Improved schema validation

* Added debug Dockerfile/Makefile
* Validation of examples in docs

Dockerfile-debug-3.4

@@ -0,0 +1,5 @@
+from python:3.4
+
+RUN pip install pytest
+ADD requirements.txt /usr/src/app/
+RUN pip install -r /usr/src/app/requirements.txt

Makefile

@@ -20,12 +20,17 @@ buildall: $(addprefix build-, $(PYVERSIONS))
 build-%: Dockerfile-%
 	docker build -t '$(REPO)/$(NAME):$(VERSION)-python$*' -f Dockerfile-$* .;
 
+build-debug-%:
+	docker build -t '$(NAME)-debug' -f Dockerfile-debug-$* .;
+
 test: $(addprefix test-,$(PYMAIN))
 
 testall: $(addprefix test-,$(PYVERSIONS))
 
-debug-%: build-%
+debug-%: build-debug-%
-	docker run --rm -w /usr/src/app/ -v $$PWD:/usr/src/app --entrypoint=/bin/bash -ti '$(REPO)/$(NAME):$(VERSION)-python$*' ;
+	docker run --rm -w /usr/src/app/ -v $$PWD:/usr/src/app --entrypoint=/bin/bash -ti $(NAME)-debug ;
+
+debug: debug-$(PYMAIN)
 
 test-%: build-%
 	docker run --rm -w /usr/src/app/ --entrypoint=/usr/local/bin/python -ti '$(REPO)/$(NAME):$(VERSION)-python$*' setup.py test --addopts "-vvv -s" ;
@@ -51,6 +56,7 @@ upload: testall $(addprefix upload-,$(PYVERSIONS))
 clean:
 	@docker ps -a | awk '/$(REPO)\/$(NAME)/{ split($$2, vers, "-"); if(vers[1] != "${VERSION}"){ print $$1;}}' | xargs docker rm 2>/dev/null|| true
 	@docker images | awk '/$(REPO)\/$(NAME)/{ split($$2, vers, "-"); if(vers[1] != "${VERSION}"){ print $$1":"$$2;}}' | xargs docker rmi 2>/dev/null|| true
+	@docker rmi $(NAME)-debug 2>/dev/null || true
 
 upload_git:
 	git commit -a

docs/bad-examples/plugins/noplugins.json

@@ -0,0 +1,4 @@
+{
+  "plugins": [
+    ]
+}

docs/bad-examples/results/example-basic-FAIL.json

@@ -0,0 +1,18 @@
+{
+  "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
+  "@id": "http://example.com#NIFExample",
+  "@type": "results",
+  "analysis": [
+  ],
+  "entries": [
+    {
+      "@type": [
+        "nif:RFC5147String",
+        "nif:Context"
+      ],
+      "nif:beginIndex": 0,
+      "nif:endIndex": 40,
+      "nif:isString": "My favourite actress is Natalie Portman"
+    }
+  ]
+}

docs/examples/plugins/noplugins.json

@@ -0,0 +1,5 @@
+{
+  "@type": "plugins",
+  "plugins": [
+    ]
+}

docs/examples/results/example-basic.json

@@ -1,6 +1,7 @@
 {
   "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
   "@id": "http://example.com#NIFExample",
+  "@type": "results",
   "analysis": [
   ],
   "entries": [

docs/examples/results/example-complete.json

@@ -1,6 +1,7 @@
 {
   "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
   "@id": "me:Result1",
+  "@type": "results",
   "analysis": [
     {
       "@id": "me:SAnalysis1",

docs/examples/results/example-emotion.json

@@ -1,6 +1,7 @@
 {
   "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
   "@id": "me:Result1",
+  "@type": "results",
   "analysis": [
     {
       "@id": "me:EmotionAnalysis1",

docs/examples/results/example-ner.json

@@ -1,6 +1,7 @@
 {
   "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
   "@id": "me:Result1",
+  "@type": "results",
   "analysis": [
     {
       "@id": "me:NER1",

docs/examples/results/example-pad.json

@@ -0,0 +1,46 @@
+{
+"@context": [
+             "http://mixedemotions-project.eu/ns/context.jsonld",
+             {
+               "emovoc": "http://www.gsi.dit.upm.es/ontologies/onyx/vocabularies/emotionml/ns#"
+             }
+  ],
+  "@id": "me:Result1",
+  "@type": "results",
+  "analysis": [
+    {
+      "@id": "me:HesamsAnalysis",
+      "@type": "onyx:EmotionAnalysis",
+      "onyx:usesEmotionModel": "emovoc:pad-dimensions"
+    }
+  ],
+  "entries": [
+    {
+      "@id": "Entry1",
+      "@type": [
+        "nif:RFC5147String",
+        "nif:Context"
+      ],
+      "nif:isString": "This is a test string",
+      "entities": [
+      ],
+      "suggestions": [
+      ],
+      "sentiments": [
+      ],
+      "emotions": [
+        {
+          "@id": "Entry1#char=0,21",
+          "nif:anchorOf": "This is a test string",
+          "prov:wasGeneratedBy": "me:HesamAnalysis",
+          "onyx:hasEmotion": [
+            {
+                "emovoc:pleasure": 0.5,
+                "emovoc:arousal": 0.7
+            }
+          ]
+        }
+      ]
+    }
+  ]
+}

docs/examples/results/example-sentiment.json

@@ -1,6 +1,7 @@
 {
   "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
   "@id": "me:Result1",
+  "@type": "results",
   "analysis": [
     {
       "@id": "me:SAnalysis1",

docs/examples/results/example-suggestion.json

@@ -1,6 +1,7 @@
 {
   "@context": "http://mixedemotions-project.eu/ns/context.jsonld",
   "@id": "me:Result1",
+  "@type": "results",
   "analysis": [
     {
       "@id": "me:SgAnalysis1",

senpy/VERSION

@@ -1 +1 @@
-0.6.2
+pre-0.7.0

senpy/models.py

@@ -36,7 +36,6 @@ def read_schema(schema_file, absolute=False):
         return jsonref.load(f, base_uri=schema_uri)
     
         
-
 base_schema = read_schema(DEFINITIONS_FILE)
 logging.debug(base_schema)
 
@@ -157,18 +156,16 @@ class SenpyModel(SenpyMixin, dict):
 
         temp = dict(*args, **kwargs)
 
+        for obj in [self.schema,]+self.schema.get('allOf', []):
+            for k, v in obj.get('properties', {}).items():
+                if 'default' in v:
+                    temp[k] = copy.deepcopy(v['default'])
+
         for i in temp:
             nk = self._get_key(i)
             if nk != i:
                 temp[nk] = temp[i]
                 del temp[i]
-
-        reqs = self.schema.get('required', [])
-        for i in reqs:
-            if i not in temp:
-                prop = self.schema['properties'][i]
-                if 'default' in prop:
-                    temp[i] = copy.deepcopy(prop['default'])
         if 'context' in temp:
             context = temp['context']
             del temp['context']
@@ -226,12 +223,21 @@ class EmotionSet(SenpyModel):
 class Emotion(SenpyModel):
     schema = read_schema('emotion.json')
 
+class EmotionModel(SenpyModel):
+    schema = read_schema('emotionModel.json')
+
 class Suggestion(SenpyModel):
     schema = read_schema('suggestion.json')
 
 class PluginModel(SenpyModel):
     schema = read_schema('plugin.json')
 
+class EmotionPluginModel(SenpyModel):
+    schema = read_schema('plugin.json')
+
+class SentimentPluginModel(SenpyModel):
+    schema = read_schema('plugin.json')
+
 class Plugins(SenpyModel):
     schema = read_schema('plugins.json')
 

senpy/schemas/dimensions.json

@@ -0,0 +1,9 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "properties": {
+    "name": {"type": "string"},
+    "maxValue": {"type": "number"},
+    "minValue": {"type": "number"}
+  },
+  "required": ["name", "maxValue", "minValue"]
+}

senpy/schemas/emotionAnalysis.json

@@ -0,0 +1,18 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "description": "Senpy Emotion analysis",
+  "type": "object",
+  "allOf": [
+    {"$ref": "analysis.json"},
+    {"properties":
+     {
+       "onyx:hasEmotionModel": {
+         "anyOf": [
+           {"type": "string"},
+           {"$ref": "emotionModel.json"}
+         ]
+       }
+     },
+     "required": ["onyx:hasEmotionModel"]
+    }]
+}

senpy/schemas/emotionModel.json

@@ -8,17 +8,20 @@
       "description": "Piece of context that contains the Sentiment",
       "type": "string"
     },
-    "onyx:hasEmotion": {
+    "onyx:hasDimension": {
+      "type": "array",
+      "items": {
+        "$ref": "dimensions.json"
+      },
+      "uniqueItems": true
+    },
+    "onyx:hasEmotionCategory": {
       "type": "array",
       "items": {
         "$ref": "emotion.json"
       },
       "default": []
-    },
-    "prov:wasGeneratedBy": {
-      "type": "string",
-      "description": "The ID of the analysis that generated this Emotion. The full object should be included in the \"analysis\" property of the root object"
     }
   },
-  "required": ["@id", "prov:wasGeneratedBy", "onyx:hasEmotion"]
+  "required": ["@id", "onyx:hasEmotion"]
 }

senpy/schemas/plugins.json

@@ -1,11 +1,18 @@
 {
   "$schema": "http://json-schema.org/draft-04/schema#",
-  "properties": {
+  "allOf": [
-    "plugins": {
+    {"$ref": "response.json"},
-      "type": "array",
+    {
-      "items": {
+      "properties": {
-        "$ref": "plugin.json"
+        "plugins": {
+          "type": "array",
+          "items": {
+            "$ref": "plugin.json"
+          }
+        },
+        "@type": {
+        }
       }
     }
-  }
+  ]
 }

senpy/schemas/response.json

@@ -1,4 +1,9 @@
 {
   "$schema": "http://json-schema.org/draft-04/schema#",
-  "type": "object"
+  "type": "object",
+  "properties": {
+    "@type": {"type": "string"}
+  },
+  "required": ["@type"]
+
 }

senpy/schemas/results.json

@@ -1,31 +1,39 @@
 {
   "$schema": "http://json-schema.org/draft-04/schema#",
-  "title": "Results",
+  "allOf": [
-  "description": "The results of an analysis",
+    {"$ref": "response.json"},
-  "type": "object",
+    {
-  "properties": {
+      "title": "Results",
-    "@context": {
+      "description": "The results of an analysis",
-      "$ref": "context.json"
+      "type": "object",
-    },
+      "properties": {
-    "@id": {
+        "@context": {
-      "description": "ID of the analysis",
+          "$ref": "context.json"
-      "type": "string"
+        },
-    },
+        "@type": {
-    "analysis": {
+          "default": "results"
-      "type": "array",
+        },
-      "default": [],
+        "@id": {
-      "items": {
+          "description": "ID of the analysis",
-        "$ref": "analysis.json"
+          "type": "string"
-      }
+        },
-    },
+        "analysis": {
-    "entries": {
+          "type": "array",
-      "type": "array",
+          "default": [],
-      "default": [],
+          "items": {
-      "items": {
+            "$ref": "analysis.json"
-        "$ref": "entry.json"
+          }
-      }
+        },
-    }
+        "entries": {
+          "type": "array",
+          "default": [],
+          "items": {
+            "$ref": "entry.json"
+          }
+        }
 
-  },
+      },
-  "required": ["@id", "analysis", "entries"]
+      "required": ["@id", "analysis", "entries"]
+    }
+  ]
 }

tests/test_schemas.py

@@ -0,0 +1,52 @@
+from __future__ import print_function
+
+import json
+import unittest
+import os
+from os import path
+from fnmatch import fnmatch
+
+import pyld
+from jsonschema import validate, RefResolver, Draft4Validator, ValidationError
+
+root_path = path.join(path.dirname(path.realpath(__file__)), '..')
+schema_folder = path.join(root_path, 'senpy', 'schemas')
+examples_path = path.join(root_path, 'docs', 'examples')
+bad_examples_path = path.join(root_path, 'docs', 'bad-examples')
+
+class JSONSchemaTests(unittest.TestCase):
+    pass
+
+def do_create_(jsfile, success):
+    def do_expected(self):
+        with open(jsfile) as f:
+            js = json.load(f)
+        try:
+            assert '@type' in js
+            schema_name = js['@type']
+            with open(os.path.join(schema_folder, schema_name+".json")) as file_object:
+                schema = json.load(file_object)
+            resolver = RefResolver('file://' + schema_folder + '/', schema)
+            validator = Draft4Validator(schema, resolver=resolver)
+            validator.validate(js)
+        except (AssertionError, ValidationError, KeyError) as ex:
+            if success:
+                raise
+    return do_expected
+
+def add_examples(dirname, success):
+    for dirpath, dirnames, filenames in os.walk(dirname):
+        for i in filenames:
+            if fnmatch(i, '*.json'):
+                filename = path.join(dirpath, i)
+                test_method = do_create_(filename, success)
+                test_method.__name__ = 'test_file_%s_success_%s' % (filename, success)
+                test_method.__doc__ = '%s should %svalidate' % (filename, '' if success else 'not' )
+                setattr(JSONSchemaTests, test_method.__name__, test_method)
+                del test_method
+
+add_examples(examples_path, True)
+add_examples(bad_examples_path, False)
+
+if __name__ == '__main__':
+    unittest.main()