TYX TestBase
Version 2.5.2
14 November 2003
1. Overview
This is a maintenance release that corrects several problems.
The release includes the following main items:
- TestBase installation package, including a complete set of sample files, documentation and tutorial slides
- Redistributables, including:
- TYX License Manager
- Third-party run-time engines
- Instrument drivers needed for running some of the samples
- Installation Guide
2. Detailed Description
2.1 Critical items
2.1.1 Security setting for Visio 2002
Visio 2002 displays a warning message each time it opens a flowchart that contains macros. Because the TestBase IDE makes use of Visio macros, this message will appear during IDE operation. To disable the display of the warning message, open the Visio application, select the menu entry Tools | Macros | Security and choose the security level Low.
2.1.2 Environment variable setting for LabVIEW Adapter
The LabVIEW Adapter loads the LabVIEW engine when it is loaded by TestBase, and unloads the engine when it is unloaded. It appears that the repeated unloading and reloading of the LabVIEW engine causes unreliable operation. Thus, for a reliable operation of the LabVIEW test procedures it is recommended to set the value of the environment variable TB_TCA_UNLOAD to 0 (zero). See the Help System for details.
2.1.3 Error when generating Diagnostic Procedure HTML Reports
When generating an HTML report for a diagnostic strategy, you may get the error message "Unable to export CFD". The report files containing the Visio diagram are not generated. The report file containing the text description of the diagnostic strategy is generated, but its hyperlinks will not operate.
This problem occurs if Visio Professional was installed on the system but its HTML export feature was never used. To fix the problem, open an existing flowchart in Visio or create a new one (menu File | New), then export it as HTML (menu File | Save As, select file type "HTML Files", select a temporary folder and press OK).
2.1.4 Operating system compatibility
The product works properly with Windows NT 4.0 SP 6, Windows 2000 SP 2 and Windows XP.
2.1.5 Internet Explorer compatibility
The product works properly with Microsoft Internet Explorer 5.5 and 6.0. It does not work with Internet Explorer 5.0.
2.1.6 Microsoft Office compatibility
The display of Excel reports works properly with Microsoft Office 2000 Service Release 1 and Microsoft Office XP.
2.1.7 Microsoft Visio compatibility
The product works properly with Microsoft Visio 2000 and Visio 2002.
2.1.8 Oracle compatibility
The MTI Controller distributed with the current version of the product was tested with the following combinations of versions :
- Oracle 8.1.5 and 8.1.7, under Windows NT 4.0
- Oracle 8.1.5 and 8.1.7, under Windows 2000
- Oracle 9.0.1.1, under Windows NT 4.0.
A known compatibility problem between the Oracle software and the Microsoft libraries used internally by TestBase may be fixed, for Oracle 8 versions, by configuring the system registry as described in the document Connectivity Issue with MDAC and Oracle8i.pdf. For Oracle 9, use the information provided for version 8.1 in the above document, performing the following replacements in the strings to be entered in the registry:
- replace oraclient8.dll with oraclient9.dll
- replace orasql8.dll with orasql9.dll
2.1.9 LabWindows/CVI compatibility
The Adapter for LabWindows/CVI works properly with the following versions: 5.0, 5.5 and 6.0. Minor changes to the sample test code must be performed in order to compile it under version 5.0. The TestBase installation package redistributes version 6.0 of the LabWindows/CVI Run Time Engine. You may obtain other versions of the Run Time Engine from National Instruments.
2.1.10 Case sensitivity of Outcome comparisons
When a Test block is followed by decision blocks, the compiler compares the outcome values tested by the decision blocks against the set of possible outcomes returned by the called Test procedure. In previous releases, the above comparison was case insensitive. For consistency with other comparisons in TestBase, the comparison algorithm was changed to case sensitive, starting with release 2.4.0. This change does not affect the operation of existing projects and does not require their recompilation.
2.1.11 Rebuilding samples of Custom Data Type Editors and samples using Custom Data Type Editors
When rebuilding the Custom Data Type (CDT) Editor samples (or a modified version of these samples), overwrite the existing .OCX file (also included in the distribution). Do not delete the existing file before generating the new one. Overwriting the original file preserves the version used by Visual Basic to record binary compatibility. If you generate a new .OCX file (without overwriting), you will need to re-build all test procedures using that CDT Editor.
To rebuild the LabWindows/CVI test procedure samples that use Custom Data Type (CDT) Editors (or a modified version of these samples), you must first re-generate the wrapper files for the CDT Editor components. When re-generating, overwrite the original files.
2.2 Known Limitations
2.2.1 TestBase IDE
The "Undo" command used during flowchart editing does not operate properly in some situations. It is recommended to avoid using it.
The "Undo" command for the addition of an Off-Page Reference block does not delete the "pair" block that was automatically appended. Workaround: delete manually the "pair" block.
When running a test strategy in debug mode, with execution stopped between steps, if the user clicks on a different Control Flow Diagram (flowchart) then clicks the Run, Step, SimulatedStep or SimulatedStepWithUI buttons, an error message may be displayed: "Automation Error. Illegal to call out while inside message filter". This is normal. Press OK, click on the Control Flow Diagram that is currently debugged and continue using the IDE. The error indicated before has no impact on subsequent operation.
In some situations accelerator keys (Ctrl-O for File | Open, F5 for Debug | Go, etc.) do not work. Use menus or toolbar buttons instead.
A General Protection Fault occurs when closing the IDE or a Diagnostic Controller user interface after an Abort operation. Because it occurs only when the application is terminated, this behavior does not have harmful effects.
2.2.2 LabWindows/CVI Adapter
A limited subset of scalar data types is currently supported in the TestBase Support Library for LabWindows/CVI. Use direct access to VARIANT fields for other data types.
The execution of samples that use Custom Data Type Editors as ActiveX controls requires a LabWindows/CVI installation (the run-time engine is not sufficient).
2.2.3 PAWS Adapter
Array parameters are not supported.
Debugging of ATLAS TPSs launched from TestBase is not supported. Workaround: debug TPSs in standalone execution mode.
2.2.4 LabVIEW Adapter
The documentation for developing test procedures with LabVIEW does not cover the use of Custom Data Type Editors as ActiveX controls on LabVIEW panels.
2.2.5 VEE Adapter
The documentation for developing test procedures with VEE does not cover the use of Custom Data Type Editors.
2.2.6 Permissions for installation
The user performing the installation must have Administrator permissions for the Windows operating system. After installing TestBase, this user must start the IDE once, to perform a Registry initialization. After that, users with more restricted permissions can use TestBase modules. The user performing security administration from the IDE must have also Administrator permissions for the Windows operating system.
2.2.7 MTI error on Global Parameters Values with the same name
If Global Parameter Values with identical names are defined at multiple level of the Project tree and the storage of test results in MTI databases is enabled, an error will be generated at run-time. Workaround: avoid using identical names, until this problem is fixed.
2.2.8 Custom Data Types
The use of CDT Editors is currently not supported in test procedures developed with ATLAS and Agilent VEE.
The generic reports included with TestBase display the values of CDTs as follows:
- The Diagnostic Procedure report displays the generic text "<Value for Custom Data Type Editor with ProgID: ", followed by the ProgID of the CDT Editor.
- The sample Excel report "Sample_MTI_offline.xlt" displays the generic text "<Value for Custom Data Type Editor with ProgID: ", followed by the ProgID of the CDT Editor.
The values of Global Parameters with Custom Data Types are not stored in Oracle and Access MTI databases. You may retrieve these values from the Input Parameters to whom the Global Parameters are assigned.
2.2.9 Diagnostic Procedure HTML Reports
The internal links of text-based reports will not operate if the set of files generated for a report is moved to a different location in the file system or on a web server.
Graphical reports are not generated if Visio 2000 Standard is installed. This is due to a functional limitation of the Standard edition of Visio. The text-based reports are generated but their internal links will not operate.
Graphical reports are not generated if Visio 2002 is installed. This functionality will be added in a future release of TestBase. The text-based reports are generated but their internal links will not operate.
2.2.10 Generic Procedures
The release contains a Generic Procedure that is not described in the Help System. Provisional documentation is provided in Section 2.5.1. below.
2.2.11. Inconsistent behavior when compiling Diagnostic Procedures
An inconsistent behavior exists when compiling Diagnostic Procedures (DPs) in IDE. For details, see notes for Problem Report 03-126, in Section 2.4.
2.3 Enhancements
none
2.4 Problem Reports
This section addresses problem reports identified for release 2.5.1.
03-126 Test Strategy marked as "not built" reports "nothing to build"
Trying to run a Test Strategy in FTC. The error message "Strategy not build" is displayed. Going back to IDE and building the Test Strategy generates the message "nothing to build". After that, the Test Strategy runs in FTC.
The Test Strategy was not built. The error message displayed by the IDE when building the Test Strategy was wrong. Changed to "Nothing to compile. Building successfully finished.".
Note: An inconsistent behavior exists when compiling Diagnostic Procedures (DPs) in IDE. If a DP is modified, then saved (menu File | Document | Save, or toolbar button Save Document), then compiled, the Test Strategies (TSs) that include calls to that DP are marked as "not built". These TSs must be built in order to be executed. On the other hand, if a DP is modified, then compiled (without being explicitly saved), the Test Strategies (TSs) that include calls to that DP are not marked as "not built". These TSs can be executed without a rebuild. Because, in some rare cases, this may cause errors at run-time, it is recommended to explicitly save DPs before compiling them. This inconsistency will be corrected in the next major release.
03-128 Various error messages in IDE
The following error messages appear occasionally in IDE: "Unable to get Element for ID", "Unable to set Diagnosis property", "Unable to append element to DB collection."
Several such error conditions were detected and corrected.
Note: This includes the occurrence of the error message "Unable to get element ID" on Diagnostic Procedure import. The problem occurred when clicking on a block in the imported flowchart, while the import operation is still in progress.
Note: The error message "Unable to set Diagnosis property" occurs in the context described in Section 2.2.7.. This limitation will be addressed in the next major release.
03-129 FTC error related to expanded DP call steps
Open a Test Strategy that contains DP call blocks and expand the DP call steps. Select in grid the DP call step, but do not select any steps inside the call. Run the selected steps. The error "Subscript out of range" will be generated.
The problem was corrected. In the situation described above, the FTC automatically deselects the DP call step.
Note: To execute all steps inside a DP call, you must explicitly select these steps. Due to the above fix, selecting the DP call step does not have that effect. The handling of special steps (DP call and END) will be improved in the next major release.
Fix for "File Download" problem does not work with foreign-language Windows versions
Release 2.5.1 included a fix for a problem introduced by one of the new security updates for Internet Explorer (also included in Internet Explorer Service Pack 1). For details, see the Release Notes of version 2.5.1. The fix works only for English versions of Windows.
In this release, the fix works for French and German versions of Windows.
Array editor stores wrong value on close.
When the Array Editor is closed without the contents being explicitly saved, the parameter received the value of the last array that was edited and explicitly saved.
Corrected.
Test Editor error "Unable to store test procedure attributes"
The above error occurs in some circumstances when switching between the Test Editor and the main IDE Window.
Corrected.
Error during debugging of "DP call" blocks.
An error occurs when stepping into a DP call block, if the flowchart of that DP is not already open.
The problem is a side effect of the fix implemented in release 2.5.1 for the error caused by the Internet Explorer security patch. The problem was corrected.
Selection in tree area after adding a UUT Model.
After creating a UUT Model node, the tree selection moves to the DPs folder located under the new UUT Model node. This is misleading to users, because the next phase of a typical development process consists in the addition of a Test Strategy, which in turn requires the selection of the UUT Model node in the tree. To add a Test Strategy, the user has to manually change the selection.
After creating a UUT Model node, the selection stays on that node.
Display of breakpoints on "DP call" blocks.
Breakpoints placed on DP call blocks are no longer displayed, after performing an execution through those breakpoints.
This was only a display problem (the breakpoint was still considered by the debugger). Corrected.
MTI error during FTC execution
An error occurs when issuing a Stop Debug command in FTC, if the MTI Controller for Oracle/Access is in use.
Corrected.
Improper display in Array Editor after array import.
The error occurs if, at the time of the import, the Array Editor bookmark is located on a row, and the index of that row is greater that the number of rows of the imported array.
Corrected.
Menu entries and buttons improperly enabled in IDE.
Menus File | Project | Close and File | Project | Save are enabled when no project is open. Selecting the first of the above menus produces an error.
The sub-menus of the Window menu are enabled when no Control Flow Diagram is open.
The Import button of the tree area is enabled after deleting a Test Strategy, while at least one Control Flow Diagram is open (after the deletion, the tree selection moves to the Diagnostic Procedure of the active Control Flow Diagram; in this situation, the Import button should be disabled). Pressing the button causes an error.
Corrected - menus and buttons are properly enabled.
Input value lost in Array Editor
If the File | Export menu is selected while editing a cell (with the cursor inside the cell), the value entered in that cell is lost (the exported data contains the value of the cell before editing).
Corrected.
Error after deleting the Main Diagnostic Procedure.
If a Diagnostic Procedure, which is assigned as the "Main Diagnostic Procedure" of a Test Strategy, is deleted, an error occurs on a subsequent IDE operation.
Corrected by preventing users to delete a Diagnostic Procedure, if it is assigned as the "Main Diagnostic Procedure" of a Test Strategy. A detailed message is displayed, including instructions for deleting the Diagnostic Procedure (by first assigning a different Main Diagnostic Procedure, then repeating the Delete operation).
Error when opening non-existing files.
The File Open dialog enables the user to enter the name of a file that does not exist. A fatal error occurs after pressing the OK button of the dialog.
Corrected. The Open operation can be completed only when an existing file is selected.
2.5 Additional Documentation
2.5.1 Generic Procedure "EvaluateExpression"
_EvaluateExpression – calculates the result of an expression
Input Parameters
- Arg0…Arg9 – expression arguments; have the type "double"
- Expression – string representing the expression to be evaluated; details are provided in the following
Expression syntax
The expression string can contain:
- immediate numeric constants, as described below
- operators, as described below
- functions, as described below
- the arguments Arg0 ... Arg9
Spaces can be freely introduced to visually separate expression elements. Parentheses '(' and ')' can be used to override operator precedence as necessary.
Numeric constants
You can specify constant numbers exactly like in programming languages, using the "e" or the "E" to separate mantissa from exponent and the point "." as decimal separator. Do not use the comma "," as decimal separator as it's used as parameter separator. Do not use the space " " as decimal separator as it could represent implied multiplication. Examples:
|
Correct |
Incorrect |
|
1 |
1,000,000 |
|
1.0 |
1,5 |
|
1e5 |
1e+0.5 |
|
1000000 |
1 000 000 |
|
1e+5 |
.05 |
Operators
|
Operator |
Name |
Description |
|
&
AND |
logical AND operators |
Combine multiple conditions formed using relational or equality expressions. Return the value 1 if both operands are nonzero; otherwise, return 0. |
|
|
OR |
logical OR operators |
Combine multiple conditions formed using relational or equality expressions. Return the value 1 if both operands are nonzero; otherwise, return 0. |
|
NOT |
logical negation |
Returns the value 0 if its operand is nonzero; returns the value 1 if its operand is 0. |
|
>
<
=
>=
<= |
binary relational and equality operators |
Compare their first operand to their second operand to test the validity of the specified relationship. Return 1 if the tested relationship is true and 0 if it is false. |
|
<>
><
!= |
binary inequality operators |
Compare their first operand to their second operand to test whether they are different. Return 1 if they are not equal, 0 if they are equal. |
|
+ |
binary addition operator |
Causes its two operands to be added. |
|
- |
binary subtraction operator |
Subtracts the second operand from the first. |
|
– |
unary minus operator |
Causes the following operand to undergo a change in sign. |
|
* |
binary multiplication operator |
Causes its two operands to be multiplied. |
|
/ |
binary division operator |
Causes the first operand to be divided by the second. |
|
^ |
binary power operator |
Calculates the first operand raised to the power of the second operand. |
|
! |
unary factorial operator |
Calculates the factorial of the operand. |
The table below contains all the operators, in order of increasing evaluation precedence:
|
&, AND, |, OR |
|
NOT |
|
>=, <=, >, =, <, <>, ><, != |
|
+, - |
|
- |
|
*, / |
|
^ |
|
! |
Functions
|
Function |
Description |
|
ABS(x) |
Returns the absolute value of x. Example: ABS(-5) returns 5. |
|
ACOS(x) |
Returns the arc-cosine of x. The return value is an angle specified in radians. |
|
ASIN(x) |
Returns the arc-sine of x. The return value is an angle specified in radians. |
|
ATAN(x) |
Returns the arc-tangent of x. The return value is an angle specified in radians. |
|
CEIL(x) |
Returns a floating-point value representing the smallest integer that is greater than or equal to x. Example: CEIL(4.3) returns 5. |
|
COS(x) |
Returns the cosine of x. The x value is an angle expressed in radians. |
|
COSH(x) |
Returns the hyperbolic cosine of x. The x value is an angle expressed in radians. |
|
DIV(x,y) |
Returns the quotient of the integer division between x and y. The x and y values must be integer values, otherwise unpredictable results will be generated. |
|
MOD(x,y) |
Returns the remainder of the integer division between x and y. The x and y values must be integer values, otherwise unpredictable results will be generated. |
|
EXP(x,y) |
Returns the exponential of x, that is to say the neperian constant e raised to the x-th power. |
|
FLOOR(x,y) |
Returns a floating-point value representing the largest integer that is less than or equal to x. Example: FLOOR(6.7) returns 6. |
|
HYPOT(x,y) |
Returns the hypotenuse of a right triangle with the given cateti x and y. Example: HYPOT(3,4) returns 5. |
|
LN(x) |
Returns the natural logarithm of x. If x is negative, returns an indefinite. If x is 0, returns INF (infinity). |
|
LOG(x) |
Returns the base-10 logarithm of x. If x is negative, returns an indefinite. If x is 0, returns INF (infinity). |
|
MAX(x,y) |
Returns the maximum of x and y. |
|
MIN(x,y) |
Returns the minimum of x and y. |
|
RAND(x) |
Returns a pseudorandom integer in the range 0 to x. |
|
SIN(x) |
Returns the sine of x. The x value is an angle expressed in radians. |
|
SINH(x) |
Returns the hyperbolic sine of x. The x value is an angle expressed in radians. |
|
SQRT(x) |
Returns the square root of x. |
|
TAN(x) |
Returns the tangent of x. The x value is an angle expressed in radians. |
|
TANH(x) |
Returns the hyperbolic tangent of x. The x value is an angle expressed in radians. |
Internal number representation
The evaluation algorithm makes uses the type double (8 bytes) for internal number representation. The range for this data type is +/-1.7E308, with at least 15 digits of precision.
Examples
Arg1*(Arg1-Arg3)
1.2*SIN(Arg3)
Output Parameters
- Result – contains the expression evaluation result; has the type "double"
Outcome Values
- PASS – returned when the expression was successfully evaluated
- FAIL – returned when expression evaluation is impossible, although the expression’s syntax is correct; this may occur, for example, in the case of a division by zero
Error conditions
An error condition is returned if the expression has an incorrect syntax. The following error messages may be returned:
|
Error Message |
Description |
|
Empty expression |
This error arises when the expression string is empty, or composed of white spaces only |
|
Wrong number of arguments |
This error arises when you call a function with a bad number of arguments, for example when you try to evaluate the expression: sin(45, 2), when the sin() is known to accept one only argument. |
|
Wrong use of reserved word |
This error occur when you try to use a name of a function as if it were a variable or a constant or something else. In practice this is true whenever the parentheses which indicate a function call are not used. For example the following expressions generate this type of error:
1 + sin
(COS + TAN)
sin 45
|
|
An argument was expected |
This error usually occurs when an expression is truncated or when you forget an operand or an argument; this error is also generated when you forget a parenthesis at the end of a correct expression, rendering it an unusable sub-expression. For example the following expressions generate this sort of error:
Sin(
(1+5)/(4+)
(1+5)/(4+5
|
|
Undefined function |
This error is reported when you use an identifier name (a name that can be used to name a variable) immediately before an open parenthesis: it seems to be a function name but it is not (probably because you misspelled the name, or perhaps because you intended to multiply a variable by something that is inside the parenthesis). Examples:
x(5+4)+y - you intended: x*(5+4)+y
sine(3.14) - you intended: sin(3.14)
|
|
Undefined identifier |
This error is reported when you use an identifier name (a name that can be used to name a variable) different from Arg0 ... Arg9, probably because you misspelled the name. |
|
