teradataml 20.0.0.6__py3-none-any.whl → 20.0.0.7__py3-none-any.whl

teradataml/data/jsons/sqle/20.00/TD_API_VertexAI.json

@@ -0,0 +1,183 @@
+{
+  "json_schema_major_version": "1",
+  "json_schema_minor_version": "1",
+  "json_content_version": "1",
+  "function_name": "TD_API_VertexAI",
+  "function_version": "1.0",
+  "function_type": "fastpath",
+  "function_category": "Text Analytics AI",
+  "function_alias_name": "TDAPIVertexAI",
+  "function_r_name": "aa.td_apivertexai",
+  "supports_view": false,
+  "short_description": "This function sends an API request to a machine learning model deployed on Vertex AI, using input data from a Teradata table. The model processes the data and returns predictions, which are then provided as the output of the function for users to consume.",
+  "long_description": "This function sends an API request to a machine learning model deployed on Vertex AI, using input data from a Teradata table. The model processes the data and returns predictions, which are then provided as the output of the function for users to consume.",
+  "input_tables": [
+    {
+      "isOrdered": false,
+      "partitionByOne": false,
+      "name": "InputTable",
+      "alternateNames": [],
+      "isRequired": true,
+      "rDescription": "Specifies the table containing the input data.",
+      "description": "Specifies the table containing the input data.",
+      "datatype": "TABLE_ALIAS",
+      "allowsLists": false,
+      "rName": "data",
+      "useInR": true,
+      "rOrderNum": 1
+    }
+  ],
+  "argument_clauses": [
+      {
+        "name": "Endpoint",
+        "alternateNames": [],
+        "isRequired": false,
+        "rOrderNum": 2,
+        "allowNaN": false,
+        "description": "Specifies the Endpoint of the model deployed on Vertex AI.",
+        "rDescription": "Specifies the Endpoint of the model deployed on Vertex AI.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "allowPadding": true,
+        "rName": "endpoint",
+        "useInR": true,
+        "rFormulaUsage": false
+      },
+      {
+        "name": "Region",
+        "alternateNames": [],
+        "isRequired": false,
+        "rOrderNum": 3,
+        "allowNaN": false,
+        "description": "Specifies the name of the Google Cloud service region.",
+        "rDescription": "Specifies the name of the Google Cloud service region.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "allowPadding": true,
+        "rName": "region",
+        "useInR": true,
+        "rFormulaUsage": false
+      },
+      {
+        "name": "AccessToken",
+        "alternateNames": [],
+        "isRequired": false,
+        "rOrderNum": 4,
+        "allowNaN": false,
+        "description": "Specifies the session token for google cloud.",
+        "rDescription": "Specifies the session token for google cloud.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "allowPadding": true,
+        "rName": "access.token",
+        "useInR": true,
+        "rFormulaUsage": false
+      },
+      {
+        "name": "Project",
+        "alternateNames": [],
+        "isRequired": false,
+        "rOrderNum": 5,
+        "allowNaN": false,
+        "description": "Specify the name of the gcp project.",
+        "rDescription": "Specify the name of the gcp project.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "allowPadding": true,
+        "rName": "project",
+        "useInR": true,
+        "rFormulaUsage": false
+      },
+      {
+        "name": "EnableSafety",
+        "alternateNames": [],
+        "isRequired": true,
+        "rOrderNum": 6,
+        "permittedValues": ["TRUE", "FALSE"],
+        "allowNaN": false,
+        "description": "Specifies if the safety feature needs to be enabled.",
+        "rDescription": "Specifies if the safety feature needs to be enabled.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "allowPadding": true,
+        "rName": "enable.safety",
+        "useInR": true,
+        "rFormulaUsage": false
+      },
+      {
+          "targetTable": [
+            "InputTable"
+          ],
+          "checkDuplicate": true,
+          "allowedTypes": [],
+          "allowedTypeGroups": [
+            "ALL"
+          ],
+          "matchLengthOfArgument": "",
+          "allowPadding": false,
+          "name": "InputColumns",
+          "alternateNames": [],
+          "isRequired": true,
+          "rDescription": "Specifies the input table columns to be sent to the model hosted on vertex for inference.",
+          "description": "Specifies the input table columns to be sent to the model hosted on vertex for inference.",
+          "datatype": "COLUMNS",
+          "allowsLists": true,
+          "rName": "input.columns",
+          "useInR": true,
+          "rOrderNum": 7
+      },
+      {
+        "permittedValues": ["TRUE", "FALSE"],
+        "isOutputColumn": false,
+        "name": "isDebug",
+        "alternateNames": [],
+        "isRequired": false,
+        "rDescription": "Specify whether error logging is required.",
+        "description": "Specify whether error logging is required.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "rName": "is.debug",
+        "useInR": true,
+        "rOrderNum": 8
+      },
+      {
+          "targetTable": [
+            "InputTable"
+          ],
+          "checkDuplicate": true,
+          "allowedTypes": [],
+          "allowedTypeGroups": [
+            "ALL"
+          ],
+          "matchLengthOfArgument": "",
+          "allowPadding": false,
+          "name": "Accumulate",
+          "alternateNames": [],
+          "isRequired": false,
+          "rDescription": "Specifies the input table columns to copy to the output table. By default, the function copy all input table columns to the output table.",
+          "description": "Specifies the input table columns to copy to the output table. By default, the function copy all any input table columns to the output table.",
+          "datatype": "COLUMNS",
+          "allowsLists": true,
+          "rName": "accumulate",
+          "useInR": true,
+          "rOrderNum": 9
+      },
+      {
+        "name": "AUTHORIZATION",
+        "alternateNames": [],
+        "isRequired": false,
+        "rOrderNum":10,
+        "allowNaN": false,
+        "description": "Specifies the authorization object name containing the credentials, optionally prefixed by the database name.",
+        "rDescription": "Specifies the authorization object name containing the credentials, optionally prefixed by the database name.",
+        "datatype": "STRING",
+        "allowsLists": false,
+        "allowPadding": true,
+        "rName": "authorization",
+        "useInR": true,
+        "rFormulaUsage": false
+      }
+  ]
+}
+
+    

teradataml/data/load_example_data.py

@@ -13,6 +13,7 @@ import csv
 import json
 import os
 import datetime
+from teradataml.common.constants import TeradataReservedKeywords
 from teradataml.common.exceptions import TeradataMlException
 from teradataml.common.messages import Messages
 from teradataml.common.messagecodes import MessageCodes
@@ -189,16 +190,18 @@ def __create_table_insert_data(tablename):
     date_time_varbyte = {}
     pti_table = False
     pti_clause = ""
+    td_number_of_columns= ''
+    column_names = ''
     
     '''
     Create column datatype string required to create a table.
     EXAMPLE:
         id integer,model varchar(30)
     '''
-    column_count = 0
     for column in col_types_dict.keys():
         if column in ["TD_TIMECODE", "TD_SEQNO"]:
-            column_count = column_count + 1
+            td_number_of_columns += '?,'
+            column_names += f"{column},"
             continue
 
         if column == "<PTI_CLAUSE>":
@@ -212,11 +215,22 @@ def __create_table_insert_data(tablename):
         for column_type in ["date", "timestamp", "varbyte"]:
             if column_type in col_types_dict[column]:
                 date_time_varbyte.setdefault(column_type, []).append(column)
+        quoted_column = f'"{column}"' if column.upper()  in TeradataReservedKeywords.TERADATA_RESERVED_WORDS.value  else column
+        column_dtypes ="{0}{1} {2},\n" .format(column_dtypes, quoted_column, col_types_dict[column])
+        if "PERIOD" in col_types_dict[column].upper() and tablename in ["Employee", "Employee_roles", "Employee_Address"]:
+            # Extract the type passed in PERIOD, e.g., PERIOD(DATE), PERIOD(TIMESTAMP)
+            if "AS TRANSACTIONTIME" in col_types_dict[column].upper():
+                continue
+            period_type = col_types_dict[column].upper().split("PERIOD(")[1].rsplit(")", 1)[0]
+            td_number_of_columns += f'PERIOD(CAST(? AS {period_type}),CAST(? AS {period_type})),'
+        else:
+            td_number_of_columns += '?,'
 
-        column_dtypes ="{0}{1} {2},\n" .format(column_dtypes, column, col_types_dict[column])
-        column_count = column_count + 1
-
-    td_number_of_columns = '?,' * column_count
+         # Dynamically build column_names
+        if column != "<PTI_CLAUSE>" and "AS TRANSACTIONTIME" not in col_types_dict[column].upper():
+            column_names += f"{quoted_column},"
+    
+    column_names = column_names.rstrip(',')
     # Deriving global connection using context.get_context()
     con = get_connection()
     # Get temporary database.
@@ -237,14 +251,14 @@ def __create_table_insert_data(tablename):
             UtilFuncs._create_table_using_columns(tablename, column_dtypes[:-2])
 
         try:
-            __insert_into_table_from_csv(tablename, td_number_of_columns[:-1], csv_file, date_time_varbyte)
+            __insert_into_table_from_csv(tablename, td_number_of_columns[:-1], csv_file, date_time_varbyte, column_names)
         except:
             # Drop the table, as we have created the same.
             UtilFuncs._drop_table(tablename)
             raise
 
 
-def __insert_into_table_from_csv(tablename, column_markers, file, date_time_varbyte_columns):
+def __insert_into_table_from_csv(tablename, column_markers, file, date_time_varbyte_columns, column_names):
         """
         Builds and executes a prepared statement with parameter markers for a table. 
         
@@ -268,11 +282,16 @@ def __insert_into_table_from_csv(tablename, column_markers, file, date_time_varb
                 Required Argument.
                 Dictionary containing date, time and varbyte columns.
                 Types: Dictionary
+
+            column_names
+                Required Argument.
+                Comma separated string of column names to be inserted into table.
+                Types: str
             
         EXAMPLES:
             date_time_varbyte_columns = {'date':['orderdate']}
             preparedstmt = __insert_into_table_from_csv(
-                            'mytab', '?, ?','file.csv', date_time_varbyte_columns )
+                            'mytab', '?, ?','file.csv', date_time_varbyte_columns, column_names)
 
         RETURNS:
              None
@@ -281,7 +300,7 @@ def __insert_into_table_from_csv(tablename, column_markers, file, date_time_varb
             Database error if an error occurred while executing the DDL statement.
         
         """
-        insert_stmt = SQLBundle._build_insert_into_table_records(tablename, column_markers)
+        insert_stmt = SQLBundle._build_insert_into_table_records(tablename, column_markers, column_names)
 
         # Defining the formatter.
         formatter = {
@@ -322,7 +341,6 @@ def __insert_into_table_from_csv(tablename, column_markers, file, date_time_varb
                                         pass
 
                         insert_list.append(tuple(new_row))
-
                     # Batch Insertion (using DBAPI's executeMany) used here to insert list of dictionaries
                     execute_sql(insert_stmt, insert_list)