Merge pull request #346 from yskale/end-point-program-name-id

End point program name

Dockerfile

@@ -11,7 +11,9 @@ RUN apk update && \
     apk add g++ make
 
 #upgrade openssl \
-#RUN apk  add openssl=3.1.4-r4
+
+RUN apk  add openssl=3.1.4-r5
+
 
 RUN pip install --upgrade pip
 # Create a non-root user.

docker-compose.yaml

@@ -40,11 +40,9 @@ services:
       REDIS_PASSWORD: "$REDIS_PASSWORD"
       FLASK_ENV: "development"
       PYTHONUNBUFFERED: "TRUE"
-    entrypoint: [ "gunicorn",
-                     "--workers=$API_WORKERS", "--name=dug",
-                     "--bind=0.0.0.0:$API_PORT", "--timeout=$API_TIMEOUT",
-                     "--log-level=DEBUG", "--enable-stdio-inheritance",
-                     "-k", "uvicorn.workers.UvicornWorker", "--reload", "dug.server:APP" ]
+    entrypoint: [ "uvicorn",
+                     "--host", "0.0.0.0" , "--port" , "$API_PORT",
+                     "--log-level=debug",  "--reload-dir", "/home/dug/dug/",  "--reload", "dug.server:APP" ]
     volumes:
       - ./src:/home/dug/dug/
     ports:

src/dug/core/async_search.py

@@ -690,3 +690,93 @@ async def search_kg(self, unique_id, query, offset=0, size=None,
         )
         search_results.update({'total_items': total_items['count']})
         return search_results
+
+    async def search_study(self, study_id=None, study_name=None, offset=0, size=None):
+        """
+        Search for studies by unique_id (ID or name) and/or study_name.
+        """
+        # Define the base query
+         # Define the base query
+        query_body = {
+            "bool": {
+                "must": []
+            }
+        }
+
+        # Add conditions based on user input
+        if study_id:
+            query_body["bool"]["must"].append({
+                "match": {"collection_id": study_id}
+            })
+
+        if study_name:
+            query_body["bool"]["must"].append({
+                "match": {"collection_name": study_name}
+            })
+
+        print("query_body",query_body)
+        body = {'query': query_body}
+        total_items = await self.es.count(body=body, index="variables_index")
+        search_results = await self.es.search(
+            index="variables_index",
+            body=body,
+            filter_path=['hits.hits._id', 'hits.hits._type', 'hits.hits._source'],
+            from_=offset,
+            size=size
+        )
+        search_results.update({'total_items': total_items['count']})
+        return search_results
+
+
+    async def search_program(self, program_name=None, offset=0, size=None):
+        """
+        Search for studies by unique_id (ID or name) and/or study_name.
+        """
+
+        query_body = {
+            "query": {
+                "bool": {
+                    "must": []
+                }
+            },
+            "aggs": {
+                "unique_collection_ids": {
+                    "terms": {
+                        "field": "collection_id.keyword"
+                    }
+                }
+            }
+        }
+
+        # specify the fields to be returned
+        query_body["_source"] = ["collection_id", "collection_name", "collection_action"]
+
+        # search for program_name based on uses input
+        if program_name:
+            query_body["query"]["bool"]["must"].append({
+                "match": {"data_type": program_name}
+            })
+
+        print("query_body", query_body)
+
+        # Prepare the query body for execution
+        body = query_body
+        #print(body)
+
+        # Execute the search query
+        search_results = await self.es.search(
+            index="variables_index",
+            body=body,
+            filter_path=['hits.hits._id', 'hits.hits._type', 'hits.hits._source', 'aggregations.unique_collection_ids.buckets'],
+            from_=offset,
+            size=size
+        )
+
+        # The unique collection_ids will be in the 'aggregations' field of the response
+        unique_collection_ids = search_results['aggregations']['unique_collection_ids']['buckets']
+
+        #print("Unique collection_ids:", unique_collection_ids)
+
+
+        #print(search_results)
+        return search_results

src/dug/server.py

@@ -8,12 +8,13 @@
 from dug.core.async_search import Search
 from pydantic import BaseModel
 import asyncio
+from typing import Optional
 
 logger = logging.getLogger (__name__)
 
 APP = FastAPI(
     title="Dug Search API",
-    root_path=os.environ.get("ROOT_PATH", "/"),
+    # root_path=os.environ.get("ROOT_PATH", "/"),
 )
 
 APP.add_middleware(
@@ -49,6 +50,19 @@ class SearchKgQuery(BaseModel):
     index: str = "kg_index"
     size:int = 100
 
+class SearchStudyQuery(BaseModel):
+    #query: str
+    study_id: Optional[str] = None
+    study_name: Optional[str] = None
+    #index: str = "variables_index"
+    size:int = 100
+class SearchProgramQuery(BaseModel):
+    #query: str
+    program_id: Optional[str] = None
+    program_name: Optional[str] = None
+    #index: str = "variables_index"
+    size:int = 100   
+
 search = Search(Config.from_env())
 
 @APP.on_event("shutdown")
@@ -107,5 +121,36 @@ async def search_var(search_query: SearchVariablesQuery):
     }
 
 
+
+@APP.get('/search_study')
+async def search_study(study_id: Optional[str] = None, study_name: Optional[str] = None):
+    """
+    Search for studies by unique_id (ID or name) and/or study_name.
+    """
+    result = await search.search_study(study_id=study_id, study_name=study_name)
+    return {
+        "message": "Search result",
+        # Although index in provided by the query we will keep it around for backward compatibility, but
+        # search concepts should always search against "variables_index"
+        "result": result,
+        "status": "success"
+    }
+
+
+@APP.get('/search_program')
+async def search_program( program_name: Optional[str] = None):
+    """
+    Search for studies by unique_id (ID or name) and/or study_name.
+    """
+    result = await search.search_program(program_name=program_name)
+    return {
+        "message": "Search result",
+        # Although index in provided by the query we will keep it around for backward compatibility, but
+        # search concepts should always search against "variables_index"
+        "result": result,
+        "status": "success"
+    }
+
+
 if __name__ == '__main__':
     uvicorn.run(APP)