chore: use importlib to refer to datafiles and create entrypoint

README.md

@@ -14,6 +14,13 @@ According to [World History Encyclopedia](https://www.worldhistory.org/article/7
 
 In the same way the Domus TDD API does not just offer a standards-conformant interface as specified in the link above, but also allows for flexible and scalable deployment, and has the possibility for extensions to carry out a bit more than just store & retrieve operations.
 
+## Install from pypi
+
+```
+pip install domus-tdd-api
+domus-tdd-api run
+```
+
 ## Configuration
 
 The TDD API can be configured using two methods.
@@ -26,7 +33,7 @@ The TDD API can be configured using two methods.
      export TDD__CHECK_SCHEMA=True
    ```
 
-2. Editing the `config.toml` file using the direct name of the variable
+2. Create and edit a `config.toml` file at the root of the folder where you run the API using the direct name of the variable
 
    ```
      SPARQLENDPOINT_URL="http://my-new-sparql.endpoint/address"
@@ -55,9 +62,6 @@ The `config.toml` file can also be used to define FLask server configuration (c.
 | [TDD__]MANDATE_TTL            | False                                     | Boolean value, if set to True, it will only upload TDs having a time-to-live (ttl) value. The server will send a 400 HTTP code if the TD does not contain one. |
 | [TDD__]LIMIT_BATCH_TDS        | 25                                        | Default limit of returned TDs by batch (used for pagination)                                                                                                   |
 | [TDD__]ENDPOINT_TYPE          | None                                      | Special configuration to workaround SPARQL endpoints which do not follow the SPARQL standard. Possible values: `GRAPHDB` or `VIRTUOSO`                         |
-| [TDD__]TD_JSONSCHEMA          | ./tdd/data/td-json-schema-validation.json | The path to the file containing JSON-Schema to validate the TDs                                                                                                |
-| [TDD__]TD_ONTOLOGY            | ./tdd/data/td.ttl                         | The path to the file containing the TD OWL Ontology (only used for SHACL validation)                                                                           |
-| [TDD__]TD_SHACL_VALIDATOR     | ./tdd/data/td-validation.ttl              | The path to the file containing the SHACL shapes (only used for SHACL validation)                                                                              |
 | [TDD__]PERIOD_CLEAR_EXPIRE_TD | 3600                                      | The number of seconds between each clearing of expired TDs (0 to disable clearing expired TD)                                                                  |
 | [TDD__]OVERWRITE_DISCOVERY    | False                                     | Use custom discovery context (for offline purposes)                                                                                                            |
 

setup.py

@@ -66,4 +66,7 @@
             "requests",
         ],
     },
+    entry_points={
+        "console_scripts": ["domus-tdd-api = tdd:cli"],
+    },
 )

tdd/__init__.py

@@ -13,15 +13,19 @@
  * SPDX-License-Identifier: EPL-2.0 OR W3C-20150513
  ********************************************************************************"""
 
+from importlib import resources
 import sys
 import time
 from threading import Thread
+import click
 from flask import Flask, request, Response, stream_with_context
 import json
+from flask.cli import FlaskGroup
 import json_merge_patch
 import httpx
 import toml
 from importlib_metadata import entry_points
+from pathlib import Path
 
 
 from tdd.errors import (
@@ -95,7 +99,8 @@ def thread_clear_expire_td():
 
 def create_app():
     app = Flask(__name__)
-    app.config.from_file("../config.toml", load=toml.load)
+    if Path("config.toml").is_file():
+        app.config.from_file("../config.toml", load=toml.load)
     wait_for_sparqlendpoint()
     register_error_handler(app)
     register_routes(app)
@@ -127,6 +132,11 @@ def create_app():
     return app
 
 
+@click.group(cls=FlaskGroup, create_app=create_app)
+def cli():
+    """Management script for the API application."""
+
+
 def register_error_handler(app):
     @app.errorhandler(AppException)
     def error_response(e):
@@ -155,7 +165,7 @@ def add_cors_headers(response):
 
     @app.route("/", methods=["GET"])
     def directory_description():
-        with open("tdd/data/tdd-description.json", "r") as f:
+        with resources.open_text("tdd.data", "tdd-description.json") as f:
             tdd_description = json.loads(f.read())
             tdd_description["base"] = CONFIG["TD_REPO_URL"]
             return Response(

tdd/common.py

@@ -13,6 +13,7 @@
  * SPDX-License-Identifier: EPL-2.0 OR W3C-20150513
  ********************************************************************************"""
 
+from importlib import resources
 import subprocess
 import json
 import re
@@ -29,7 +30,6 @@
 from tdd.metadata import insert_metadata, delete_metadata
 from tdd.errors import IDNotFound
 from tdd.config import CONFIG
-from tdd.paths import LIB_PATH
 
 
 def get_check_schema_from_url_params(request):
@@ -53,11 +53,16 @@ def delete_id(uri):
 
 
 def json_ld_to_ntriples(ld_content):
-    p = subprocess.Popen(
-        ["node", LIB_PATH / "transform-to-nt.js", json.dumps(ld_content)],
-        stdout=subprocess.PIPE,
-    )
-    nt_content = p.stdout.read()
+    with resources.path("tdd.lib", "transform-to-nt.js") as transform_lib_path:
+        p = subprocess.Popen(
+            [
+                "node",
+                transform_lib_path,
+                json.dumps(ld_content),
+            ],
+            stdout=subprocess.PIPE,
+        )
+        nt_content = p.stdout.read()
     return nt_content.decode("utf-8")
 
 
@@ -96,11 +101,17 @@ def put_rdf_in_sparql(g, uri, context, delete_if_exists, ontology, forced_type=N
 
 
 def frame_nt_content(id, nt_content, frame):
-    p = subprocess.Popen(
-        ["node", LIB_PATH / "frame-jsonld.js", nt_content, json.dumps(frame)],
-        stdout=subprocess.PIPE,
-    )
-    json_ld_compacted = p.stdout.read()
+    with resources.path("tdd.lib", "frame-jsonld.js") as frame_lib_path:
+        p = subprocess.Popen(
+            [
+                "node",
+                frame_lib_path,
+                nt_content,
+                json.dumps(frame),
+            ],
+            stdout=subprocess.PIPE,
+        )
+        json_ld_compacted = p.stdout.read()
     return json_ld_compacted
 
 

tdd/config.py

@@ -15,15 +15,12 @@
 
 from config import config_from_env, config_from_toml, config_from_dict
 from config.configuration_set import ConfigurationSet
+from pathlib import Path
 
-from tdd.paths import DATA_PATH
 
 _default_config = {
     "TD_REPO_URL": "http://localhost:5000",
     "SPARQLENDPOINT_URL": "http://127.0.0.1:3030/things",
-    "TD_JSONSCHEMA": DATA_PATH / "td-json-schema-validation.json",
-    "TD_ONTOLOGY": DATA_PATH / "td.ttl",
-    "TD_SHACL_VALIDATOR": DATA_PATH / "td-validation.ttl",
     "ENDPOINT_TYPE": None,
     "LIMIT_BATCH_TDS": 25,
     "CHECK_SCHEMA": False,
@@ -33,11 +30,18 @@
     "OVERWRITE_DISCOVERY": False,
 }
 
-CONFIG = ConfigurationSet(
-    config_from_env(prefix="TDD", interpolate=True),
-    config_from_toml("config.toml", read_from_file=True),
-    config_from_dict(_default_config),
-)
+
+if Path("config.toml").is_file():
+    CONFIG = ConfigurationSet(
+        config_from_env(prefix="TDD", interpolate=True),
+        config_from_toml("config.toml", read_from_file=True),
+        config_from_dict(_default_config),
+    )
+else:
+    CONFIG = ConfigurationSet(
+        config_from_env(prefix="TDD", interpolate=True),
+        config_from_dict(_default_config),
+    )
 
 # Remove trailing /
 if CONFIG["SPARQLENDPOINT_URL"][-1] == "/":

tdd/context.py

@@ -17,7 +17,6 @@
 import json
 
 
-from tdd.paths import DATA_PATH
 from tdd.config import CONFIG
 from tdd.utils import DEFAULT_THING_CONTEXT_URI, DEFAULT_DISCOVERY_CONTEXT_URI
 from tdd.sparql import (
@@ -26,6 +25,7 @@
     GET_ALL_CONTEXTS,
     query,
 )
+from importlib import resources
 
 
 def convert_context_to_array(ld_content):
@@ -45,7 +45,7 @@ def overwrite_thing_context(ld_content):
         return
     if type(ld_content["@context"]) not in (tuple, list):
         return
-    with open(DATA_PATH / "fixed-ctx.json") as fp:
+    with resources.open_text("tdd.data", "fixed-ctx.json") as fp:
         fixed_ctx = fp.read()
         try:
             index_wot_ctx = ld_content["@context"].index(DEFAULT_THING_CONTEXT_URI)
@@ -61,7 +61,7 @@ def overwrite_discovery_context(ld_content):
         return
     if type(ld_content["@context"]) not in (tuple, list):
         return
-    with open(DATA_PATH / "fixed-discovery-ctx.json") as fp:
+    with resources.open_text("tdd.data", "fixed-discovery-ctx.json") as fp:
         fixed_discovery_ctx = fp.read()
         try:
             index_discovery_ctx = ld_content["@context"].index(

tdd/paths.py

@@ -1,4 +0,0 @@
-from pathlib import Path
-
-DATA_PATH = Path(__file__).parent / "data"
-LIB_PATH = Path(__file__).parent / "lib"

tdd/td.py

@@ -16,6 +16,7 @@
 import concurrent.futures
 from copy import copy
 from datetime import datetime
+from importlib import resources
 import json
 from jsonschema import Draft7Validator
 import uuid
@@ -71,7 +72,7 @@
 )
 
 
-with open(CONFIG["TD_JSONSCHEMA"]) as fp:
+with resources.open_text("tdd.data", "td-json-schema-validation.json") as fp:
     schema = json.load(fp)
 
 validator = Draft7Validator(schema=schema)
@@ -183,10 +184,11 @@ def put_td_rdf_in_sparql(
 
     if check_schema:
         ontology_graph = create_binded_graph()
-        ontology_graph.parse(location=CONFIG["TD_ONTOLOGY"], format="turtle")
-
-        shacl_shapes_graph = create_binded_graph()
-        shacl_shapes_graph.parse(location=CONFIG["TD_SHACL_VALIDATOR"], format="turtle")
+        with resources.path("tdd.data", "td.ttl") as onto_path:
+            ontology_graph.parse(location=onto_path, format="turtle")
+        with resources.path("tdd.data", "td-validation.ttl") as shacl_path:
+            shacl_shapes_graph = create_binded_graph()
+            shacl_shapes_graph.parse(location=shacl_path, format="turtle")
 
         conforms, graph_reports, text_reports = pyshacl.validate(
             g,