C Test Report Generator - attolpostpro

Purpose

The C Test Report Generator processes a trace file produced during test execution, and generates a test report.

Syntax

attolpostpro <trace_filename> <report_filename> [<options>]

attolpostpro @<option file>

where:

  1. <trace_filename> specifies the root (filename without extension) of the trace file that is generated when the program runs.

    <report_filename> specifies the name of the .rod compact report file produced by the Test Report Generator.

    <options> can be any of the optional parameters specified below.

    <option_file> is the name of a plain-text file containing a list of options.

Description

The Test Report Generator uses <trace_filename> to find the names of both the .rio trace file and the .tdc table of correspondence file that are generated by the Test Script Compiler.

If <report_filename> is provided without an extension, the Test Report Generator attaches .rod.

If either <trace_filename> or <report_filename> are omitted, the Test Report Generator produces an error message and terminates.

If the either <trace_filename>.rio or <trace_filename>.tdc  do not exist, cannot be read, or contain synchronization errors, the Test Report Generator produces an error message and terminates.

If the Test Report Generator cannot create the .rod compact report file, generation of the report is terminated. If the file already exists, the newly generated file replaces the existing report.

The .rod compact report file is an intermediate low-footprint format that can be stored on remote targets. The .rod files must be converted to the .xrd report file format to be displayed by the OneTest Embedded GUI with the rod2xrd command line tool.

Options

Options can be in any order. They may be upper or lowercase and can be abbreviated to their shortest unambiguous number of characters.

 

-cio=<coverage result file>

This option allows you to insert coverage results in the report file. This option must be used only in conjunction with the Code Coverage feature.

 

-compare[=strict]

This option lets you compare the results from two test runs. A trace file generated during the first run has a .rio extension, and the one generated during the second run has a .ri2 extension.

When making a comparison, the Test Report Generator generates the test report from:

  1. The .tdc table of correspondence file

    The .rio trace file generated during the first run

    The .ri2 trace file generated during the second run

The same root name is used for the names of the three files.

When comparing values, a variable will only be deemed correct if the two obtained values are the same as the expected value, or within the specified validity interval for that variable. With the compare=strict option, the two results must have the same value.

 

-ra[=test | error]

This option specifies the form of the output report generated by the Test Report Generator.

Use -ra with no parameter, to display ALL test variables and mark any variables that are incorrect for a given test. This option is used by default.

Use the -ra=test option to display ALL test variables, with incorrect variables marked. This option provides a comprehensive display of variables for an incorrect test, which can prove useful in a complex test environment.

Use -ra=error to to display only erroneous test variables.

For both -ra=test and -ra=error, if no errors are detected in the test, only general information about the test is produced.

 

 

-va=eval | noeval | combined

This option lets you specify the way in which initial and expected values of each variable are displayed in the test report.

Use -va=eval if you want the test report to show the initial and expected value of each variable evaluated during execution of the test. This is only relevant for variables whose initial or expected value expressions are not reducible in the test script.

Note: For arrays and structures in which one of the members is an array, the initial values are not evaluated. For the expected values, only incorrect elements are evaluated.

Use -va=noeval if you want the test report to show the initial and expected values described in the test script.

The -va=combined option combines both eval and noeval parameters. For each variable, the Report Generator includes the initial and expected values described in the test script, as well as the initial and expected values evaluated during execution, if these values differ.

By default, the -va=eval parameter is used.

 

-studio_log

This option is for internal usage only.

Return Codes

After execution, the program exits with the following return codes

 

Code

Description

0

End of execution with no errors

3

End of execution with one or more warning messages

5

End of execution with one or more errors

7

End of execution because of fatal error

9

End of execution because of internal error

 

All messages are sent to the standard error output device.