db.session.commit()

app/corpora/cqi_over_sio/__init__.py

@@ -1,13 +1,14 @@
 from cqi import CQiClient
 from cqi.errors import CQiException
 from cqi.status import CQiStatus
-from flask import session
+from docker.models.containers import Container
+from flask import current_app, session
 from flask_login import current_user
 from flask_socketio import Namespace
 from inspect import signature
 from threading import Lock
 from typing import Callable, Dict, List, Optional
-from app import db, hashids, socketio
+from app import db, docker_client, hashids, socketio
 from app.decorators import socketio_login_required
 from app.models import Corpus, CorpusStatus
 from . import extensions
@@ -121,10 +122,11 @@ class CQiNamespace(Namespace):
             socketio.sleep(3)
             retry_counter -= 1
             db.session.refresh(db_corpus)
-        cqi_client: CQiClient = CQiClient(
-            f'cqpserver_{db_corpus_id}',
-            timeout=float('inf')
-        )
+        # cqi_client: CQiClient = CQiClient(f'cqpserver_{db_corpus_id}')
+        cqpserver_container_name: str = f'cqpserver_{db_corpus_id}'
+        cqpserver_container: Container = docker_client.containers.get(cqpserver_container_name)
+        cqpserver_host: str = cqpserver_container.attrs['NetworkSettings']['Networks'][current_app.config['NOPAQUE_DOCKER_NETWORK_NAME']]['IPAddress']
+        cqi_client: CQiClient = CQiClient(cqpserver_host)
         session['cqi_over_sio'] = {
             'cqi_client': cqi_client,
             'cqi_client_lock': Lock(),

app/corpora/cqi_over_sio/extensions.py

@@ -1,6 +1,7 @@
 from collections import Counter
 from cqi import CQiClient
 from cqi.models.corpora import Corpus as CQiCorpus
+from cqi.models.subcorpora import Subcorpus as CQiSubcorpus
 from cqi.models.attributes import (
     PositionalAttribute as CQiPositionalAttribute,
     StructuralAttribute as CQiStructuralAttribute
@@ -11,7 +12,6 @@ from typing import Dict, List
 import gzip
 import json
 import math
-import os
 from app import db
 from app.models import Corpus
 from .utils import lookups_by_cpos, partial_export_subcorpus, export_subcorpus
@@ -40,161 +40,132 @@ def ext_corpus_update_db(corpus: str) -> CQiStatusOk:
 def ext_corpus_static_data(corpus: str) -> Dict:
     db_corpus_id: int = session['cqi_over_sio']['db_corpus_id']
     db_corpus: Corpus = Corpus.query.get(db_corpus_id)
-    cache_file_path: str = os.path.join(db_corpus.path, 'cwb', 'static.json.gz')
-    if os.path.exists(cache_file_path):
-        with open(cache_file_path, 'rb') as f:
+
+    static_data_file_path = db_corpus.path / 'cwb' / 'static.json.gz'
+    if static_data_file_path.exists():
+        with static_data_file_path.open('rb') as f:
             return f.read()
+
     cqi_client: CQiClient = session['cqi_over_sio']['cqi_client']
     cqi_corpus: CQiCorpus = cqi_client.corpora.get(corpus)
-    cqi_p_attrs: Dict[str, CQiPositionalAttribute] = {
-        p_attr.name: p_attr
-        for p_attr in cqi_corpus.positional_attributes.list()
-    }
-    cqi_s_attrs: Dict[str, CQiStructuralAttribute] = {
-        s_attr.name: s_attr
-        for s_attr in cqi_corpus.structural_attributes.list()
-    }
-    static_corpus_data = {
+    cqi_p_attrs: List[CQiPositionalAttribute] = cqi_corpus.positional_attributes.list()
+    cqi_s_attrs: List[CQiStructuralAttribute] = cqi_corpus.structural_attributes.list()
+
+    static_data = {
         'corpus': {
             'bounds': [0, cqi_corpus.size - 1],
-            'counts': {
-                'token': cqi_corpus.size
-            },
             'freqs': {}
         },
         'p_attrs': {},
         's_attrs': {},
         'values': {'p_attrs': {}, 's_attrs': {}}
     }
-    for p_attr in cqi_p_attrs.values():
-        static_corpus_data['corpus']['freqs'][p_attr.name] = {}
-        chunk_size = 10000
-        p_attr_id_list = list(range(p_attr.lexicon_size))
-        chunks = [p_attr_id_list[i:i+chunk_size] for i in range(0, len(p_attr_id_list), chunk_size)]
+
+    for p_attr in cqi_p_attrs:
+        print(f'corpus.freqs.{p_attr.name}')
+        static_data['corpus']['freqs'][p_attr.name] = []
+        p_attr_id_list: List[int] = list(range(p_attr.lexicon_size))
+        static_data['corpus']['freqs'][p_attr.name].extend(p_attr.freqs_by_ids(p_attr_id_list))
         del p_attr_id_list
-        for chunk in chunks:
-            # print(f'corpus.freqs.{p_attr.name}: {chunk[0]} - {chunk[-1]}')
-            static_corpus_data['corpus']['freqs'][p_attr.name].update(
-                dict(zip(chunk, p_attr.freqs_by_ids(chunk)))
-            )
-        del chunks
-        static_corpus_data['p_attrs'][p_attr.name] = {}
-        cpos_list = list(range(cqi_corpus.size))
-        chunks = [cpos_list[i:i+chunk_size] for i in range(0, len(cpos_list), chunk_size)]
+
+        print(f'p_attrs.{p_attr.name}')
+        static_data['p_attrs'][p_attr.name] = []
+        cpos_list: List[int] = list(range(cqi_corpus.size))
+        static_data['p_attrs'][p_attr.name].extend(p_attr.ids_by_cpos(cpos_list))
         del cpos_list
-        for chunk in chunks:
-            # print(f'p_attrs.{p_attr.name}: {chunk[0]} - {chunk[-1]}')
-            static_corpus_data['p_attrs'][p_attr.name].update(
-                dict(zip(chunk, p_attr.ids_by_cpos(chunk)))
-            )
-        del chunks
-        static_corpus_data['values']['p_attrs'][p_attr.name] = {}
-        p_attr_id_list = list(range(p_attr.lexicon_size))
-        chunks = [p_attr_id_list[i:i+chunk_size] for i in range(0, len(p_attr_id_list), chunk_size)]
+
+        print(f'values.p_attrs.{p_attr.name}')
+        static_data['values']['p_attrs'][p_attr.name] = []
+        p_attr_id_list: List[int] = list(range(p_attr.lexicon_size))
+        static_data['values']['p_attrs'][p_attr.name].extend(p_attr.values_by_ids(p_attr_id_list))
         del p_attr_id_list
-        for chunk in chunks:
-            # print(f'values.p_attrs.{p_attr.name}: {chunk[0]} - {chunk[-1]}')
-            static_corpus_data['values']['p_attrs'][p_attr.name].update(
-                dict(zip(chunk, p_attr.values_by_ids(chunk)))
-            )
-        del chunks
-    for s_attr in cqi_s_attrs.values():
+
+    for s_attr in cqi_s_attrs:
         if s_attr.has_values:
             continue
-        static_corpus_data['corpus']['counts'][s_attr.name] = s_attr.size
-        static_corpus_data['s_attrs'][s_attr.name] = {'lexicon': {}, 'values': None}
-        static_corpus_data['values']['s_attrs'][s_attr.name] = {}
-        ##########################################################################
-        # A faster way to get cpos boundaries for smaller s_attrs                #
-        ##########################################################################
-        # if s_attr.name in ['s', 'ent']:
-        #     cqi_corpus.query('Last', f'<{s_attr.name}> []* </{s_attr.name}>;')
-        #     cqi_subcorpus = cqi_corpus.subcorpora.get('Last')
-        #     first_match = 0
-        #     last_match = cqi_subcorpus.size - 1
-        #     match_boundaries = zip(
-        #         range(first_match, last_match + 1),
-        #         cqi_subcorpus.dump(cqi_subcorpus.fields['match'], first_match, last_match),
-        #         cqi_subcorpus.dump(cqi_subcorpus.fields['matchend'], first_match, last_match)
-        #     )
-        #     for id, lbound, rbound in match_boundaries:
-        #         static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id] = {}
-        #         static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['bounds'] = [lbound, rbound]
-        #         static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['counts'] = {}
-        #         static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['counts']['token'] = rbound - lbound + 1
-        #     cqi_subcorpus.drop()
+
+        static_data['s_attrs'][s_attr.name] = {'lexicon': [], 'values': None}
+
+        if s_attr.name in ['s', 'ent']:
+            ##############################################################
+            # A faster way to get cpos boundaries for smaller s_attrs    #
+            # Note: Needs more testing, don't use it in production       #
+            ##############################################################
+            cqi_corpus.query('Last', f'<{s_attr.name}> []* </{s_attr.name}>;')
+            cqi_subcorpus: CQiSubcorpus = cqi_corpus.subcorpora.get('Last')
+            first_match: int = 0
+            last_match: int = cqi_subcorpus.size - 1
+            match_boundaries = zip(
+                range(first_match, last_match + 1),
+                cqi_subcorpus.dump(
+                    cqi_subcorpus.fields['match'],
+                    first_match,
+                    last_match
+                ),
+                cqi_subcorpus.dump(
+                    cqi_subcorpus.fields['matchend'],
+                    first_match,
+                    last_match
+                )
+            )
+            cqi_subcorpus.drop()
+            del cqi_subcorpus, first_match, last_match
+            for id, lbound, rbound in match_boundaries:
+                static_data['s_attrs'][s_attr.name]['lexicon'].append({})
+                print(f's_attrs.{s_attr.name}.lexicon.{id}.bounds')
+                static_data['s_attrs'][s_attr.name]['lexicon'][id]['bounds'] = [lbound, rbound]
+            del match_boundaries
+
+        if s_attr.name != 'text':
+            continue
+
         for id in range(0, s_attr.size):
-            # print(f's_attrs.{s_attr.name}.lexicon.{id}')
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id] = {
-                'bounds': None,
-                'counts': None,
-                'freqs': None
-            }
-            if s_attr.name != 'text':
-                continue
+            static_data['s_attrs'][s_attr.name]['lexicon'].append({})
+            # This is a very slow operation, thats why we only use it for
+            # the text attribute
             lbound, rbound = s_attr.cpos_by_id(id)
-            # print(f's_attrs.{s_attr.name}.lexicon.{id}.bounds')
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['bounds'] = [lbound, rbound]
-            # print(f's_attrs.{s_attr.name}.lexicon.{id}.counts')
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['counts'] = {}
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['counts']['token'] = rbound - lbound + 1
-            cpos_list = list(range(lbound, rbound + 1))
-            chunks = [cpos_list[i:i+chunk_size] for i in range(0, len(cpos_list), chunk_size)]
-            del cpos_list
-            ent_ids = set()
-            for chunk in chunks:
-                # print(f'Gather ent_ids from cpos: {chunk[0]} - {chunk[-1]}')
-                ent_ids.update({x for x in cqi_s_attrs['ent'].ids_by_cpos(chunk) if x != -1})
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['counts']['ent'] = len(ent_ids)
-            del ent_ids
-            s_ids = set()
-            for chunk in chunks:
-                # print(f'Gather s_ids from cpos: {chunk[0]} - {chunk[-1]}')
-                s_ids.update({x for x in cqi_s_attrs['s'].ids_by_cpos(chunk) if x != -1})
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['counts']['s'] = len(s_ids)
-            del s_ids
-            # print(f's_attrs.{s_attr.name}.lexicon.{id}.freqs')
-            static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['freqs'] = {}
-            for p_attr in cqi_p_attrs.values():
-                p_attr_ids = []
-                for chunk in chunks:
-                    # print(f'Gather p_attr_ids from cpos: {chunk[0]} - {chunk[-1]}')
-                    p_attr_ids.extend(p_attr.ids_by_cpos(chunk))
-                static_corpus_data['s_attrs'][s_attr.name]['lexicon'][id]['freqs'][p_attr.name] = dict(Counter(p_attr_ids))
+            print(f's_attrs.{s_attr.name}.lexicon.{id}.bounds')
+            static_data['s_attrs'][s_attr.name]['lexicon'][id]['bounds'] = [lbound, rbound]
+            static_data['s_attrs'][s_attr.name]['lexicon'][id]['freqs'] = {}
+            cpos_list: List[int] = list(range(lbound, rbound + 1))
+            for p_attr in cqi_p_attrs:
+                p_attr_ids: List[int] = []
+                p_attr_ids.extend(p_attr.ids_by_cpos(cpos_list))
+                print(f's_attrs.{s_attr.name}.lexicon.{id}.freqs.{p_attr.name}')
+                static_data['s_attrs'][s_attr.name]['lexicon'][id]['freqs'][p_attr.name] = dict(Counter(p_attr_ids))
                 del p_attr_ids
-            del chunks
-        sub_s_attrs = cqi_corpus.structural_attributes.list(filters={'part_of': s_attr})
-        s_attr_value_names: List[str] = [
+            del cpos_list
+
+        sub_s_attrs: List[CQiStructuralAttribute] = cqi_corpus.structural_attributes.list(filters={'part_of': s_attr})
+        print(f's_attrs.{s_attr.name}.values')
+        static_data['s_attrs'][s_attr.name]['values'] = [
             sub_s_attr.name[(len(s_attr.name) + 1):]
             for sub_s_attr in sub_s_attrs
         ]
-        s_attr_id_list = list(range(s_attr.size))
-        chunks = [s_attr_id_list[i:i+chunk_size] for i in range(0, len(s_attr_id_list), chunk_size)]
-        del s_attr_id_list
-        sub_s_attr_values = []
+        s_attr_id_list: List[int] = list(range(s_attr.size))
+        sub_s_attr_values: List[str] = []
         for sub_s_attr in sub_s_attrs:
             tmp = []
-            for chunk in chunks:
-                tmp.extend(sub_s_attr.values_by_ids(chunk))
+            tmp.extend(sub_s_attr.values_by_ids(s_attr_id_list))
             sub_s_attr_values.append(tmp)
             del tmp
-        del chunks
-        # print(f's_attrs.{s_attr.name}.values')
-        static_corpus_data['s_attrs'][s_attr.name]['values'] = s_attr_value_names
-        # print(f'values.s_attrs.{s_attr.name}')
-        static_corpus_data['values']['s_attrs'][s_attr.name] = {
-            s_attr_id: {
-                s_attr_value_name: sub_s_attr_values[s_attr_value_name_idx][s_attr_id_idx]
+        del s_attr_id_list
+        print(f'values.s_attrs.{s_attr.name}')
+        static_data['values']['s_attrs'][s_attr.name] = [
+            {
+                s_attr_value_name: sub_s_attr_values[s_attr_value_name_idx][s_attr_id]
                 for s_attr_value_name_idx, s_attr_value_name in enumerate(
-                    static_corpus_data['s_attrs'][s_attr.name]['values']
+                    static_data['s_attrs'][s_attr.name]['values']
                 )
-            } for s_attr_id_idx, s_attr_id in enumerate(range(0, s_attr.size))
-        }
+            } for s_attr_id in range(0, s_attr.size)
+        ]
         del sub_s_attr_values
-    with gzip.open(cache_file_path, 'wt') as f:
-        json.dump(static_corpus_data, f)
-    del static_corpus_data
-    with open(cache_file_path, 'rb') as f:
+    print('Saving static data to file')
+    with gzip.open(static_data_file_path, 'wt') as f:
+        json.dump(static_data, f)
+    del static_data
+    print('Sending static data to client')
+    with open(static_data_file_path, 'rb') as f:
         return f.read()
 
 

app/corpora/cqi_over_sio/utils.py

@@ -9,8 +9,7 @@ def lookups_by_cpos(corpus: CQiCorpus, cpos_list: List[int]) -> Dict:
     for attr in corpus.positional_attributes.list():
         cpos_attr_values: List[str] = attr.values_by_cpos(cpos_list)
         for i, cpos in enumerate(cpos_list):
-            lookups['cpos_lookup'][cpos][attr.attrs['name']] = \
-                cpos_attr_values[i]
+            lookups['cpos_lookup'][cpos][attr.name] = cpos_attr_values[i]
     for attr in corpus.structural_attributes.list():
         # We only want to iterate over non subattributes, identifiable by
         # attr.has_values == False
@@ -20,19 +19,19 @@ def lookups_by_cpos(corpus: CQiCorpus, cpos_list: List[int]) -> Dict:
         for i, cpos in enumerate(cpos_list):
             if cpos_attr_ids[i] == -1:
                 continue
-            lookups['cpos_lookup'][cpos][attr.attrs['name']] = cpos_attr_ids[i]
+            lookups['cpos_lookup'][cpos][attr.name] = cpos_attr_ids[i]
         occured_attr_ids = [x for x in set(cpos_attr_ids) if x != -1]
-        if not occured_attr_ids:
+        if len(occured_attr_ids) == 0:
             continue
         subattrs = corpus.structural_attributes.list(filters={'part_of': attr})
-        if not subattrs:
+        if len(subattrs) == 0:
             continue
-        lookup_name: str = f'{attr.attrs["name"]}_lookup'
+        lookup_name: str = f'{attr.name}_lookup'
         lookups[lookup_name] = {}
         for attr_id in occured_attr_ids:
             lookups[lookup_name][attr_id] = {}
         for subattr in subattrs:
-            subattr_name = subattr.attrs['name'][(len(attr.attrs['name']) + 1):]  # noqa
+            subattr_name = subattr.name[(len(attr.name) + 1):]  # noqa
             for i, subattr_value in enumerate(subattr.values_by_ids(occured_attr_ids)):  # noqa
                 lookups[lookup_name][occured_attr_ids[i]][subattr_name] = subattr_value  # noqa
     return lookups

app/corpora/files/routes.py

@@ -7,7 +7,6 @@ from flask import (
     url_for
 )
 from flask_breadcrumbs import register_breadcrumb
-import os
 from app import db
 from app.models import Corpus, CorpusFile, CorpusStatus
 from ..decorators import corpus_follower_permission_required
@@ -92,8 +91,8 @@ def corpus_file(corpus_id, corpus_file_id):
 def download_corpus_file(corpus_id, corpus_file_id):
     corpus_file = CorpusFile.query.filter_by(corpus_id=corpus_id, id=corpus_file_id).first_or_404()
     return send_from_directory(
-        os.path.dirname(corpus_file.path),
-        os.path.basename(corpus_file.path),
+        corpus_file.path.parent,
+        corpus_file.path.name,
         as_attachment=True,
         attachment_filename=corpus_file.filename,
         mimetype=corpus_file.mimetype

app/corpora/followers/json_routes.py

@@ -12,65 +12,65 @@ from ..decorators import corpus_follower_permission_required
 from . import bp
 
 
-# @bp.route('/<hashid:corpus_id>/followers', methods=['POST'])
-# @corpus_follower_permission_required('MANAGE_FOLLOWERS')
-# @content_negotiation(consumes='application/json', produces='application/json')
-# def create_corpus_followers(corpus_id):
-#     usernames = request.json
-#     if not (isinstance(usernames, list) or all(isinstance(u, str) for u in usernames)):
-#         abort(400)
-#     corpus = Corpus.query.get_or_404(corpus_id)
-#     for username in usernames:
-#         user = User.query.filter_by(username=username, is_public=True).first_or_404()
-#         user.follow_corpus(corpus)
-#     db.session.commit()
-#     response_data = {
-#         'message': f'Users are now following "{corpus.title}"',
-#         'category': 'corpus'
-#     }
-#     return response_data, 200
+@bp.route('/<hashid:corpus_id>/followers', methods=['POST'])
+@corpus_follower_permission_required('MANAGE_FOLLOWERS')
+@content_negotiation(consumes='application/json', produces='application/json')
+def create_corpus_followers(corpus_id):
+    usernames = request.json
+    if not (isinstance(usernames, list) or all(isinstance(u, str) for u in usernames)):
+        abort(400)
+    corpus = Corpus.query.get_or_404(corpus_id)
+    for username in usernames:
+        user = User.query.filter_by(username=username, is_public=True).first_or_404()
+        user.follow_corpus(corpus)
+    db.session.commit()
+    response_data = {
+        'message': f'Users are now following "{corpus.title}"',
+        'category': 'corpus'
+    }
+    return response_data, 200
 
 
-# @bp.route('/<hashid:corpus_id>/followers/<hashid:follower_id>/role', methods=['PUT'])
-# @corpus_follower_permission_required('MANAGE_FOLLOWERS')
-# @content_negotiation(consumes='application/json', produces='application/json')
-# def update_corpus_follower_role(corpus_id, follower_id):
-#     role_name = request.json
-#     if not isinstance(role_name, str):
-#         abort(400)
-#     cfr = CorpusFollowerRole.query.filter_by(name=role_name).first()
-#     if cfr is None:
-#         abort(400)
-#     cfa = CorpusFollowerAssociation.query.filter_by(corpus_id=corpus_id, follower_id=follower_id).first_or_404()
-#     cfa.role = cfr
-#     db.session.commit()
-#     response_data = {
-#         'message': f'User "{cfa.follower.username}" is now {cfa.role.name}',
-#         'category': 'corpus'
-#     }
-#     return response_data, 200
+@bp.route('/<hashid:corpus_id>/followers/<hashid:follower_id>/role', methods=['PUT'])
+@corpus_follower_permission_required('MANAGE_FOLLOWERS')
+@content_negotiation(consumes='application/json', produces='application/json')
+def update_corpus_follower_role(corpus_id, follower_id):
+    role_name = request.json
+    if not isinstance(role_name, str):
+        abort(400)
+    cfr = CorpusFollowerRole.query.filter_by(name=role_name).first()
+    if cfr is None:
+        abort(400)
+    cfa = CorpusFollowerAssociation.query.filter_by(corpus_id=corpus_id, follower_id=follower_id).first_or_404()
+    cfa.role = cfr
+    db.session.commit()
+    response_data = {
+        'message': f'User "{cfa.follower.username}" is now {cfa.role.name}',
+        'category': 'corpus'
+    }
+    return response_data, 200
 
 
-# @bp.route('/<hashid:corpus_id>/followers/<hashid:follower_id>', methods=['DELETE'])
-# def delete_corpus_follower(corpus_id, follower_id):
-#     cfa = CorpusFollowerAssociation.query.filter_by(corpus_id=corpus_id, follower_id=follower_id).first_or_404()
-#     if not (
-#         current_user.id == follower_id
-#         or current_user == cfa.corpus.user 
-#         or CorpusFollowerAssociation.query.filter_by(corpus_id=corpus_id, follower_id=current_user.id).first().role.has_permission('MANAGE_FOLLOWERS')
-#         or current_user.is_administrator()):
-#         abort(403)
-#     if current_user.id == follower_id:
-#         flash(f'You are no longer following "{cfa.corpus.title}"', 'corpus')
-#         response = make_response()
-#         response.status_code = 204
-#     else:
-#         response_data = {
-#             'message': f'"{cfa.follower.username}" is not following "{cfa.corpus.title}" anymore',
-#             'category': 'corpus'
-#         }
-#         response = jsonify(response_data)
-#         response.status_code = 200
-#     cfa.follower.unfollow_corpus(cfa.corpus)
-#     db.session.commit()
-#     return response
+@bp.route('/<hashid:corpus_id>/followers/<hashid:follower_id>', methods=['DELETE'])
+def delete_corpus_follower(corpus_id, follower_id):
+    cfa = CorpusFollowerAssociation.query.filter_by(corpus_id=corpus_id, follower_id=follower_id).first_or_404()
+    if not (
+        current_user.id == follower_id
+        or current_user == cfa.corpus.user 
+        or CorpusFollowerAssociation.query.filter_by(corpus_id=corpus_id, follower_id=current_user.id).first().role.has_permission('MANAGE_FOLLOWERS')
+        or current_user.is_administrator()):
+        abort(403)
+    if current_user.id == follower_id:
+        flash(f'You are no longer following "{cfa.corpus.title}"', 'corpus')
+        response = make_response()
+        response.status_code = 204
+    else:
+        response_data = {
+            'message': f'"{cfa.follower.username}" is not following "{cfa.corpus.title}" anymore',
+            'category': 'corpus'
+        }
+        response = jsonify(response_data)
+        response.status_code = 200
+    cfa.follower.unfollow_corpus(cfa.corpus)
+    db.session.commit()
+    return response

app/corpora/json_routes.py

@@ -61,7 +61,7 @@ def build_corpus(corpus_id):
 @bp.route('/stopwords')
 @content_negotiation(produces='application/json')
 def get_stopwords():
-    nltk.download('stopwords')
+    nltk.download('stopwords', quiet=True)
     languages = ["german", "english", "catalan", "greek", "spanish", "french", "italian", "russian", "chinese"]
     stopwords = {}
     for language in languages:
@@ -71,55 +71,55 @@ def get_stopwords():
     response_data = stopwords
     return response_data, 202
 
-# @bp.route('/<hashid:corpus_id>/generate-share-link', methods=['POST'])
-# @corpus_follower_permission_required('MANAGE_FOLLOWERS')
-# @content_negotiation(consumes='application/json', produces='application/json')
-# def generate_corpus_share_link(corpus_id):
-#     data = request.json
-#     if not isinstance(data, dict):
-#         abort(400)
-#     expiration = data.get('expiration')
-#     if not isinstance(expiration, str):
-#         abort(400)
-#     role_name = data.get('role')
-#     if not isinstance(role_name, str):
-#         abort(400)
-#     expiration_date = datetime.strptime(expiration, '%b %d, %Y')
-#     cfr = CorpusFollowerRole.query.filter_by(name=role_name).first()
-#     if cfr is None:
-#         abort(400)
-#     corpus = Corpus.query.get_or_404(corpus_id)
-#     token = current_user.generate_follow_corpus_token(corpus.hashid, role_name, expiration_date)
-#     corpus_share_link = url_for(
-#         'corpora.follow_corpus',
-#         corpus_id=corpus_id,
-#         token=token,
-#         _external=True
-#     )
-#     response_data = {
-#         'message': 'Corpus share link generated',
-#         'category': 'corpus',
-#         'corpusShareLink': corpus_share_link
-#     }
-#     return response_data, 200
+@bp.route('/<hashid:corpus_id>/generate-share-link', methods=['POST'])
+@corpus_follower_permission_required('MANAGE_FOLLOWERS')
+@content_negotiation(consumes='application/json', produces='application/json')
+def generate_corpus_share_link(corpus_id):
+    data = request.json
+    if not isinstance(data, dict):
+        abort(400)
+    expiration = data.get('expiration')
+    if not isinstance(expiration, str):
+        abort(400)
+    role_name = data.get('role')
+    if not isinstance(role_name, str):
+        abort(400)
+    expiration_date = datetime.strptime(expiration, '%b %d, %Y')
+    cfr = CorpusFollowerRole.query.filter_by(name=role_name).first()
+    if cfr is None:
+        abort(400)
+    corpus = Corpus.query.get_or_404(corpus_id)
+    token = current_user.generate_follow_corpus_token(corpus.hashid, role_name, expiration_date)
+    corpus_share_link = url_for(
+        'corpora.follow_corpus',
+        corpus_id=corpus_id,
+        token=token,
+        _external=True
+    )
+    response_data = {
+        'message': 'Corpus share link generated',
+        'category': 'corpus',
+        'corpusShareLink': corpus_share_link
+    }
+    return response_data, 200
     
 
 
-# @bp.route('/<hashid:corpus_id>/is_public', methods=['PUT'])
-# @corpus_owner_or_admin_required
-# @content_negotiation(consumes='application/json', produces='application/json')
-# def update_corpus_is_public(corpus_id):
-#     is_public = request.json
-#     if not isinstance(is_public, bool):
-#         abort(400)
-#     corpus = Corpus.query.get_or_404(corpus_id)
-#     corpus.is_public = is_public
-#     db.session.commit()
-#     response_data = {
-#         'message': (
-#             f'Corpus "{corpus.title}" is now'
-#             f' {"public" if is_public else "private"}'
-#         ),
-#         'category': 'corpus'
-#     }
-#     return response_data, 200
+@bp.route('/<hashid:corpus_id>/is_public', methods=['PUT'])
+@corpus_owner_or_admin_required
+@content_negotiation(consumes='application/json', produces='application/json')
+def update_corpus_is_public(corpus_id):
+    is_public = request.json
+    if not isinstance(is_public, bool):
+        abort(400)
+    corpus = Corpus.query.get_or_404(corpus_id)
+    corpus.is_public = is_public
+    db.session.commit()
+    response_data = {
+        'message': (
+            f'Corpus "{corpus.title}" is now'
+            f' {"public" if is_public else "private"}'
+        ),
+        'category': 'corpus'
+    }
+    return response_data, 200

app/corpora/routes.py

@@ -68,20 +68,19 @@ def corpus(corpus_id):
             corpus=corpus,
             cfr=cfr,
             cfrs=cfrs,
-            users = users
+            users=users
         )
     if (current_user.is_following_corpus(corpus) or corpus.is_public):
-        abort(404)
-        # cfas = CorpusFollowerAssociation.query.filter(Corpus.id == corpus_id, CorpusFollowerAssociation.follower_id != corpus.user.id).all()
-        # return render_template(
-        #     'corpora/public_corpus.html.j2',
-        #     title=corpus.title,
-        #     corpus=corpus,
-        #     cfrs=cfrs,
-        #     cfr=cfr,
-        #     cfas=cfas,
-        #     users = users
-        # )
+        cfas = CorpusFollowerAssociation.query.filter(Corpus.id == corpus_id, CorpusFollowerAssociation.follower_id != corpus.user.id).all()
+        return render_template(
+            'corpora/public_corpus.html.j2',
+            title=corpus.title,
+            corpus=corpus,
+            cfrs=cfrs,
+            cfr=cfr,
+            cfas=cfas,
+            users=users
+        )
     abort(403)
 
 
@@ -98,14 +97,14 @@ def analysis(corpus_id):
     )
 
 
-# @bp.route('/<hashid:corpus_id>/follow/<token>')
-# def follow_corpus(corpus_id, token):
-#     corpus = Corpus.query.get_or_404(corpus_id)
-#     if current_user.follow_corpus_by_token(token):
-#         db.session.commit()
-#         flash(f'You are following "{corpus.title}" now', category='corpus')
-#         return redirect(url_for('corpora.corpus', corpus_id=corpus_id))
-#     abort(403)
+@bp.route('/<hashid:corpus_id>/follow/<token>')
+def follow_corpus(corpus_id, token):
+    corpus = Corpus.query.get_or_404(corpus_id)
+    if current_user.follow_corpus_by_token(token):
+        db.session.commit()
+        flash(f'You are following "{corpus.title}" now', category='corpus')
+        return redirect(url_for('corpora.corpus', corpus_id=corpus_id))
+    abort(403)
 
 
 @bp.route('/import', methods=['GET', 'POST'])

app/daemon/corpus_utils.py

@@ -45,7 +45,7 @@ def _create_build_corpus_service(corpus):
     ''' ## Constraints ## '''
     constraints = ['node.role==worker']
     ''' ## Image ## '''
-    image = f'{current_app.config["NOPAQUE_DOCKER_IMAGE_PREFIX"]}cwb:r1702'
+    image = f'{current_app.config["NOPAQUE_DOCKER_IMAGE_PREFIX"]}cwb:r1879'
     ''' ## Labels ## '''
     labels = {
         'origin': current_app.config['SERVER_NAME'],
@@ -139,11 +139,11 @@ def _create_cqpserver_container(corpus):
     ''' ## Entrypoint ## '''
     entrypoint = ['bash', '-c']
     ''' ## Image ## '''
-    image = f'{current_app.config["NOPAQUE_DOCKER_IMAGE_PREFIX"]}cwb:r1702'
+    image = f'{current_app.config["NOPAQUE_DOCKER_IMAGE_PREFIX"]}cwb:r1879'
     ''' ## Name ## '''
     name = f'cqpserver_{corpus.id}'
     ''' ## Network ## '''
-    network = f'{current_app.config["DOCKER_NETWORK_NAME"]}'
+    network = f'{current_app.config["NOPAQUE_DOCKER_NETWORK_NAME"]}'
     ''' ## Volumes ## '''
     volumes = []
     ''' ### Corpus data volume ### '''

app/ext/flask_sqlalchemy/__init__.py

@@ -0,0 +1,2 @@
+from .container_column import ContainerColumn
+from .int_enum_column import IntEnumColumn

app/ext/flask_sqlalchemy/container_column.py

@@ -0,0 +1,21 @@
+import json
+from app import db
+
+
+class ContainerColumn(db.TypeDecorator):
+    impl = db.String
+
+    def __init__(self, container_type, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.container_type = container_type
+
+    def process_bind_param(self, value, dialect):
+        if isinstance(value, self.container_type):
+            return json.dumps(value)
+        elif isinstance(value, str) and isinstance(json.loads(value), self.container_type):
+            return value
+        else:
+            return TypeError()
+
+    def process_result_value(self, value, dialect):
+        return json.loads(value)

app/ext/flask_sqlalchemy/int_enum_column.py

@@ -0,0 +1,22 @@
+from app import db
+
+
+class IntEnumColumn(db.TypeDecorator):
+    impl = db.Integer
+
+    def __init__(self, enum_type, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.enum_type = enum_type
+
+    def process_bind_param(self, value, dialect):
+        if isinstance(value, self.enum_type) and isinstance(value.value, int):
+            return value.value
+        elif isinstance(value, int):
+            return self.enum_type(value).value
+        elif isinstance(value, str):
+            return self.enum_type[value].value
+        else:
+            return TypeError()
+
+    def process_result_value(self, value, dialect):
+        return self.enum_type(value)

app/jobs/json_routes.py

@@ -1,7 +1,6 @@
 from flask import abort, current_app
 from flask_login import current_user
 from threading import Thread
-import os
 from app import db
 from app.decorators import admin_required, content_negotiation
 from app.models import Job, JobStatus
@@ -39,7 +38,7 @@ def job_log(job_id):
     if job.status not in [JobStatus.COMPLETED, JobStatus.FAILED]:
         response = {'errors': {'message': 'Job status is not completed or failed'}}
         return response, 409
-    with open(os.path.join(job.path, 'pipeline_data', 'logs', 'pyflow_log.txt')) as log_file:
+    with open(job.path / 'pipeline_data' / 'logs' / 'pyflow_log.txt') as log_file:
         log = log_file.read()
     response_data = {
         'jobLog': log

app/jobs/routes.py

@@ -7,7 +7,6 @@ from flask import (
 )
 from flask_breadcrumbs import register_breadcrumb
 from flask_login import current_user
-import os
 from app.models import Job, JobInput, JobResult
 from . import bp
 from .utils import job_dynamic_list_constructor as job_dlc
@@ -38,8 +37,8 @@ def download_job_input(job_id, job_input_id):
     if not (job_input.job.user == current_user or current_user.is_administrator()):
         abort(403)
     return send_from_directory(
-        os.path.dirname(job_input.path),
-        os.path.basename(job_input.path),
+        job_input.path.parent,
+        job_input.path.name,
         as_attachment=True,
         attachment_filename=job_input.filename,
         mimetype=job_input.mimetype
@@ -52,8 +51,8 @@ def download_job_result(job_id, job_result_id):
     if not (job_result.job.user == current_user or current_user.is_administrator()):
         abort(403)
     return send_from_directory(
-        os.path.dirname(job_result.path),
-        os.path.basename(job_result.path),
+        job_result.path.parent,
+        job_result.path.name,
         as_attachment=True,
         attachment_filename=job_result.filename,
         mimetype=job_result.mimetype

app/main/cli.py

@@ -1,6 +1,7 @@
 from flask import current_app
 from flask_migrate import upgrade
-import os
+from pathlib import Path
+from typing import List
 from app.models import (
     CorpusFollowerRole,
     Role,
@@ -17,16 +18,15 @@ def deploy():
     # Make default directories
     print('Make default directories')
     base_dir = current_app.config['NOPAQUE_DATA_DIR']
-    default_dirs = [
-        os.path.join(base_dir, 'tmp'),
-        os.path.join(base_dir, 'users')
+    default_dirs: List[Path] = [
+        base_dir / 'tmp',
+        base_dir / 'users'
     ]
-    for dir in default_dirs:
-        if os.path.exists(dir):
-            if not os.path.isdir(dir):
-                raise NotADirectoryError(f'{dir} is not a directory')
-        else:
-            os.mkdir(dir)
+    for default_dir in default_dirs:
+        if not default_dir.exists():
+            default_dir.mkdir()
+        if not default_dir.is_dir():
+            raise NotADirectoryError(f'{default_dir} is not a directory')
 
     # migrate database to latest revision
     print('Migrate database to latest revision')
@@ -43,3 +43,5 @@ def deploy():
     SpaCyNLPPipelineModel.insert_defaults()
     print('Insert/Update default TesseractOCRPipelineModels')
     TesseractOCRPipelineModel.insert_defaults()
+
+    # TODO: Implement checks for if the nopaque network exists

app/main/routes.py

@@ -45,12 +45,6 @@ def dashboard():
     )
 
 
-# @bp.route('/user_manual')
-# @register_breadcrumb(bp, '.user_manual', '<i class="material-icons left">help</i>User manual')
-# def user_manual():
-#     return render_template('main/user_manual.html.j2', title='User manual')
-
-
 @bp.route('/news')
 @register_breadcrumb(bp, '.news', '<i class="material-icons left">email</i>News')
 def news():
@@ -78,15 +72,17 @@ def terms_of_use():
     )
 
 
-# @bp.route('/social-area')
-# @register_breadcrumb(bp, '.social_area', '<i class="material-icons left">group</i>Social Area')
-# @login_required
-# def social_area():
-#     corpora = Corpus.query.filter(Corpus.is_public == True, Corpus.user != current_user).all()
-#     users = User.query.filter(User.is_public == True, User.id != current_user.id).all()
-#     return render_template(
-#         'main/social_area.html.j2',
-#         title='Social Area',
-#         corpora=corpora,
-#         users=users
-#     )
+@bp.route('/social-area')
+@register_breadcrumb(bp, '.social_area', '<i class="material-icons left">group</i>Social Area')
+@login_required
+def social_area():
+    print('test')
+    corpora = Corpus.query.filter(Corpus.is_public == True, Corpus.user != current_user).all()
+    print(corpora)
+    users = User.query.filter(User.is_public == True, User.id != current_user.id).all()
+    return render_template(
+        'main/social_area.html.j2',
+        title='Social Area',
+        corpora=corpora,
+        users=users
+    )

app/models/__init__.py

@@ -0,0 +1,19 @@
+from .avatar import *
+from .corpus_file import *
+from .corpus_follower_association import *
+from .corpus_follower_role import *
+from .corpus import *
+from .job_input import *
+from .job_result import *
+from .job import *
+from .role import *
+from .spacy_nlp_pipeline_model import *
+from .tesseract_ocr_pipeline_model import *
+from .token import *
+from .user import *
+from app import login
+
+
+@login.user_loader
+def load_user(user_id):
+    return User.query.get(int(user_id))

app/models/avatar.py

@@ -0,0 +1,40 @@
+from flask import current_app
+from flask_hashids import HashidMixin
+from pathlib import Path
+from app import db
+from .file_mixin import FileMixin
+
+
+class Avatar(HashidMixin, FileMixin, db.Model):
+    __tablename__ = 'avatars'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    user_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    # Relationships
+    user = db.relationship('User', back_populates='avatar')
+
+    @property
+    def path(self) -> Path:
+        return self.user.path / 'avatar'
+        # return os.path.join(self.user.path, 'avatar')
+
+    def delete(self):
+        try:
+            self.path.unlink(missing_ok=True)
+        except OSError as e:
+            current_app.logger.error(e)
+            raise
+        db.session.delete(self)
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            **self.file_mixin_to_json_serializeable()
+        }
+        if backrefs:
+            json_serializeable['user'] = \
+                self.user.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable

app/models/corpus.py

@@ -0,0 +1,200 @@
+from datetime import datetime
+from enum import IntEnum
+from flask import current_app, url_for
+from flask_hashids import HashidMixin
+from sqlalchemy.ext.associationproxy import association_proxy
+from typing import Union
+from pathlib import Path
+import shutil
+import xml.etree.ElementTree as ET
+from app import db
+from app.converters.vrt import normalize_vrt_file
+from app.ext.flask_sqlalchemy import IntEnumColumn
+from .corpus_follower_association import CorpusFollowerAssociation
+
+
+class CorpusStatus(IntEnum):
+    UNPREPARED = 1
+    SUBMITTED = 2
+    QUEUED = 3
+    BUILDING = 4
+    BUILT = 5
+    FAILED = 6
+    STARTING_ANALYSIS_SESSION = 7
+    RUNNING_ANALYSIS_SESSION = 8
+    CANCELING_ANALYSIS_SESSION = 9
+
+    @staticmethod
+    def get(corpus_status: Union['CorpusStatus', int, str]) -> 'CorpusStatus':
+        if isinstance(corpus_status, CorpusStatus):
+            return corpus_status
+        if isinstance(corpus_status, int):
+            return CorpusStatus(corpus_status)
+        if isinstance(corpus_status, str):
+            return CorpusStatus[corpus_status]
+        raise TypeError('corpus_status must be CorpusStatus, int, or str')
+
+
+class Corpus(HashidMixin, db.Model):
+    '''
+    Class to define a corpus.
+    '''
+    __tablename__ = 'corpora'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    user_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    # Fields
+    creation_date = db.Column(db.DateTime(), default=datetime.utcnow)
+    description = db.Column(db.String(255))
+    status = db.Column(
+        IntEnumColumn(CorpusStatus),
+        default=CorpusStatus.UNPREPARED
+    )
+    title = db.Column(db.String(32))
+    num_analysis_sessions = db.Column(db.Integer, default=0)
+    num_tokens = db.Column(db.Integer, default=0)
+    is_public = db.Column(db.Boolean, default=False)
+    # Relationships
+    files = db.relationship(
+        'CorpusFile',
+        back_populates='corpus',
+        lazy='dynamic',
+        cascade='all, delete-orphan'
+    )
+    corpus_follower_associations = db.relationship(
+        'CorpusFollowerAssociation',
+        back_populates='corpus',
+        cascade='all, delete-orphan'
+    )
+    followers = association_proxy(
+        'corpus_follower_associations',
+        'follower',
+        creator=lambda u: CorpusFollowerAssociation(follower=u)
+    )
+    user = db.relationship('User', back_populates='corpora')
+    # "static" attributes
+    max_num_tokens = 2_147_483_647
+
+    def __repr__(self):
+        return f'<Corpus {self.title}>'
+
+    @property
+    def analysis_url(self):
+        return url_for('corpora.analysis', corpus_id=self.id)
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.user.jsonpatch_path}/corpora/{self.hashid}'
+
+    @property
+    def path(self) -> Path:
+        return self.user.path / 'corpora' / f'{self.id}'
+
+    @property
+    def url(self):
+        return url_for('corpora.corpus', corpus_id=self.id)
+
+    @property
+    def user_hashid(self):
+        return self.user.hashid
+
+    @staticmethod
+    def create(**kwargs):
+        corpus = Corpus(**kwargs)
+        db.session.add(corpus)
+        db.session.flush(objects=[corpus])
+        db.session.refresh(corpus)
+        corpus_files_dir = corpus.path / 'files'
+        corpus_cwb_dir = corpus.path / 'cwb'
+        corpus_cwb_data_dir = corpus_cwb_dir / 'data'
+        corpus_cwb_registry_dir = corpus_cwb_dir / 'registry'
+        try:
+            corpus.path.mkdir()
+            corpus_files_dir.mkdir()
+            corpus_cwb_dir.mkdir()
+            corpus_cwb_data_dir.mkdir()
+            corpus_cwb_registry_dir.mkdir()
+        except OSError as e:
+            # TODO: Potential leftover cleanup
+            current_app.logger.error(e)
+            db.session.rollback()
+            raise
+        return corpus
+
+    def build(self):
+        corpus_cwb_dir = self.path / 'cwb'
+        corpus_cwb_data_dir = corpus_cwb_dir / 'data'
+        corpus_cwb_registry_dir = corpus_cwb_dir / 'registry'
+        try:
+            shutil.rmtree(corpus_cwb_dir, ignore_errors=True)
+            corpus_cwb_dir.mkdir()
+            corpus_cwb_data_dir.mkdir()
+            corpus_cwb_registry_dir.mkdir()
+        except OSError as e:
+            current_app.logger.error(e)
+            self.status = CorpusStatus.FAILED
+            raise
+        corpus_element = ET.fromstring('<corpus>\n</corpus>')
+        for corpus_file in self.files:
+            normalized_vrt_path = corpus_cwb_dir / f'{corpus_file.id}.norm.vrt'
+            try:
+                normalize_vrt_file(corpus_file.path, normalized_vrt_path)
+            except:
+                self.status = CorpusStatus.FAILED
+                return
+            element_tree = ET.parse(normalized_vrt_path)
+            text_element = element_tree.getroot()
+            text_element.set('author', corpus_file.author)
+            text_element.set('title', corpus_file.title)
+            text_element.set(
+                'publishing_year',
+                f'{corpus_file.publishing_year}'
+            )
+            text_element.set('address', corpus_file.address or 'NULL')
+            text_element.set('booktitle', corpus_file.booktitle or 'NULL')
+            text_element.set('chapter', corpus_file.chapter or 'NULL')
+            text_element.set('editor', corpus_file.editor or 'NULL')
+            text_element.set('institution', corpus_file.institution or 'NULL')
+            text_element.set('journal', corpus_file.journal or 'NULL')
+            text_element.set('pages', f'{corpus_file.pages}' or 'NULL')
+            text_element.set('publisher', corpus_file.publisher or 'NULL')
+            text_element.set('school', corpus_file.school or 'NULL')
+            text_element.tail = '\n'
+            # corpus_element.insert(1, text_element)
+            corpus_element.append(text_element)
+        ET.ElementTree(corpus_element).write(
+            corpus_cwb_dir / 'corpus.vrt',
+            encoding='utf-8'
+        )
+        self.status = CorpusStatus.SUBMITTED
+
+    def delete(self):
+        shutil.rmtree(self.path, ignore_errors=True)
+        db.session.delete(self)
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'creation_date': f'{self.creation_date.isoformat()}Z',
+            'description': self.description,
+            'max_num_tokens': self.max_num_tokens,
+            'num_analysis_sessions': self.num_analysis_sessions,
+            'num_tokens': self.num_tokens,
+            'status': self.status.name,
+            'title': self.title,
+            'is_public': self.is_public
+        }
+        if backrefs:
+            json_serializeable['user'] = \
+                self.user.to_json_serializeable(backrefs=True)
+        if relationships:
+            json_serializeable['corpus_follower_associations'] = {
+                x.hashid: x.to_json_serializeable()
+                for x in self.corpus_follower_associations
+            }
+            json_serializeable['files'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.files
+            }
+        return json_serializeable

app/models/corpus_file.py

@@ -0,0 +1,102 @@
+from flask import current_app, url_for
+from flask_hashids import HashidMixin
+from pathlib import Path
+from app import db
+from .corpus import CorpusStatus
+from .file_mixin import FileMixin
+
+
+class CorpusFile(FileMixin, HashidMixin, db.Model):
+    __tablename__ = 'corpus_files'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    corpus_id = db.Column(db.Integer, db.ForeignKey('corpora.id'))
+    # Fields
+    author = db.Column(db.String(255))
+    description = db.Column(db.String(255))
+    publishing_year = db.Column(db.Integer)
+    title = db.Column(db.String(255))
+    address = db.Column(db.String(255))
+    booktitle = db.Column(db.String(255))
+    chapter = db.Column(db.String(255))
+    editor = db.Column(db.String(255))
+    institution = db.Column(db.String(255))
+    journal = db.Column(db.String(255))
+    pages = db.Column(db.String(255))
+    publisher = db.Column(db.String(255))
+    school = db.Column(db.String(255))
+    # Relationships
+    corpus = db.relationship(
+        'Corpus',
+        back_populates='files'
+    )
+
+    @property
+    def download_url(self):
+        return url_for(
+            'corpora.download_corpus_file',
+            corpus_id=self.corpus_id,
+            corpus_file_id=self.id
+        )
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.corpus.jsonpatch_path}/files/{self.hashid}'
+
+    @property
+    def path(self) -> Path:
+        return self.corpus.path / 'files' / f'{self.id}'
+
+    @property
+    def url(self):
+        return url_for(
+            'corpora.corpus_file',
+            corpus_id=self.corpus_id,
+            corpus_file_id=self.id
+        )
+
+    @property
+    def user_hashid(self):
+        return self.corpus.user.hashid
+
+    @property
+    def user_id(self):
+        return self.corpus.user_id
+
+    def delete(self):
+        try:
+            self.path.unlink(missing_ok=True)
+        except OSError as e:
+            current_app.logger.error(e)
+            raise
+        db.session.delete(self)
+        self.corpus.status = CorpusStatus.UNPREPARED
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'address': self.address,
+            'author': self.author,
+            'description': self.description,
+            'booktitle': self.booktitle,
+            'chapter': self.chapter,
+            'editor': self.editor,
+            'institution': self.institution,
+            'journal': self.journal,
+            'pages': self.pages,
+            'publisher': self.publisher,
+            'publishing_year': self.publishing_year,
+            'school': self.school,
+            'title': self.title,
+            **self.file_mixin_to_json_serializeable(
+                backrefs=backrefs,
+                relationships=relationships
+            )
+        }
+        if backrefs:
+            json_serializeable['corpus'] = \
+                self.corpus.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable

app/models/corpus_follower_association.py

@@ -0,0 +1,47 @@
+from flask_hashids import HashidMixin
+from app import db
+from .corpus_follower_role import CorpusFollowerRole
+
+
+class CorpusFollowerAssociation(HashidMixin, db.Model):
+    __tablename__ = 'corpus_follower_associations'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    corpus_id = db.Column(db.Integer, db.ForeignKey('corpora.id'))
+    follower_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    role_id = db.Column(db.Integer, db.ForeignKey('corpus_follower_roles.id'))
+    # Relationships
+    corpus = db.relationship(
+        'Corpus',
+        back_populates='corpus_follower_associations'
+    )
+    follower = db.relationship(
+        'User',
+        back_populates='corpus_follower_associations'
+    )
+    role = db.relationship(
+        'CorpusFollowerRole',
+        back_populates='corpus_follower_associations'
+    )
+
+    def __init__(self, **kwargs):
+        if 'role' not in kwargs:
+            kwargs['role'] = CorpusFollowerRole.query.filter_by(default=True).first()
+        super().__init__(**kwargs)
+
+    def __repr__(self):
+        return f'<CorpusFollowerAssociation {self.follower.__repr__()} ~ {self.role.__repr__()} ~ {self.corpus.__repr__()}>'
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'corpus': self.corpus.to_json_serializeable(backrefs=True),
+            'follower': self.follower.to_json_serializeable(),
+            'role': self.role.to_json_serializeable()
+        }
+        if backrefs:
+            pass
+        if relationships:
+            pass
+        return json_serializeable

app/models/corpus_follower_role.py

@@ -0,0 +1,107 @@
+from flask_hashids import HashidMixin
+from enum import IntEnum
+from typing import Union
+from app import db
+
+
+class CorpusFollowerPermission(IntEnum):
+    VIEW = 1
+    MANAGE_FILES = 2
+    MANAGE_FOLLOWERS = 4
+    MANAGE_CORPUS = 8
+
+    @staticmethod
+    def get(corpus_follower_permission: Union['CorpusFollowerPermission', int, str]) -> 'CorpusFollowerPermission':
+        if isinstance(corpus_follower_permission, CorpusFollowerPermission):
+            return corpus_follower_permission
+        if isinstance(corpus_follower_permission, int):
+            return CorpusFollowerPermission(corpus_follower_permission)
+        if isinstance(corpus_follower_permission, str):
+            return CorpusFollowerPermission[corpus_follower_permission]
+        raise TypeError('corpus_follower_permission must be CorpusFollowerPermission, int, or str')
+
+
+class CorpusFollowerRole(HashidMixin, db.Model):
+    __tablename__ = 'corpus_follower_roles'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Fields
+    name = db.Column(db.String(64), unique=True)
+    default = db.Column(db.Boolean, default=False, index=True)
+    permissions = db.Column(db.Integer, default=0)
+    # Relationships
+    corpus_follower_associations = db.relationship(
+        'CorpusFollowerAssociation',
+        back_populates='role'
+    )
+
+    def __repr__(self):
+        return f'<CorpusFollowerRole {self.name}>'
+
+    def has_permission(self, permission: Union[CorpusFollowerPermission, int, str]):
+        perm = CorpusFollowerPermission.get(permission)
+        return self.permissions & perm.value == perm.value
+    
+    def add_permission(self, permission: Union[CorpusFollowerPermission, int, str]):
+        perm = CorpusFollowerPermission.get(permission)
+        if not self.has_permission(perm):
+            self.permissions += perm.value
+    
+    def remove_permission(self, permission: Union[CorpusFollowerPermission, int, str]):
+        perm = CorpusFollowerPermission.get(permission)
+        if self.has_permission(perm):
+            self.permissions -= perm.value
+
+    def reset_permissions(self):
+        self.permissions = 0
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'default': self.default,
+            'name': self.name,
+            'permissions': [
+                x.name
+                for x in CorpusFollowerPermission
+                if self.has_permission(x)
+            ]
+        }
+        if backrefs:
+            pass
+        if relationships:
+            json_serializeable['corpus_follower_association'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.corpus_follower_association
+            }
+        return json_serializeable
+
+    @staticmethod
+    def insert_defaults():
+        roles = {
+            'Anonymous': [],
+            'Viewer': [
+                CorpusFollowerPermission.VIEW
+            ],
+            'Contributor': [
+                CorpusFollowerPermission.VIEW,
+                CorpusFollowerPermission.MANAGE_FILES
+            ],
+            'Administrator': [
+                CorpusFollowerPermission.VIEW,
+                CorpusFollowerPermission.MANAGE_FILES,
+                CorpusFollowerPermission.MANAGE_FOLLOWERS,
+                CorpusFollowerPermission.MANAGE_CORPUS
+
+            ]
+        }
+        default_role_name = 'Viewer'
+        for role_name, permissions in roles.items():
+            role = CorpusFollowerRole.query.filter_by(name=role_name).first()
+            if role is None:
+                role = CorpusFollowerRole(name=role_name)
+            role.reset_permissions()
+            for permission in permissions:
+                role.add_permission(permission)
+            role.default = role.name == default_role_name
+            db.session.add(role)
+        db.session.commit()

app/models/default_records/spacy_nlp_pipeline_model.yml

@@ -8,7 +8,7 @@
   pipeline_name: 'ca_core_news_md'
   version: '3.2.0'
   compatible_service_versions:
-    - '0.1.0' 
+    - '0.1.0'
 - title: 'German'
   description: 'German pipeline optimized for CPU. Components: tok2vec, tagger, morphologizer, parser, senter, ner, attribute_ruler, lemmatizer.'
   url: 'https://github.com/explosion/spacy-models/releases/download/de_core_news_md-3.2.0/de_core_news_md-3.2.0.tar.gz'
@@ -19,7 +19,7 @@
   pipeline_name: 'de_core_news_md'
   version: '3.2.0'
   compatible_service_versions:
-    - '0.1.0' 
+    - '0.1.0'
 - title: 'Greek'
   description: 'Greek pipeline optimized for CPU. Components: tok2vec, morphologizer, parser, senter, ner, attribute_ruler, lemmatizer.'
   url: 'https://github.com/explosion/spacy-models/releases/download/el_core_news_md-3.2.0/el_core_news_md-3.2.0.tar.gz'
@@ -180,7 +180,7 @@
   version: '3.4.0'
   compatible_service_versions:
     - '0.1.1'
-    - '0.1.2' 
+    - '0.1.2'
 - title: 'Italian'
   description: 'Italian pipeline optimized for CPU. Components: tok2vec, morphologizer, tagger, parser, lemmatizer (trainable_lemmatizer), senter, ner'
   url: 'https://github.com/explosion/spacy-models/releases/download/it_core_news_md-3.4.0/it_core_news_md-3.4.0.tar.gz'

app/models/default_records/tesseract_ocr_pipeline_model.yml

@@ -9,6 +9,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Amharic'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/amh.traineddata'
@@ -20,6 +21,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Arabic'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ara.traineddata'
@@ -31,6 +33,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Assamese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/asm.traineddata'
@@ -42,6 +45,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Azerbaijani'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/aze.traineddata'
@@ -53,6 +57,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Azerbaijani - Cyrillic'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/aze_cyrl.traineddata'
@@ -64,6 +69,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Belarusian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/bel.traineddata'
@@ -75,6 +81,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Bengali'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ben.traineddata'
@@ -86,6 +93,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Tibetan'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/bod.traineddata'
@@ -97,6 +105,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Bosnian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/bos.traineddata'
@@ -108,6 +117,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Bulgarian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/bul.traineddata'
@@ -119,6 +129,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Catalan; Valencian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/cat.traineddata'
@@ -130,6 +141,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Cebuano'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ceb.traineddata'
@@ -141,6 +153,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Czech'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ces.traineddata'
@@ -152,6 +165,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Chinese - Simplified'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/chi_sim.traineddata'
@@ -163,6 +177,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Chinese - Traditional'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/chi_tra.traineddata'
@@ -174,6 +189,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Cherokee'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/chr.traineddata'
@@ -185,6 +201,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Welsh'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/cym.traineddata'
@@ -196,6 +213,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Danish'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/dan.traineddata'
@@ -207,6 +225,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'German'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/deu.traineddata'
@@ -218,6 +237,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Dzongkha'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/dzo.traineddata'
@@ -229,6 +249,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Greek, Modern (1453-)'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ell.traineddata'
@@ -240,6 +261,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'English'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/eng.traineddata'
@@ -251,6 +273,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'English, Middle (1100-1500)'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/enm.traineddata'
@@ -262,6 +285,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Esperanto'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/epo.traineddata'
@@ -273,6 +297,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Estonian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/est.traineddata'
@@ -284,6 +309,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Basque'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/eus.traineddata'
@@ -295,6 +321,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Persian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/fas.traineddata'
@@ -306,6 +333,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Finnish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/fin.traineddata'
@@ -317,6 +345,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'French'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/fra.traineddata'
@@ -328,6 +357,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'German Fraktur'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/frk.traineddata'
@@ -339,6 +369,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'French, Middle (ca. 1400-1600)'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/frm.traineddata'
@@ -350,6 +381,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Irish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/gle.traineddata'
@@ -361,6 +393,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Galician'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/glg.traineddata'
@@ -372,6 +405,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Greek, Ancient (-1453)'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/grc.traineddata'
@@ -383,6 +417,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Gujarati'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/guj.traineddata'
@@ -394,6 +429,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Haitian; Haitian Creole'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/hat.traineddata'
@@ -405,6 +441,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Hebrew'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/heb.traineddata'
@@ -416,6 +453,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Hindi'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/hin.traineddata'
@@ -427,6 +465,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Croatian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/hrv.traineddata'
@@ -438,6 +477,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Hungarian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/hun.traineddata'
@@ -449,6 +489,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Inuktitut'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/iku.traineddata'
@@ -460,6 +501,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Indonesian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ind.traineddata'
@@ -471,6 +513,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Icelandic'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/isl.traineddata'
@@ -482,6 +525,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Italian'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ita.traineddata'
@@ -493,6 +537,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'Italian - Old'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ita_old.traineddata'
@@ -504,6 +549,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Javanese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/jav.traineddata'
@@ -515,6 +561,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Japanese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/jpn.traineddata'
@@ -526,6 +573,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Kannada'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kan.traineddata'
@@ -537,6 +585,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Georgian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kat.traineddata'
@@ -548,6 +597,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Georgian - Old'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kat_old.traineddata'
@@ -559,6 +609,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Kazakh'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kaz.traineddata'
@@ -570,6 +621,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Central Khmer'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/khm.traineddata'
@@ -581,6 +633,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Kirghiz; Kyrgyz'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kir.traineddata'
@@ -592,6 +645,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Korean'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kor.traineddata'
@@ -603,6 +657,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Kurdish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/kur.traineddata'
@@ -614,6 +669,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Lao'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/lao.traineddata'
@@ -625,6 +681,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Latin'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/lat.traineddata'
@@ -636,6 +693,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Latvian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/lav.traineddata'
@@ -647,6 +705,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Lithuanian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/lit.traineddata'
@@ -658,6 +717,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Malayalam'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/mal.traineddata'
@@ -669,6 +729,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Marathi'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/mar.traineddata'
@@ -680,6 +741,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Macedonian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/mkd.traineddata'
@@ -691,6 +753,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Maltese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/mlt.traineddata'
@@ -702,6 +765,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Malay'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/msa.traineddata'
@@ -713,6 +777,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Burmese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/mya.traineddata'
@@ -724,6 +789,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Nepali'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/nep.traineddata'
@@ -735,6 +801,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Dutch; Flemish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/nld.traineddata'
@@ -746,6 +813,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Norwegian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/nor.traineddata'
@@ -757,6 +825,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Oriya'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ori.traineddata'
@@ -768,6 +837,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Panjabi; Punjabi'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/pan.traineddata'
@@ -779,6 +849,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Polish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/pol.traineddata'
@@ -790,6 +861,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Portuguese'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/por.traineddata'
@@ -801,6 +873,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Pushto; Pashto'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/pus.traineddata'
@@ -812,6 +885,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Romanian; Moldavian; Moldovan'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ron.traineddata'
@@ -823,6 +897,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Russian'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/rus.traineddata'
@@ -834,6 +909,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Sanskrit'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/san.traineddata'
@@ -845,6 +921,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Sinhala; Sinhalese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/sin.traineddata'
@@ -856,6 +933,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Slovak'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/slk.traineddata'
@@ -867,6 +945,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Slovenian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/slv.traineddata'
@@ -878,6 +957,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 - title: 'Spanish; Castilian'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/spa.traineddata'
@@ -889,6 +969,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 - title: 'Spanish; Castilian - Old'
   description: ''
   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/spa_old.traineddata'
@@ -900,6 +981,7 @@
   compatible_service_versions:
     - '0.1.0'
     - '0.1.1'
+    - '0.1.2'
 # - title: 'Albanian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/sqi.traineddata'
@@ -911,6 +993,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Serbian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/srp.traineddata'
@@ -922,6 +1005,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Serbian - Latin'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/srp_latn.traineddata'
@@ -933,6 +1017,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Swahili'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/swa.traineddata'
@@ -944,6 +1029,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Swedish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/swe.traineddata'
@@ -955,6 +1041,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Syriac'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/syr.traineddata'
@@ -966,6 +1053,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Tamil'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tam.traineddata'
@@ -977,6 +1065,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Telugu'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tel.traineddata'
@@ -988,6 +1077,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Tajik'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tgk.traineddata'
@@ -999,6 +1089,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Tagalog'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tgl.traineddata'
@@ -1010,6 +1101,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Thai'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tha.traineddata'
@@ -1021,6 +1113,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Tigrinya'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tir.traineddata'
@@ -1032,6 +1125,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Turkish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/tur.traineddata'
@@ -1043,6 +1137,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Uighur; Uyghur'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/uig.traineddata'
@@ -1054,6 +1149,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Ukrainian'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/ukr.traineddata'
@@ -1065,6 +1161,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Urdu'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/urd.traineddata'
@@ -1076,6 +1173,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Uzbek'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/uzb.traineddata'
@@ -1087,6 +1185,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Uzbek - Cyrillic'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/uzb_cyrl.traineddata'
@@ -1098,6 +1197,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Vietnamese'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/vie.traineddata'
@@ -1109,6 +1209,7 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'
 # - title: 'Yiddish'
 #   description: ''
 #   url: 'https://github.com/tesseract-ocr/tessdata/raw/4.1.0/yid.traineddata'
@@ -1120,3 +1221,4 @@
 #   compatible_service_versions:
 #     - '0.1.0'
 #     - '0.1.1'
+#     - '0.1.2'

app/models/event_listeners.py

@@ -0,0 +1,133 @@
+from datetime import datetime
+from enum import Enum
+from app import db, mail, socketio
+from app.email import create_message
+from .corpus_file import CorpusFile
+from .corpus_follower_association import CorpusFollowerAssociation
+from .corpus import Corpus
+from .job_input import JobInput
+from .job_result import JobResult
+from .job import Job, JobStatus
+from .spacy_nlp_pipeline_model import SpaCyNLPPipelineModel
+from .tesseract_ocr_pipeline_model import TesseractOCRPipelineModel
+from .user import UserSettingJobStatusMailNotificationLevel
+
+
+def register_event_listeners():
+    resources = [
+        Corpus,
+        CorpusFile,
+        Job,
+        JobInput,
+        JobResult,
+        SpaCyNLPPipelineModel,
+        TesseractOCRPipelineModel
+    ]
+
+    for resource in resources:
+        db.event.listen(resource, 'after_delete', resource_after_delete)
+        db.event.listen(resource, 'after_insert', resource_after_insert)
+        db.event.listen(resource, 'after_update', resource_after_update)
+
+    db.event.listen(CorpusFollowerAssociation, 'after_delete', cfa_after_delete)
+    db.event.listen(CorpusFollowerAssociation, 'after_insert', cfa_after_insert)
+
+    db.event.listen(Job, 'after_update', job_after_update)
+
+
+def resource_after_delete(mapper, connection, resource):
+    jsonpatch = [
+        {
+            'op': 'remove',
+            'path': resource.jsonpatch_path
+        }
+    ]
+    room = f'/users/{resource.user_hashid}'
+    socketio.emit('PATCH', jsonpatch, room=room)
+
+
+def cfa_after_delete(mapper, connection, cfa):
+    jsonpatch_path = f'/users/{cfa.corpus.user.hashid}/corpora/{cfa.corpus.hashid}/corpus_follower_associations/{cfa.hashid}'
+    jsonpatch = [
+        {
+            'op': 'remove',
+            'path': jsonpatch_path
+        }
+    ]
+    room = f'/users/{cfa.corpus.user.hashid}'
+    socketio.emit('PATCH', jsonpatch, room=room)
+
+
+def resource_after_insert(mapper, connection, resource):
+    jsonpatch_value = resource.to_json_serializeable()
+    for attr in mapper.relationships:
+        jsonpatch_value[attr.key] = {}
+    jsonpatch = [
+        {
+            'op': 'add',
+            'path': resource.jsonpatch_path,
+            'value': jsonpatch_value
+        }
+    ]
+    room = f'/users/{resource.user_hashid}'
+    socketio.emit('PATCH', jsonpatch, room=room)
+
+
+def cfa_after_insert(mapper, connection, cfa):
+    jsonpatch_value = cfa.to_json_serializeable()
+    jsonpatch_path = f'/users/{cfa.corpus.user.hashid}/corpora/{cfa.corpus.hashid}/corpus_follower_associations/{cfa.hashid}'
+    jsonpatch = [
+        {
+            'op': 'add',
+            'path': jsonpatch_path,
+            'value': jsonpatch_value
+        }
+    ]
+    room = f'/users/{cfa.corpus.user.hashid}'
+    socketio.emit('PATCH', jsonpatch, room=room)
+
+
+def resource_after_update(mapper, connection, resource):
+    jsonpatch = []
+    for attr in db.inspect(resource).attrs:
+        if attr.key in mapper.relationships:
+            continue
+        if not attr.load_history().has_changes():
+            continue
+        jsonpatch_path = f'{resource.jsonpatch_path}/{attr.key}'
+        if isinstance(attr.value, datetime):
+            jsonpatch_value = f'{attr.value.isoformat()}Z'
+        elif isinstance(attr.value, Enum):
+            jsonpatch_value = attr.value.name
+        else:
+            jsonpatch_value = attr.value
+        jsonpatch.append(
+            {
+                'op': 'replace',
+                'path': jsonpatch_path,
+                'value': jsonpatch_value
+            }
+        )
+    if jsonpatch:
+        room = f'/users/{resource.user_hashid}'
+        socketio.emit('PATCH', jsonpatch, room=room)
+
+
+def job_after_update(mapper, connection, job):
+    for attr in db.inspect(job).attrs:
+        if attr.key != 'status':
+            continue
+        if not attr.load_history().has_changes():
+            return
+        if job.user.setting_job_status_mail_notification_level == UserSettingJobStatusMailNotificationLevel.NONE:
+            return
+        if job.user.setting_job_status_mail_notification_level == UserSettingJobStatusMailNotificationLevel.END:
+            if job.status not in [JobStatus.COMPLETED, JobStatus.FAILED]:
+                return
+        msg = create_message(
+            job.user.email,
+            f'Status update for your Job "{job.title}"',
+            'tasks/email/notification',
+            job=job
+        )
+        mail.send(msg)

app/models/file_mixin.py

@@ -0,0 +1,40 @@
+from datetime import datetime
+from flask import current_app
+from werkzeug.utils import secure_filename
+from app import db
+
+
+class FileMixin:
+    '''
+    Mixin for db.Model classes. All file related models should use this.
+    '''
+    creation_date = db.Column(db.DateTime, default=datetime.utcnow)
+    filename = db.Column(db.String(255))
+    mimetype = db.Column(db.String(255))
+
+    def file_mixin_to_json_serializeable(self, backrefs=False, relationships=False):
+        return {
+            'creation_date': f'{self.creation_date.isoformat()}Z',
+            'filename': self.filename,
+            'mimetype': self.mimetype
+        }
+    
+    @classmethod
+    def create(cls, file_storage, **kwargs):
+        filename = kwargs.pop('filename', file_storage.filename)
+        mimetype = kwargs.pop('mimetype', file_storage.mimetype)
+        obj = cls(
+            filename=secure_filename(filename),
+            mimetype=mimetype,
+            **kwargs
+        )
+        db.session.add(obj)
+        db.session.flush(objects=[obj])
+        db.session.refresh(obj)
+        try:
+            file_storage.save(obj.path)
+        except (AttributeError, OSError) as e:
+            current_app.logger.error(e)
+            db.session.rollback()
+            raise e
+        return obj

app/models/job.py

@@ -0,0 +1,172 @@
+from datetime import datetime
+from enum import IntEnum
+from flask import current_app, url_for
+from flask_hashids import HashidMixin
+from time import sleep
+from typing import Union
+from pathlib import Path
+import shutil
+from app import db
+from app.ext.flask_sqlalchemy import ContainerColumn, IntEnumColumn
+
+
+class JobStatus(IntEnum):
+    INITIALIZING = 1
+    SUBMITTED = 2
+    QUEUED = 3
+    RUNNING = 4
+    CANCELING = 5
+    CANCELED = 6
+    COMPLETED = 7
+    FAILED = 8
+
+    @staticmethod
+    def get(job_status: Union['JobStatus', int, str]) -> 'JobStatus':
+        if isinstance(job_status, JobStatus):
+            return job_status
+        if isinstance(job_status, int):
+            return JobStatus(job_status)
+        if isinstance(job_status, str):
+            return JobStatus[job_status]
+        raise TypeError('job_status must be JobStatus, int, or str')
+
+
+class Job(HashidMixin, db.Model):
+    '''
+    Class to define Jobs.
+    '''
+    __tablename__ = 'jobs'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    user_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    # Fields
+    creation_date = \
+        db.Column(db.DateTime(), default=datetime.utcnow)
+    description = db.Column(db.String(255))
+    end_date = db.Column(db.DateTime())
+    service = db.Column(db.String(64))
+    service_args = db.Column(ContainerColumn(dict, 255))
+    service_version = db.Column(db.String(16))
+    status = db.Column(
+        IntEnumColumn(JobStatus),
+        default=JobStatus.INITIALIZING
+    )
+    title = db.Column(db.String(32))
+    # Relationships
+    inputs = db.relationship(
+        'JobInput',
+        back_populates='job',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+    results = db.relationship(
+        'JobResult',
+        back_populates='job',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+    user = db.relationship(
+        'User',
+        back_populates='jobs'
+    )
+
+    def __repr__(self):
+        return f'<Job {self.title}>'
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.user.jsonpatch_path}/jobs/{self.hashid}'
+
+    @property
+    def path(self) -> Path:
+        return self.user.path / 'jobs' / f'{self.id}'
+
+    @property
+    def url(self):
+        return url_for('jobs.job', job_id=self.id)
+
+    @property
+    def user_hashid(self):
+        return self.user.hashid
+
+    @staticmethod
+    def create(**kwargs):
+        job = Job(**kwargs)
+        db.session.add(job)
+        db.session.flush(objects=[job])
+        db.session.refresh(job)
+        job_inputs_dir = job.path / 'inputs'
+        job_pipeline_data_dir = job.path / 'pipeline_data'
+        job_results_dir = job.path / 'results'
+        try:
+            job.path.mkdir()
+            job_inputs_dir.mkdir()
+            job_pipeline_data_dir.mkdir()
+            job_results_dir.mkdir()
+        except OSError as e:
+            # TODO: Potential leftover cleanup
+            current_app.logger.error(e)
+            db.session.rollback()
+            raise
+        return job
+
+    def delete(self):
+        ''' Delete the job and its inputs and results from the database. '''
+        if self.status not in [JobStatus.COMPLETED, JobStatus.FAILED]:  # noqa
+            self.status = JobStatus.CANCELING
+            db.session.commit()
+            while self.status != JobStatus.CANCELED:
+                # In case the daemon handled a job in any way
+                if self.status != JobStatus.CANCELING:
+                    self.status = JobStatus.CANCELING
+                    db.session.commit()
+                sleep(1)
+                db.session.refresh(self)
+        try:
+            shutil.rmtree(self.path)
+        except OSError as e:
+            current_app.logger.error(e)
+            db.session.rollback()
+            raise e
+        db.session.delete(self)
+
+    def restart(self):
+        ''' Restart a job - only if the status is failed '''
+        if self.status != JobStatus.FAILED:
+            raise Exception('Job status is not "failed"')
+        shutil.rmtree(self.path / 'results', ignore_errors=True)
+        shutil.rmtree(self.path / 'pyflow.data', ignore_errors=True)
+        for result in self.results:
+            db.session.delete(result)
+        self.end_date = None
+        self.status = JobStatus.SUBMITTED
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'creation_date': f'{self.creation_date.isoformat()}Z',
+            'description': self.description,
+            'end_date': (
+                None if self.end_date is None
+                else f'{self.end_date.isoformat()}Z'
+            ),
+            'service': self.service,
+            'service_args': self.service_args,
+            'service_version': self.service_version,
+            'status': self.status.name,
+            'title': self.title
+        }
+        if backrefs:
+            json_serializeable['user'] = \
+                self.user.to_json_serializeable(backrefs=True)
+        if relationships:
+            json_serializeable['inputs'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.inputs
+            }
+            json_serializeable['results'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.results
+            }
+        return json_serializeable

app/models/job_input.py

@@ -0,0 +1,65 @@
+from flask import url_for
+from flask_hashids import HashidMixin
+from pathlib import Path
+from app import db
+from .file_mixin import FileMixin
+
+
+class JobInput(FileMixin, HashidMixin, db.Model):
+    __tablename__ = 'job_inputs'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    job_id = db.Column(db.Integer, db.ForeignKey('jobs.id'))
+    # Relationships
+    job = db.relationship(
+        'Job',
+        back_populates='inputs'
+    )
+
+    def __repr__(self):
+        return f'<JobInput {self.filename}>'
+
+    @property
+    def content_url(self):
+        return url_for(
+            'jobs.download_job_input',
+            job_id=self.job.id,
+            job_input_id=self.id
+        )
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.job.jsonpatch_path}/inputs/{self.hashid}'
+
+    @property
+    def path(self) -> Path:
+        return self.job.path / 'inputs' / f'{self.id}'
+
+    @property
+    def url(self):
+        return url_for(
+            'jobs.job',
+            job_id=self.job_id,
+            _anchor=f'job-{self.job.hashid}-input-{self.hashid}'
+        )
+
+    @property
+    def user_hashid(self):
+        return self.job.user.hashid
+
+    @property
+    def user_id(self):
+        return self.job.user.id
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            **self.file_mixin_to_json_serializeable()
+        }
+        if backrefs:
+            json_serializeable['job'] = \
+                self.job.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable

app/models/job_result.py

@@ -0,0 +1,71 @@
+from flask import url_for
+from flask_hashids import HashidMixin
+from pathlib import Path
+from app import db
+from .file_mixin import FileMixin
+
+
+class JobResult(FileMixin, HashidMixin, db.Model):
+    __tablename__ = 'job_results'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    job_id = db.Column(db.Integer, db.ForeignKey('jobs.id'))
+    # Fields
+    description = db.Column(db.String(255))
+    # Relationships
+    job = db.relationship(
+        'Job',
+        back_populates='results'
+    )
+
+    def __repr__(self):
+        return f'<JobResult {self.filename}>'
+
+    @property
+    def download_url(self):
+        return url_for(
+            'jobs.download_job_result',
+            job_id=self.job_id,
+            job_result_id=self.id
+        )
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.job.jsonpatch_path}/results/{self.hashid}'
+
+    @property
+    def path(self) -> Path:
+        return self.job.path / 'results' / f'{self.id}'
+
+    @property
+    def url(self):
+        return url_for(
+            'jobs.job',
+            job_id=self.job_id,
+            _anchor=f'job-{self.job.hashid}-result-{self.hashid}'
+        )
+
+    @property
+    def user_hashid(self):
+        return self.job.user.hashid
+
+    @property
+    def user_id(self):
+        return self.job.user.id
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'description': self.description,
+            **self.file_mixin_to_json_serializeable(
+                backrefs=backrefs,
+                relationships=relationships
+            )
+        }
+        if backrefs:
+            json_serializeable['job'] = \
+                self.job.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable

app/models/role.py

@@ -0,0 +1,100 @@
+from enum import IntEnum
+from flask_hashids import HashidMixin
+from typing import Union
+from app import db
+
+
+class Permission(IntEnum):
+    '''
+    Defines User permissions as integers by the power of 2. User permission
+    can be evaluated using the bitwise operator &.
+    '''
+    ADMINISTRATE = 1
+    CONTRIBUTE = 2
+    USE_API = 4
+
+    @staticmethod
+    def get(permission: Union['Permission', int, str]) -> 'Permission':
+        if isinstance(permission, Permission):
+            return permission
+        if isinstance(permission, int):
+            return Permission(permission)
+        if isinstance(permission, str):
+            return Permission[permission]
+        raise TypeError('permission must be Permission, int, or str')
+
+
+class Role(HashidMixin, db.Model):
+    __tablename__ = 'roles'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Fields
+    name = db.Column(db.String(64), unique=True)
+    default = db.Column(db.Boolean, default=False, index=True)
+    permissions = db.Column(db.Integer, default=0)
+    # Relationships
+    users = db.relationship('User', back_populates='role', lazy='dynamic')
+
+    def __repr__(self):
+        return f'<Role {self.name}>'
+
+    def has_permission(self, permission: Union[Permission, int, str]):
+        p = Permission.get(permission)
+        return self.permissions & p.value == p.value
+    
+    def add_permission(self, permission: Union[Permission, int, str]):
+        p = Permission.get(permission)
+        if not self.has_permission(p):
+            self.permissions += p.value
+    
+    def remove_permission(self, permission: Union[Permission, int, str]):
+        p = Permission.get(permission)
+        if self.has_permission(p):
+            self.permissions -= p.value
+
+    def reset_permissions(self):
+        self.permissions = 0
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'default': self.default,
+            'name': self.name,
+            'permissions': [
+                x.name for x in Permission
+                if self.has_permission(x.value)
+            ]
+        }
+        if backrefs:
+            pass
+        if relationships:
+            json_serializeable['users'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.users
+            }
+        return json_serializeable
+
+    @staticmethod
+    def insert_defaults():
+        roles = {
+            'User': [],
+            'API user': [Permission.USE_API],
+            'Contributor': [Permission.CONTRIBUTE],
+            'Administrator': [
+                Permission.ADMINISTRATE,
+                Permission.CONTRIBUTE,
+                Permission.USE_API
+            ],
+            'System user': []
+        }
+        default_role_name = 'User'
+        for role_name, permissions in roles.items():
+            role = Role.query.filter_by(name=role_name).first()
+            if role is None:
+                role = Role(name=role_name)
+            role.reset_permissions()
+            for permission in permissions:
+                role.add_permission(permission)
+            role.default = role.name == default_role_name
+            db.session.add(role)
+        db.session.commit()

app/models/spacy_nlp_pipeline_model.py

@@ -0,0 +1,136 @@
+from flask import current_app, url_for
+from flask_hashids import HashidMixin
+from tqdm import tqdm
+from pathlib import Path
+import requests
+import yaml
+from app import db
+from app.ext.flask_sqlalchemy import ContainerColumn
+from .file_mixin import FileMixin
+from .user import User
+
+
+class SpaCyNLPPipelineModel(FileMixin, HashidMixin, db.Model):
+    __tablename__ = 'spacy_nlp_pipeline_models'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    user_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    # Fields
+    title = db.Column(db.String(64))
+    description = db.Column(db.String(255))
+    version = db.Column(db.String(16))
+    compatible_service_versions = db.Column(ContainerColumn(list, 255))
+    publisher = db.Column(db.String(128))
+    publisher_url = db.Column(db.String(512))
+    publishing_url = db.Column(db.String(512))
+    publishing_year = db.Column(db.Integer)
+    pipeline_name = db.Column(db.String(64))
+    is_public = db.Column(db.Boolean, default=False)
+    # Relationships
+    user = db.relationship('User', back_populates='spacy_nlp_pipeline_models')
+
+    @property
+    def path(self) -> Path:
+        return self.user.path / 'spacy_nlp_pipeline_models' / f'{self.id}'
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.user.jsonpatch_path}/spacy_nlp_pipeline_models/{self.hashid}'
+
+    @property
+    def url(self):
+        return url_for(
+            'contributions.spacy_nlp_pipeline_model',
+            spacy_nlp_pipeline_model_id=self.id
+        )
+
+    @property
+    def user_hashid(self):
+        return self.user.hashid
+
+    @staticmethod
+    def insert_defaults(force_download=False):
+        nopaque_user = User.query.filter_by(username='nopaque').first()
+        default_records_file = Path(__file__).parent / 'default_records' / 'spacy_nlp_pipeline_model.yml'
+        with default_records_file.open('r') as f:
+            default_records = yaml.safe_load(f)
+        for m in default_records:
+            model = SpaCyNLPPipelineModel.query.filter_by(title=m['title'], version=m['version']).first()  # noqa
+            if model is not None:
+                model.compatible_service_versions = m['compatible_service_versions']
+                model.description = m['description']
+                model.filename = m['url'].split('/')[-1]
+                model.publisher = m['publisher']
+                model.publisher_url = m['publisher_url']
+                model.publishing_url = m['publishing_url']
+                model.publishing_year = m['publishing_year']
+                model.is_public = True
+                model.title = m['title']
+                model.version = m['version']
+                model.pipeline_name = m['pipeline_name']
+            else:
+                model = SpaCyNLPPipelineModel(
+                    compatible_service_versions=m['compatible_service_versions'],
+                    description=m['description'],
+                    filename=m['url'].split('/')[-1],
+                    publisher=m['publisher'],
+                    publisher_url=m['publisher_url'],
+                    publishing_url=m['publishing_url'],
+                    publishing_year=m['publishing_year'],
+                    is_public=True,
+                    title=m['title'],
+                    user=nopaque_user,
+                    version=m['version'],
+                    pipeline_name=m['pipeline_name']
+                )
+                db.session.add(model)
+                db.session.flush(objects=[model])
+                db.session.refresh(model)
+            if not model.path.exists() or force_download:
+                r = requests.get(m['url'], stream=True)
+                pbar = tqdm(
+                    desc=f'{model.title} ({model.filename})',
+                    unit="B",
+                    unit_scale=True,
+                    unit_divisor=1024,
+                    total=int(r.headers['Content-Length'])
+                )
+                pbar.clear()
+                with open(model.path, 'wb') as f:
+                    for chunk in r.iter_content(chunk_size=1024):
+                        if chunk:  # filter out keep-alive new chunks
+                            pbar.update(len(chunk))
+                            f.write(chunk)
+                    pbar.close()
+        db.session.commit()
+    
+    def delete(self):
+        try:
+            self.path.unlink(missing_ok=True)
+        except OSError as e:
+            current_app.logger.error(e)
+            raise
+        db.session.delete(self)
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'compatible_service_versions': self.compatible_service_versions,
+            'description': self.description,
+            'publisher': self.publisher,
+            'publisher_url': self.publisher_url,
+            'publishing_url': self.publishing_url,
+            'publishing_year': self.publishing_year,
+            'pipeline_name': self.pipeline_name,
+            'is_public': self.is_public,
+            'title': self.title,
+            'version': self.version,
+            **self.file_mixin_to_json_serializeable()
+        }
+        if backrefs:
+            json_serializeable['user'] = \
+                self.user.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable

app/models/tesseract_ocr_pipeline_model.py

@@ -0,0 +1,132 @@
+from flask import current_app, url_for
+from flask_hashids import HashidMixin
+from tqdm import tqdm
+from pathlib import Path
+import requests
+import yaml
+from app import db
+from app.ext.flask_sqlalchemy import ContainerColumn
+from .file_mixin import FileMixin
+from .user import User
+
+
+class TesseractOCRPipelineModel(FileMixin, HashidMixin, db.Model):
+    __tablename__ = 'tesseract_ocr_pipeline_models'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    user_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    # Fields
+    title = db.Column(db.String(64))
+    description = db.Column(db.String(255))
+    version = db.Column(db.String(16))
+    compatible_service_versions = db.Column(ContainerColumn(list, 255))
+    publisher = db.Column(db.String(128))
+    publisher_url = db.Column(db.String(512))
+    publishing_url = db.Column(db.String(512))
+    publishing_year = db.Column(db.Integer)
+    is_public = db.Column(db.Boolean, default=False)
+    # Relationships
+    user = db.relationship('User', back_populates='tesseract_ocr_pipeline_models')
+
+    @property
+    def path(self) -> Path:
+        return self.user.path / 'tesseract_ocr_pipeline_models' / f'{self.id}'
+
+    @property
+    def jsonpatch_path(self):
+        return f'{self.user.jsonpatch_path}/tesseract_ocr_pipeline_models/{self.hashid}'
+
+    @property
+    def url(self):
+        return url_for(
+            'contributions.tesseract_ocr_pipeline_model',
+            tesseract_ocr_pipeline_model_id=self.id
+        )
+
+    @property
+    def user_hashid(self):
+        return self.user.hashid
+
+    @staticmethod
+    def insert_defaults(force_download=False):
+        nopaque_user = User.query.filter_by(username='nopaque').first()
+        default_records_file = Path(__file__).parent / 'default_records' / 'tesseract_ocr_pipeline_model.yml'
+        with default_records_file.open('r') as f:
+            default_records = yaml.safe_load(f)
+        for m in default_records:
+            model = TesseractOCRPipelineModel.query.filter_by(title=m['title'], version=m['version']).first()  # noqa
+            if model is not None:
+                model.compatible_service_versions = m['compatible_service_versions']
+                model.description = m['description']
+                model.filename = f'{model.id}.traineddata'
+                model.publisher = m['publisher']
+                model.publisher_url = m['publisher_url']
+                model.publishing_url = m['publishing_url']
+                model.publishing_year = m['publishing_year']
+                model.is_public = True
+                model.title = m['title']
+                model.version = m['version']
+            else:
+                model = TesseractOCRPipelineModel(
+                    compatible_service_versions=m['compatible_service_versions'],
+                    description=m['description'],
+                    publisher=m['publisher'],
+                    publisher_url=m['publisher_url'],
+                    publishing_url=m['publishing_url'],
+                    publishing_year=m['publishing_year'],
+                    is_public=True,
+                    title=m['title'],
+                    user=nopaque_user,
+                    version=m['version']
+                )
+                db.session.add(model)
+                db.session.flush(objects=[model])
+                db.session.refresh(model)
+                model.filename = f'{model.id}.traineddata'
+            if not model.path.exists() or force_download:
+                r = requests.get(m['url'], stream=True)
+                pbar = tqdm(
+                    desc=f'{model.title} ({model.filename})',
+                    unit="B",
+                    unit_scale=True,
+                    unit_divisor=1024,
+                    total=int(r.headers['Content-Length'])
+                )
+                pbar.clear()
+                with open(model.path, 'wb') as f:
+                    for chunk in r.iter_content(chunk_size=1024):
+                        if chunk:  # filter out keep-alive new chunks
+                            pbar.update(len(chunk))
+                            f.write(chunk)
+                    pbar.close()
+        db.session.commit()
+
+    def delete(self):
+        try:
+            self.path.unlink(missing_ok=True)
+        except OSError as e:
+            current_app.logger.error(e)
+            raise
+        db.session.delete(self)
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'compatible_service_versions': self.compatible_service_versions,
+            'description': self.description,
+            'publisher': self.publisher,
+            'publisher_url': self.publisher_url,
+            'publishing_url': self.publishing_url,
+            'publishing_year': self.publishing_year,
+            'is_public': self.is_public,
+            'title': self.title,
+            'version': self.version,
+            **self.file_mixin_to_json_serializeable()
+        }
+        if backrefs:
+            json_serializeable['user'] = \
+                self.user.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable

app/models/token.py

@@ -0,0 +1,48 @@
+from datetime import datetime, timedelta
+from app import db
+
+
+class Token(db.Model):
+    __tablename__ = 'tokens'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    user_id = db.Column(db.Integer, db.ForeignKey('users.id'))
+    # Fields
+    access_token = db.Column(db.String(64), index=True)
+    access_expiration = db.Column(db.DateTime)
+    refresh_token = db.Column(db.String(64), index=True)
+    refresh_expiration = db.Column(db.DateTime)
+    # Relationships
+    user = db.relationship('User', back_populates='tokens')
+
+    def expire(self):
+        self.access_expiration = datetime.utcnow()
+        self.refresh_expiration = datetime.utcnow()
+
+    def to_json_serializeable(self, backrefs=False, relationships=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'access_token': self.access_token,
+            'access_expiration': (
+                None if self.access_expiration is None
+                else f'{self.access_expiration.isoformat()}Z'
+            ),
+            'refresh_token': self.refresh_token,
+            'refresh_expiration': (
+                None if self.refresh_expiration is None
+                else f'{self.refresh_expiration.isoformat()}Z'
+            )
+        }
+        if backrefs:
+            json_serializeable['user'] = \
+                self.user.to_json_serializeable(backrefs=True)
+        if relationships:
+            pass
+        return json_serializeable
+
+    @staticmethod
+    def clean():
+        """Remove any tokens that have been expired for more than a day."""
+        yesterday = datetime.utcnow() - timedelta(days=1)
+        Token.query.filter(Token.refresh_expiration < yesterday).delete()

app/models/user.py

@@ -0,0 +1,452 @@
+from datetime import datetime, timedelta
+from enum import IntEnum
+from flask import current_app, url_for
+from flask_hashids import HashidMixin
+from flask_login import UserMixin
+from sqlalchemy.ext.associationproxy import association_proxy
+from pathlib import Path
+from typing import Union
+from werkzeug.security import generate_password_hash, check_password_hash
+import jwt
+import re
+import secrets
+import shutil
+from app import db, hashids
+from app.ext.flask_sqlalchemy import IntEnumColumn
+from .corpus import Corpus
+from .corpus_follower_association import CorpusFollowerAssociation
+from .corpus_follower_role import CorpusFollowerRole
+from .role import Permission, Role
+from .token import Token
+
+
+class ProfilePrivacySettings(IntEnum):
+    SHOW_EMAIL = 1
+    SHOW_LAST_SEEN = 2
+    SHOW_MEMBER_SINCE = 4
+
+    @staticmethod
+    def get(profile_privacy_setting: Union['ProfilePrivacySettings', int, str]) -> 'ProfilePrivacySettings':
+        if isinstance(profile_privacy_setting, ProfilePrivacySettings):
+            return profile_privacy_setting
+        if isinstance(profile_privacy_setting, int):
+            return ProfilePrivacySettings(profile_privacy_setting)
+        if isinstance(profile_privacy_setting, str):
+            return ProfilePrivacySettings[profile_privacy_setting]
+        raise TypeError('profile_privacy_setting must be ProfilePrivacySettings, int, or str')
+
+
+class UserSettingJobStatusMailNotificationLevel(IntEnum):
+    NONE = 1
+    END = 2
+    ALL = 3
+
+
+class User(HashidMixin, UserMixin, db.Model):
+    __tablename__ = 'users'
+    # Primary key
+    id = db.Column(db.Integer, primary_key=True)
+    # Foreign keys
+    role_id = db.Column(db.Integer, db.ForeignKey('roles.id'))
+    # Fields
+    email = db.Column(db.String(254), index=True, unique=True)
+    username = db.Column(db.String(64), index=True, unique=True)
+    username_pattern = re.compile(r'^[A-Za-zÄÖÜäöüß0-9_.]*$')
+    password_hash = db.Column(db.String(128))
+    confirmed = db.Column(db.Boolean, default=False)
+    terms_of_use_accepted = db.Column(db.Boolean, default=False)
+    member_since = db.Column(db.DateTime(), default=datetime.utcnow)
+    setting_job_status_mail_notification_level = db.Column(
+        IntEnumColumn(UserSettingJobStatusMailNotificationLevel),
+        default=UserSettingJobStatusMailNotificationLevel.END
+    )
+    last_seen = db.Column(db.DateTime())
+    full_name = db.Column(db.String(64))
+    about_me = db.Column(db.String(256))
+    location = db.Column(db.String(64))
+    website = db.Column(db.String(128))
+    organization = db.Column(db.String(128))
+    is_public = db.Column(db.Boolean, default=False)
+    profile_privacy_settings = db.Column(db.Integer(), default=0)
+    # Relationships
+    avatar = db.relationship(
+        'Avatar',
+        back_populates='user',
+        cascade='all, delete-orphan',
+        uselist=False
+    )
+    corpora = db.relationship(
+        'Corpus',
+        back_populates='user',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+    corpus_follower_associations = db.relationship(
+        'CorpusFollowerAssociation',
+        back_populates='follower',
+        cascade='all, delete-orphan'
+    )
+    followed_corpora = association_proxy(
+        'corpus_follower_associations',
+        'corpus',
+        creator=lambda c: CorpusFollowerAssociation(corpus=c)
+    )
+    jobs = db.relationship(
+        'Job',
+        back_populates='user',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+    role = db.relationship(
+        'Role',
+        back_populates='users'
+    )
+    spacy_nlp_pipeline_models = db.relationship(
+        'SpaCyNLPPipelineModel',
+        back_populates='user',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+    tesseract_ocr_pipeline_models = db.relationship(
+        'TesseractOCRPipelineModel',
+        back_populates='user',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+    tokens = db.relationship(
+        'Token',
+        back_populates='user',
+        cascade='all, delete-orphan',
+        lazy='dynamic'
+    )
+
+    def __init__(self, **kwargs):
+        if 'role' not in kwargs:
+            kwargs['role'] = (
+                Role.query.filter_by(name='Administrator').first()
+                if kwargs['email'] == current_app.config['NOPAQUE_ADMIN']
+                else Role.query.filter_by(default=True).first()
+            )
+        super().__init__(**kwargs)
+
+    def __repr__(self):
+        return f'<User {self.username}>'
+
+    @property
+    def jsonpatch_path(self):
+        return f'/users/{self.hashid}'
+
+    @property
+    def password(self):
+        raise AttributeError('password is not a readable attribute')
+
+    @password.setter
+    def password(self, password):
+        self.password_hash = generate_password_hash(password)
+
+    @property
+    def path(self) -> Path:
+        return current_app.config.get('NOPAQUE_DATA_DIR') / 'users' / f'{self.id}'
+
+    @staticmethod
+    def create(**kwargs):
+        user = User(**kwargs)
+        db.session.add(user)
+        db.session.flush(objects=[user])
+        db.session.refresh(user)
+        user_spacy_nlp_pipeline_models_dir = user.path / 'spacy_nlp_pipeline_models'
+        user_tesseract_ocr_pipeline_models_dir = user.path / 'tesseract_ocr_pipeline_models'
+        user_corpora_dir = user.path / 'corpora'
+        user_jobs_dir = user.path / 'jobs'
+        try:
+            user.path.mkdir()
+            user_spacy_nlp_pipeline_models_dir.mkdir()
+            user_tesseract_ocr_pipeline_models_dir.mkdir()
+            user_corpora_dir.mkdir()
+            user_jobs_dir.mkdir()
+        except OSError as e:
+            # TODO: Potential leftover cleanup
+            current_app.logger.error(e)
+            db.session.rollback()
+            raise
+        return user
+
+    @staticmethod
+    def insert_defaults():
+        nopaque_user = User.query.filter_by(username='nopaque').first()
+        system_user_role = Role.query.filter_by(name='System user').first()
+        if nopaque_user is None:
+            nopaque_user = User.create(
+                username='nopaque',
+                role=system_user_role
+            )
+            db.session.add(nopaque_user)
+        elif nopaque_user.role != system_user_role:
+            nopaque_user.role = system_user_role
+        db.session.commit()
+
+    @staticmethod
+    def reset_password(token, new_password):
+        try:
+            payload = jwt.decode(
+                token,
+                current_app.config['SECRET_KEY'],
+                algorithms=['HS256'],
+                issuer=current_app.config['SERVER_NAME'],
+                options={'require': ['exp', 'iat', 'iss', 'purpose', 'sub']}
+            )
+        except jwt.PyJWTError:
+            return False
+        if payload.get('purpose') != 'User.reset_password':
+            return False
+        user_hashid = payload.get('sub')
+        user_id = hashids.decode(user_hashid)
+        user = User.query.get(user_id)
+        if user is None:
+            return False
+        user.password = new_password
+        db.session.add(user)
+        return True
+
+    @staticmethod
+    def verify_access_token(access_token, refresh_token=None):
+        token = Token.query.filter(Token.access_token == access_token).first()
+        if token is not None:
+            if token.access_expiration > datetime.utcnow():
+                token.user.ping()
+                db.session.commit()
+                if token.user.role.name != 'System user':
+                    return token.user
+
+    @staticmethod
+    def verify_refresh_token(refresh_token, access_token):
+        token = Token.query.filter((Token.refresh_token == refresh_token) & (Token.access_token == access_token)).first()
+        if token is not None:
+            if token.refresh_expiration > datetime.utcnow():
+                return token
+            # someone tried to refresh with an expired token
+            # revoke all tokens from this user as a precaution
+            token.user.revoke_auth_tokens()
+            db.session.commit()
+
+    def can(self, permission):
+        return self.role is not None and self.role.has_permission(permission)
+
+    def confirm(self, confirmation_token):
+        try:
+            payload = jwt.decode(
+                confirmation_token,
+                current_app.config['SECRET_KEY'],
+                algorithms=['HS256'],
+                issuer=current_app.config['SERVER_NAME'],
+                options={'require': ['exp', 'iat', 'iss', 'purpose', 'sub']}
+            )
+        except jwt.PyJWTError:
+            return False
+        if payload.get('purpose') != 'user.confirm':
+            return False
+        if payload.get('sub') != self.hashid:
+            return False
+        self.confirmed = True
+        db.session.add(self)
+        return True
+
+    def delete(self):
+        shutil.rmtree(self.path, ignore_errors=True)
+        db.session.delete(self)
+
+    def generate_auth_token(self):
+        return Token(
+            access_token=secrets.token_urlsafe(),
+            access_expiration=datetime.utcnow() + timedelta(minutes=15),
+            refresh_token=secrets.token_urlsafe(),
+            refresh_expiration=datetime.utcnow() + timedelta(days=7),
+            user=self
+        )
+
+    def generate_confirm_token(self, expiration=3600):
+        now = datetime.utcnow()
+        payload = {
+            'exp': now + timedelta(seconds=expiration),
+            'iat': now,
+            'iss': current_app.config['SERVER_NAME'],
+            'purpose': 'user.confirm',
+            'sub': self.hashid
+        }
+        return jwt.encode(
+            payload,
+            current_app.config['SECRET_KEY'],
+            algorithm='HS256'
+        )
+
+    def generate_reset_password_token(self, expiration=3600):
+        now = datetime.utcnow()
+        payload = {
+            'exp': now + timedelta(seconds=expiration),
+            'iat': now,
+            'iss': current_app.config['SERVER_NAME'],
+            'purpose': 'User.reset_password',
+            'sub': self.hashid
+        }
+        return jwt.encode(
+            payload,
+            current_app.config['SECRET_KEY'],
+            algorithm='HS256'
+        )
+
+    def is_administrator(self):
+        return self.can(Permission.ADMINISTRATE)
+
+    def ping(self):
+        self.last_seen = datetime.utcnow()
+
+    def revoke_auth_tokens(self):
+        for token in self.tokens:
+            db.session.delete(token)
+
+    def verify_password(self, password):
+        if self.role.name == 'System user':
+            return False
+        return check_password_hash(self.password_hash, password)
+
+    #region Profile Privacy settings
+    def has_profile_privacy_setting(self, setting):
+        s = ProfilePrivacySettings.get(setting)
+        return self.profile_privacy_settings & s.value == s.value
+    
+    def add_profile_privacy_setting(self, setting):
+        s = ProfilePrivacySettings.get(setting)
+        if not self.has_profile_privacy_setting(s):
+            self.profile_privacy_settings += s.value
+
+    def remove_profile_privacy_setting(self, setting):
+        s = ProfilePrivacySettings.get(setting)
+        if self.has_profile_privacy_setting(s):
+            self.profile_privacy_settings -= s.value
+
+    def reset_profile_privacy_settings(self):
+        self.profile_privacy_settings = 0
+    #endregion Profile Privacy settings
+
+    def follow_corpus(self, corpus, role=None):
+        if role is None:
+            cfr = CorpusFollowerRole.query.filter_by(default=True).first()
+        else:
+            cfr = role
+        if self.is_following_corpus(corpus):
+            cfa = CorpusFollowerAssociation.query.filter_by(corpus=corpus, follower=self).first()
+            if cfa.role != cfr:
+                cfa.role = cfr
+        else:
+            cfa = CorpusFollowerAssociation(corpus=corpus, role=cfr, follower=self)
+            db.session.add(cfa)
+
+    def unfollow_corpus(self, corpus):
+        if not self.is_following_corpus(corpus):
+            return
+        self.followed_corpora.remove(corpus)
+
+    def is_following_corpus(self, corpus):
+        return corpus in self.followed_corpora
+    
+    def generate_follow_corpus_token(self, corpus_hashid, role_name, expiration=7):
+        now = datetime.utcnow()
+        payload = {
+            'exp': expiration,
+            'iat': now,
+            'iss': current_app.config['SERVER_NAME'],
+            'purpose': 'User.follow_corpus',
+            'role_name': role_name,
+            'sub': corpus_hashid
+        }
+        return jwt.encode(
+            payload,
+            current_app.config['SECRET_KEY'],
+            algorithm='HS256'
+        )
+    
+    def follow_corpus_by_token(self, token):
+        try:
+            payload = jwt.decode(
+                token,
+                current_app.config['SECRET_KEY'],
+                algorithms=['HS256'],
+                issuer=current_app.config['SERVER_NAME'],
+                options={'require': ['exp', 'iat', 'iss', 'purpose', 'role_name', 'sub']}
+            )
+        except jwt.PyJWTError:
+            return False
+        if payload.get('purpose') != 'User.follow_corpus':
+            return False
+        corpus_hashid = payload.get('sub')
+        corpus_id = hashids.decode(corpus_hashid)
+        corpus = Corpus.query.get_or_404(corpus_id)
+        if corpus is None:
+            return False
+        role_name = payload.get('role_name')
+        role = CorpusFollowerRole.query.filter_by(name=role_name).first()
+        if role is None:
+            return False
+        self.follow_corpus(corpus, role)
+        # db.session.add(self)
+        return True
+
+    def to_json_serializeable(self, backrefs=False, relationships=False, filter_by_privacy_settings=False):
+        json_serializeable = {
+            'id': self.hashid,
+            'confirmed': self.confirmed,
+            'avatar': url_for('users.user_avatar', user_id=self.id),
+            'email': self.email,
+            'last_seen': (
+                None if self.last_seen is None
+                else f'{self.last_seen.isoformat()}Z'
+            ),
+            'member_since': f'{self.member_since.isoformat()}Z',
+            'username': self.username,
+            'full_name': self.full_name,
+            'about_me': self.about_me,
+            'website': self.website,
+            'location': self.location,
+            'organization': self.organization,
+            'job_status_mail_notification_level': \
+                    self.setting_job_status_mail_notification_level.name,
+            'profile_privacy_settings': {
+                'is_public': self.is_public,
+                'show_email': self.has_profile_privacy_setting(ProfilePrivacySettings.SHOW_EMAIL),
+                'show_last_seen': self.has_profile_privacy_setting(ProfilePrivacySettings.SHOW_LAST_SEEN),
+                'show_member_since': self.has_profile_privacy_setting(ProfilePrivacySettings.SHOW_MEMBER_SINCE)
+            }
+        }
+        if backrefs:
+            json_serializeable['role'] = \
+                self.role.to_json_serializeable(backrefs=True)
+        if relationships:
+            json_serializeable['corpus_follower_associations'] = {
+                x.hashid: x.to_json_serializeable()
+                for x in self.corpus_follower_associations
+            }
+            json_serializeable['corpora'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.corpora
+            }
+            json_serializeable['jobs'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.jobs
+            }
+            json_serializeable['tesseract_ocr_pipeline_models'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.tesseract_ocr_pipeline_models
+            }
+            json_serializeable['spacy_nlp_pipeline_models'] = {
+                x.hashid: x.to_json_serializeable(relationships=True)
+                for x in self.spacy_nlp_pipeline_models
+            }
+
+        if filter_by_privacy_settings:
+            if not self.has_profile_privacy_setting(ProfilePrivacySettings.SHOW_EMAIL):
+                json_serializeable.pop('email')
+            if not self.has_profile_privacy_setting(ProfilePrivacySettings.SHOW_LAST_SEEN):
+                json_serializeable.pop('last_seen')
+            if not self.has_profile_privacy_setting(ProfilePrivacySettings.SHOW_MEMBER_SINCE):
+                json_serializeable.pop('member_since')
+        return json_serializeable

app/services/__init__.py

@@ -1,12 +1,11 @@
 from flask import Blueprint
 from flask_login import login_required
-import os
+from pathlib import Path
 import yaml
 
 
-services_file = \
-    os.path.join(os.path.dirname(os.path.abspath(__file__)), 'services.yml')
-with open(services_file, 'r') as f:
+services_file = Path(__file__).parent / 'services.yml'
+with services_file.open('r') as f:
     SERVICES = yaml.safe_load(f)
 
 bp = Blueprint('services', __name__)

app/services/routes.py

@@ -6,6 +6,7 @@ from app import db, hashids
 from app.models import (
     Job,
     JobInput,
+    JobResult,
     JobStatus,
     TesseractOCRPipelineModel,
     SpaCyNLPPipelineModel
@@ -74,6 +75,8 @@ def tesseract_ocr_pipeline():
     version = request.args.get('version', service_manifest['latest_version'])
     if version not in service_manifest['versions']:
         abort(404)
+    job_results = JobResult.query.all()
+    choosable_job_ids = [job_result.job.hashid for job_result in job_results if job_result.job.service == "file-setup-pipeline" and job_result.filename.endswith('.pdf')]
     form = CreateTesseractOCRPipelineJobForm(prefix='create-job-form', version=version)
     if form.is_submitted():
         if not form.validate():
@@ -111,6 +114,7 @@ def tesseract_ocr_pipeline():
     return render_template(
         'services/tesseract_ocr_pipeline.html.j2',
         title=service_manifest['name'],
+        choosable_job_ids=choosable_job_ids,
         form=form,
         tesseract_ocr_pipeline_models=tesseract_ocr_pipeline_models,
         user_tesseract_ocr_pipeline_models_count=user_tesseract_ocr_pipeline_models_count

app/services/services.yml

@@ -10,7 +10,7 @@ file-setup-pipeline:
 tesseract-ocr-pipeline:
   name: 'Tesseract OCR Pipeline'
   publisher: 'Bielefeld University - CRC 1288 - INF'
-  latest_version: '0.1.1'
+  latest_version: '0.1.2'
   versions:
     0.1.0:
       methods:
@@ -23,6 +23,12 @@ tesseract-ocr-pipeline:
         - 'ocropus_nlbin_threshold'
       publishing_year: 2022
       url: 'https://gitlab.ub.uni-bielefeld.de/sfb1288inf/tesseract-ocr-pipeline/-/releases/v0.1.1'
+    0.1.2:
+      methods:
+        - 'binarization'
+        - 'ocropus_nlbin_threshold'
+      publishing_year: 2023
+      url: 'https://gitlab.ub.uni-bielefeld.de/sfb1288inf/tesseract-ocr-pipeline/-/releases/v0.1.2'
 transkribus-htr-pipeline:
   name: 'Transkribus HTR Pipeline'
   publisher: 'Bielefeld University - CRC 1288 - INF'
@@ -41,7 +47,7 @@ transkribus-htr-pipeline:
 spacy-nlp-pipeline:
   name: 'SpaCy NLP Pipeline'
   publisher: 'Bielefeld University - CRC 1288 - INF'
-  latest_version: '0.1.2'
+  latest_version: '0.1.1'
   versions:
     0.1.0:
       methods:
@@ -56,5 +62,5 @@ spacy-nlp-pipeline:
     0.1.2:
       methods:
         - 'encoding_detection'
-      publishing_year: 2022
+      publishing_year: 2024
       url: 'https://gitlab.ub.uni-bielefeld.de/sfb1288inf/spacy-nlp-pipeline/-/releases/v0.1.2'

app/static/css/queryBuilder.css

@@ -1,3 +1,13 @@
+#corpus-analysis-concordance-query-builder-input-field {
+  border-bottom: #9E9E9E 1px solid;
+  min-height: 38px;
+  margin-top: 23px;
+}
+
+#corpus-analysis-concordance-query-builder-input-field-placeholder {
+  color: #9E9E9E;
+}
+
 .modal-content {
   overflow-x: hidden;
 }
@@ -75,40 +85,24 @@
   color: black;
 }
 
-#corpus-analysis-concordance-incidence-modifiers-button a{
-  background-color: #2FBBAB;
+[data-toggle-area="input-field-options"] a {
+  margin-right: 10px;
 }
 
-#corpus-analysis-concordance-incidence-modifiers-dropdown a{
-  background-color: white;
+[data-target="corpus-analysis-concordance-character-incidence-modifiers-dropdown"], [data-target="corpus-analysis-concordance-token-incidence-modifiers-dropdown"] {
+  background-color: #2FBBAB !important;
 }
 
-#corpus-analysis-concordance-ignore-case-checkbox {
-  margin-left: 5px;
+#corpus-analysis-concordance-exactly-n-token-modal, #corpus-analysis-concordance-between-nm-token-modal {
+  width: 30%;
 }
 
-#corpus-analysis-concordance-or, #corpus-analysis-concordance-and {
-  background-color: #fc0;
+[data-modal-id="corpus-analysis-concordance-exactly-n-token-modal"], [data-modal-id="corpus-analysis-concordance-between-nm-token-modal"] {
+  margin-top: 15px !important;
 }
 
-#corpus-analysis-concordance-betweenNM {
-  width: 60%;
-}
-
-#corpus-analysis-concordance-query-builder-tutorial-modal {
-  width: 60%;
-}
-
-#corpus-analysis-concordance-query-builder-tutorial-modal ul {
-  margin-top: 10px;
-}
-
-#corpus-analysis-concordance-query-builder-tutorial {
-  padding:15px;
-}
-
-#corpus-analysis-concordance-scroll-up-button-query-builder-tutorial {
-  background-color: #28B3D1;
+[data-options-action="and"], [data-options-action="or"] {
+  background-color: #fc0 !important;
 }
 
 [data-type="start-sentence"], [data-type="end-sentence"] {
@@ -116,13 +110,18 @@
 }
 
 [data-type="start-empty-entity"], [data-type="start-entity"], [data-type="end-entity"] {
-  background-color: #A6E22D;
+  background-color: #a6e22d;
 }
 
-[data-type="start-text-annotation"]{
+[data-type="text-annotation"]{
   background-color: #2FBBAB;
 }
 
 [data-type="token"] {
   background-color: #28B3D1;
 }
+
+[data-type="token-incidence-modifier"] {
+  background-color: #4db6ac;
+  color: white;
+}

app/static/js/App.js

@@ -1,104 +0,0 @@
-class App {
-  constructor() {
-    this.data = {
-      promises: {getUser: {}, subscribeUser: {}},
-      users: {},
-    };
-    this.socket = io({transports: ['websocket'], upgrade: false});
-    this.socket.on('PATCH', (patch) => {this.onPatch(patch);});
-  }
-
-  getUser(userId) {
-    if (userId in this.data.promises.getUser) {
-      return this.data.promises.getUser[userId];
-    }
-
-    this.data.promises.getUser[userId] = new Promise((resolve, reject) => {
-      this.socket.emit('GET /users/<user_id>', userId, (response) => {
-        if (response.status === 200) {
-          this.data.users[userId] = response.body;
-          resolve(this.data.users[userId]);
-        } else {
-          reject(`[${response.status}] ${response.statusText}`);
-        }
-      });
-    });
-
-    return this.data.promises.getUser[userId];
-  }
-
-  subscribeUser(userId) {
-    if (userId in this.data.promises.subscribeUser) {
-      return this.data.promises.subscribeUser[userId];
-    }
-
-    this.data.promises.subscribeUser[userId] = new Promise((resolve, reject) => {
-      this.socket.emit('SUBSCRIBE /users/<user_id>', userId, (response) => {
-        if (response.status !== 200) {
-          reject(response);
-          return;
-        }
-        resolve(response);
-      });
-    });
-
-    return this.data.promises.subscribeUser[userId];
-  }
-
-  flash(message, category) {
-    let iconPrefix = '';
-    switch (category) {
-      case 'corpus': {
-        iconPrefix = '<i class="left material-icons">book</i>';
-        break;
-      }
-      case 'error': {
-        iconPrefix = '<i class="error-color-text left material-icons">error</i>';
-        break;
-      }
-      case 'job': {
-        iconPrefix = '<i class="left nopaque-icons">J</i>';
-        break;
-      }
-      case 'settings': {
-        iconPrefix = '<i class="left material-icons">settings</i>';
-        break;
-      }
-      default: {
-        iconPrefix = '<i class="left material-icons">notifications</i>';
-        break;
-      }
-    }
-    let toast = M.toast(
-      {
-        html: `
-          <span>${iconPrefix}${message}</span>
-          <button class="action-button btn-flat toast-action white-text" data-action="close">
-            <i class="material-icons">close</i>
-          </button>
-        `.trim()
-      }
-    );
-    let toastCloseActionElement = toast.el.querySelector('.action-button[data-action="close"]');
-    toastCloseActionElement.addEventListener('click', () => {toast.dismiss();});
-  }
-  
-  onPatch(patch) {
-    // Filter Patch to only include operations on users that are initialized
-    let regExp = new RegExp(`^/users/(${Object.keys(this.data.users).join('|')})`);
-    let filteredPatch = patch.filter(operation => regExp.test(operation.path));
-
-    // Handle job status updates
-    let subRegExp = new RegExp(`^/users/([A-Za-z0-9]*)/jobs/([A-Za-z0-9]*)/status$`);
-    let subFilteredPatch = filteredPatch
-      .filter((operation) => {return operation.op === 'replace';})
-      .filter((operation) => {return subRegExp.test(operation.path);});
-    for (let operation of subFilteredPatch) {
-      let [match, userId, jobId] = operation.path.match(subRegExp);
-      this.flash(`[<a href="/jobs/${jobId}">${this.data.users[userId].jobs[jobId].title}</a>] New status: <span class="job-status-text" data-status="${operation.value}"></span>`, 'job');
-    }
-
-    // Apply Patch
-    jsonpatch.applyPatch(this.data, filteredPatch);
-  }
-}

app/static/js/CorpusAnalysis/QueryBuilder.js

@@ -1,54 +0,0 @@
-class ConcordanceQueryBuilder {
-
-  constructor() {
-    
-    this.elements = new ElementReferencesQueryBuilder();
-    this.generalFunctions = new GeneralFunctionsQueryBuilder(this.elements);
-    this.tokenAttributeBuilder = new TokenAttributeBuilderFunctionsQueryBuilder(this.elements);
-    this.structuralAttributeBuilder = new StructuralAttributeBuilderFunctionsQueryBuilder(this.elements);
-
-    // Event listener for structural attribute modal
-    document.querySelectorAll('[data-structural-attr-modal-action-button]').forEach(button => {
-      button.addEventListener('click', () => {
-        this.structuralAttributeBuilder.actionButtonHandler(button.dataset.structuralAttrModalActionButton);
-      });
-    });
-
-    // Event listener for token attribute modal
-    this.elements.positionalAttrSelection.addEventListener('change', (event) => {
-      this.generalFunctions.toggleClass(['word', 'lemma', 'english-pos', 'german-pos', 'simple-pos'], 'hide', 'add');
-      this.generalFunctions.toggleClass([event.target.value], 'hide', 'remove');
-      this.generalFunctions.toggleClass(['incidence-modifiers', 'or', 'and'], 'disabled', 'add');
-    });
-    // Options for positional attribute selection
-    document.querySelectorAll('.positional-attr-options-action-button[data-options-action]').forEach(button => {
-      button.addEventListener('click', () => {this.tokenAttributeBuilder.actionButtonHandler(button.dataset.optionsAction);});
-    });
-    document.querySelectorAll('.incidence-modifier-selection[data-incidence-modifier]').forEach(button => {
-      button.addEventListener('click', () => {this.tokenAttributeBuilder.incidenceModifierHandler(button);});
-    });
-    document.querySelectorAll('.n-m-submit-button').forEach(button => {
-      button.addEventListener('click', () => {
-        this.tokenAttributeBuilder.nmSubmitHandler(button.dataset.modalId);
-      });
-    });
-
-    // Initializing and styling the Materialize Chip components
-    M.Chips.init(
-      this.elements.queryInputField, 
-      {
-        placeholder: 'Add your query here'
-      }
-    );
-    document.querySelector('#corpus-analysis-concordance-form-query-builder input').style.setProperty('width', '150px', 'important');
-
-    this.elements.positionalAttrModal = M.Modal.init(
-      document.querySelector('#corpus-analysis-concordance-positional-attr-modal'), 
-      {
-        onOpenStart: () => {
-          this.tokenAttributeBuilder.optionToggleHandler();
-        }
-      }
-    );
-  }
-}

app/static/js/CorpusAnalysis/QueryBuilder/ElementReferencesQueryBuilder.js

@@ -1,31 +0,0 @@
-class ElementReferencesQueryBuilder {
-  constructor() {
-    // General Elements
-    this.counter = 0;
-    this.queryInputField = document.querySelector('#corpus-analysis-concordance-form-query-builder');
-    this.queryInputFieldInstance = M.Chips.getInstance(this.queryInputField);
-    this.queryInputFieldContent = [];
-
-    // Structural Attribute Builder Elements
-    this.structuralAttrModalInstance = document.querySelector('#corpus-analysis-concordance-structural-attr-modal');
-    this.sentenceElement = document.querySelector('[data-structural-attr-modal-action-button="sentence"]');
-    this.entityElement = document.querySelector('[data-structural-attr-modal-action-button="entity"]');
-    this.textAnnotationElement = document.querySelector('[data-structural-attr-modal-action-button="text-annotation"]');
-
-    // Token Attribute Builder Elements
-    this.positionalAttrModal = M.Modal.getInstance(document.querySelector('#corpus-analysis-concordance-positional-attr-modal'));
-    this.positionalAttrSelection = document.querySelector('#corpus-analysis-concordance-positional-attr-selection');
-    this.tokenQuery = document.querySelector('#corpus-analysis-concordance-token-query');
-    this.tokenSubmitButton = document.querySelector('#corpus-analysis-concordance-token-submit');
-    this.noValueMessage = document.querySelector('#corpus-analysis-concordance-no-value-message');
-    this.isTokenQueryInvalid = false;
-
-    this.wordInput = document.querySelector('#corpus-analysis-concordance-word-input');
-    this.lemmaInput = document.querySelector('#corpus-analysis-concordance-lemma-input');
-    this.englishPosSelection = document.querySelector('#corpus-analysis-concordance-english-pos-selection');
-    this.germanPosSelection = document.querySelector('#corpus-analysis-concordance-german-pos-selection');
-    this.simplePosSelection = document.querySelector('#corpus-analysis-concordance-simple-pos-selection');
-
-    this.ignoreCaseCheckbox = document.querySelector('#corpus-analysis-concordance-ignore-case-checkbox');
-  }
-}

app/static/js/CorpusAnalysis/QueryBuilder/GeneralFunctionsQueryBuilder.js

@@ -1,125 +0,0 @@
-class GeneralFunctionsQueryBuilder {
-  constructor(elements) {
-    this.elements = elements;
-  }
-
-  toggleClass(elements, className, action){
-    elements.forEach(element => {
-      document.querySelector('[data-toggle-area="' + element + '"]').classList[action](className);
-    });
-  }
-
-  queryChipFactory(dataType, prettyQueryText, queryText) {
-    this.elements.counter++;
-    this.elements.queryInputFieldInstance.addChip({
-      tag: prettyQueryText
-    });
-
-    let queryChipElementIndex = this.elements.queryInputFieldInstance.$chips.length - 1;
-    let queryChipElement = this.elements.queryInputFieldInstance.$chips[queryChipElementIndex];
-    queryChipElement.classList.add('query-component');
-    queryChipElement.setAttribute('data-type', dataType);
-    queryChipElement.setAttribute('data-query', queryText);
-    queryChipElement.setAttribute('draggable', 'true');
-
-    queryChipElement.addEventListener('click', () => {this.deleteAttr(queryChipElement);});
-    queryChipElement.addEventListener('dragstart', (event) => {
-      // selects all nodes without target class
-      let queryChips = this.elements.queryInputField.querySelectorAll('.query-component');
-
-      // Adds a target chip in front of all draggable childnodes
-      setTimeout(() => {
-        let targetChipElement = Utils.HTMLToElement('<span class="chip drop-target">Drop here</span>');
-        for (let element of queryChips) {
-          if (element === queryChipElement.nextSibling) {continue;}
-          let targetChipClone = targetChipElement.cloneNode(true);
-          if (element === queryChipElement) {
-            // If the dragged element is not at the very end, a target chip is also inserted at the end
-            if (queryChips[queryChips.length - 1] !== element) {
-              queryChips[queryChips.length - 1].insertAdjacentElement('afterend', targetChipClone);
-            }
-          } else {
-            element.insertAdjacentElement('beforebegin', targetChipClone);
-          }
-          targetChipClone.addEventListener('dragover', (event) => {
-            event.preventDefault();
-          });
-          targetChipClone.addEventListener('dragenter', (event) => {
-            event.preventDefault();
-            event.target.style.borderStyle = 'solid dotted';
-          });
-          targetChipClone.addEventListener('dragleave', (event) => {
-            event.preventDefault();
-            event.target.style.borderStyle = 'hidden';
-          });
-          targetChipClone.addEventListener('drop', (event) => {
-            let dropzone = event.target;
-            dropzone.parentElement.replaceChild(queryChipElement, dropzone);
-            this.queryPreviewBuilder();
-          });
-        }
-      }, 0);
-    });
-
-    queryChipElement.addEventListener('dragend', (event) => {
-      let targets = document.querySelectorAll('.drop-target');
-      for (let target of targets) {
-        target.remove();
-      }
-    });
-
-    // Ensures that metadata is always at the end of the query:
-    // const lastChild = this.elements.queryInputFieldInstance.lastChild;
-    // const isLastChildTextAnnotation = lastChild && lastChild.dataset.type === 'text-annotation';
-    // console.log(isLastChildTextAnnotation);
-    // console.log(lastChild);
-
-    // if (!isLastChildTextAnnotation) {
-    //   this.elements.queryInputFieldInstance.appendChild(queryChipElement);
-    // } else {
-    //   this.elements.queryInputFieldInstance.insertBefore(queryChipElement, lastChild);
-    // }
-
-    this.queryPreviewBuilder();
-
-    // Shows a hint about possible functions for editing the query at the first added element in the query
-    // if (this.elements.queryInputFieldInstance.childNodes.length === 1) {
-    //   app.flash('You can edit your query by deleting individual elements or moving them via drag and drop.');
-    // }
-  }
-
-  queryPreviewBuilder() {
-    let queryPreview = document.querySelector('#corpus-analysis-concordance-query-preview');
-    let queryChipElements = Array.from(Object.values(this.elements.queryInputFieldInstance.$chips));
-    queryChipElements.pop();
-    this.elements.queryInputFieldContent = [];
-    queryChipElements.forEach(element => {
-
-      let queryElement = element.dataset.query;
-      if (queryElement !== undefined) {
-        queryElement = Utils.escape(queryElement);
-        this.elements.queryInputFieldContent.push(queryElement);
-      }
-    });
-  
-    let queryString = this.elements.queryInputFieldContent.join(' ');
-    queryString += ';';
-  
-    queryPreview.innerHTML = queryString;
-    queryPreview.parentNode.classList.toggle('hide', queryString === ';');
-  }
-
-  deleteAttr(attr) {
-    if (attr.dataset.type === "start-sentence") {
-      this.elements.sentenceElement.innerHTML = 'Sentence';
-    } else if (attr.dataset.type === "start-entity" || attr.dataset.type === "start-empty-entity") {
-      this.elements.entityElement.innerHTML = 'Entity';
-    }
-    let queryChipElements = Array.from(Object.values(this.elements.queryInputFieldInstance.$chips));
-    queryChipElements.pop();
-    this.elements.counter -= 1;
-    this.elements.queryInputFieldInstance.deleteChip(queryChipElements.indexOf(attr));
-    this.queryPreviewBuilder();
-  }
-}
-

app/static/js/CorpusAnalysis/QueryBuilder/StructuralAttributeBuilderFunctionsQueryBuilder.js

@@ -1,23 +0,0 @@
-class StructuralAttributeBuilderFunctionsQueryBuilder extends GeneralFunctionsQueryBuilder {
-  constructor(elements) {
-    super(elements);
-    this.elements = elements;
-  }
-
-  actionButtonHandler(action) {
-    switch (action) {
-      case 'sentence':
-        break;
-      case 'entity':
-        this.toggleClass(['entity-builder'], 'hide', 'toggle');
-        this.toggleClass(['text-annotation-builder'], 'hide', 'add');
-        break;
-      case 'meta-data':
-        this.toggleClass(['text-annotation-builder'], 'hide', 'toggle');
-        this.toggleClass(['entity-builder'], 'hide', 'add');
-        break;
-      default:
-        break;
-    }
-  }
-}

app/static/js/CorpusAnalysis/QueryBuilder/TokenAttributeBuilderFunctionsQueryBuilder.js

@@ -1,300 +0,0 @@
-class TokenAttributeBuilderFunctionsQueryBuilder extends GeneralFunctionsQueryBuilder {
-  constructor(elements) {
-    super(elements);
-    this.elements = elements;
-
-    this.elements.tokenSubmitButton.addEventListener('click', () => {this.addTokenToQuery();});
-    this.elements.wordInput.addEventListener('input', () => {this.optionToggleHandler();});
-    this.elements.lemmaInput.addEventListener('input', () => {this.optionToggleHandler();});
-  }
-
-  lemmaOrWordInputCheck() {
-    let input;
-
-    if (document.querySelector('[data-toggle-area="word"]').classList.contains('hide') === false) {
-      input = this.elements.wordInput;
-    } else {
-      input = this.elements.lemmaInput;
-    }
-
-    return input;
-  }
-
-  optionToggleHandler() {
-    let input = this.lemmaOrWordInputCheck()
-
-    if (input.value === '') {
-      this.toggleClass(['incidence-modifiers', 'or', 'and'], 'disabled', 'add');
-    } else {
-      this.toggleClass(['incidence-modifiers', 'or', 'and'], 'disabled', 'remove');
-    }
-  }
-
-  disableTokenSubmit() {
-    this.elements.isTokenQueryInvalid = true;
-    this.elements.tokenSubmitButton.classList.add('red');
-    this.elements.noValueMessage.classList.remove('hide');
-    setTimeout(() => {
-      this.elements.tokenSubmitButton.classList.remove('red');
-    }, 500);
-    setTimeout(() => {
-      this.elements.noValueMessage.classList.add('hide');
-    }, 3000);
-  }
-
-  tokenChipFactory(prettyQueryText, tokenText) {
-    tokenText = encodeURI(tokenText);
-    let builderElement;
-    let queryChipElement;
-    builderElement = document.createElement('div');
-    builderElement.innerHTML = `
-    <div class='chip col s2 l2' style='margin-top:20px;' data-tokentext='${tokenText}'>
-      ${prettyQueryText}
-      <i class='material-icons close'>close</i>
-    </div>`;
-    queryChipElement = builderElement.firstElementChild;
-    this.elements.tokenQuery.appendChild(queryChipElement);
-  }
-
-  addTokenToQuery() {
-    let c = this.elements.ignoreCaseCheckbox.checked ? ' %c' : '';
-    let tokenQueryContent = '';
-    let tokenQueryText = '';
-    this.elements.isTokenQueryInvalid = false;
-    let tokenIsEmpty = false;
-
-    this.elements.tokenQuery.childNodes.forEach(element => {
-      tokenQueryContent += ' ' + element.firstChild.data + ' ';
-      tokenQueryText += decodeURI(element.dataset.tokentext);
-      if (element.innerText.indexOf('empty token') !== -1) {
-        tokenIsEmpty = true;
-      }
-    });
-
-    switch (this.elements.positionalAttrSelection.value) {
-      case 'word':
-        if (this.elements.wordInput.value === '') {
-          this.disableTokenSubmit();
-        } else {
-          tokenQueryContent += `word=${this.elements.wordInput.value}${c}`;
-          tokenQueryText += `word="${this.elements.wordInput.value}"${c}`;
-          this.elements.wordInput.value = '';
-        }
-        break;
-      case 'lemma':
-        if (this.elements.lemmaInput.value === '') {
-          this.disableTokenSubmit();
-        } else {
-          tokenQueryContent += `lemma=${this.elements.lemmaInput.value}${c}`;
-          tokenQueryText += `lemma="${this.elements.lemmaInput.value}"${c}`;
-          this.elements.lemmaInput.value = '';
-        }
-        break;
-      case 'english-pos':
-        if (this.elements.englishPos.value === 'default') {
-          this.disableTokenSubmit();
-        } else {
-          tokenQueryContent += `pos=${this.elements.englishPosSelection.value}`;
-          tokenQueryText += `pos="${this.elements.englishPosSelection.value}"`;
-          this.elements.englishPosSelection.value = '';
-        }
-        break;
-      case 'german-pos':
-        if (this.elements.germanPosSelection.value === 'default') {
-          this.disableTokenSubmit();
-        } else {
-          tokenQueryContent += `pos=${this.elements.germanPosSelection.value}`;
-          tokenQueryText += `pos="${this.elements.germanPosSelection.value}"`;
-          this.elements.germanPosSelection.value = '';
-        }
-        break;
-      case 'simple-pos':
-        if (this.elements.simplePosSelection.value === 'default') {
-          this.disableTokenSubmit();
-        } else {
-          tokenQueryContent += `simple_pos=${this.elements.simplePosSelection.value}`;
-          tokenQueryText += `simple_pos="${this.elements.simplePosSelection.value}"`;
-          this.elements.simplePosSelection.value = '';
-        }
-        break;
-      default:
-        break;
-    }
-
-    // isTokenQueryInvalid looks if a valid value is passed. If the input fields/dropdowns are empty (isTokenQueryInvalid === true), no token is added.
-    if (this.elements.isTokenQueryInvalid === false) {
-      // Square brackets are added only if it is not an empty token (where they are already present).
-      if (tokenIsEmpty === false) {
-        tokenQueryText = '[' + tokenQueryText + ']';
-        }
-      this.queryChipFactory('token', tokenQueryContent, tokenQueryText);
-      this.resetPositionalAttrModal();
-      this.elements.positionalAttrModal.close();
-
-    }
-  }
-
-  resetPositionalAttrModal() {
-    let originalSelectionList = 
-      `
-        <option value="word" selected>word</option>
-        <option value="lemma" >lemma</option>
-        <option value="english-pos">english pos</option>
-        <option value="german-pos">german pos</option>
-        <option value="simple-pos">simple_pos</option>
-        <option value="empty-token">empty token</option>
-      `;
-    document.querySelector('#corpus-analysis-concordance-positional-attr-selection').innerHTML = originalSelectionList;
-    this.elements.tokenQuery.innerHTML = '';
-    this.toggleClass(['word', 'lemma', 'english-pos', 'german-pos', 'simple-pos'], 'hide', 'add');
-    this.toggleClass(['word'], 'hide', 'remove');
-    this.toggleClass(['incidence-modifiers', 'or', 'and'], 'disabled', 'add');
-
-    document.querySelector(`#corpus-analysis-concordance-positional-attr-selection option[value="word"]`).selected = true;
-
-    let instance = M.FormSelect.getInstance(document.getElementById('positional-attr-selection'));
-    instance.destroy();
-    M.FormSelect.init(document.getElementById('positional-attr-selection'));
-  }
-
-  actionButtonHandler(elem) {
-    let input = this.lemmaOrWordInputCheck();
-    switch (elem) {
-      case 'option-group':
-        input.value  += '(option1|option2)';
-        let firstIndex = input.value.indexOf('option1');
-        let lastIndex = firstIndex + 'option1'.length;
-        input.focus();
-        input.setSelectionRange(firstIndex, lastIndex);
-        break;
-      case 'wildcard-char':
-        input.value  += '.';
-        break;
-      case 'and':
-        this.conditionHandler('and', " & ");
-        break;
-      case 'or':
-        this.conditionHandler('or', " | ");
-        break;
-      default:
-        break;
-    }
-    this.optionToggleHandler();
-  }
-
-  incidenceModifierHandler(elem) {
-    // For word and lemma, the incidence modifiers are inserted in the input field. For the others, one or two chips are created which contain the respective value of the token and the incidence modifier.
-    switch (this.elements.positionalAttrSelection.value) {
-      case 'empty-token':
-        this.tokenChipFactory(elem.innerText, elem.dataset.token);
-        break;
-      case 'english-pos':
-        this.tokenChipFactory(`pos=${this.elements.englishPos.value}`, `pos="${this.elements.englishPos.value}"`);
-        this.tokenChipFactory(elem.innerText, elem.dataset.token);
-        this.elements.tokenQueryFilled = true;
-        break;
-      case 'german-pos':
-        this.tokenChipFactory(`pos=${this.elements.germanPos.value}`, `pos="${this.elements.germanPos.value}"`);
-        this.tokenChipFactory(elem.innerText, elem.dataset.token);
-        this.elements.tokenQueryFilled = true;
-        break;
-      case 'simple-pos':
-        this.tokenChipFactory(`simple_pos=${this.elements.simplePos.value}`, `simple_pos="${this.elements.simplePos.value}"`);
-        this.tokenChipFactory(elem.innerText, elem.dataset.token);
-        this.elements.tokenQueryFilled = true;
-        break;     
-      default:
-        let input = this.lemmaOrWordInputCheck();
-        input.value += elem.dataset.token;
-        break;
-    }
-  }
-
-  nmSubmitHandler(modalId) {
-    let modal = document.querySelector(`#${modalId}`);
-    let input_n = modal.querySelector('.n-m-input[data-value-type="n"]').value;
-    let input_m = modalId === 'between-nm-modal' ? ',' + modal.querySelector('.n-m-input[data-value-type="m"]').value : '';
-    let input = `${input_n}${input_m}`;
-
-    let instance = M.Modal.getInstance(modal);
-    instance.close();
-    
-    switch (this.elements.positionalAttrSelection.value) {
-      case 'word':
-        this.elements.wordInput.value += '{' + input + '}';
-        break;
-      case 'lemma':
-        this.elements.lemmaInput.value += '{' + input + '}';
-        break;
-      default:
-        break;
-    }
-  }
-
-  conditionHandler(conditionText, conditionQueryContent) {
-    let tokenQueryContent;
-    let tokenQueryText;
-    let c = this.elements.ignoreCaseCheckbox.checked ? ' %c' : '';
-    let selectionDefault = "word";
-    let optionDeleteList = [];
-
-    switch (this.elements.positionalAttrSelection.value) {
-      case 'word':
-        tokenQueryContent = `word=${this.elements.wordInput.value}${c}`;
-        tokenQueryText = `word="${this.elements.wordInput.value}"${c}`;
-        this.elements.wordInput.value = '';
-        if (conditionText === 'and') {
-          selectionDefault = "english-pos";
-          optionDeleteList = ['word', 'lemma', 'empty-token'];
-        }
-        break;
-      case 'lemma':
-        tokenQueryContent = `lemma=${this.elements.lemmaInput.value}${c}`;
-        tokenQueryText = `lemma="${this.elements.lemmaInput.value}"${c}`;
-        this.elements.lemmaInput.value = '';
-        if (conditionText === 'and') {
-          selectionDefault = "english-pos";
-          optionDeleteList = ['word', 'lemma', 'empty-token'];
-        }
-        break;
-      case 'english-pos':
-        tokenQueryContent = `pos=${this.elements.englishPos.value}`;
-        tokenQueryText = `pos="${this.elements.englishPos.value}"`;
-        this.elements.englishPos.value = '';
-        break;
-      case 'german-pos':
-        tokenQueryContent = `pos=${this.elements.germanPos.value}`;
-        tokenQueryText = `pos="${this.elements.germanPos.value}"`;
-        this.elements.germanPos.value = '';
-        break;
-      case 'simple-pos':
-        tokenQueryContent = `simple_pos=${this.elements.simplePos.value}`;
-        tokenQueryText = `simple_pos="${this.elements.simplePos.value}"`;
-        this.elements.simplePos.value = '';
-        break;
-      default:
-        break;
-    }
-
-    this.tokenChipFactory(tokenQueryContent, tokenQueryText);
-    this.tokenChipFactory(conditionText, conditionQueryContent);
-    this.setTokenSelection(selectionDefault, optionDeleteList);
-  }
-
-  setTokenSelection(selection, optionDeleteList) {
-    optionDeleteList.forEach(option => {
-      document.querySelector(`#corpus-analysis-concordance-positional-attr-selection option[value=${option}]`).remove();
-    });
-
-    document.querySelector(`#corpus-analysis-concordance-positional-attr-selection option[value=${selection}]`).selected = true;
-
-    let instance = M.FormSelect.getInstance(document.getElementById('positional-attr-selection'));
-    instance.destroy();
-    M.FormSelect.init(document.getElementById('positional-attr-selection'));
-
-    this.toggleClass(['word', 'lemma', 'english-pos', 'german-pos', 'simple-pos'], 'hide', 'add');
-    this.toggleClass([selection], 'hide', 'remove');
-    this.toggleClass(['incidence-modifiers', 'or', 'and'], 'disabled', 'add');
-  }
-
-}

app/static/js/Forms/CreateContributionForm.js

@@ -1,18 +0,0 @@
-class CreateContributionForm extends Form {
-  static autoInit() {
-    let createContributionFormElements = document.querySelectorAll('.create-contribution-form');
-    for (let createContributionFormElement of createContributionFormElements) {
-      new CreateContributionForm(createContributionFormElement);
-    }
-  }
-
-  constructor(formElement) {
-    super(formElement);
-
-    this.addEventListener('requestLoad', (event) => {
-      if (event.target.status === 201) {
-        window.location.href = event.target.getResponseHeader('Location');
-      }
-    });
-  }
-}

app/static/js/Forms/CreateCorpusFileForm.js

@@ -1,18 +0,0 @@
-class CreateCorpusFileForm extends Form {
-  static autoInit() {
-    let createCorpusFileFormElements = document.querySelectorAll('.create-corpus-file-form');
-    for (let createCorpusFileFormElement of createCorpusFileFormElements) {
-      new CreateCorpusFileForm(createCorpusFileFormElement);
-    }
-  }
-
-  constructor(formElement) {
-    super(formElement);
-
-    this.addEventListener('requestLoad', (event) => {
-      if (event.target.status === 201) {
-        window.location.href = event.target.getResponseHeader('Location');
-      }
-    });
-  }
-}

app/static/js/Requests/admin/admin.js

@@ -1,20 +0,0 @@
-/*****************************************************************************
-* Admin                                                                      *
-* Fetch requests for /admin routes                                           *
-*****************************************************************************/
-Requests.admin = {};
-
-Requests.admin.users = {};
-
-Requests.admin.users.entity = {};
-
-Requests.admin.users.entity.confirmed = {};
-
-Requests.admin.users.entity.confirmed.update = (userId, value) => {
-  let input = `/admin/users/${userId}/confirmed`;
-  let init = {
-    method: 'PUT',
-    body: JSON.stringify(value)
-  };
-  return Requests.JSONfetch(input, init);
-};

app/static/js/Requests/contributions/contributions.js

@@ -1,5 +0,0 @@
-/*****************************************************************************
-* Contributions                                                              *
-* Fetch requests for /contributions routes                                   *
-*****************************************************************************/
-Requests.contributions = {};

app/static/js/Requests/contributions/spacy_nlp_pipeline_models.js

@@ -1,26 +0,0 @@
-/*****************************************************************************
-* SpaCy NLP Pipeline Models                                                  *
-* Fetch requests for /contributions/spacy-nlp-pipeline-models routes         *
-*****************************************************************************/
-Requests.contributions.spacy_nlp_pipeline_models = {};
-
-Requests.contributions.spacy_nlp_pipeline_models.entity = {};
-
-Requests.contributions.spacy_nlp_pipeline_models.entity.delete = (spacyNlpPipelineModelId) => {
-  let input = `/contributions/spacy-nlp-pipeline-models/${spacyNlpPipelineModelId}`;
-  let init = {
-    method: 'DELETE'
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.contributions.spacy_nlp_pipeline_models.entity.isPublic = {};
-
-Requests.contributions.spacy_nlp_pipeline_models.entity.isPublic.update = (spacyNlpPipelineModelId, value) => {
-  let input = `/contributions/spacy-nlp-pipeline-models/${spacyNlpPipelineModelId}/is_public`;
-  let init = {
-    method: 'PUT',
-    body: JSON.stringify(value)
-  };
-  return Requests.JSONfetch(input, init);
-};

app/static/js/Requests/contributions/tesseract_ocr_pipeline_models.js

@@ -1,26 +0,0 @@
-/*****************************************************************************
-* Tesseract OCR Pipeline Models                                              *
-* Fetch requests for /contributions/tesseract-ocr-pipeline-models routes     *
-*****************************************************************************/
-Requests.contributions.tesseract_ocr_pipeline_models = {};
-
-Requests.contributions.tesseract_ocr_pipeline_models.entity = {};
-
-Requests.contributions.tesseract_ocr_pipeline_models.entity.delete = (tesseractOcrPipelineModelId) => {
-  let input = `/contributions/tesseract-ocr-pipeline-models/${tesseractOcrPipelineModelId}`;
-  let init = {
-    method: 'DELETE'
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.contributions.tesseract_ocr_pipeline_models.entity.isPublic = {};
-
-Requests.contributions.tesseract_ocr_pipeline_models.entity.isPublic.update = (tesseractOcrPipelineModelId, value) => {
-  let input = `/contributions/tesseract-ocr-pipeline-models/${tesseractOcrPipelineModelId}/is_public`;
-  let init = {
-    method: 'PUT',
-    body: JSON.stringify(value)
-  };
-  return Requests.JSONfetch(input, init);
-};

app/static/js/Requests/corpora/corpora.js

@@ -1,53 +0,0 @@
-/*****************************************************************************
-* Corpora                                                                    *
-* Fetch requests for /corpora routes                                         *
-*****************************************************************************/
-Requests.corpora = {};
-
-Requests.corpora.entity = {};
-
-Requests.corpora.entity.delete = (corpusId) => {
-  let input = `/corpora/${corpusId}`;
-  let init = {
-    method: 'DELETE'
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.corpora.entity.build = (corpusId) => {
-  let input = `/corpora/${corpusId}/build`;
-  let init = {
-    method: 'POST',
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.corpora.entity.generateShareLink = (corpusId, role, expiration) => {
-  let input = `/corpora/${corpusId}/generate-share-link`;
-  let init = {
-    method: 'POST',
-    body: JSON.stringify({role: role, expiration: expiration})
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.corpora.entity.getStopwords = () => {
-  let input = `/corpora/stopwords`;
-  let init = {
-    method: 'GET'
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.corpora.entity.isPublic = {};
-
-Requests.corpora.entity.isPublic.update = (corpusId, isPublic) => {
-  let input = `/corpora/${corpusId}/is_public`;
-  let init = {
-    method: 'PUT',
-    body: JSON.stringify(isPublic)
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-

app/static/js/Requests/corpora/files.js

@@ -1,15 +0,0 @@
-/*****************************************************************************
-* Corpora                                                                    *
-* Fetch requests for /corpora/<entity>/files routes                          *
-*****************************************************************************/
-Requests.corpora.entity.files = {};
-
-Requests.corpora.entity.files.ent = {};
-
-Requests.corpora.entity.files.ent.delete = (corpusId, corpusFileId) => {
-  let input = `/corpora/${corpusId}/files/${corpusFileId}`;
-  let init = {
-    method: 'DELETE',
-  };
-  return Requests.JSONfetch(input, init);
-};

app/static/js/Requests/corpora/followers.js

@@ -1,35 +0,0 @@
-/*****************************************************************************
-* Corpora                                                                    *
-* Fetch requests for /corpora/<entity>/followers routes                      *
-*****************************************************************************/
-Requests.corpora.entity.followers = {};
-
-Requests.corpora.entity.followers.add = (corpusId, usernames) => {
-  let input = `/corpora/${corpusId}/followers`;
-  let init = {
-    method: 'POST',
-    body: JSON.stringify(usernames)
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.corpora.entity.followers.entity = {};
-
-Requests.corpora.entity.followers.entity.delete = (corpusId, followerId) => {
-  let input = `/corpora/${corpusId}/followers/${followerId}`;
-  let init = {
-    method: 'DELETE',
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.corpora.entity.followers.entity.role = {};
-
-Requests.corpora.entity.followers.entity.role.update = (corpusId, followerId, value) => {
-  let input = `/corpora/${corpusId}/followers/${followerId}/role`;
-  let init = {
-    method: 'PUT',
-    body: JSON.stringify(value)
-  };
-  return Requests.JSONfetch(input, init);
-};

app/static/js/Requests/jobs/jobs.js

@@ -1,31 +0,0 @@
-/*****************************************************************************
-* Jobs                                                              *
-* Fetch requests for /jobs routes                                   *
-*****************************************************************************/
-Requests.jobs = {};
-
-Requests.jobs.entity = {};
-
-Requests.jobs.entity.delete = (jobId) => {
-  let input = `/jobs/${jobId}`;
-  let init = {
-    method: 'DELETE'
-  };
-  return Requests.JSONfetch(input, init);
-} 
-
-Requests.jobs.entity.log = (jobId) => {
-  let input = `/jobs/${jobId}/log`;
-  let init = {
-    method: 'GET'
-  };
-  return Requests.JSONfetch(input, init);
-}
-
-Requests.jobs.entity.restart = (jobId) => {
-  let input = `/jobs/${jobId}/restart`;
-  let init = {
-    method: 'POST'
-  };
-  return Requests.JSONfetch(input, init);
-} 

app/static/js/Requests/users/settings.js

@@ -1,17 +0,0 @@
-/*****************************************************************************
-* Settings                                                                   *
-* Fetch requests for /users/<entity>/settings routes                         *
-*****************************************************************************/
-Requests.users.entity.settings = {};
-
-Requests.users.entity.settings.profilePrivacy = {};
-
-Requests.users.entity.settings.profilePrivacy.update = (userId, profilePrivacySetting, enabled) => {
-  let input = `/users/${userId}/settings/profile-privacy/${profilePrivacySetting}`;
-  let init = {
-    method: 'PUT',
-    body: JSON.stringify(enabled)
-  };
-  return Requests.JSONfetch(input, init);
-};
-

app/static/js/Requests/users/users.js

@@ -1,35 +0,0 @@
-/*****************************************************************************
-* Users                                                                      *
-* Fetch requests for /users routes                                           *
-*****************************************************************************/
-Requests.users = {};
-
-Requests.users.entity = {};
-
-Requests.users.entity.delete = (userId) => {
-  let input = `/users/${userId}`;
-  let init = {
-    method: 'DELETE'
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-Requests.users.entity.acceptTermsOfUse = () => {
-  let input = `/users/accept-terms-of-use`;
-  let init = {
-    method: 'POST'
-  };
-  return Requests.JSONfetch(input, init);
-};
-
-
-Requests.users.entity.avatar = {};
-
-Requests.users.entity.avatar.delete = (userId) => {
-  let input = `/users/${userId}/avatar`;
-  let init = {
-    method: 'DELETE'
-  };
-  return Requests.JSONfetch(input, init);
-}
-

app/static/js/app.js

@@ -0,0 +1,204 @@
+nopaque.App = class App {
+  constructor() {
+    this.data = {
+      promises: {getUser: {}, subscribeUser: {}},
+      users: {},
+    };
+    this.socket = io({transports: ['websocket'], upgrade: false});
+    this.socket.on('PATCH', (patch) => {this.onPatch(patch);});
+  }
+
+  getUser(userId) {
+    if (userId in this.data.promises.getUser) {
+      return this.data.promises.getUser[userId];
+    }
+
+    this.data.promises.getUser[userId] = new Promise((resolve, reject) => {
+      this.socket.emit('GET /users/<user_id>', userId, (response) => {
+        if (response.status === 200) {
+          this.data.users[userId] = response.body;
+          resolve(this.data.users[userId]);
+        } else {
+          reject(`[${response.status}] ${response.statusText}`);
+        }
+      });
+    });
+
+    return this.data.promises.getUser[userId];
+  }
+
+  subscribeUser(userId) {
+    if (userId in this.data.promises.subscribeUser) {
+      return this.data.promises.subscribeUser[userId];
+    }
+
+    this.data.promises.subscribeUser[userId] = new Promise((resolve, reject) => {
+      this.socket.emit('SUBSCRIBE /users/<user_id>', userId, (response) => {
+        if (response.status !== 200) {
+          reject(response);
+          return;
+        }
+        resolve(response);
+      });
+    });
+
+    return this.data.promises.subscribeUser[userId];
+  }
+
+  flash(message, category) {
+    let iconPrefix = '';
+    switch (category) {
+      case 'corpus': {
+        iconPrefix = '<i class="left material-icons">book</i>';
+        break;
+      }
+      case 'error': {
+        iconPrefix = '<i class="error-color-text left material-icons">error</i>';
+        break;
+      }
+      case 'job': {
+        iconPrefix = '<i class="left nopaque-icons">J</i>';
+        break;
+      }
+      case 'settings': {
+        iconPrefix = '<i class="left material-icons">settings</i>';
+        break;
+      }
+      default: {
+        iconPrefix = '<i class="left material-icons">notifications</i>';
+        break;
+      }
+    }
+    let toast = M.toast(
+      {
+        html: `
+          <span>${iconPrefix}${message}</span>
+          <button class="action-button btn-flat toast-action white-text" data-action="close">
+            <i class="material-icons">close</i>
+          </button>
+        `.trim()
+      }
+    );
+    let toastCloseActionElement = toast.el.querySelector('.action-button[data-action="close"]');
+    toastCloseActionElement.addEventListener('click', () => {toast.dismiss();});
+  }
+  
+  onPatch(patch) {
+    // Filter Patch to only include operations on users that are initialized
+    let regExp = new RegExp(`^/users/(${Object.keys(this.data.users).join('|')})`);
+    let filteredPatch = patch.filter(operation => regExp.test(operation.path));
+
+    // Handle job status updates
+    let subRegExp = new RegExp(`^/users/([A-Za-z0-9]*)/jobs/([A-Za-z0-9]*)/status$`);
+    let subFilteredPatch = filteredPatch
+      .filter((operation) => {return operation.op === 'replace';})
+      .filter((operation) => {return subRegExp.test(operation.path);});
+    for (let operation of subFilteredPatch) {
+      let [match, userId, jobId] = operation.path.match(subRegExp);
+      this.flash(`[<a href="/jobs/${jobId}">${this.data.users[userId].jobs[jobId].title}</a>] New status: <span class="job-status-text" data-status="${operation.value}"></span>`, 'job');
+    }
+
+    // Apply Patch
+    jsonpatch.applyPatch(this.data, filteredPatch);
+  }
+
+  init() {
+    this.initUi();
+  }
+
+  initUi() {
+    /* Pre-Initialization fixes */
+    // #region
+
+    // Flask-WTF sets the standard HTML maxlength Attribute on input/textarea
+    // elements to specify their maximum length (in characters). Unfortunatly
+    // Materialize won't recognize the maxlength Attribute, instead it uses
+    // the data-length Attribute. It's conversion time :)
+    for (let elem of document.querySelectorAll('input[maxlength], textarea[maxlength]')) {
+      elem.dataset.length = elem.getAttribute('maxlength');
+      elem.removeAttribute('maxlength');
+    }
+
+    // To work around some limitations with the Form setup of Flask-WTF.
+    // HTML option elements with an empty value are considered as placeholder
+    // elements. The user should not be able to actively select these options.
+    // So they get the disabled attribute.
+    for (let optionElement of document.querySelectorAll('option[value=""]')) {
+      optionElement.disabled = true;
+    }
+
+    // TODO: Check why we are doing this.
+    for (let optgroupElement of document.querySelectorAll('optgroup[label=""]')) {
+      for (let c of optgroupElement.children) {
+        optgroupElement.parentElement.insertAdjacentElement('afterbegin', c);
+      }
+      optgroupElement.remove();
+    }
+    // #endregion
+
+
+    /* Initialize Materialize Components */
+    // #region
+  
+    // Automatically initialize Materialize Components that do not require
+    // additional configuration.
+    M.AutoInit();
+
+    // CharacterCounters
+    // Materialize didn't include the CharacterCounter plugin within the
+    // AutoInit method (maybe they forgot it?). Anyway... We do it here. :)
+    M.CharacterCounter.init(document.querySelectorAll('input[data-length]:not(.no-autoinit), textarea[data-length]:not(.no-autoinit)'));
+
+    // Header navigation "more" Dropdown.
+    M.Dropdown.init(
+      document.querySelector('#nav-more-dropdown-trigger'),
+      {
+        alignment: 'right',
+        constrainWidth: false,
+        coverTrigger: false
+      }
+    );
+
+    // Manual modal
+    M.Modal.init(
+      document.querySelector('#manual-modal'),
+      {
+        onOpenStart: (modalElement, modalTriggerElement) => {
+          if ('manualModalChapter' in modalTriggerElement.dataset) {
+            let manualModalTocElement = document.querySelector('#manual-modal-toc');
+            let manualModalToc = M.Tabs.getInstance(manualModalTocElement);
+            manualModalToc.select(modalTriggerElement.dataset.manualModalChapter);
+            // TODO: Make this work.
+            // if ('manualModalChapterAnchor' in modalTriggerElement.dataset) {
+            //   let manualModalChapterAnchor = document.querySelector(`#${modalTriggerElement.dataset.manualModalChapterAnchor}`);
+            //   let xCoord = manualModalChapterAnchor.getBoundingClientRect().left;
+            //   let yCoord = manualModalChapterAnchor.getBoundingClientRect().top;
+            //   let modalContentElement = modalElement.querySelector('.modal-content');
+            //   modalContentElement.scroll(xCoord, yCoord);
+            // }
+          }
+        }
+      }
+    );
+
+    // Terms of use modal
+    M.Modal.init(
+      document.querySelector('#terms-of-use-modal'),
+      {
+        dismissible: false,
+        onCloseEnd: (modalElement) => {
+          nopaque.requests.users.entity.acceptTermsOfUse();
+        }
+      }
+    );
+    // #endregion
+
+
+    /* Initialize nopaque Components */
+    // #region 
+    nopaque.resource_displays.AutoInit();
+    nopaque.resource_lists.AutoInit();
+    nopaque.forms.AutoInit();
+    // #endregion
+  }
+};

app/static/js/corpus-analysis/app.js

@@ -1,4 +1,4 @@
-class CorpusAnalysisApp {
+nopaque.corpus_analysis.App = class App {
   constructor(corpusId) {
     this.corpusId = corpusId;
 
@@ -25,12 +25,12 @@ class CorpusAnalysisApp {
   async init() {
     this.disableActionElements();
     this.elements.m.initModal.open();
-  
+
     try {
       // Setup CQi over SocketIO connection and gather data from the CQPServer
       const statusTextElement = this.elements.initModal.querySelector('.status-text');
       statusTextElement.innerText = 'Creating CQi over SocketIO client...';
-      const cqiClient = new cqi.CQiClient('/cqi_over_sio');
+      const cqiClient = new nopaque.corpus_analysis.cqi.Client('/cqi_over_sio');
       statusTextElement.innerText += ' Done';
       statusTextElement.innerHTML = 'Waiting for the CQP server...';
       const response = await cqiClient.api.socket.emitWithAck('init', this.corpusId);
@@ -93,7 +93,6 @@ class CorpusAnalysisApp {
           actionElement.disabled = true;
           break;
         case 'SELECT':
-          console.log(actionElement);
           actionElement.parentNode.querySelector('input.select-dropdown').disabled = true;
           break;
         default:

app/static/js/corpus-analysis/concordance-extension.js

@@ -1,4 +1,4 @@
-class CorpusAnalysisConcordance {
+nopaque.corpus_analysis.ConcordanceExtension = class ConcordanceExtension {
   name = 'Concordance';
 
   constructor(app) {
@@ -10,7 +10,8 @@ class CorpusAnalysisConcordance {
       container: document.querySelector(`#corpus-analysis-concordance-container`),
       error: document.querySelector(`#corpus-analysis-concordance-error`),
       userInterfaceForm: document.querySelector(`#corpus-analysis-concordance-user-interface-form`),
-      form: document.querySelector(`#corpus-analysis-concordance-form`),
+      expertModeForm: document.querySelector(`#corpus-analysis-concordance-expert-mode-form`),
+      queryBuilderForm: document.querySelector(`#corpus-analysis-concordance-query-builder-form`),
       progress: document.querySelector(`#corpus-analysis-concordance-progress`),
       subcorpusInfo: document.querySelector(`#corpus-analysis-concordance-subcorpus-info`),
       subcorpusActions: document.querySelector(`#corpus-analysis-concordance-subcorpus-actions`),
@@ -30,10 +31,14 @@ class CorpusAnalysisConcordance {
     this.app.registerExtension(this);
   }
 
-  async submitForm() {
+  async submitForm(queryModeId) {
     this.app.disableActionElements();
-    let query = this.elements.form.query.value.trim();
-    let subcorpusName = this.elements.form['subcorpus-name'].value;
+    let queryBuilderQuery = nopaque.Utils.unescape(document.querySelector('#corpus-analysis-concordance-query-preview').innerHTML.trim());
+    let expertModeQuery = this.elements.expertModeForm.query.value.trim();
+    let query = queryModeId === 'corpus-analysis-concordance-expert-mode-form' ? expertModeQuery : queryBuilderQuery;
+    let form = queryModeId === 'corpus-analysis-concordance-expert-mode-form' ? this.elements.expertModeForm : this.elements.queryBuilderForm;
+
+    let subcorpusName = form['subcorpus-name'].value;
     this.elements.error.innerText = '';
     this.elements.error.classList.add('hide');
     this.elements.progress.classList.remove('hide');
@@ -72,16 +77,17 @@ class CorpusAnalysisConcordance {
     this.data.corpus = this.app.data.corpus;
     this.data.subcorpora = {};
     // Add event listeners
-    this.elements.form.addEventListener('submit', (event) => {
+    this.elements.expertModeForm.addEventListener('submit', (event) => {
       event.preventDefault();
-      this.submitForm();
+      this.submitForm(this.elements.expertModeForm.id);
+    });
+    this.elements.queryBuilderForm.addEventListener('submit', (event) => {
+      event.preventDefault();
+      this.submitForm(this.elements.queryBuilderForm.id);
     });
     this.elements.userInterfaceForm.addEventListener('change', (event) => {
       if (event.target === this.elements.userInterfaceForm['context']) {
-        console.log(this.settings.context);
-        console.log(parseInt(this.elements.userInterfaceForm['context'].value));
         this.settings.context = parseInt(this.elements.userInterfaceForm['context'].value);
-        console.log(this.settings.context);
         this.submitForm();
       }
       if (event.target === this.elements.userInterfaceForm['per-page']) {
@@ -165,11 +171,11 @@ class CorpusAnalysisConcordance {
     this.elements.subcorpusActions.querySelector('.subcorpus-export-trigger').addEventListener('click', (event) => {
       event.preventDefault();
       let subcorpus = this.data.subcorpora[this.settings.selectedSubcorpus];
-      let modalElementId = Utils.generateElementId('export-subcorpus-modal-');
-      let exportFormatSelectElementId = Utils.generateElementId('export-format-select-');
-      let exportSelectedMatchesOnlyCheckboxElementId = Utils.generateElementId('export-selected-matches-only-checkbox-');
-      let exportFileNameInputElementId = Utils.generateElementId('export-file-name-input-');
-      let modalElement = Utils.HTMLToElement(
+      let modalElementId = nopaque.Utils.generateElementId('export-subcorpus-modal-');
+      let exportFormatSelectElementId = nopaque.Utils.generateElementId('export-format-select-');
+      let exportSelectedMatchesOnlyCheckboxElementId = nopaque.Utils.generateElementId('export-selected-matches-only-checkbox-');
+      let exportFileNameInputElementId = nopaque.Utils.generateElementId('export-file-name-input-');
+      let modalElement = nopaque.Utils.HTMLToElement(
         `
           <div class="modal" id="${modalElementId}">
             <div class="modal-content">

app/static/js/corpus-analysis/cqi/api/client.js

@@ -1,4 +1,4 @@
-cqi.api.APIClient = class APIClient {
+nopaque.corpus_analysis.cqi.api.Client = class Client {
   /**
    * @param {string} host
    * @param {number} [timeout=60] timeout
@@ -30,10 +30,10 @@ cqi.api.APIClient = class APIClient {
     } else if (response.code === 500) {
       throw new Error(`[${response.code}] ${response.msg}`);
     } else if (response.code === 502) {
-      if (response.payload.code in cqi.errors.lookup) {
-        throw new cqi.errors.lookup[response.payload.code]();
+      if (response.payload.code in nopaque.corpus_analysis.cqi.errors.lookup) {
+        throw new nopaque.corpus_analysis.cqi.errors.lookup[response.payload.code]();
       } else {
-        throw new cqi.errors.CQiError();
+        throw new nopaque.corpus_analysis.cqi.errors.CQiError();
       }
     }
   }
@@ -41,22 +41,22 @@ cqi.api.APIClient = class APIClient {
   /**
    * @param {string} username
    * @param {string} password
-   * @returns {Promise<cqi.status.StatusConnectOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusConnectOk>}
    */
   async ctrl_connect(username, password) {
     const fn_name = 'ctrl_connect';
     const fn_args = {username: username, password: password};
     let payload = await this.#request(fn_name, fn_args);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusByeOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusByeOk>}
    */
   async ctrl_bye() {
     const fn_name = 'ctrl_bye';
     let payload = await this.#request(fn_name);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
@@ -68,12 +68,12 @@ cqi.api.APIClient = class APIClient {
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusPingOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusPingOk>}
    */
   async ctrl_ping() {
     const fn_name = 'ctrl_ping';
     let payload = await this.#request(fn_name);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
@@ -208,13 +208,13 @@ cqi.api.APIClient = class APIClient {
    * try to unload a corpus and all its attributes from memory
    * 
    * @param {string} corpus
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async corpus_drop_corpus(corpus) {
     const fn_name = 'corpus_drop_corpus';
     const fn_args = {corpus: corpus};
     let payload = await this.#request(fn_name, fn_args);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
@@ -250,13 +250,13 @@ cqi.api.APIClient = class APIClient {
    * unload attribute from memory
    * 
    * @param {string} attribute
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async cl_drop_attribute(attribute) {
     const fn_name = 'cl_drop_attribute';
     const fn_args = {attribute: attribute};
     let payload = await this.#request(fn_name, fn_args);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
@@ -482,13 +482,13 @@ cqi.api.APIClient = class APIClient {
    * @param {string} mother_corpus
    * @param {string} subcorpus_name
    * @param {string} query
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async cqp_query(mother_corpus, subcorpus_name, query) {
     const fn_name = 'cqp_query';
     const fn_args = {mother_corpus: mother_corpus, subcorpus_name: subcorpus_name, query: query};
     let payload = await this.#request(fn_name, fn_args);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
@@ -524,7 +524,7 @@ cqi.api.APIClient = class APIClient {
 
   /**
    * Dump the values of <field> for match ranges <first> .. <last>
-   * in <subcorpus>. <field> is one of the CQI_CONST_FIELD_* constants.
+   * in <subcorpus>. <field> is one of the nopaque.corpus_analysis.cqi.constants.FIELD_* constants.
    * 
    * @param {string} subcorpus
    * @param {number} field
@@ -542,13 +542,13 @@ cqi.api.APIClient = class APIClient {
    * delete a subcorpus from memory
    * 
    * @param {string} subcorpus
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async cqp_drop_subcorpus(subcorpus) {
     const fn_name = 'cqp_drop_subcorpus';
     const fn_args = {subcorpus: subcorpus};
     let payload = await this.#request(fn_name, fn_args);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**
@@ -561,9 +561,9 @@ cqi.api.APIClient = class APIClient {
    *
    * returns <n> (id, frequency) pairs flattened into a list of size 2*<n>
    * field is one of
-   * - CQI_CONST_FIELD_MATCH
-   * - CQI_CONST_FIELD_TARGET
-   * - CQI_CONST_FIELD_KEYWORD
+   * - nopaque.corpus_analysis.cqi.constants.FIELD_MATCH
+   * - nopaque.corpus_analysis.cqi.constants.FIELD_TARGET
+   * - nopaque.corpus_analysis.cqi.constants.FIELD_KEYWORD
    *
    * NB: pairs are sorted by frequency desc.
    * 
@@ -610,13 +610,13 @@ cqi.api.APIClient = class APIClient {
 
   /**
    * @param {string} corpus
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async ext_corpus_update_db(corpus) {
     const fn_name = 'ext_corpus_update_db';
     const fn_args = {corpus: corpus};
     let payload = await this.#request(fn_name, fn_args);
-    return new cqi.status.lookup[payload.code]();
+    return new nopaque.corpus_analysis.cqi.status.lookup[payload.code]();
   }
 
   /**

app/static/js/corpus-analysis/cqi/api/index.js

@@ -0,0 +1 @@
+nopaque.corpus_analysis.cqi.api = {};

app/static/js/corpus-analysis/cqi/client.js

@@ -1,23 +1,23 @@
-cqi.CQiClient = class CQiClient {
+nopaque.corpus_analysis.cqi.Client = class Client {
   /**
    * @param {string} host
    * @param {number} [timeout=60] timeout
    * @param {string} [version=0.1] version
    */
   constructor(host, timeout = 60, version = '0.1') {
-     /** @type {cqi.api.APIClient} */
-    this.api = new cqi.api.APIClient(host, timeout, version);
+     /** @type {nopaque.corpus_analysis.cqi.api.Client} */
+    this.api = new nopaque.corpus_analysis.cqi.api.Client(host, timeout, version);
   }
 
   /**
-   * @returns {cqi.models.corpora.CorpusCollection}
+   * @returns {nopaque.corpus_analysis.cqi.models.corpora.CorpusCollection}
    */
   get corpora() {
-    return new cqi.models.corpora.CorpusCollection(this);
+    return new nopaque.corpus_analysis.cqi.models.corpora.CorpusCollection(this);
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusByeOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusByeOk>}
    */
   async bye() {
     return await this.api.ctrl_bye();
@@ -26,14 +26,14 @@ cqi.CQiClient = class CQiClient {
   /**
    * @param {string} username
    * @param {string} password
-   * @returns {Promise<cqi.status.StatusConnectOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusConnectOk>}
    */
   async connect(username, password) {
     return await this.api.ctrl_connect(username, password);
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusPingOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusPingOk>}
    */
   async ping() {
     return await this.api.ctrl_ping();
@@ -49,7 +49,7 @@ cqi.CQiClient = class CQiClient {
   /**
    * Alias for "bye" method
    * 
-   * @returns {Promise<cqi.status.StatusByeOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusByeOk>}
    */
   async disconnect() {
     return await this.api.ctrl_bye();

app/static/js/corpus-analysis/cqi/constants.js

@@ -0,0 +1,43 @@
+nopaque.corpus_analysis.cqi.constants = {};
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_KEYWORD = 9;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_MATCH = 16;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_MATCHEND = 17;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET = 0;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_0 = 0;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_1 = 1;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_2 = 2;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_3 = 3;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_4 = 4;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_5 = 5;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_6 = 6;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_7 = 7;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_8 = 8;
+
+/** @type {number} */
+nopaque.corpus_analysis.cqi.constants.FIELD_TARGET_9 = 9;

app/static/js/corpus-analysis/cqi/errors.js

@@ -0,0 +1,185 @@
+nopaque.corpus_analysis.cqi.errors = {};
+
+
+/**
+ * A base class from which all other errors inherit.
+ * If you want to catch all errors that the CQi package might throw,
+ * catch this base error.
+ */
+nopaque.corpus_analysis.cqi.errors.CQiError = class CQiError extends Error {
+  constructor(message) {
+    super(message);
+    this.code = undefined;
+    this.description = undefined;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.Error = class Error extends nopaque.corpus_analysis.cqi.errors.CQiError {
+  constructor(message) {
+    super(message);
+    this.code = 2;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.ErrorGeneralError = class ErrorGeneralError extends nopaque.corpus_analysis.cqi.errors.Error {
+  constructor(message) {
+    super(message);
+    this.code = 513;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.ErrorConnectRefused = class ErrorConnectRefused extends nopaque.corpus_analysis.cqi.errors.Error {
+  constructor(message) {
+    super(message);
+    this.code = 514;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.ErrorUserAbort = class ErrorUserAbort extends nopaque.corpus_analysis.cqi.errors.Error {
+  constructor(message) {
+    super(message);
+    this.code = 515;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.ErrorSyntaxError = class ErrorSyntaxError extends nopaque.corpus_analysis.cqi.errors.Error {
+  constructor(message) {
+    super(message);
+    this.code = 516;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLError = class Error extends nopaque.corpus_analysis.cqi.errors.CQiError {
+  constructor(message) {
+    super(message);
+    this.code = 4;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorNoSuchAttribute = class CLErrorNoSuchAttribute extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1025;
+    this.description = "CQi server couldn't open attribute";
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorWrongAttributeType = class CLErrorWrongAttributeType extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1026;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorOutOfRange = class CLErrorOutOfRange extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1027;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorRegex = class CLErrorRegex extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1028;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorCorpusAccess = class CLErrorCorpusAccess extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1029;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorOutOfMemory = class CLErrorOutOfMemory extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1030;
+    this.description = 'CQi server has run out of memory; try discarding some other corpora and/or subcorpora';
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CLErrorInternal = class CLErrorInternal extends nopaque.corpus_analysis.cqi.errors.CLError {
+  constructor(message) {
+    super(message);
+    this.code = 1031;
+    this.description = "The classical 'please contact technical support' error";
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CQPError = class Error extends nopaque.corpus_analysis.cqi.errors.CQiError {
+  constructor(message) {
+    super(message);
+    this.code = 5;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CQPErrorGeneral = class CQPErrorGeneral extends nopaque.corpus_analysis.cqi.errors.CQPError {
+  constructor(message) {
+    super(message);
+    this.code = 1281;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CQPErrorNoSuchCorpus = class CQPErrorNoSuchCorpus extends nopaque.corpus_analysis.cqi.errors.CQPError {
+  constructor(message) {
+    super(message);
+    this.code = 1282;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CQPErrorInvalidField = class CQPErrorInvalidField extends nopaque.corpus_analysis.cqi.errors.CQPError {
+  constructor(message) {
+    super(message);
+    this.code = 1283;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.CQPErrorOutOfRange = class CQPErrorOutOfRange extends nopaque.corpus_analysis.cqi.errors.CQPError {
+  constructor(message) {
+    super(message);
+    this.code = 1284;
+    this.description = 'A number is out of range';
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.errors.lookup = {
+  2: nopaque.corpus_analysis.cqi.errors.Error,
+  513: nopaque.corpus_analysis.cqi.errors.ErrorGeneralError,
+  514: nopaque.corpus_analysis.cqi.errors.ErrorConnectRefused,
+  515: nopaque.corpus_analysis.cqi.errors.ErrorUserAbort,
+  516: nopaque.corpus_analysis.cqi.errors.ErrorSyntaxError,
+  4: nopaque.corpus_analysis.cqi.errors.CLError,
+  1025: nopaque.corpus_analysis.cqi.errors.CLErrorNoSuchAttribute,
+  1026: nopaque.corpus_analysis.cqi.errors.CLErrorWrongAttributeType,
+  1027: nopaque.corpus_analysis.cqi.errors.CLErrorOutOfRange,
+  1028: nopaque.corpus_analysis.cqi.errors.CLErrorRegex,
+  1029: nopaque.corpus_analysis.cqi.errors.CLErrorCorpusAccess,
+  1030: nopaque.corpus_analysis.cqi.errors.CLErrorOutOfMemory,
+  1031: nopaque.corpus_analysis.cqi.errors.CLErrorInternal,
+  5: nopaque.corpus_analysis.cqi.errors.CQPError,
+  1281: nopaque.corpus_analysis.cqi.errors.CQPErrorGeneral,
+  1282: nopaque.corpus_analysis.cqi.errors.CQPErrorNoSuchCorpus,
+  1283: nopaque.corpus_analysis.cqi.errors.CQPErrorInvalidField,
+  1284: nopaque.corpus_analysis.cqi.errors.CQPErrorOutOfRange
+};

app/static/js/corpus-analysis/cqi/index.js

@@ -0,0 +1 @@
+nopaque.corpus_analysis.cqi = {};

app/static/js/corpus-analysis/cqi/models/attributes.js

@@ -1,7 +1,7 @@
-cqi.models.attributes = {};
+nopaque.corpus_analysis.cqi.models.attributes = {};
 
 
-cqi.models.attributes.Attribute = class Attribute extends cqi.models.resource.Model {
+nopaque.corpus_analysis.cqi.models.attributes.Attribute = class Attribute extends nopaque.corpus_analysis.cqi.models.resource.Model {
   /**
    * @returns {string}
    */
@@ -24,7 +24,7 @@ cqi.models.attributes.Attribute = class Attribute extends cqi.models.resource.Mo
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async drop() {
     return await this.client.api.cl_drop_attribute(this.apiName);
@@ -32,17 +32,17 @@ cqi.models.attributes.Attribute = class Attribute extends cqi.models.resource.Mo
 };
 
 
-cqi.models.attributes.AttributeCollection = class AttributeCollection extends cqi.models.resource.Collection {
-   /** @type{typeof cqi.models.attributes.Attribute} */
-  static model = cqi.models.attributes.Attribute;
+nopaque.corpus_analysis.cqi.models.attributes.AttributeCollection = class AttributeCollection extends nopaque.corpus_analysis.cqi.models.resource.Collection {
+   /** @type{typeof nopaque.corpus_analysis.cqi.models.attributes.Attribute} */
+  static model = nopaque.corpus_analysis.cqi.models.attributes.Attribute;
 
   /**
-   * @param {cqi.CQiClient} client
-   * @param {cqi.models.corpora.Corpus} corpus
+   * @param {nopaque.corpus_analysis.cqi.Client} client
+   * @param {nopaque.corpus_analysis.cqi.models.corpora.Corpus} corpus
    */
   constructor(client, corpus) {
     super(client);
-     /** @type {cqi.models.corpora.Corpus} */
+     /** @type {nopaque.corpus_analysis.cqi.models.corpora.Corpus} */
     this.corpus = corpus;
   }
 
@@ -62,7 +62,7 @@ cqi.models.attributes.AttributeCollection = class AttributeCollection extends cq
 
   /**
    * @param {string} attributeName
-   * @returns {Promise<cqi.models.attributes.Attribute>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.attributes.Attribute>}
    */
   async get(attributeName) {
     return this.prepareModel(await this._get(attributeName));
@@ -70,7 +70,7 @@ cqi.models.attributes.AttributeCollection = class AttributeCollection extends cq
 };
 
 
-cqi.models.attributes.AlignmentAttribute = class AlignmentAttribute extends cqi.models.attributes.Attribute {
+nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttribute = class AlignmentAttribute extends nopaque.corpus_analysis.cqi.models.attributes.Attribute {
   /**
    * @param {number} id 
    * @returns {Promise<[number, number, number, number]>}
@@ -89,17 +89,17 @@ cqi.models.attributes.AlignmentAttribute = class AlignmentAttribute extends cqi.
 };
 
 
-cqi.models.attributes.AlignmentAttributeCollection = class AlignmentAttributeCollection extends cqi.models.attributes.AttributeCollection {
-   /** @type{typeof cqi.models.attributes.AlignmentAttribute} */
-  static model = cqi.models.attributes.AlignmentAttribute;
+nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttributeCollection = class AlignmentAttributeCollection extends nopaque.corpus_analysis.cqi.models.attributes.AttributeCollection {
+   /** @type{typeof nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttribute} */
+  static model = nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttribute;
 
   /**
-   * @returns {Promise<cqi.models.attributes.AlignmentAttribute[]>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttribute[]>}
    */
   async list() {
      /** @type {string[]} */
      let alignmentAttributeNames = await this.client.api.corpus_alignment_attributes(this.corpus.apiName);
-     /** @type {cqi.models.attributes.AlignmentAttribute[]} */
+     /** @type {nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttribute[]} */
     let alignmentAttributes = [];
     for (let alignmentAttributeName of alignmentAttributeNames) {
       alignmentAttributes.push(await this.get(alignmentAttributeName));
@@ -109,7 +109,7 @@ cqi.models.attributes.AlignmentAttributeCollection = class AlignmentAttributeCol
 };
 
 
-cqi.models.attributes.PositionalAttribute = class PositionalAttribute extends cqi.models.attributes.Attribute {
+nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute = class PositionalAttribute extends nopaque.corpus_analysis.cqi.models.attributes.Attribute {
   /**
    * @returns {number}
    */
@@ -183,9 +183,9 @@ cqi.models.attributes.PositionalAttribute = class PositionalAttribute extends cq
 };
 
 
-cqi.models.attributes.PositionalAttributeCollection = class PositionalAttributeCollection extends cqi.models.attributes.AttributeCollection {
-   /** @type{typeof cqi.models.attributes.PositionalAttribute} */
-  static model = cqi.models.attributes.PositionalAttribute;
+nopaque.corpus_analysis.cqi.models.attributes.PositionalAttributeCollection = class PositionalAttributeCollection extends nopaque.corpus_analysis.cqi.models.attributes.AttributeCollection {
+   /** @type{typeof nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute} */
+  static model = nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute;
 
   /**
    * @param {string} positionalAttributeName
@@ -198,7 +198,7 @@ cqi.models.attributes.PositionalAttributeCollection = class PositionalAttributeC
   }
 
   /**
-   * @returns {Promise<cqi.models.attributes.PositionalAttribute[]>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute[]>}
    */
   async list() {
     let positionalAttributeNames = await this.client.api.corpus_positional_attributes(this.corpus.apiName);
@@ -211,7 +211,7 @@ cqi.models.attributes.PositionalAttributeCollection = class PositionalAttributeC
 };
 
 
-cqi.models.attributes.StructuralAttribute = class StructuralAttribute extends cqi.models.attributes.Attribute {
+nopaque.corpus_analysis.cqi.models.attributes.StructuralAttribute = class StructuralAttribute extends nopaque.corpus_analysis.cqi.models.attributes.Attribute {
   /**
    * @returns {boolean}
    */
@@ -261,9 +261,9 @@ cqi.models.attributes.StructuralAttribute = class StructuralAttribute extends cq
 };
 
 
-cqi.models.attributes.StructuralAttributeCollection = class StructuralAttributeCollection extends cqi.models.attributes.AttributeCollection {
-   /** @type{typeof cqi.models.attributes.StructuralAttribute} */
-  static model = cqi.models.attributes.StructuralAttribute;
+nopaque.corpus_analysis.cqi.models.attributes.StructuralAttributeCollection = class StructuralAttributeCollection extends nopaque.corpus_analysis.cqi.models.attributes.AttributeCollection {
+   /** @type{typeof nopaque.corpus_analysis.cqi.models.attributes.StructuralAttribute} */
+  static model = nopaque.corpus_analysis.cqi.models.attributes.StructuralAttribute;
 
   /**
    * @param {string} structuralAttributeName
@@ -276,7 +276,7 @@ cqi.models.attributes.StructuralAttributeCollection = class StructuralAttributeC
   }
 
   /**
-   * @returns {Promise<cqi.models.attributes.StructuralAttribute[]>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.attributes.StructuralAttribute[]>}
    */
   async list() {
     let structuralAttributeNames = await this.client.api.corpus_structural_attributes(this.corpus.apiName);

app/static/js/corpus-analysis/cqi/models/corpora.js

@@ -1,7 +1,7 @@
-cqi.models.corpora = {};
+nopaque.corpus_analysis.cqi.models.corpora = {};
 
 
-cqi.models.corpora.Corpus = class Corpus extends cqi.models.resource.Model {
+nopaque.corpus_analysis.cqi.models.corpora.Corpus = class Corpus extends nopaque.corpus_analysis.cqi.models.resource.Model {
   /**
    * @returns {string}
    */
@@ -38,35 +38,35 @@ cqi.models.corpora.Corpus = class Corpus extends cqi.models.resource.Model {
   }
 
   /**
-   * @returns {cqi.models.attributes.AlignmentAttributeCollection}
+   * @returns {nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttributeCollection}
    */
   get alignmentAttributes() {
-    return new cqi.models.attributes.AlignmentAttributeCollection(this.client, this);
+    return new nopaque.corpus_analysis.cqi.models.attributes.AlignmentAttributeCollection(this.client, this);
   }
 
   /**
-   * @returns {cqi.models.attributes.PositionalAttributeCollection}
+   * @returns {nopaque.corpus_analysis.cqi.models.attributes.PositionalAttributeCollection}
    */
   get positionalAttributes() {
-    return new cqi.models.attributes.PositionalAttributeCollection(this.client, this);
+    return new nopaque.corpus_analysis.cqi.models.attributes.PositionalAttributeCollection(this.client, this);
   }
 
   /**
-   * @returns {cqi.models.attributes.StructuralAttributeCollection}
+   * @returns {nopaque.corpus_analysis.cqi.models.attributes.StructuralAttributeCollection}
    */
   get structuralAttributes() {
-    return new cqi.models.attributes.StructuralAttributeCollection(this.client, this);
+    return new nopaque.corpus_analysis.cqi.models.attributes.StructuralAttributeCollection(this.client, this);
   }
 
   /**
-   * @returns {cqi.models.subcorpora.SubcorpusCollection}
+   * @returns {nopaque.corpus_analysis.cqi.models.subcorpora.SubcorpusCollection}
    */
   get subcorpora() {
-    return new cqi.models.subcorpora.SubcorpusCollection(this.client, this);
+    return new nopaque.corpus_analysis.cqi.models.subcorpora.SubcorpusCollection(this.client, this);
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async drop() {
     return await this.client.api.corpus_drop_corpus(this.apiName);
@@ -75,7 +75,7 @@ cqi.models.corpora.Corpus = class Corpus extends cqi.models.resource.Model {
   /**
    * @param {string} subcorpusName
    * @param {string} query
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async query(subcorpusName, query) {
     return await this.client.api.cqp_query(this.apiName, subcorpusName, query);
@@ -96,7 +96,7 @@ cqi.models.corpora.Corpus = class Corpus extends cqi.models.resource.Model {
   }
 
   /**
-   * @returns {cqi.status.StatusOk}
+   * @returns {nopaque.corpus_analysis.cqi.status.StatusOk}
    */
   async updateDb() {
     return await this.client.api.ext_corpus_update_db(this.apiName);
@@ -113,9 +113,9 @@ cqi.models.corpora.Corpus = class Corpus extends cqi.models.resource.Model {
 };
 
 
-cqi.models.corpora.CorpusCollection = class CorpusCollection extends cqi.models.resource.Collection {
-   /** @type {typeof cqi.models.corpora.Corpus} */
-  static model = cqi.models.corpora.Corpus;
+nopaque.corpus_analysis.cqi.models.corpora.CorpusCollection = class CorpusCollection extends nopaque.corpus_analysis.cqi.models.resource.Collection {
+   /** @type {typeof nopaque.corpus_analysis.cqi.models.corpora.Corpus} */
+  static model = nopaque.corpus_analysis.cqi.models.corpora.Corpus;
 
   /**
    * @param {string} corpusName
@@ -144,19 +144,19 @@ cqi.models.corpora.CorpusCollection = class CorpusCollection extends cqi.models.
 
   /**
    * @param {string} corpusName
-   * @returns {Promise<cqi.models.corpora.Corpus>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.corpora.Corpus>}
    */
   async get(corpusName) {
     return this.prepareModel(await this._get(corpusName));
   }
 
   /**
-   * @returns {Promise<cqi.models.corpora.Corpus[]>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.corpora.Corpus[]>}
    */
   async list() {
      /** @type {string[]} */
     let corpusNames = await this.client.api.corpus_list_corpora();
-     /** @type {cqi.models.corpora.Corpus[]} */
+     /** @type {nopaque.corpus_analysis.cqi.models.corpora.Corpus[]} */
     let corpora = [];
     for (let corpusName of corpusNames) {
       corpora.push(await this.get(corpusName));

app/static/js/corpus-analysis/cqi/models/index.js

@@ -0,0 +1 @@
+nopaque.corpus_analysis.cqi.models = {};

app/static/js/corpus-analysis/cqi/models/resource.js

@@ -1,26 +1,26 @@
-cqi.models.resource = {};
+nopaque.corpus_analysis.cqi.models.resource = {};
 
 
 /**
  * A base class for representing a single object on the server.
  */
-cqi.models.resource.Model = class Model {
+nopaque.corpus_analysis.cqi.models.resource.Model = class Model {
   /**
    * @param {object} attrs
-   * @param {cqi.CQiClient} client
-   * @param {cqi.models.resource.Collection} collection
+   * @param {nopaque.corpus_analysis.cqi.CQiClient} client
+   * @param {nopaque.corpus_analysis.cqi.models.resource.Collection} collection
    */
   constructor(attrs, client, collection) {
      /**
       * A client pointing at the server that this object is on.
       *
-      * @type {cqi.CQiClient}
+      * @type {nopaque.corpus_analysis.cqi.CQiClient}
       */
     this.client = client;
      /**
       * The collection that this model is part of.
       *
-      * @type {cqi.models.resource.Collection}
+      * @type {nopaque.corpus_analysis.cqi.models.resource.Collection}
       */
     this.collection = collection;
      /**
@@ -50,22 +50,22 @@ cqi.models.resource.Model = class Model {
 /**
  * A base class for representing all objects of a particular type on the server.
  */
-cqi.models.resource.Collection = class Collection {
+nopaque.corpus_analysis.cqi.models.resource.Collection = class Collection {
    /** 
     * The type of object this collection represents, set by subclasses
     * 
-    * @type {typeof cqi.models.resource.Model}
+    * @type {typeof nopaque.corpus_analysis.cqi.models.resource.Model}
     */
   static model;
 
   /**
-   * @param {cqi.CQiClient} client
+   * @param {nopaque.corpus_analysis.cqi.CQiClient} client
    */
   constructor(client) {
      /**
       * A client pointing at the server that this object is on.
       *
-      * @type {cqi.CQiClient}
+      * @type {nopaque.corpus_analysis.cqi.CQiClient}
       */
      this.client = client;
   }
@@ -82,7 +82,7 @@ cqi.models.resource.Collection = class Collection {
    * Create a model from a set of attributes.
    * 
    * @param {object} attrs
-   * @returns {cqi.models.resource.Model}
+   * @returns {nopaque.corpus_analysis.cqi.models.resource.Model}
    */
   prepareModel(attrs) {
     return new this.constructor.model(attrs, this.client, this);

app/static/js/corpus-analysis/cqi/models/subcorpora.js

@@ -1,7 +1,7 @@
-cqi.models.subcorpora = {};
+nopaque.corpus_analysis.cqi.models.subcorpora = {};
 
 
-cqi.models.subcorpora.Subcorpus = class Subcorpus extends cqi.models.resource.Model {
+nopaque.corpus_analysis.cqi.models.subcorpora.Subcorpus = class Subcorpus extends nopaque.corpus_analysis.cqi.models.resource.Model {
   /**
    * @returns {string}
    */
@@ -31,7 +31,7 @@ cqi.models.subcorpora.Subcorpus = class Subcorpus extends cqi.models.resource.Mo
   }
 
   /**
-   * @returns {Promise<cqi.status.StatusOk>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.status.StatusOk>}
    */
   async drop() {
     return await this.client.api.cqp_drop_subcorpus(this.apiName);
@@ -55,7 +55,7 @@ cqi.models.subcorpora.Subcorpus = class Subcorpus extends cqi.models.resource.Mo
   /**
    * @param {number} cutoff
    * @param {number} field
-   * @param {cqi.models.attributes.PositionalAttribute} attribute
+   * @param {nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute} attribute
    * @returns {Promise<number[]>}
    */
   async fdist1(cutoff, field, attribute) {
@@ -70,9 +70,9 @@ cqi.models.subcorpora.Subcorpus = class Subcorpus extends cqi.models.resource.Mo
   /**
    * @param {number} cutoff
    * @param {number} field1
-   * @param {cqi.models.attributes.PositionalAttribute} attribute1
+   * @param {nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute} attribute1
    * @param {number} field2
-   * @param {cqi.models.attributes.PositionalAttribute} attribute2
+   * @param {nopaque.corpus_analysis.cqi.models.attributes.PositionalAttribute} attribute2
    * @returns {Promise<number[]>}
    */ 
   async fdist2(cutoff, field1, attribute1, field2, attribute2) {
@@ -122,17 +122,17 @@ cqi.models.subcorpora.Subcorpus = class Subcorpus extends cqi.models.resource.Mo
 };
 
 
-cqi.models.subcorpora.SubcorpusCollection = class SubcorpusCollection extends cqi.models.resource.Collection {
-   /** @type {typeof cqi.models.subcorpora.Subcorpus} */
-  static model = cqi.models.subcorpora.Subcorpus;
+nopaque.corpus_analysis.cqi.models.subcorpora.SubcorpusCollection = class SubcorpusCollection extends nopaque.corpus_analysis.cqi.models.resource.Collection {
+   /** @type {typeof nopaque.corpus_analysis.cqi.models.subcorpora.Subcorpus} */
+  static model = nopaque.corpus_analysis.cqi.models.subcorpora.Subcorpus;
 
   /**
-   * @param {cqi.CQiClient} client
-   * @param {cqi.models.corpora.Corpus} corpus
+   * @param {nopaque.corpus_analysis.cqi.CQiClient} client
+   * @param {nopaque.corpus_analysis.cqi.models.corpora.Corpus} corpus
    */
   constructor(client, corpus) {
     super(client);
-     /** @type {cqi.models.corpora.Corpus} */
+     /** @type {nopaque.corpus_analysis.cqi.models.corpora.Corpus} */
     this.corpus = corpus;
   }
 
@@ -145,17 +145,17 @@ cqi.models.subcorpora.SubcorpusCollection = class SubcorpusCollection extends cq
     let apiName = `${this.corpus.apiName}:${subcorpusName}`;
      /** @type {object} */
     let fields = {};
-    if (await this.client.api.cqp_subcorpus_has_field(apiName, cqi.CONST_FIELD_MATCH)) {
-      fields.match = cqi.CONST_FIELD_MATCH;
+    if (await this.client.api.cqp_subcorpus_has_field(apiName, nopaque.corpus_analysis.cqi.constants.FIELD_MATCH)) {
+      fields.match = nopaque.corpus_analysis.cqi.constants.FIELD_MATCH;
     }
-    if (await this.client.api.cqp_subcorpus_has_field(apiName, cqi.CONST_FIELD_MATCHEND)) {
-      fields.matchend = cqi.CONST_FIELD_MATCHEND
+    if (await this.client.api.cqp_subcorpus_has_field(apiName, nopaque.corpus_analysis.cqi.constants.FIELD_MATCHEND)) {
+      fields.matchend = nopaque.corpus_analysis.cqi.constants.FIELD_MATCHEND
     }
-    if (await this.client.api.cqp_subcorpus_has_field(apiName, cqi.CONST_FIELD_TARGET)) {
-      fields.target = cqi.CONST_FIELD_TARGET
+    if (await this.client.api.cqp_subcorpus_has_field(apiName, nopaque.corpus_analysis.cqi.constants.FIELD_TARGET)) {
+      fields.target = nopaque.corpus_analysis.cqi.constants.FIELD_TARGET
     }
-    if (await this.client.api.cqp_subcorpus_has_field(apiName, cqi.CONST_FIELD_KEYWORD)) {
-      fields.keyword = cqi.CONST_FIELD_KEYWORD
+    if (await this.client.api.cqp_subcorpus_has_field(apiName, nopaque.corpus_analysis.cqi.constants.FIELD_KEYWORD)) {
+      fields.keyword = nopaque.corpus_analysis.cqi.constants.FIELD_KEYWORD
     }
     return {
       api_name: apiName,
@@ -167,19 +167,19 @@ cqi.models.subcorpora.SubcorpusCollection = class SubcorpusCollection extends cq
 
   /**
    * @param {string} subcorpusName
-   * @returns {Promise<cqi.models.subcorpora.Subcorpus>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.subcorpora.Subcorpus>}
    */
   async get(subcorpusName) {
     return this.prepareModel(await this._get(subcorpusName));
   }
 
   /**
-   * @returns {Promise<cqi.models.subcorpora.Subcorpus[]>}
+   * @returns {Promise<nopaque.corpus_analysis.cqi.models.subcorpora.Subcorpus[]>}
    */
   async list() {
      /** @type {string[]} */
     let subcorpusNames = await this.client.api.cqp_list_subcorpora(this.corpus.apiName);
-     /** @type {cqi.models.subcorpora.Subcorpus[]} */
+     /** @type {nopaque.corpus_analysis.cqi.models.subcorpora.Subcorpus[]} */
     let subcorpora = [];
     for (let subcorpusName of subcorpusNames) {
       subcorpora.push(await this.get(subcorpusName));

app/static/js/corpus-analysis/cqi/status.js

@@ -0,0 +1,51 @@
+nopaque.corpus_analysis.cqi.status = {};
+
+
+/**
+ * A base class from which all other status inherit.
+ */
+nopaque.corpus_analysis.cqi.status.CQiStatus = class CQiStatus {
+  constructor() {
+    this.code = undefined;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.status.StatusOk = class StatusOk extends nopaque.corpus_analysis.cqi.status.CQiStatus {
+  constructor() {
+    super();
+    this.code = 257;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.status.StatusConnectOk = class StatusConnectOk extends nopaque.corpus_analysis.cqi.status.CQiStatus {
+  constructor() {
+    super();
+    this.code = 258;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.status.StatusByeOk = class StatusByeOk extends nopaque.corpus_analysis.cqi.status.CQiStatus {
+  constructor() {
+    super();
+    this.code = 259;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.status.StatusPingOk = class StatusPingOk extends nopaque.corpus_analysis.cqi.status.CQiStatus {
+  constructor() {
+    super();
+    this.code = 260;
+  }
+};
+
+
+nopaque.corpus_analysis.cqi.status.lookup = {
+  257: nopaque.corpus_analysis.cqi.status.StatusOk,
+  258: nopaque.corpus_analysis.cqi.status.StatusConnectOk,
+  259: nopaque.corpus_analysis.cqi.status.StatusByeOk,
+  260: nopaque.corpus_analysis.cqi.status.StatusPingOk
+};

app/static/js/corpus-analysis/index.js

@@ -0,0 +1 @@
+nopaque.corpus_analysis = {};

app/static/js/corpus-analysis/query-builder/element-references.js

@@ -0,0 +1,28 @@
+nopaque.corpus_analysis.query_builder.ElementReferences = class ElementReferences {
+  constructor() {
+    // General Elements
+    this.queryInputField = document.querySelector('#corpus-analysis-concordance-query-builder-input-field');
+    this.queryChipElements = [];
+    this.queryElementTarget = document.querySelector('.query-element-target')
+    this.editingModusOn = false;
+    this.editedQueryChipElementIndex = undefined;
+    this.deleteQueryButton = document.querySelector('#corpus-analysis-concordance-delete-query-button');
+
+    // Structural Attribute Builder Elements
+    this.structuralAttrModal = M.Modal.getInstance(document.querySelector('#corpus-analysis-concordance-structural-attr-modal'));
+    this.englishEntTypeSelection = document.querySelector('#corpus-analysis-concordance-english-ent-type-selection');
+    this.germanEntTypeSelection = document.querySelector('#corpus-analysis-concordance-german-ent-type-selection');
+    
+    // Token Attribute Builder Elements
+    this.positionalAttrModal = M.Modal.getInstance(document.querySelector('#corpus-analysis-concordance-positional-attr-modal'));
+    this.positionalAttrSelection = document.querySelector('#corpus-analysis-concordance-positional-attr-selection');
+    this.tokenBuilderContent = document.querySelector('#corpus-analysis-concordance-token-builder-content');
+    this.tokenQuery = document.querySelector('#corpus-analysis-concordance-token-query');
+    this.tokenQueryTemplate = document.querySelector('#corpus-analysis-concordance-token-query-template');
+    this.tokenSubmitButton = document.querySelector('#corpus-analysis-concordance-token-submit');
+    this.noValueMessage = document.querySelector('#corpus-analysis-concordance-no-value-message');
+    this.isTokenQueryInvalid = false;
+
+    this.ignoreCaseCheckbox = document.querySelector('#corpus-analysis-concordance-ignore-case-checkbox');
+  }
+};

app/static/js/corpus-analysis/query-builder/index.js

@@ -0,0 +1 @@
+nopaque.corpus_analysis.query_builder = {};

app/static/js/corpus-analysis/query-builder/query-builder.js

@@ -0,0 +1,500 @@
+nopaque.corpus_analysis.query_builder.QueryBuilder = class QueryBuilder {
+  constructor() {
+    this.elements = new nopaque.corpus_analysis.query_builder.ElementReferences();
+
+    this.addEventListenersToQueryElementTarget();
+    this.addEventListenersToIncidenceModifier();
+    this.addEventListenersToNAndMInputSubmit();
+
+    this.elements.deleteQueryButton.addEventListener('click', () => {this.resetQueryInputField()});
+    this.expertModeQueryBuilderSwitchHandler();
+
+    this.extensions = {
+      structuralAttributeBuilderFunctions: new nopaque.corpus_analysis.query_builder.StructuralAttributeBuilderFunctions(this),
+      tokenAttributeBuilderFunctions: new nopaque.corpus_analysis.query_builder.TokenAttributeBuilderFunctions(this),
+    };
+
+    this.dropdown = M.Dropdown.init(
+      document.querySelector('.dropdown-trigger[data-toggle-area="token-incidence-modifiers"]'), 
+      {
+        onCloseStart: () => {
+          this.unselectChipElement(this.elements.queryInputField.querySelector('.chip.teal'));
+        }
+      }
+    )
+  }
+
+  addEventListenersToQueryElementTarget() {
+    this.elements.queryElementTarget.addEventListener('click', () => {
+      this.elements.positionalAttrModal.open();
+    });
+    this.elements.queryElementTarget.addEventListener('dragstart', this.handleDragStart.bind(this, this.elements.queryElementTarget));
+    this.elements.queryElementTarget.addEventListener('dragend', this.handleDragEnd);
+  }
+
+  addEventListenersToIncidenceModifier() {
+    // Eventlisteners for the incidence modifiers. There are two different types of incidence modifiers: token and character incidence modifiers.
+    document.querySelectorAll('.incidence-modifier-selection').forEach(button => {
+      let dropdownId = button.parentNode.parentNode.id;
+      if (dropdownId === 'corpus-analysis-concordance-token-incidence-modifiers-dropdown') {
+        button.addEventListener('click', () => this.tokenIncidenceModifierHandler(button.dataset.token, button.innerHTML));
+      } else if (dropdownId === 'corpus-analysis-concordance-character-incidence-modifiers-dropdown') {
+        button.addEventListener('click', () => this.extensions.tokenAttributeBuilderFunctions.characterIncidenceModifierHandler(button));
+      }
+    });
+  }
+
+  addEventListenersToNAndMInputSubmit() {
+    // Eventlisteners for the submit of n- and m-values of the incidence modifier modal for "exactly n" or "between n and m".
+    document.querySelectorAll('.n-m-submit-button').forEach(button => {
+      let modalId = button.dataset.modalId;
+      if (modalId === 'corpus-analysis-concordance-exactly-n-token-modal' || modalId === 'corpus-analysis-concordance-between-nm-token-modal') {
+        button.addEventListener('click', () => this.tokenNMSubmitHandler(modalId));
+      } else if (modalId === 'corpus-analysis-concordance-exactly-n-character-modal' || modalId === 'corpus-analysis-concordance-between-nm-character-modal') {
+        button.addEventListener('click', () => this.extensions.tokenAttributeBuilderFunctions.characterNMSubmitHandler(modalId));
+      }
+    });
+  }
+
+  toggleClass(elements, className, action) {
+    elements.forEach(element => {
+      document.querySelector(`[data-toggle-area="${element}"]`).classList[action](className);
+    });
+  }
+
+  resetQueryInputField() {
+    this.elements.queryInputField.innerHTML = '';
+    this.addQueryElementTarget();
+    this.updateChipList();
+    this.queryPreviewBuilder();
+  }
+
+  addQueryElementTarget() {
+    let queryElementTarget = nopaque.Utils.HTMLToElement(
+      `
+      <a class="query-element-target btn-floating btn-small blue-grey lighten-4 waves-effect waves-light tooltipped" style="margin-bottom:10px; margin-right:5px;" draggable="true" data-position="bottom" data-tooltip="Add an Element to your query">
+        <i class="material-icons">add</i>
+      </a>
+      `
+    );
+    this.elements.queryInputField.appendChild(queryElementTarget);
+    this.elements.queryElementTarget = queryElementTarget;
+    this.addEventListenersToQueryElementTarget();
+  }
+
+  updateChipList() {
+    this.elements.queryChipElements = this.elements.queryInputField.querySelectorAll('.query-component');
+  }
+  
+  resetMaterializeSelection(selectionElements, value = "default") {
+    selectionElements.forEach(selectionElement => {
+      if (selectionElement.querySelector(`option[value=${value}]`) !== null) {
+        selectionElement.querySelector(`option[value=${value}]`).selected = true;
+      }
+      let instance = M.FormSelect.getInstance(selectionElement);
+      instance.destroy();
+      M.FormSelect.init(selectionElement);
+    })
+  }
+
+  submitQueryChipElement(dataType=undefined, prettyQueryText=undefined, queryText=undefined, index=null, isClosingTag=false, isEditable=false) {
+    if (this.elements.editingModusOn) {
+      let editedQueryChipElement = this.elements.queryChipElements[this.elements.editedQueryChipElementIndex];
+      editedQueryChipElement.dataset.type = dataType;
+      editedQueryChipElement.dataset.query = queryText;
+      editedQueryChipElement.firstChild.textContent = prettyQueryText;
+      this.updateChipList();
+      this.queryPreviewBuilder();
+    } else {
+      this.queryChipFactory(dataType, prettyQueryText, queryText, index, isClosingTag, isEditable);
+    }
+  }
+
+  queryChipFactory(dataType, prettyQueryText, queryText, index=null, isClosingTag=false, isEditable=false) {
+    // Creates a new query chip element, adds Eventlisteners for selection, deletion and drag and drop and appends it to the query input field.
+    queryText = nopaque.Utils.escape(queryText);
+    prettyQueryText = nopaque.Utils.escape(prettyQueryText);
+    let queryChipElement = nopaque.Utils.HTMLToElement(
+      `
+        <span class="chip query-component" data-type="${dataType}" data-query="${queryText}" draggable="true"">
+          ${prettyQueryText}${isEditable ? '<i class="material-icons chip-action-button" data-chip-action="edit" style="padding-left:5px; font-size:18px; cursor:pointer;">edit</i>': ''}
+          ${isClosingTag ? '' : '<i class="material-icons close chip-action-button" data-chip-action="delete">close</i>'}
+        </span>
+      `
+    );
+    this.addActionListeners(queryChipElement);
+    queryChipElement.addEventListener('dragstart', this.handleDragStart.bind(this, queryChipElement));
+    queryChipElement.addEventListener('dragend', this.handleDragEnd);
+    // If an index is given, inserts the query chip after the given index (only relevant for Incidence Modifier) and if there is a closing tag, inserts the query chip before the closing tag.
+    if (index !== null) {
+      this.updateChipList();
+      this.elements.queryChipElements[index].after(queryChipElement);
+    } else {
+      this.elements.queryInputField.insertBefore(queryChipElement, this.elements.queryElementTarget);
+    }
+    if (isClosingTag) {
+      this.moveQueryElementTarget(queryChipElement);
+    }
+
+    this.updateChipList();
+    this.queryPreviewBuilder();
+  }
+
+  moveQueryElementTarget(element) {
+    this.elements.queryInputField.insertBefore(this.elements.queryElementTarget, element);
+  }
+
+  addActionListeners(queryChipElement) {
+    let notQuantifiableDataTypes = ['start-sentence', 'end-sentence', 'start-entity', 'start-empty-entity', 'end-entity', 'token-incidence-modifier'];
+    queryChipElement.addEventListener('click', (event) => {
+      if (event.target.classList.contains('chip')) {
+        if (!notQuantifiableDataTypes.includes(queryChipElement.dataset.type)) {
+          this.selectChipElement(queryChipElement);
+        }
+      }
+    });
+    let chipActionButtons = queryChipElement.querySelectorAll('.chip-action-button');
+    chipActionButtons.forEach(button => {
+      button.addEventListener('click', (event) => {
+      if (event.target.dataset.chipAction === 'delete') {
+        this.deleteChipElement(queryChipElement);
+      } else if (event.target.dataset.chipAction === 'edit') {
+        this.editChipElement(queryChipElement);
+      }
+      });
+    });
+  }
+
+  editChipElement(queryChipElement) {
+    this.elements.editingModusOn = true;
+    this.elements.editedQueryChipElementIndex = Array.from(this.elements.queryInputField.children).indexOf(queryChipElement);
+    switch (queryChipElement.dataset.type) {
+      case 'start-entity':
+        this.extensions.structuralAttributeBuilderFunctions.editStartEntityChipElement(queryChipElement);
+        break;
+      case 'token':
+        let queryElementsContent = this.extensions.tokenAttributeBuilderFunctions.prepareTokenQueryElementsContent(queryChipElement);
+        this.extensions.tokenAttributeBuilderFunctions.editTokenChipElement(queryElementsContent);
+        break;
+      default:
+        break;
+    }
+  }
+
+  deleteChipElement(attr) {
+    let elementIndex = Array.from(this.elements.queryInputField.children).indexOf(attr);
+    switch (attr.dataset.type) {
+      case 'start-sentence':
+        this.deleteClosingTagHandler(elementIndex, 'end-sentence');
+        break;
+      case 'start-empty-entity':
+      case 'start-entity':
+        this.deleteClosingTagHandler(elementIndex, 'end-entity');
+        break;
+      case 'token':
+        let nextElement = Array.from(this.elements.queryInputField.children)[elementIndex+1];
+        if (nextElement !== undefined && nextElement.dataset.type === 'token-incidence-modifier') {
+          this.deleteChipElement(nextElement);
+        }
+      default:
+        break;
+    }
+    this.elements.queryInputField.removeChild(attr);
+    this.updateChipList();
+    this.queryPreviewBuilder();
+  }
+
+  deleteClosingTagHandler(elementIndex, closingTagType) {
+    let closingTags = this.elements.queryInputField.querySelectorAll(`[data-type="${closingTagType}"]`);
+    for (let i = 0; i < closingTags.length; i++) {
+      let closingTag = closingTags[i];
+    
+      if (Array.from(this.elements.queryInputField.children).indexOf(closingTag) > elementIndex) {
+        this.deleteChipElement(closingTag);
+        break;
+      }
+    }
+  }
+
+  handleDragStart(queryChipElement) {
+    // is called when a query chip is dragged. It creates a dropzone (in form of a chip) for the dragged chip and adds it to the query input field.
+    let queryChips = this.elements.queryInputField.querySelectorAll('.query-component');
+    if (queryChipElement.dataset.type === 'token-incidence-modifier') {
+      queryChips = this.elements.queryInputField.querySelectorAll('.query-component[data-type="token"]');
+    }
+    setTimeout(() => {
+      let targetChipElement = nopaque.Utils.HTMLToElement('<span class="chip drop-target">Drop here</span>');
+      for (let element of queryChips) {
+        if (element === this.elements.queryInputField.querySelectorAll('.query-component')[0]) {
+          let secondTargetChipClone = targetChipElement.cloneNode(true);
+          element.insertAdjacentElement('beforebegin', secondTargetChipClone);
+          this.addDragDropListeners(secondTargetChipClone, queryChipElement);
+        }
+        if (element === queryChipElement || element.nextSibling === queryChipElement) {continue;}
+
+        let targetChipClone = targetChipElement.cloneNode(true);
+        element.insertAdjacentElement('afterend', targetChipClone);
+        //TODO: Change to two different functions for drag and drop
+        this.addDragDropListeners(targetChipClone, queryChipElement);
+      }
+    }, 0);
+  }
+
+  handleDragEnd(event) {
+    // is called when a query chip is dropped. It removes the dropzones and initializes the tooltips if the dragged element is the query element target.
+    if (event.target.classList.contains('query-element-target')) {
+      M.Tooltip.init(event.target);
+    }
+    document.querySelectorAll('.drop-target').forEach(target => target.remove());
+  }
+
+  addDragDropListeners(targetChipClone, queryChipElement) {
+    targetChipClone.addEventListener('dragover', (event) => {
+      event.preventDefault();
+    });
+    targetChipClone.addEventListener('dragenter', (event) => {
+      event.preventDefault();
+      event.target.style.borderStyle = 'solid dotted';
+    });
+    targetChipClone.addEventListener('dragleave', (event) => {
+      event.preventDefault();
+      event.target.style.borderStyle = 'hidden';
+    });
+    targetChipClone.addEventListener('drop', (event) => {
+      let dropzone = event.target;
+      dropzone.parentElement.replaceChild(queryChipElement, dropzone);
+      this.updateChipList();
+      this.queryPreviewBuilder();
+    });
+  }
+
+  queryPreviewBuilder() {
+    // Builds the query preview in the form of pure CQL and displays it in the query preview field.
+    let queryPreview = document.querySelector('#corpus-analysis-concordance-query-preview');
+    let queryInputFieldContent = [];
+    this.elements.queryChipElements.forEach(element => {
+      let queryElement = element.dataset.query;
+      if (queryElement !== undefined) {
+        queryElement = nopaque.Utils.escape(queryElement);
+      }
+      queryInputFieldContent.push(queryElement);
+    });
+  
+    let queryString = queryInputFieldContent.join(' ');
+    let replacements = {
+      ' +': '+',
+      ' *': '*',
+      ' ?': '?',
+      ' {': '{'
+    };
+
+    for (let key in replacements) {
+      queryString = queryString.replace(key, replacements[key]);
+    }
+    queryString += ';';
+  
+    queryPreview.innerHTML = queryString;
+    queryPreview.parentNode.classList.toggle('hide', queryString === ';');
+  }
+
+  selectChipElement(attr) {
+    if (attr.classList.contains('teal')) {
+      return;
+    }
+
+    this.toggleClass(['token-incidence-modifiers'], 'disabled', 'toggle');
+    attr.classList.toggle('teal');
+    attr.classList.toggle('lighten-5');
+
+    M.Dropdown.getInstance(document.querySelector('.dropdown-trigger[data-toggle-area="token-incidence-modifiers"]')).open();
+
+  }
+
+  unselectChipElement(attr) {
+    let nModalInstance = M.Modal.getInstance(document.querySelector('#corpus-analysis-concordance-exactly-n-token-modal'));
+    let nmModalInstance = M.Modal.getInstance(document.querySelector('#corpus-analysis-concordance-between-nm-token-modal'));
+    if (nModalInstance.isOpen || nmModalInstance.isOpen) {
+      return;
+    }
+    attr.classList.remove('teal', 'lighten-5');
+    this.toggleClass(['token-incidence-modifiers'], 'disabled', 'add');
+  }
+
+  tokenIncidenceModifierHandler(incidenceModifier, incidenceModifierPretty, nOrNM = false) {
+    // Adds a token incidence modifier to the query input field.
+    let selectedChip = this.elements.queryInputField.querySelector('.chip.teal');
+    let selectedChipIndex = Array.from(this.elements.queryChipElements).indexOf(selectedChip);
+    if (nOrNM) {
+      this.unselectChipElement(selectedChip);
+    }
+    this.submitQueryChipElement('token-incidence-modifier', incidenceModifierPretty, incidenceModifier, selectedChipIndex);
+  }
+
+  tokenNMSubmitHandler(modalId) {
+    // Adds a token incidence modifier (exactly n or between n and m) to the query input field.
+    let modal = document.querySelector(`#${modalId}`);
+    let input_n = modal.querySelector('.n-m-input[data-value-type="n"]').value;
+    let input_m = modal.querySelector('.n-m-input[data-value-type="m"]') || undefined;
+    input_m = input_m !== undefined ? input_m.value : '';
+    let input = `{${input_n}${input_m !== '' ? ',' : ''}${input_m}}`;
+    let pretty_input = `between ${input_n} and ${input_m} (${input})`;
+    if (input_m === '') {
+     pretty_input = `exactly ${input_n} (${input})`;
+    }
+
+    let instance = M.Modal.getInstance(modal);
+    instance.close();
+
+    this.tokenIncidenceModifierHandler(input, pretty_input, true);
+  }
+
+  expertModeQueryBuilderSwitchHandler() {
+    let queryBuilderDisplay = document.querySelector("#corpus-analysis-concordance-query-builder-display");
+    let expertModeDisplay = document.querySelector("#corpus-analysis-concordance-expert-mode-display");
+    let expertModeSwitch = document.querySelector("#corpus-analysis-concordance-expert-mode-switch");
+    let submitModal = M.Modal.getInstance(document.querySelector('#corpus-analysis-concordance-switch-to-query-builder-submit-modal'));
+    
+    let confirmSwitchToQueryBuilderButton = document.querySelector('.switch-action[data-switch-action="confirm"]');
+    confirmSwitchToQueryBuilderButton.addEventListener("click", () => {
+      queryBuilderDisplay.classList.remove("hide");
+      expertModeDisplay.classList.add("hide");
+      this.switchToQueryBuilderParser();
+    });
+
+    expertModeSwitch.addEventListener("change", () => {
+      const isChecked = expertModeSwitch.checked;
+      if (isChecked) {
+        queryBuilderDisplay.classList.add("hide");
+        expertModeDisplay.classList.remove("hide");
+        this.switchToExpertModeParser();
+      } else {
+        submitModal.open();
+      }
+    });
+  }
+
+  switchToExpertModeParser() {
+    let expertModeInputField = document.querySelector('#corpus-analysis-concordance-form-query');
+    expertModeInputField.value = '';
+    let queryBuilderInputFieldValue = nopaque.Utils.unescape(document.querySelector('#corpus-analysis-concordance-query-preview').innerHTML.trim());
+    if (queryBuilderInputFieldValue !== "" && queryBuilderInputFieldValue !== ";") {
+      expertModeInputField.value = queryBuilderInputFieldValue;
+    }
+  }
+
+  switchToQueryBuilderParser() {
+    this.resetQueryInputField();
+    let expertModeInputFieldValue = document.querySelector('#corpus-analysis-concordance-form-query').value;
+    let chipElements = this.parseTextToChip(expertModeInputFieldValue);
+    let editableElements = ['start-entity', 'token'];
+    for (let chipElement of chipElements) {
+      let isEditable = editableElements.includes(chipElement['type']);
+      if (chipElement['query'] === '[]'){
+        isEditable = false;
+      }
+      this.submitQueryChipElement(chipElement['type'], chipElement['pretty'], chipElement['query'], null, false, isEditable);
+    }
+  }
+
+  parseTextToChip(query) {
+    const parsingElementDict = {
+      '<s>': {
+        pretty: 'Sentence Start',
+        type: 'start-sentence'
+      },
+      '<\/s>': {
+        pretty: 'Sentence End',
+        type: 'end-sentence'
+      },
+      '<ent>': {
+        pretty: 'Entity Start',
+        type: 'start-empty-entity'
+      },
+      '<ent_type="([A-Z]+)">': {
+        pretty: '',
+        type: 'start-entity'
+      },
+      '<\\\/ent(_type)?>': {
+        pretty: 'Entity End',
+        type: 'end-entity'
+      },
+      '\\[(word|lemma|pos|simple_pos)=("(?:[^"\\\\]|\\\\")*") ?(%c)? ?((\\&|\\|) ?(word|lemma|pos|simple_pos)=("(?:[^"\\\\]|\\\\")*") ?(%c)? ?)*\\]': {
+        pretty: '',
+        type: 'token'
+      },
+      '\\[\\]': {
+        pretty: 'Empty Token',
+        type: 'token'
+      },
+      '(?<!\\[) ?\\+ ?(?![^\\]]\\])': {
+        pretty: ' one or more (+)',
+        type: 'token-incidence-modifier'
+      },
+      '(?<!\\[) ?\\* ?(?![^\\]]\\])': {
+        pretty: 'zero or more (*)',
+        type: 'token-incidence-modifier'
+      },
+      '(?<!\\[) ?\\? ?(?![^\\]]\\])': {
+        pretty: 'zero or one (?)',
+        type: 'token-incidence-modifier'
+      },
+      '(?<!\\[) ?\\{[0-9]+} ?(?![^\\]]\\])': {
+        pretty: '',
+        type: 'token-incidence-modifier'
+      },
+      '(?<!\\[) ?\\{[0-9]+(,[0-9]+)?} ?(?![^\\]]\\])': {
+        pretty: '',
+        type: 'token-incidence-modifier'
+      }
+    }
+  
+    let chipElements = [];
+    let regexPattern = Object.keys(parsingElementDict).map(pattern => `(${pattern})`).join('|');
+    const regex = new RegExp(regexPattern, 'gi');
+    let match;
+  
+    while ((match = regex.exec(query)) !== null) {
+      // this is necessary to avoid infinite loops with zero-width matches
+      if (match.index === regex.lastIndex) {
+        regex.lastIndex++;
+      }
+      let stringElement = match[0];
+      for (let [pattern, chipElement] of Object.entries(parsingElementDict)) {
+        const parsingRegex = new RegExp(pattern, 'gi');
+        if (parsingRegex.exec(stringElement)) {
+          // Creating the pretty text for the chip element
+          let prettyText;
+          switch (pattern) {
+            case '<ent_type="([A-Z]+)">':
+              prettyText = `Entity Type=${stringElement.replace(/<ent_type="|">/g, '')}`;
+              break;
+            case ':: ?match\\.text_[A-Za-z]+="[^"]+"':
+              prettyText = stringElement.replace(/:: ?match\.text_|"|"/g, '');
+              break;
+            case '\\[(word|lemma|pos|simple_pos)=("(?:[^"\\\\]|\\\\")*") ?(%c)? ?((\\&|\\|) ?(word|lemma|pos|simple_pos)=("(?:[^"\\\\]|\\\\")*") ?(%c)? ?)*\\]':
+              prettyText = stringElement.replace(/^\[|\]$|(?<!\\)"/g, '');
+              prettyText = prettyText.replace(/\&/g, ' and ').replace(/\|/g, ' or ');
+              break;
+            case '(?<!\\[) ?\\{[0-9]+} ?(?![^\\]]\\])':
+              prettyText = `exactly ${stringElement.replace(/{|}/g, '')} (${stringElement})`;
+              break;
+            case '(?<!\\[) ?\\{[0-9]+(,[0-9]+)?} ?(?![^\\]]\\])':
+              prettyText = `between${stringElement.replace(/{|}/g, ' ').replace(',', ' and ')}(${stringElement})`;
+              break;
+            default:
+              prettyText = chipElement.pretty;
+              break;
+          }
+          chipElements.push({
+            type: chipElement.type,
+            pretty: prettyText,
+            query: stringElement
+          });
+          break;
+        }
+      }
+    }
+  
+    return chipElements;
+  }
+};

app/static/js/corpus-analysis/query-builder/structural-attribute-builder-functions.js

@@ -0,0 +1,82 @@
+nopaque.corpus_analysis.query_builder.StructuralAttributeBuilderFunctions = class StructuralAttributeBuilderFunctions {
+  constructor(app) {
+    this.app = app;
+    this.elements = app.elements;
+
+    this.structuralAttrModalEventlisteners();
+
+    this.elements.structuralAttrModal = M.Modal.init(
+      document.querySelector('#corpus-analysis-concordance-structural-attr-modal'), 
+      {
+        onCloseStart: () => {
+          this.resetStructuralAttrModal();
+        }
+      }
+    );
+  }
+
+  structuralAttrModalEventlisteners() {
+    document.querySelectorAll('[data-structural-attr-modal-action-button]').forEach(button => {
+      button.addEventListener('click', () => {
+        this.actionButtonInStrucAttrModalHandler(button.dataset.structuralAttrModalActionButton);
+      });
+    });
+    document.querySelector('.ent-type-selection-action[data-ent-type="any"]').addEventListener('click', () => {
+      this.app.submitQueryChipElement('start-empty-entity', 'Entity Start', '<ent>');
+      this.app.submitQueryChipElement('end-entity', 'Entity End', '</ent>', null, true);
+      this.elements.structuralAttrModal.close();
+    });
+    document.querySelector('.ent-type-selection-action[data-ent-type="english"]').addEventListener('change', (event) => {
+      this.app.submitQueryChipElement('start-entity', `Entity Type=${event.target.value}`, `<ent_type="${event.target.value}">`, null, false, true);
+      if (!this.elements.editingModusOn) {
+        this.app.submitQueryChipElement('end-entity', 'Entity End', '</ent_type>', null, true);
+      }
+      this.elements.structuralAttrModal.close();
+    });
+    document.querySelector('.ent-type-selection-action[data-ent-type="german"]').addEventListener('change', (event) => {
+      this.app.submitQueryChipElement('start-entity', `Entity Type=${event.target.value}`, `<ent_type="${event.target.value}">`, null, false, true); 
+      if (!this.elements.editingModusOn) {
+        this.app.submitQueryChipElement('end-entity', 'Entity End', '</ent_type>', null, true);
+      }
+      this.elements.structuralAttrModal.close();
+    });
+  }
+
+  resetStructuralAttrModal() {
+    this.app.resetMaterializeSelection([this.elements.englishEntTypeSelection, this.elements.germanEntTypeSelection]);
+    this.app.toggleClass(['entity-builder'], 'hide', 'add');
+    this.toggleEditingAreaStructuralAttrModal('remove');
+    this.elements.editingModusOn = false;
+    this.elements.editedQueryChipElementIndex = undefined;
+  }
+
+  actionButtonInStrucAttrModalHandler(action) {
+    switch (action) {
+      case 'sentence':
+        this.app.submitQueryChipElement('start-sentence', 'Sentence Start', '<s>');
+        this.app.submitQueryChipElement('end-sentence', 'Sentence End', '</s>', null, true);
+        this.elements.structuralAttrModal.close();
+        break;
+      case 'entity':
+        this.app.toggleClass(['entity-builder'], 'hide', 'toggle');
+        break;
+      default:
+        break;
+    }
+  }
+
+  toggleEditingAreaStructuralAttrModal(action) {
+    // If the user edits a query chip element, the corresponding editing area is displayed and the other areas are hidden or disabled.
+    this.app.toggleClass(['sentence-button', 'entity-button', 'any-type-entity-button'], 'disabled', action);
+  }
+
+  editStartEntityChipElement(queryChipElement) {
+    this.elements.structuralAttrModal.open();
+    this.app.toggleClass(['entity-builder'], 'hide', 'remove');
+    this.toggleEditingAreaStructuralAttrModal('add');
+    let entType = queryChipElement.dataset.query.replace(/<ent_type="|">/g, '');
+    let isEnglishEntType = this.elements.englishEntTypeSelection.querySelector(`option[value=${entType}]`) !== null;
+    let selection = isEnglishEntType ? this.elements.englishEntTypeSelection : this.elements.germanEntTypeSelection;
+    this.app.resetMaterializeSelection([selection], entType);
+  }
+}