added string as a a data type for check categorical columns for, and …

…better outputs in the CLI

defog/cli.py

@@ -209,7 +209,6 @@ def init():
         print(
             "Do you want to automatically scan these tables to determine which column might be categorical? The distinct values in each categorical column will be sent to our server. (y/n)"
         )
-
         scan_option = prompt().strip()
         if scan_option.lower() == "y" or scan_option.lower() == "yes":
             scan = True
@@ -280,18 +279,29 @@ def gen():
         print(
             "defog gen requires a list of tables to generate. Please enter the names of the tables whose schema you would like to generate, separated by a space:"
         )
-        print(
-            "If you would like to index all of your tables, just leave this blank and hit enter (Supported for postgres + redshift only)."
-        )
         table_names = prompt().strip()
         table_name_list = re.split(r"\s+", table_names.strip())
     else:
         table_name_list = sys.argv[2:]
-    filename = df.generate_db_schema(table_name_list)
+
+    if table_name_list == [""] or table_name_list == []:
+        print("No tables were registered. Exiting.")
+        sys.exit(0)
+    print(
+        "Do you want to automatically scan these tables to determine which column might be categorical? The distinct values in each column likely to be a categorical column will be sent to our server. (y/n)"
+    )
+    scan_option = prompt().strip()
+    if scan_option.lower() == "y" or scan_option.lower() == "yes":
+        scan = True
+    else:
+        scan = False
+
+    filename = df.generate_db_schema(table_name_list, scan=scan)
+    pwd = os.getcwd()
     print(
         "Your schema has been generated and is available at the following CSV file in this folder:\n"
     )
-    print(f"\033[1m{filename}\033[0m\n")
+    print(f"\033[1m{pwd}/{filename}\033[0m\n")
 
     print("We are now uploading this auto-generated schema to Defog.")
     df.update_db_schema_csv(filename)

defog/util.py

@@ -73,14 +73,15 @@ def identify_categorical_columns(
     # if it is a categorical variable, then we want to get the distinct values and their counts
     # we will then send this to the defog servers so that we can generate a column description
     # for each categorical variable
-    print(f"Identifying categorical columns in {table_name}...")
+    print(f"Identifying categorical columns in {table_name}. This might take a while if you have many rows in your table.")
     for idx, row in enumerate(rows):
         if row["data_type"].lower() in [
             "character varying",
             "text",
             "character",
             "varchar",
             "char",
+            "string"
         ]:
             # get the total number of rows and number of distinct values in the table for this column
             cur.execute(
@@ -96,6 +97,7 @@ def identify_categorical_columns(
                 top_values = cur.fetchall()
                 top_values = [i[0] for i in top_values if i[0] is not None]
                 rows[idx]["top_values"] = top_values
+            print(f"Identified {row['column_name']} as a likely categorical column.")
     return rows
 
 

setup.py

@@ -25,7 +25,7 @@ def package_files(directory):
     name="defog",
     packages=find_packages(),
     package_data={"defog": ["gcp/*", "aws/*"] + next_static_files},
-    version="0.56.2",
+    version="0.56.3",
     description="Defog is a Python library that helps you generate data queries from natural language questions.",
     author="Full Stack Data Pte. Ltd.",
     license="MIT",