From 61347769a2d2f2296e605518b053dea3189ea24e Mon Sep 17 00:00:00 2001
From: Hugh Greenberg <hng@lanl.gov>
Date: Thu, 1 Aug 2024 10:06:27 -0600
Subject: [PATCH] Added function and a test for importing a json file

---
 dsi/backends/sqlite.py            | 51 ++++++++++++++++++++++++++++++-
 dsi/backends/tests/test_sqlite.py | 13 ++++++++
 2 files changed, 63 insertions(+), 1 deletion(-)

diff --git a/dsi/backends/sqlite.py b/dsi/backends/sqlite.py
index 7e5d86a9..0cae3444 100644
--- a/dsi/backends/sqlite.py
+++ b/dsi/backends/sqlite.py
@@ -1,5 +1,7 @@
 import csv
 import sqlite3
+import json
+import re
 
 from dsi.backends.filesystem import Filesystem
 
@@ -9,6 +11,7 @@
 STRING = "VARCHAR"
 FLOAT = "FLOAT"
 INT = "INT"
+JSON = "TEXT"
 
 # Holds table name and data properties
 
@@ -83,7 +86,6 @@ def put_artifact_type(self, types, isVerbose=False):
         if isVerbose:
             print(str_query)
 
-        print(str_query)
         self.cur.execute(str_query)
         self.con.commit()
 
@@ -219,6 +221,53 @@ def put_artifacts_lgcy(self,artifacts, isVerbose=False):
         self.cur.execute(str_query)
         self.con.commit()
 
+    def put_artifacts_json(self, fname, tname, isVerbose=False):
+        """
+        Function for insertion of Artifact metadata into a defined schema by using a JSON file
+        `fname`: filepath to the .json file to be read and inserted into the database
+
+        `tname`: String name of the table to be inserted
+
+        `return`: none
+        """
+
+        json_str = None
+        try:
+            j = open(fname)
+            data = json.load(j)
+            json_str = json.dumps(data)
+            json_str = "'" + json_str + "'"
+            j.close()
+        except IOError as i:
+            print(i)
+            return
+        except ValueError as v:
+            print(v)
+            return
+
+        types = DataType()
+        types.properties = {}
+        types.name = tname
+        
+        # Check if this has been defined from helper function
+        if self.types != None:
+            types.name = self.types.name
+
+        col_name = re.sub(r'.json', '', fname)
+        col_name = re.sub(r'.*/', '', col_name)
+        col_name = "'" + col_name + "'"
+        types.properties[col_name] = JSON
+           
+        self.put_artifact_type(types)
+        col_names = ', '.join(types.properties.keys())
+        str_query = "INSERT INTO {} ({}) VALUES ({});".format(str(types.name), col_names, json_str)
+        if isVerbose:
+            print(str_query)
+
+        self.types = types
+        self.cur.execute(str_query)
+        self.con.commit()
+
     # Adds columns and rows automaticallly based on a csv file
     #[NOTE 3] This method should be deprecated in favor of put_artifacts.
     def put_artifacts_csv(self, fname, tname, isVerbose=False):
diff --git a/dsi/backends/tests/test_sqlite.py b/dsi/backends/tests/test_sqlite.py
index dbe02553..54ecc0dd 100644
--- a/dsi/backends/tests/test_sqlite.py
+++ b/dsi/backends/tests/test_sqlite.py
@@ -46,6 +46,16 @@ def test_wildfiredata_artifact_put_t():
    # No error implies success
    assert True
 
+#Data from: https://microsoftedge.github.io/Demos/json-dummy-data/64KB.json
+def test_jsondata_artifact_put():
+   jsonpath = '/'.join([get_git_root('.'), 'dsi/data/64KB.json'])
+   dbpath = "jsondata.db"
+   store = Sqlite(dbpath)
+   store.put_artifacts_json(jsonpath, tname="JSONData")
+   store.close()
+   # No error implies success
+   assert True
+
 def test_yosemite_data_csv_artifact():
     csvpath = '/'.join([get_git_root('.'), 'dsi/data/yosemite5.csv'])
     dbpath = "yosemite.db"
@@ -68,3 +78,6 @@ def test_artifact_query():
     store.close()
     # No error implies success
     assert True
+
+
+test_jsondata_artifact_put()