FAQs

It's possible to add a button on the standard toolbar, which will trigger a macro. However it can be a little bit tricky to put this button on a customized toolbar. This is how you do it: 1. Press the small down-arrow on the Standard toolbar and select "Add or Remove buttons" -> "Standard". 2. Select Macro 1 to add this button to the Standard toolbar. 3. Create a new toolbar (View - Toolbars - Customize...). 4. Leave the Customize window open and drag and drop the Macro 1 button from the Standard toolbar to your Custom Toolbar. Close the Customize window

Yes. Use the automation method IDocumentProperties.SuppressScriptProgressButtons. If you set this to True through a macro, the buttons will not be available when reloading the document. See the current APIguideQVn.qvw for descriptions and examples.

Analyzing truly massive datasets requires a robust hardware platform to support this analysis. In cases with extremely large data volumes our clients have chosen 64-bit solutions. Regardless of the production hardware platform, the evaluation and development of these applications are typically performed on more modest hardware with limited datasets. Datasets can be limited by Year, Geography, Product, Business Unit or any other appropriate criteria. The load testing and performance evaluation aspects of evaluating QlikView against truly massive datasets (terabyte and greater) typically require more planning. Please contact QlikTech sales support.

Yes, you can. Don't believe us? When the German Association of Compulsory Health Insurance Physicians (KBV – Kassenärztliche Bundesvereinigung) required a solution that could analyze terabytes of data they turned to QlikTech. “Everything spoke clearly in favor of the project team at QlikTech: in the proof-of-concept, we saw that this solution would allow us to quickly and easily analyze a data volume of even 15 to 20 terabytes across two billion records. There is simply no other supplier or system that could even come close to handling these large data volumes,� said Eva Obermeyer, head of the KBV Accounting Department, explaining the reason for their decision. For the full press release see: http://www.qliktech.com/ArticleDisplay.asp?id=1306

The list of enhancements available in version 7 is quite lengthy. A few highlights include 64 bit platform support, ajax based zero footprint client, printable reports, alerting, block/heat charting, parallelization of chart calculation across processors, caching of chart result sets, incremental reload using QVD, and hundreds of additional features.

Local QlikView only runs under Microsoft Windows. If you are running under a different OS, you may be able to utilize the Zero Footprint Client connected to QlikView Server.

Yes there is, as part of the QlikView Installation.. If you have already installed QlikView without the Examples component, simply run the installer again and you will be given the option of modifying your installation and can add the Examples component at that time. You can also reference the Shared QlikView section of this website to download various QlikView examples from QlikTech and other users.

In general, QlikView Enterprise allows you to control all aspects of developing and maintaining QlikView documents. QlikView Professional excludes creating new documents and editing the script, although some script manipulation involving local text files is possible with Professional as of 7.50. QlikView Analyzer+ prohibits any access to the load script or layout properties – it is basically just for using an existing document.

Yes there is, as part of the QlikView Installation. If you have already installed QlikView without the Tutorial component, simply run the installer again and you will be given the option of modifying your installation and can add the Tutorial component at that time.

No. Currently, there is no support for hot-key mappings.

Yes. Please refer to the QlikView Reference Manual, and look for the section titled "Starting QlikView."

Yes. Use the SQLColumns, SQLTables, and SQLTypes statements in a load script to load data about your database. Note that some ODBC drivers do not support one or more of these commands.

The online downloads may contain patches not available on the install CD

10 (%). A typical (bar/pie/line) chart takes up around 5-10KB - corresponding to some 15000 cached objects in a 10% of memory of a 2GB machine.

This feature was introduced in QlikView 7.20.2767 (October 2005). QlikView will utilize some parts of RAM for caching of calculated charts. As the user (or another user on the same QlikView Server) returns to a logical state that has been calculated before, QlikView will use the cached calculations rather than wasting CPU time for doing the same work again. The cache contents will be emptied whenever the document is closed or when the script is re-executed.

No.

Suppress When Value Is Null for a Dimension will suppress the display of a null value (typically represented by “-“) in a chart. This is usually caused by one or more transaction records that do not link properly to a dimension. For example, if a transaction links to CustomerID = 9, but the Customer table only has CustomerID 1 – 8, and the chart uses CustomerName as a dimension, then the data in the transaction record cannot be shown linked to a valid CustomerName – therefore it will be included under the null value. If Suppress When Value Is Null is checked for the CustomerName field, this transaction data will not be included in the chart – either as a dimension value, or in the overall total. Suppress Zero-Values on the Presentation tab (checked by default), on the other hand, refers to the ability to disregard all combinations of the dimension fields returning zero or null for all expressions. For example, if there were no transaction records linked to CustomerID = 2, that CustomerName would have a value in the expression of 0. If Suppress Zero-Values is checked (default), there will be no display of CustomerName for CustomerID = 2 in the chart.

(taken from the QlikView Reference Manual) If this check box is marked, all combinations of the dimension fields associated with only NULL values in all fields in all expressions will be disregarded in the calculation. This option is selected by default. Turning it off can be useful only in special cases, e.g. if you want to count NULL values in a chart. This option should generally be left alone, since it is only for very special cases.

If an object name contains special characters (e.g. “/�, “?�, “*�, etc.), the Export to Excel function will not work, since the name of the temporary file produced is based on the Object ID. If you rename the Object ID to not include special characters, the Export to Excel function will work correctly.

Yes. Some tests have been made running Open Office 2.0 for Microsoft Windows together with QlikView 7.2 and Send to Excel works in the same way as when running MS Office.

QlikView is only connected to a database during the Reload process. Opening and using a QlikView document has no effect whatsoever on any external data sources. If you are concerned about affecting performance of the database during reload, you may want to consider scheduling reloads during an overnight "œbatch"� window.

In order to optimize a QlikView document for use with QlikView Publisher, the document must contain information about the available roles. This information can be found directly in the Directory Service through the appropriate ODBC driver or equivalent tool. The roles must be associated with data, creating a relationship between a role and the data set within the document. The document must also contain rules for showing and hiding layout. This is easily achieved by setting Show Conditions on sheets. The sheets are designed to present a certain view of data and the document contains several views. The aim is to represent the roles in the document in such a way that a given user can be a member of more than one Layout Group, more than one Security Group and associated with a given data set.

Generally speaking, Section Access and reduction by Section Access in combination with a QlikView Publisher distribution is a bad combination. You run the risk of performing a reduction in either of the two methods that will render the reduction in the other invalid.

Set up a root folder of Publisher Data that contains two sub-folders; Private Data and Public Data. By dividing the content into these two separate main areas, three goals are achieved; back-ups are made easier, a more intuitive architecture is created and the structure is more secure. The users are given access to the content folders but restricted to any contained documents that they are permitted to view. The subfolders of the main areas are protected according to the same policy.

Keep it contained. A solid infrastructure is crucial in any good installation. A centralized storage is the key to achieving a compartmentalized, easy-to-maintain architecture that is easy to back-up.

The folder structure rights assignment should be based on the principle of least possible level of access for its assigned users. While assigning permissions to groups is a simple way of giving multiple accounts access, it goes against the purpose of this principle. The Administration Panel account requires read/write/change permissions to one folder only; the QlikView Publisher Administration Panel log folder. This folder is, by default, located under the directory Publisher Administration Panel has been installed into.

No, but a named/session/usage CAL must exist on the server as of version 7.

Yes, as of QlikView Server release 7.20 (October 2005)

Open a document in the QVA for IE client, and press the keys Ctrl-Alt-U simultaneously. When the User Preferences dialog is displayed, click on the Change Interface Language button.

Open a document in the QVA for IE client, and press the keys Shift-Ctrl-M simultaneously.

Open a document in the QVA for IE client, and press the keys Ctrl-Alt-U simultaneously.

One way to check if the QlikView Server IE Plug-in client is installed on a system is to check under Internet Explorer | Tools | Internet Options | Temporary Internet Files Settings | View Objects. Here, you should find a Program File named QlikOCX Control (or similar). This dialog will also provide the last accessed date and the version of the Plug-in client that is installed.

Microsoft Java 1.1, Sun Java 1.3 and later or Apple Java 1.4.2 or later

No, just as with the QlikView Java clients the data stays on the server. Only what is needed to draw the screen is transferred. However, when you print and export data locally, more data will of course need to be transferred.

Approximately 20 MB for version 7.5

Dynamic HTML or DHTML is a method of creating interactive web sites by using a combination of HTML, a client-side scripting language (such as JavaScript), the presentation definition language (e.g. Cascading Style Sheets), and the Document Object Model.

Everything you can do with the QlikView Analyzer (without the +) client (i.e. you can use a QlikView document, but not change a QlikView document).

No, apart from different versions of Java (1.1, 1.4 etc) there are several different uses of the Java language. One should at least differentiate between: - JavaScript which is a scripting language for web pages etc - Java applets (e.g. QlikView Java clients) which are programs run inside a Java Virtual Machine in a browser. - Java programs which are self-contained applications written in Java as opposed to e.g. VB or VC++.

The JVM is a separate entity from the hosting computer. It has its own graphics library and its own fonts, which differ from those used under Windows. Slight differences in font sizes and rounding errors can make a scrollbar to go on or off compared to how it looks under Windows.

No, not unless you use https tunneling. The JVM does not have support for encrypted communication.

The JVM is a "sand-box" which, for security reasons, has no contact with the resources of the host computer. Therefore local printers and disks are not accessible from the Java applet.

A JVM is a plug-in to an internet browser which creates a "sand-box" for running applets written in Java inside the browser. The applets have no access to the machine on which they are running outside the JVM "sand-box".

The QlikView Server Zero Footprint Client will also run in browsers based on the Mozilla browser engine (e.g. Firefox and Netscape 7 & 8).

The QlikView Server Zero Footprint Client will work on computers running other than Windows OS, e.g. Linux and MacOS X.

No.

No, this is Windows only. Please refer to the QlikView Server Zero Footprint Client for use with these systems.

A program hosted by and running inside a web browser. Typically an ActiveX component with ocx as file extension. The name of the QlikView Plug-in component, as of release 7.51 (June 2006) is QlikOcx.ocx.

No. There are currently (as of September 2006) no plans to extend the IE plug0in client to other browser environments.

Yes. All communication is encrypted with a 128 bit key.

The QlikView Zero Footprint client is built on a state-of-the-art AJAX architecture. The major advantage of the DHTML QlikView Zero Footprint client is that no program code is downloaded or installed from the server.

Yes. It's possible to install QlikView Server in a cluster environment. Using Microsoft Windows 2003 Enterprise Edition or Microsoft Windows 2003 Data Edition, Microsoft Cluster Service can be used to monitor the QlikView Server process in case of failure. QlikView Server is however not a cluster aware application as of August 2006.

Yes. There are several ways of implementing a solution with QlikView Server and Sharepoint Portal Server. It's recommended to use a dedicated server for QVS and to use the plug-in client as a webpart. Java and ZFP can be used as well, but requires some customization on the SPS server.

Yes. The QlikView Edit Script dialog has many advantages, including statement wizards, color coding, and function syntax. There may be times, however, when you would prefer to create and edit load scripts in a different editor. To do this, simply create a text file in your preferred editor, and save it to a disk file. Then, in the QlikView Edit Script dialog, you can either insert the script at the current cursor position (Edit … Insert File), or Include a reference to the script file (Edit … Include). The main advantages to this technique are that the script file can be edited without editing the QlikView document, and you can use script modules that multiple QlikView documents can reference so that changes are maintained in a single file (assuming you use Include).

There are a few different options to accomplish this. These include Partial Reload, and manipulating QVD files. Talk to your QlikView consulting professional or contact QlikTech to find out more about how these techniques may be applied to your environment.

Another good use of the qualify statement is to avoid data joins altogether during the data discovery stage of development. If you are not sure how data links together, simply place Qualify *; as the first statement prior to any LOAD or SQL select statements. This will allow you to load all of your data with no joins in the beginning. Then you can use QlikView to explore your data and add back in the joins one by one (by adding the appropriate unqualify statements where needed and rerunning the script) as you become more familiar with the relationships between your tables.

Join relationships between tables are handled quite elegantly in QlikView. Any fields with the same name (case sensitive) form a key field that results in an outer join between the tables. Fields that do not have the same name do not create a join field. To control this behavior in QlikView all you need to do is alias field names in the script. To join two fields that have different names, alias one or both fields so that they share the same name. This will form a join. To prevent a join from forming where two fields have the same name, alias one or both fields so that they have different names. For additional information see the Lesson titled “Associating Data from Many Tables� in the tutorial manual (default location C:Program FilesQlikViewTutorialTutorial.pdf). For more advanced options see the QlikView reference manual (default location C:Program FilesQlikViewDocumentationReference Manual.pdf)

Yes. Make sure to specify the BUFFER prefix on the "final"� Load statement, since all other Load and/or Select statements produce temporary input tables, and do not allow a prefix. For example, buffer load ...; select * from ...;

This typically indicates that you have loaded multiple tables that share the same field names across multiple fields, but not all fields. QlikView sometimes needs guidance so that it will not attempt to create all of these key fields on fields that are not actually used to form joins. Try to stick to using legitimate key fields only for table joins. One quick and easy way to accomplish this is to use the Qualify and Unqualify statements. Qualify will cause a field to be named TableName.FieldName when it is loaded into QlikView. For more advanced options see the QlikView reference manual (default location C:Program FilesQlikViewDocumentationReference Manual.pdf)

Yes. Within the QlikView script editor there is a Debug button on the toolbar (or you can use File | Debug). This will bring up the QlikView debugger. The debugger provides numerous powerful features. In order to load a limited dataset select the "Limited Load"� checkbox in the debugger and select the number of records you wish to load. Select the "Run"� button in the debugger to reload your script with the limited dataset. Be sure to specify enough records to satisfy any conditional clauses you may have on your Load or Select statements. When the script has finished running simply select the "Close"� button on the bottom left of the debugger. This method is very useful to verify that your script syntax is correct and all you data loads as expected. Be careful not to try to analyze your data after doing a limited reload because the sampling of records you read in with the limited load will not be representative of your full data set. After you have verified your script is free of errors you can then execute a full reload (outside of the debugger) and analyze your full dataset.

A text file, by definition, cannot contain NULL values. What is commonly misperceived as NULL is the empty string ('') two single quotation marks without anything in between. If this is loaded into QlikView, it will appear in a list box as looking like a blank value. A search on a <blank> will not find this value, though, since it is really an empty string. It is possible, however, to define a symbol in text files to be interpreted as NULL in QlikView. Use the SET NULLINTERPRET=<sym>; statement in the load script to enable this function. If you specify SET NULLINTERPRET=''; in your script, then all subsequent text files loaded will interpret the empty string as a NULL when loaded into QlikView. You can turn this function off again by including the statement SET NULLINTERPRET=; in your load script. For more information, search on "NULL value handling"� in the QlikView Reference Manual.</sym></blank>