Readme & docs updated, test for iri collection added

README.md

@@ -4,23 +4,40 @@ Creditrisk-poc is a Hydra powered API which serves loan portfolio data using EBA
 
 ## Features
 creditrisk-poc consist following features:
-* Loan & Borrower classes.
+* Loan ,Borrower & Collateral classes.
 * Borrower class collection.
-* Loan & Borrower class are linked using a foreign key ("CounterpartyId").
+* Loan, Borrower & Collateral classes are linked using `foreign keys`.
 * Loan class can perform all the CRUD operations ( GET, PUT, POST, DELETE).
 * Borrower can perform all the CRUD operations.
 * Borrower class collection can perform all the CRUD operations.
 
+## Classes are linked in the following manner:
+![Creditrisk_class_linking](https://user-images.githubusercontent.com/49719371/125194774-4d897280-e270-11eb-95af-4242bb1bffc2.jpg)
+
+
+## NonPerformingLoan.jsonld
+The `NonPerformingLoan.jsonld` is a subset vocabulary for NonPerformingLoan portfolios,
+vocabulary is generated automatically using `vocab_generator.py` from `NonperformingLoan.owl` ontology.
+```bash
+python npl_vocab/vocab_generator.py
+```
+It will generate the JSON-LD vocabulary which can be used to create ApiDoc.
+
 ## API_DOC
-API_Doc is generated through hydra-python-core module doc_writer.
+API_Doc is generated through hydra-python-core module `doc_writer` and `nplvoac_parse.py` which automates the creation
+of classes and properties from JSON-LD vocabulary.
 
-API_Doc & doc_writer file can be found here :
+API_Doc, doc_writer & `nplvocab_parser.py` files can be found here :
 ```
 api_doc
 |
 |___ ApiDoc.jsonld
 |___ api_docwriter.py
+|___ nplvocab_parser.py
 ```
+**nplvocab_parser** parse all the classes & properties from `NonPerformingLoan.jsonld` and provide functions for converting
+them to HydraClass & HydraClassProp.
+
 `ApiDoc` is a JSON serialized object, It can be accessed as follows:
 ```python
 import json
@@ -30,6 +47,10 @@ doc = json.load(ApiDoc_file)
 ```
 you will get the doc in `python dict` format.
 
+### ApiDoc is generated with this flow:
+```
+NonPerformingLoan.owl --> ( vocab_generator.py ) NonPerformingLoan.jsonld --> ( nplvocab_parser.py) ApiDoc.jsonld 
+```
 ## Demo
 To run hydra powered creditrisk-poc API, just do the following:
 1) Clone creditrisk-poc

docs/nplvocab_parser.md

@@ -0,0 +1,28 @@
+# nplvocab_parser
+
+`nplvocab_parser` parses all the classes & properties from `NonPerformingLoan.jsonld` and convert them to HydraClass & HydraClassProp.
+
+nplvocab_parser is located in `api_doc` directory
+```python
+api_doc
+|
+|___ nplvocab_parser.py
+```
+It can be used by importing as a python module:
+```python
+import NPLVocab_parse as parser
+
+npl_vocab = parser.get_npl_vocab()
+classes = parser.get_all_classes(npl_vocab)
+hydra_classes = parser.create_hydra_classes(classes)
+```
+nplvocab_parser provide following functions:
+* `get_all_classes()` -> Return all the classes from the given Vocabulary.
+* `create_hydra_classes()` -> Return list of HydraClass objects.
+* `get_class_properties()` -> Return all the properties of the given class.
+* `create_hydra_properties()` -> Return list of HydraclasProps from the list of properties.
+* `get_class_id()` -> Returns the class id of given class.
+* `add_operations_to_class()` -> Return list of hydra properties of given class.
+
+
+

docs/vocab_generator.md

@@ -0,0 +1,17 @@
+# vocab_generator
+
+`vocab_genrator.py` generates `NonPerformingLoan.jsonld` vocabulary from the owl ontology.
+
+It is located inside the `npl_vocab` directory.
+```python
+npl_vocab
+|
+|___ vocab_generator.py
+```
+vocab_generator uses [rdflib](https://github.com/RDFLib/rdflib-jsonld) and [pyld](https://github.com/digitalbazaar/pyld) libaries to parse & serialize owl ontology 
+to jsonld with the `@context.`
+
+To generate JSON-LD voabulary:
+```python
+python npl_vocab/vocab_generator.py
+```

tests/conftest.py

@@ -4,6 +4,8 @@
 import json
 import uuid
 import os
+from os.path import abspath, dirname
+from pathlib import Path
 from hydra_python_core import doc_maker
 from hydra_python_core.doc_writer import DocUrl, HydraLink
 from sqlalchemy import create_engine
@@ -121,8 +123,8 @@ def test_doc(constants):
     """
     HYDRUS_SERVER_URL = constants['HYDRUS_SERVER_URL']
     API_NAME = constants['API_NAME']
-    API_DOC_PATH = os.path.relpath("tests/ApiDoc.jsonld")
-    print(API_DOC_PATH)
+    cwd_path = Path(dirname(dirname(abspath(__file__))))
+    API_DOC_PATH = cwd_path / "tests" / "ApiDoc.jsonld"
     doc_file = open(API_DOC_PATH, "r")
     doc = json.load(doc_file)
 

tests/test_functional.py

@@ -8,6 +8,7 @@
 import json
 import re
 import uuid
+from hydra_python_core.doc_writer import DocUrl
 from tests.conftest import gen_dummy_object
 
 
@@ -338,3 +339,30 @@ def test_Collections_member_DELETE(self, test_app_client, constants, doc):
                         delete_response = test_app_client.delete(full_endpoint)
                         assert delete_response.status_code == 200
 
+    def test_IriTemplate(self, test_app_client, constants, doc):
+        """Test structure of IriTemplates attached to parsed classes"""
+        API_NAME = constants['API_NAME']
+        index = test_app_client.get(f'/{API_NAME}')
+        assert index.status_code == 200
+        endpoints = json.loads(index.data.decode('utf-8'))
+        expanded_base_url = DocUrl.doc_url
+        for endpoint in endpoints['collections']:
+            collection_name = '/'.join(endpoint["@id"].split(f'/{API_NAME}/')[1:])
+            collection = doc.collections[collection_name]['collection']
+            class_name = collection.manages["object"].split(expanded_base_url)[1]
+            response_get = test_app_client.get(endpoint["@id"])
+            assert response_get.status_code == 200
+            response_get_data = json.loads(response_get.data.decode('utf-8'))
+            assert 'search' in response_get_data
+            assert 'hydra:mapping' in response_get_data['search']
+            class_ = doc.parsed_classes[class_name]['class']
+            class_props = [x.prop for x in class_.supportedProperty]
+            for mapping in response_get_data['search']['hydra:mapping']:
+                prop = mapping['hydra:property']
+                prop_name = mapping['hydra:variable']
+                is_valid_class_prop = prop not in ['limit', 'offset', 'pageIndex']
+                # check if IRI property is for searching through a nested_class
+                # and not this class_
+                is_nested_class_prop = "[" in prop_name and "]" in prop_name
+                if is_valid_class_prop and not is_nested_class_prop:
+                    assert prop in class_props