Import Wizard - Teradata Warehouse Miner

Teradata Warehouse Miner User Guide - Volume 1Introduction and Profiling

Product
Teradata Warehouse Miner
Release Number
5.4.4
Published
July 2017
Language
English (United States)
Last Update
2018-05-03
dita:mapPath
wbc1492033894304.ditamap
dita:ditavalPath
ft:empty
dita:id
B035-2300
Product Category
Software

The Import Wizard allows the user to import the project or projects contained within a binary file created by the Export Wizard into the current Project Window. Alternately, if there is already a project in the Project Window that is currently selected, the analyses contained in the project or projects in the binary file can be imported directly into the currently selected project. In order to invoke the Import Wizard, the user must be connected to an ODBC data source. The metadata database associated with this data source (or the one currently specified in the Connection Properties dialog) is the metadata database in which the imported projects or analyses will be stored if the user chooses to save the current project or projects.

The Import Wizard contains facilities for mapping input and output database, table and view names, as well as input column names to other values, based entirely on matching one name to another name. Depending on the type of analysis, some references to these database entities will not be remapped in this process, and may need to be changed in individual analyses after importing. For example, changing the name of an input column will not change references to the original column in a primary index or join path, and changing a database name may not change a reference to that database in the specification of a User Defined Function call.

The Import Wizard is activated by selecting File > Import, bringing up an Open File dialog from which one or more import files (ending with .bin and created using the Export menu option) can be selected. After selecting the file or files to open, the following Import Wizard dialog is displayed. On this first dialog, the Import Wizard allows you to change the database names used in the imported analyses. It does this by determining the Available and Matched Databases as displayed below.
Import Wizard

  • Import Wizard — The header for this dialog indicates that it allows the user to Match Databases while giving a reminder of the name of the import file in this case, Import Demo.
  • Available Databases — This is a list of available databases associated with the ODBC connection. This includes the databases listed as Source Databases in the Connection Properties dialog available from the Tools menu, as well as the Result Database from the Connection Properties dialog.
  • Matched Databases — On the left side of the => symbol are all the database names acquired from the analyses within the imported binary file. On the right side of the symbol is the Available database that matches the imported database by name (not case sensitive). If a database is not matched, the “NOT MATCHED” phrase will show next to it and action must be taken to resolve the conflict.

    For example, in the screen above, the financial database does not match any databases listed in the Available Databases list, so NOT MATCHED is shown. The other databases, twm_source and twm_results, were matched and no further action is required.

    If a foreign database to be accessed via QueryGrid is present in this list (distinguished by the presence of a ‘@’ symbol in the name), the Next button leading to Table and Column matching is disabled.
  • Match button — Clicking the Match button will match the database selected on the left with the database selected on the right. The same result can be obtained by double-clicking on the desired database on the left to match with the selected database on the right.
  • New Name — The name of a matching database may be entered directly in this field and it will be matched with the currently selected database on the right. This removes the need to add a Source Database using the Connection Properties dialog just to allow its use as a matching database.
    A matched database can be unmatched by entering an empty name in this field with the Enter or Return key.
  • Use connection's Result Database for all output databases — Selecting this option overrides the value or values that output databases are mapped to and maps all output databases to the Result Database on the current Connection Properties dialog. This is particularly useful when a project that accesses and saves tables into the same database is imported into an environment where tables need to be created in a separate database.
    Tables used for both output and input will need to be adjusted individually after a project is imported in this manner.
  • Merge with selected project — By default, the project or projects in the import file are imported into a new project or projects. If there is a selected project in the Project Window, however, this option is enabled. Click it to import the analyses into the currently selected project.

    Even if the import file contains more than one project, all of the analyses contained in the projects in the import file are imported into the currently selected project in the Project Window if this option is selected.

  • Import button — Clicking the Import button results in all the projects and analyses in the binary import file being imported into the current Teradata Warehouse Miner Project Window.
    These projects and/or analyses are not saved in the current metadata database until the Save or Save All button is selected in the usual manner. Also, the Import button can be selected even though additional screens are available by selecting the Next button. A validation warning is given if there are any unmatched databases, tables or columns present, even if the screens to view all of these objects have not been viewed by selecting the Next button.
  • Cancel button — Clicking the Cancel button cancels the import operation without changing the current Project Window workspace.
  • Next button — Clicking the Next button takes the user to the screen next screen, to allow the user to map imported tables to tables with a different name. The resulting dialog is shown below, followed by explanations of the fields in the dialog.

    Note that if a foreign database to be accessed via QueryGrid is present in the Matched Databases list (distinguished by the presence of a ‘@’ symbol in the name), the Next button leading to Table and Column matching is disabled.

    Import Wizard: Map Imported Tables

  • Import Wizard — The header for this dialog indicates that it allows the user to “Match Tables” while giving a reminder of the name of the import file (in this case, Import Demo).
  • Matched Databases — These are the matched databases from the previous screen.
  • Available Tables — These are the tables and views contained in the currently selected matched database in the Available Databases pull-down list.
  • Matched Tables — On the left side of the symbol are all the table names acquired from the analyses within the imported binary file. These may include the names of tables that are accessed strictly for input, tables that are created as output, and tables that are both created as output and read in as input (output/input).
    For tables that are accessed strictly for input, the right side of the symbol identifies the Available table that matches the imported table by name (not case sensitive) and that resides in the database matched previously to the original database of the imported table. If an input table is not matched, NOT MATCHED appears next to it and action must be taken to resolve the conflict.
    It is not possible to match an imported table if the database it is imported from was not matched on the previous screen.

    To make table matching more easy, when an item under Matched Tables is selected on the right, the database matched previously to the original database of the imported table is automatically selected, if possible, under Available Databases on the left and the corresponding tables are loaded under Available Tables.

    For tables that are created as output, they are labeled either as an “Output Table” or an “Output/Input Table” and matched by default to a table with the same name (i.e., the name is not changed). For example, in the screen above, the table “output1” is labeled as an Output/Input table, and “output2” is labeled as an Output table, both with the same name. The names of output tables can be changed either by matching or entering a New Name in the usual manner.

  • Match button — Clicking the Match button matches the table selected on the left with the table selected on the right. The same result can be obtained by double-clicking on the desired table on the left to match with the selected table on the right. Note, however, that table matching can only take place if both the selected table on the left and the selected table on the right are contained in the same matched database.
  • New Name — The name of a matching table may be entered directly in this field and it will be matched with the currently selected table on the right, provided the currently selected database under Available Databases on the left is the same as the database previously matched to the original database of the imported table selected under Matched Tables on the right. Note that a matched table can be unmatched by entering an empty name in this field with the Enter or Return key.)
  • Display qualified names — This option displays matched tables qualified by the name of their containing database. It is checked by default only if the same table name occurs in more than one matched database.
  • Display input only — This option displays only input tables under Matched Tables, excluding Output and Output/Input tables.
  • Merge with selected project — By default, the project or projects in the import file are imported into a new project or projects. If there is a selected project in the Project Window, however, this option is enabled. Click it to import the analyses into the currently selected project.

    Even if the import file contains more than one project, all of the analyses contained in the projects in the import file are imported into the currently selected project in the Project Window if this option is selected.

  • Import button — Clicking the Import button results in all the projects and analyses in the binary import file being imported into the current Teradata Warehouse Miner Project Window. These projects and/or analyses are not saved in the current metadata database until the Save or Save All button is explicitly selected in the usual manner.
  • Cancel button — Clicking the Cancel button cancels the import operation without changing the current Project Window workspace.
  • Next button — Clicking the Next button takes the user to the screen next screen, which is designed to allow the user to map imported columns to columns with a different name. The resulting dialog is shown below, followed by explanations of the fields in the dialog.
    Import Wizard: Map Imported Columns

  • Import Wizard — The header for this dialog indicates that it allows the user to “Match Columns” while giving a reminder of the name of the import file (in this case, Import Demo).
  • Matched Tables — These are the matched tables from the previous screen, each displayed as qualified by its containing database.
  • Available Columns — These are the columns contained in the currently selected matched table in the Matched Tables pull-down list.
  • Matched Columns — On the left side of the symbol are all the input column names acquired from the analyses within the imported binary file. On the right side of the symbol is displayed the Available column that matches the imported input column by name (not case sensitive) and that resides in the table or view matched previously to the original table or view that contains the imported column. If an input column is not matched, the “NOT MATCHED” phrase will show next to it and action must be taken to resolve the conflict. Note that it is not possible to match an imported column if the table it is imported from was not matched on the previous screen.

    To simplify column matching, when an item under Matched Columns is selected on the right, the table matched previously to the original table of the imported column is automatically selected, if possible, under Matched Tables on the left and the corresponding columns are loaded under Available Columns.

    Note that only input columns can be changed using Matched Columns, not primary index columns, join columns or anchor columns. Therefore, only input columns that do not have these special uses should be matched to other columns.

  • Match button — Clicking the Match button matches the column selected on the left with the column selected on the right. The same result can be obtained by double-clicking on the desired column on the left to match with the selected column on the right. Note, however, that column matching can only take place if both the selected column on the left and the selected column on the right are contained in the same matched table.
  • Retain alias when column changes — Column aliases are retained when a column is mapped to the same column. When a column is mapped to a different column name, the alias is dropped unless this option is selected.
  • Display qualified names — This option displays matched columns qualified by the name of their containing table and possibly their containing database also. It is checked by default only if the same column name occurs in more than one matched table. It displays database as part of the qualified names only if the same table name occurs in more than one database in the qualified matched column names.
  • Display unmatched only — This option displays only unmatched input columns under Matched Columns.
  • Merge with selected project — By default, the project or projects in the import file are imported into a new project or projects. If there is a selected project in the Project Window, however, this option is enabled. Click it to import the analyses into the currently selected project. Even if the import file contains more than one project, all of the analyses contained in the projects in the import file are imported into the currently selected project in the Project Window if this option is selected.
  • Import button — Clicking the Import button results in all the projects and analyses in the binary import file being imported into the current Teradata Warehouse Miner Project Window. These projects and/or analyses are not saved in the current metadata database until the Save or Save All button is explicitly selected in the usual manner.
  • Cancel button — Clicking the Cancel button cancels the import operation without changing the current Project Window workspace.
    If, when importing, any database, table or column is mapped to another value and a Free-Form SQL analysis is imported, a warning is given identifying the analysis by name, indicating that it should be checked for objects that may need to be manually mapped. This warning is given also for expert clauses in various type of analysis and for SQL text elements in Variable Creation analyses. If given, the warning identifies each analysis that contains freely entered SQL text.