Rework genenetwork linkingrework-genotype-linking

2 files changed, 26 insertions, 11 deletions

diff --git a/gn3/auth/authorisation/data/views.py b/gn3/auth/authorisation/data/views.py
index d84b080..6431389 100644
--- a/gn3/auth/authorisation/data/views.py
+++ b/gn3/auth/authorisation/data/views.py
@@ -175,7 +175,7 @@ def __search_genotypes__():
             limit=limit, offset=offset)
         return jsonify(with_db_connection(__ungrouped__))
 
-def __search_phenotypes__():
+def __search_xapian__():
     # launch the external process to search for phenotypes
     redisuri = app.config["REDIS_URI"]
     with redis.Redis.from_url(redisuri, decode_responses=True) as redisconn:
@@ -184,6 +184,7 @@ def __search_phenotypes__():
         command =[
             sys.executable, "-m", "scripts.search_phenotypes",
             __request_key__("species_name"),
+            __request_key__("dataset_type"),
             __request_key__("query"),
             str(job_id),
             f"--host={__request_key__('gn3_server_uri')}",
@@ -193,6 +194,10 @@ def __search_phenotypes__():
             f"--per-page={__request_key__('per_page')}"] +(
                 [f"--selected={json.dumps(selected)}"]
                 if len(selected) > 0 else [])
+        print("+++++++++++++++++++++++++++++++++++++++++++++++++++++++++")
+        import shlex
+        print(shlex.join(command))
+        print("+++++++++++++++++++++++++++++++++++++++++++++++++++++++++")
         jobs.create_job(redisconn, {
             "job_id": job_id, "command": command, "status": "queued",
             "search_results": tuple()})
@@ -203,20 +208,21 @@ def __search_phenotypes__():
             "command": command
         })
 
-@data.route("/search", methods=["GET"])
+@data.route("/search", methods=["POST"])
 @require_oauth("profile group resource")
+@require_json
 def search_unlinked_data():
     """Search for various unlinked data."""
     dataset_type = request.json["dataset_type"]
     search_fns = {
         "mrna": __search_mrna__,
-        "genotype": __search_genotypes__,
-        "phenotype": __search_phenotypes__
+        "genotype": __search_xapian__,
+        "phenotype": __search_xapian__
     }
     return search_fns[dataset_type]()
 
-@data.route("/search/phenotype/<uuid:job_id>", methods=["GET"])
-def pheno_search_results(job_id: uuid.UUID) -> Response:
+@data.route("/search/results/<uuid:job_id>", methods=["GET"])
+def search_job_results(job_id: uuid.UUID) -> Response:
     """Get the search results from the external script"""
     def __search_error__(err):
         raise NotFoundError(err["error_description"])
diff --git a/scripts/search_phenotypes.py b/scripts/search_phenotypes.py
index 38b992b..1e8d0fc 100644
--- a/scripts/search_phenotypes.py
+++ b/scripts/search_phenotypes.py
@@ -22,10 +22,16 @@ class NoSearchResults(Exception):
     """Raise when there are no results for a search."""
 
 def do_search(
-        host: str, query: str, per_page: int, page: int = 1) -> Iterable[dict[str, Any]]:
+        dataset_type: str, host: str, query: str, per_page: int,
+        page: int = 1) -> Iterable[dict[str, Any]]:
     """Do the search and return the results"""
+    search_types = {
+        "phenotype": "phenotype",
+        "genotype": "gene"
+    }
     search_uri = urljoin(host, (f"search/?page={page}&per_page={per_page}"
-                                f"&type=phenotype&query={query}"))
+                                f"&type={search_types[dataset_type]}"
+                                f"&query={query}"))
     response = requests.get(search_uri)
     results = response.json()
     if len(results) > 0:
@@ -61,6 +67,7 @@ def expire_redis_results(redisconn: redis.Redis, redisname: str):
 
 @click.command()
 @click.argument("species")
+@click.argument("dataset_type")
 @click.argument("query")
 @click.argument("job-id", type=click.UUID)
 @click.option(
@@ -76,8 +83,9 @@ def expire_redis_results(redisconn: redis.Redis, redisname: str):
     "--redis-uri", default="redis://:@localhost:6379/0",
     help="The URI to the redis server.")
 def search(# pylint: disable=[too-many-arguments, too-many-locals]
-        species: str, query: str, job_id: uuid.UUID, host: str, per_page: int,
-        selected: str, auth_db_uri: str, gn3_db_uri: str, redis_uri: str):
+        species: str, dataset_type: str, query: str, job_id: uuid.UUID,
+        host: str, per_page: int, selected: str, auth_db_uri: str,
+        gn3_db_uri: str, redis_uri: str):
     """
     Search for phenotype traits, filtering out any linked and selected traits,
     loading more and more pages until the `per_page` quota is fulfilled or the
@@ -104,7 +112,8 @@ def search(# pylint: disable=[too-many-arguments, too-many-locals]
             while count < per_page:
                 results = tuple(remove_linked(
                     remove_selected(
-                        do_search(host, search_query, per_page, page),
+                        do_search(
+                            dataset_type, host, search_query, per_page, page),
                         selected_traits),
                     linked))[0:per_page-count]
                 count = count + len(results)