Download Endpoint

api/urls.py

@@ -5,6 +5,7 @@
 urlpatterns = [
     path('<str:name>/_search/', views.index, name='index'),
     path('<str:name>/<str:id>', views.detail, name='detail'),
+    path('<str:name>/download/', views.download, name='download'),
     path('fire_api/trackhubregistry/<str:genome_id>/<str:folder>/<str:doc_id>',
          views.trackhubregistry_with_dirs_fire_api,
          name='trackhubregistry_with_dirs_fire_api'),

api/views.py

@@ -1,5 +1,6 @@
 import requests
 import json
+import csv
 
 from django.http import JsonResponse, HttpResponse
 from elasticsearch import Elasticsearch
@@ -16,6 +17,7 @@
                    'protocol_analysis', 'analysis', 'summary_organism',
                    'summary_specimen', 'summary_dataset', 'summary_file']
 
+ALLOWED_DOWNLOADS = ['file', 'organism', 'specimen', 'dataset']
 
 @csrf_exempt
 def index(request, name):
@@ -92,6 +94,102 @@ def index(request, name):
 
     return JsonResponse(data)
 
+@csrf_exempt
+def download(request, name):
+    if request.method != 'GET':
+        return HttpResponse("This method is not allowed!\n")
+    if name not in ALLOWED_DOWNLOADS:
+        return HttpResponse("This download doesn't exist!\n")
+
+    # Request params
+    SIZE = 1000000
+    file_format = request.GET.get('file_format', '')
+    field = request.GET.get('_source', '')
+    column_names = request.GET.get('columns', '[]')
+    sort = request.GET.get('sort', '')
+    filters = request.GET.get('filters', '{}')
+
+    columns = field.split(',')
+    request_fields = []
+    for col in columns:
+        cols = col.split('.')
+        if cols[0] == '_source':
+            request_fields.append('.'.join(cols[1:]))
+    request_fields = ','.join(request_fields)
+    column_names = json.loads(column_names)
+
+    # generate query for filtering
+    filter_values = []
+    not_filter_values = []
+    filters = json.loads(filters)
+    for key in filters.keys():
+        if filters[key][0] != 'false':
+            filter_values.append({"terms": {key: filters[key]}})
+        else:
+            not_filter_values.append({"terms": {key: ["true"]}})
+    filter_val = {}
+    if filter_values:
+        filter_val['must'] = filter_values
+    if not_filter_values:
+        filter_val['must_not'] = not_filter_values
+    if filter_val:
+        filters = {"query": {"bool": filter_val}}
+
+    # Get records from elasticsearch
+    es = Elasticsearch([settings.NODE1, settings.NODE2])
+    data = es.search(index=name, _source=request_fields, sort=sort, 
+                     body=filters, size=SIZE)
+    records = data['hits']['hits']
+
+    # generate response payload
+    if file_format == 'csv':
+        filename = 'faang_data.csv'
+        content_type = 'text/csv'
+    else:
+        filename = 'faang_data.txt'
+        content_type = 'text/plain'
+    response = HttpResponse(content_type=content_type)
+    response['Content-Disposition'] = 'attachment; filename=' + filename
+
+    # generate csv data
+    writer = csv.DictWriter(response, fieldnames=columns)
+    headers = {}
+    i = 0
+    for col in columns:
+        headers[col] = column_names[i]
+        i += 1
+    writer.writerow(headers)
+    for row in records:
+        record = {}
+        for col in columns:
+            cols = col.split('.')
+            record[col] = ''
+            source = row
+            for c in cols:
+                if  isinstance(source, dict) and c in source.keys():
+                    record[col] = source[c]
+                    source = source[c]
+                else:
+                    record[col] = ''
+                    break
+        writer.writerow(record)
+
+    # return formatted data
+    if file_format == 'csv':
+        return response
+    else:
+        # add padding to align with max length data in a column
+        def space(i, d):
+            max_len = len(max(list(zip(*row_data))[i], key=len))
+            return d+b' '*(max_len-len(d))
+
+        # create fixed width and '|' separated tabular text file
+        data = response.content
+        row_data = [i.rstrip(b'\r\n').split(b',') for i in filter(None, data.split(b'\n'))]
+        align = [b' | '.join(space(*c) for c in enumerate(b)) for b in row_data]
+        tab_data = b'\n'.join(align)
+        response.content = tab_data
+        return response
 
 @csrf_exempt
 def detail(request, name, id):