Implement a encoded search to json-ld wrapper.

[htsworkflow.git] / htsworkflow / submission / encoded.py

diff --git a/htsworkflow/submission/encoded.py b/htsworkflow/submission/encoded.py
index 8eec0c0ef60cd44e0123cfc8b7d28d1f8625c50d..1228d387f015a39a9da12c8ad5ca296d218c68de 100644 (file)
--- a/htsworkflow/submission/encoded.py
+++ b/htsworkflow/submission/encoded.py
@@ -64,7 +64,7 @@ ENCODED_CONTEXT = {
 }
 
 #FIXME: this needs to be initialized from rdfns
-_encoded_namespaces = {
+ENCODED_NAMESPACES = {
     # JSON-LD lets you define namespaces so you can used the shorted url syntax.
     # (instead of http://www.w3.org/2000/01/rdf-schema#label you can do
     # rdfs:label)
@@ -84,17 +84,17 @@ _encoded_namespaces = {
     # SO: available from http://www.berkeleybop.org/ontologies/so.owl
 
 }
-ENCODED_CONTEXT[None].update(_encoded_namespaces)
+
 ENCODED_SCHEMA_ROOT='/profiles/'
 
 class ENCODED:
     '''Programatic access encoded, the software powering ENCODE3's submit site.
     '''
-    def __init__(self, server, context=None):
+    def __init__(self, server, contexts=None):
         self.server = server
         self.username = None
         self.password = None
-        self.context = context if context else ENCODED_CONTEXT
+        self.contexts = contexts if contexts else ENCODED_CONTEXT
         self.schemas = {}
 
     def get_auth(self):
@@ -109,7 +109,7 @@ class ENCODED:
             self.username = authenticators[0]
             self.password = authenticators[2]
 
-    def add_jsonld_context(self, tree, contexts, base):
+    def add_jsonld_context(self, tree, default_base):
         """Add contexts to various objects in the tree.
 
         tree is a json tree returned from the DCC's encoded database.
@@ -118,11 +118,10 @@ class ENCODED:
         base, if supplied allows setting the base url that relative
             urls will be resolved against.
         """
-        tree['@context'] = contexts[None]
-        tree['@context']['@base'] = base
-        self.add_jsonld_child_context(tree, contexts)
+        self.add_jsonld_child_context(tree, default_base)
+        self.add_jsonld_namespaces(tree['@context'])
 
-    def add_jsonld_child_context(self, obj, contexts):
+    def add_jsonld_child_context(self, obj, default_base):
         '''Add JSON-LD context to the encoded JSON.
 
         This is recursive because some of the IDs were relative URLs
@@ -136,24 +135,41 @@ class ENCODED:
         if isinstance(obj, collections.Sequence):
             # how should I update lists?
             for v in obj:
-                self.add_jsonld_child_context(v, contexts)
+                self.add_jsonld_child_context(v, default_base)
             return
 
         if isinstance(obj, collections.Mapping):
             for v in obj.values():
-                self.add_jsonld_child_context(v, contexts)
+                self.add_jsonld_child_context(v, default_base)
 
         # we have an object. attach a context to it.
         if self._is_encoded_object(obj):
-            default_base = contexts[None]['@base']
-            context = {'@base': urljoin(default_base, obj['@id']),
-                       '@vocab': self.get_schema_url(obj)}
-            for t in obj['@type']:
-                if t in contexts:
-                    context.update(contexts[t])
+            context = self.create_jsonld_context(obj, default_base)
             if len(context) > 0:
                 obj.setdefault('@context', {}).update(context)
 
+    def add_jsonld_namespaces(self, context):
+        '''Add shortcut namespaces to a context
+
+        Only needs to be run on the top-most context
+        '''
+        context.update(ENCODED_NAMESPACES)
+
+    def create_jsonld_context(self, obj, default_base):
+        '''Synthesize the context for a encoded type
+
+        self.contexts[None] = default context attributes added to any type
+        self.contexts[type] = context attributes for this type.
+        '''
+        context = {'@base': urljoin(default_base, obj['@id']),
+                    '@vocab': self.get_schema_url(obj)}
+        # add in defaults
+        context.update(self.contexts[None])
+        for t in obj['@type']:
+            if t in self.contexts:
+                context.update(self.contexts[t])
+        return context
+
     def get_json(self, obj_id, **kwargs):
         '''GET an ENCODE object as JSON and return as dict
 
@@ -188,7 +204,7 @@ class ENCODED:
         '''
         url = self.prepare_url(obj_id)
         json = self.get_json(obj_id, **kwargs)
-        self.add_jsonld_context(json, self.context, url)
+        self.add_jsonld_context(json, url)
         return json
 
     def get_object_type(self, obj):
@@ -201,7 +217,7 @@ class ENCODED:
     def get_schema_url(self, obj):
         obj_type = self.get_object_type(obj)
         if obj_type:
-            return self.prepare_url(ENCODED_SCHEMA_ROOT + obj_type + '.json')
+            return self.prepare_url(ENCODED_SCHEMA_ROOT + obj_type + '.json') + '#'
 
     def _is_encoded_object(self, obj):
         '''Test to see if an object is a JSON-LD object
@@ -261,6 +277,28 @@ class ENCODED:
         url = urlunparse(url.values())
         return url
 
+    def search_jsonld(self, term, **kwargs):
+        '''Send search request to ENCODED
+        '''
+        url = self.prepare_url('/search/')
+        result = self.get_json(url, searchTerm=term, **kwargs)
+        self.convert_search_to_jsonld(result)
+        return result
+
+    def convert_search_to_jsonld(self, result):
+        '''Add the context to search result
+
+        Also remove hard to handle nested attributes
+          e.g. remove object.term when we have no id
+        '''
+        graph = result['@graph']
+        for i, obj in enumerate(graph):
+            # suppress nested attributes
+            graph[i] = {k: v for k, v in obj.items() if '.' not in k}
+
+        self.add_jsonld_context(result, self.prepare_url(result['@id']))
+        return result
+
     def validate(self, obj):
         obj_type = self.get_object_type(obj)
         schema_url = self.get_schema_url(obj)