feat(list-ids): support list with id params (#127)

* feat(list-ids): support list with id params

* feat(list): do not paginate for ids

indexd/index/blueprint.py

@@ -50,27 +50,37 @@ def get_index():
     Returns a list of records.
     '''
     limit = flask.request.args.get('limit')
-    try: limit = 100 if limit is None else int(limit)
+    start = flask.request.args.get('start')
+
+    ids = flask.request.args.get('ids')
+    if ids:
+        ids = ids.split(',')
+        if start is not None or limit is not None:
+            raise UserError(
+                'pagination is not supported when ids is provided')
+    try:
+        limit = 100 if limit is None else int(limit)
     except ValueError as err:
         raise UserError('limit must be an integer')
 
     if limit <= 0 or limit > 1024:
         raise UserError('limit must be between 1 and 1024')
 
     size = flask.request.args.get('size')
-    try: size = size if size is None else int(size)
+    try:
+        size = size if size is None else int(size)
     except ValueError as err:
         raise UserError('size must be an integer')
 
     if size is not None and size < 0:
         raise UserError('size must be > 0')
 
-    start = flask.request.args.get('start')
 
     urls = flask.request.args.getlist('url')
 
     acl = flask.request.args.getlist('acl')
 
+
     file_name = flask.request.args.get('file_name')
 
     version = flask.request.args.get('version')
@@ -87,7 +97,7 @@ def get_index():
     if limit < 0 or limit > 1024:
         raise UserError('limit must be between 0 and 1024')
 
-    ids = blueprint.index_driver.ids(
+    records = blueprint.index_driver.ids(
         start=start,
         limit=limit,
         size=size,
@@ -96,11 +106,13 @@ def get_index():
         urls=urls,
         acl=acl,
         hashes=hashes,
+        ids=ids,
         metadata=metadata,
     )
 
     base = {
         'ids': ids,
+        'records': records,
         'limit': limit,
         'start': start,
         'size': size,

indexd/index/driver.py

@@ -11,19 +11,19 @@ class IndexDriverABC(SQLAlchemyDriverBase):
     def __init__(self, conn, **config):
         super(IndexDriverABC, self).__init__(conn, **config)
 
-
     __metaclass__ = abc.ABCMeta
 
     @abc.abstractmethod
     def ids(self,
-              limit=100,
-              start=None,
-              size=None,
-              urls=None,
-              hashes=None,
-              file_name=None,
-              version=None,
-              metadata=None):
+            limit=100,
+            start=None,
+            size=None,
+            urls=None,
+            hashes=None,
+            file_name=None,
+            version=None,
+            metadata=None,
+            ids=None):
         '''
         Returns a list of records stored by the backend.
         '''

indexd/index/drivers/alchemy.py

@@ -297,7 +297,8 @@ def ids(self,
             hashes=None,
             file_name=None,
             version=None,
-            metadata=None):
+            metadata=None,
+            ids=None):
         """
         Returns list of records stored by the backend.
         """
@@ -346,9 +347,14 @@ def ids(self,
                     query = query.filter(IndexRecord.did.in_(sub.subquery()))
 
             query = query.order_by(IndexRecord.did)
-            query = query.limit(limit)
 
-            return [i.did for i in query]
+            if ids:
+                query = query.filter(IndexRecord.did.in_(ids))
+            else:
+                # only apply limit when ids is not provided
+                query = query.limit(limit)
+
+            return [i.to_document_dict() for i in query]
 
     def hashes_to_urls(self, size, hashes, start=0, limit=100):
         """

openapis/swagger.yaml

@@ -97,9 +97,41 @@ paths:
       parameters:
         - name: metadata
           in: query
-          description: metadata associated with the entry
+          description: metadata in format key:value
           required: false
           type: string
+        - name: size
+          in: query
+          description: object size
+          required: false
+          type: integer
+        - name: hash
+          in: query
+          description: hash in format hash_type:hash_value
+          required: false
+          type: string
+        - name: ids
+          in: query
+          description: |
+            comma delimited ids, when provided,
+            pagination params(start, limit) are not supported
+          required: false
+          type: string
+        - name: urls
+          in: query
+          description: comma delimited urls
+          required: false
+          type: string
+        - name: start
+          in: query
+          description: start did
+          required: false
+          type: string
+        - name: limit
+          in: query
+          description: number of records to return for this page, default to 100
+          required: false
+          type: integer
       produces:
         - application/json
       responses:
@@ -112,6 +144,10 @@ paths:
                 type: array
                 items:
                   $ref: '#/definitions/DID'
+              records:
+                type: array
+                items:
+                  $ref: "#/definitions/OutputInfo"
               size:
                 type: integer
                 format: int64

tests/test_client.py

@@ -29,32 +29,37 @@ def get_doc(
 
 def test_index_list(swg_index_client):
     r = swg_index_client.list_entries()
-    assert r.ids == []
+    assert r.records == []
 
 
-def test_index_list_with_params(client, user):
+def test_index_list_with_params(swg_index_client):
     data = get_doc()
 
-    r_1 = client.post(
-        '/index/',
-        data=json.dumps(data),
-        headers=user)
+    r_1 = swg_index_client.add_entry(data)
     data['metadata'] = {'project_id': 'other-project'}
 
-    r_2 = client.post(
-        '/index/',
-        data=json.dumps(data),
-        headers=user)
+    r_2 = swg_index_client.add_entry(data)
 
-    r = client.get('/index/?metadata=project_id%3Abpa-UChicago')
-    assert r_1.json['did'] in r.json['ids']
+    r = swg_index_client.list_entries(metadata='project_id:bpa-UChicago')
+    ids = [record.did for record in r.records]
+    assert r_1.did in ids
 
-    r = client.get('/index/?metadata=project_id%3Aother-project')
-    assert r_2.json['did'] in r.json['ids']
+    r = swg_index_client.list_entries(metadata='project_id:other-project')
+    ids = [record.did for record in r.records]
+    assert r_2.did in ids
 
-    r = client.get('/index/?hashes=md5%3A8b9942cf415384b27cadf1f4d2d682e5')
-    assert r_1.json['did'] in r.json['ids']
-    assert r_2.json['did'] in r.json['ids']
+    r = swg_index_client.list_entries(
+        hash='md5:8b9942cf415384b27cadf1f4d2d682e5')
+    ids = [record.did for record in r.records]
+    assert r_1.did in ids
+    assert r_2.did in ids
+
+    r = swg_index_client.list_entries(
+        ids=','.join(ids))
+
+    ids = [record.did for record in r.records]
+    assert r_1.did in ids
+    assert r_2.did in ids
 
 
 def test_urls_metadata(swg_index_client):