Database restrictions

The iSeries Access for Web preferred language setting is not used on database connections. Therefore, all messages received from the database server will be in the language derived from the LANGID and CNTRYID of the user profile used to start the WebSphere^® Application Server.

Insert table records restrictions

• Insert does not support binary large object (BLOB), integer with scale, or ROWID column types. Insert does support the character large object (CLOB) column type, however, entering a very large value could potentially consume all of the browser's memory. All other column types, supported by the iSeries server, are supported by Insert.
• Insert only supports setting the URL portion of a datalink.
• Insert does not support inserting a null column value, unless the default value for the field is null and the field value is left unchanged.

Update table records restrictions

• Update does not support binary large object (BLOB), integer with scale, or ROWID column types. Update does support the character large object (CLOB) column type, however, entering a very large value could potentially consume all of the browser's memory. All other column types, supported by the iSeries server, are supported by Update.
• Update only supports setting the URL portion of a datalink.
• Update does not provide a way to insert null column values. However, if a column has a null value and the field is left blank, the column value will remain null.

On Windows^® 2000 using Internet Explorer, if you have Microsoft^® Excel installed and you try to write your results to MS Excel 3, MS Excel 4 or MS Excel XML, you will be prompted to logon to the iSeries server again. This will cause an additional license to be used. This only happens the first time you try to load an Excel file into the browser. As an alternative, you could save the request without running it, run the saved request, and redirect the results to a file. This is done by right-clicking on the Run link and choosing the Save Target As option. After saving the SQL output file, you could load it using Microsoft Excel or some other application.

If you choose PDF as the output type and the SQL statement generates a very large number of columns, the resulting output might be too compressed to read, or might be a blank page. In this case, use a different page size, choose a different output type, or modify the SQL statement to return a subset of the columns.

If you are using the Opera browser and your output contains very long column data, your data may be truncated when displayed.

If you use Microsoft Internet Explorer, choose PDF as the output type, and get a blank page instead of the SQL output, try one of the following circumventions:

• Ensure you have the installed the latest version of Microsoft Internet Explorer.
• Instead of running the request directly from Run SQL, save the request and use the Run action from My Requests.
• Change your Adobe Acrobat Reader configuration to display the reader in a separate window, instead of within the browser.

• Output types with a maximum cell size, such as Microsoft Excel and Lotus^® 1-2-3^® version 1, will truncate the data if it exceeds the maximum cell size.
• Other output types will not truncate the data, however, retrieving very large values could potentially consume all of the browser's memory.

• Only single table selects are supported.
• Nested conditions are not supported.
• Building conditions is supported for the column types supported by the iSeries, with the following exceptions: Binary large objects (BLOBs), Character large objects (CLOBs), and Datalinks.

In order to copy an OpenDocument spreadsheet file, the file must be in package format and must not be compressed or encrypted.

When a saved request is edited or run, the file name is not automatically specified in the Copy Data to Table form to protect the data on your workstation. The original file name is shown beneath the File to Copy section. This file name does not include path information when the browser is run from the Linux^® operating system.

Certain Data Transfer From AS/400^® statements cannot be converted into statements that can be modified by the SQL Wizard. The SQL Wizard does not support building or editing SQL statements containing GROUP BY, HAVING or JOIN BY clauses. In this case, you must hand-edit the resulting statement on the Run SQL panel.

Data Transfer has an option for specifying whether ANSI or ASCII data is written to or read from a PC file. Requests imported into iSeries Access for Web will use the Data Transfer setting, combined with the language and character set specified by the browser to determine the encoding of the client file. This may or may not be correct. You may have to manually change this setting.

iSeries Access for Web will not differentiate between source physical and data physical files. An imported request that selects all columns (SELECT *) from a source physical file will produce output containing all columns contained within the source physical file, including the sequence and date columns. An identical request run with Client Access Express produces output containing only the data column(s).

When importing Client Access Data Transfer to AS/400 requests that copy data to a source physical file, the request must be using an FDF file. This situation cannot be detected by the import function and an error will not be issued. However, if an FDF was not being used, the resulting copy data to table request will not work correctly.

iSeries Access for Web does not support all the file types currently supported by Client Access Data Transfer. In some cases, a Data Transfer file type may be mapped to a corresponding iSeries Access for Web file type. If a corresponding file format cannot be found the import will fail.

Some output options available in Client Access Data Transfer are not available in iSeries Access for Web. These options will be ignored.