-Setupengine Log Parser Documentation- by Risa Newyear-Ramirez

Introduction

This document is meant to be a guide for the SAP Install Team to using the setupengine.log parsing tools for the purposes of testing. It describes how the tools can be used to generate visualizations of the time taken to install certain components, and to gather all the 'error:' tags in the log to sort them by known or unknown errors.

What is included with this project

-compareIssues.py -CoordinateParser.py -coordinates.csv -ErrorReportingToolGUI.py -Known_Install_log_Issues.xlsx (Latest version from shared Install Automation folder) -KnownErrors.txt -main.py -step_duration.csv -StepParser.py -steps.txt -UnknownErrors.txt

The project will also require a setupengine.log to be in the same directory.

The python add-on wxpython (http://www.wxpython.org/) will need to be downloaded for ErrorReportingGUI.py to work.

Python 2.7 or later is required to run wxpython.

Using the setupengine parser to create steps and Lumira Visualization (from Salman Tariq's documentation)

The StepParser.py python class is responsible for opening the steps.txt file, parsing the steps listed and using them to get the minutes needed to execute each step. Each line in the steps.txt file represents a step. For example the last step currently is represented by the following line:

*** END CHECKING PRODUCT SEQUENCE *, UNCACHING DUs...,Third Product Sequence

We can break the above line into 3 parts as divided by the commas: -The first part represents the starting line for that step -The second part represents the ending line for that step -The third part is just a name to represent that step

The steps.txt file can be modified to add/remove as many steps you want, given that each line has a unique start/end pair i.e the pair is not repeated anywhere else in the setupengine log.

The main.py is the entry point for the entire program. It is executed via the shell (cmd.exe or windows powershell) with a single argument, which is the path to the setupengine.log. With the above folder structure i.e the setupengine.log and the main.py file in the same folder, the following command needs to be run:

python.exe .\main.py .\setupengine.log

The main.py script parses the setupengine.log file and passes it to the StepParser.py, which uses it to compile a list of steps and their respective durations in the form of the step_duration.csv file. As a side effect, the StepParser.py uses CoordinateParser.py to compile a list of all the coordinates in the setpengine.log file in their respective duration in the form of coordinates.csv.

In order to create a visualization in Lumira:

1. Start Lumira, either click on the Acquire Data or the New Dataset button
2. Load the step_duration.csv file
3. Create dataset
4. Set Duration as the measure
5. Select steps as a Dimension
6. Set legends (Start Time and End Time)
7. You can use different charts from the panel on the right (e.g Line Chart, Pie Chart).

Using the setupengine parser to search for errors

The compareIssues.py python class is responsible for opening up a setupengine.log file, parsing for lines marked with 'error:', and saving them to either a Known Errors or an Unknown Errors text file depending on whether the error matches keywords in the Known_Errors_log_Issues.xlsx file.

The script looks for the setupengine.log and Known_Errors_log_Issues.xlsx in the same directory as the project. If text files for Known Errors and Unknown Errors are not present in the same directory, they are created when the script is run for the first time.

You can run the script from the command line by typing:

python compareIssues.py

The script parses the setupengine.log file and saves any lines that contain 'error:' or 'Error:' to ErrorArray. Then it parses Known_Errors_log_Issues.xlsx for the strings in the keyword column of the document and saves them to a variable called keywordColumn. During the compare() method, the values in ErrorArray are cross-referenced with the values in keywordColumn. If any matching strings are found, the line in which it was found is written to KnownErrors.txt. If no matching strings are found, the lines that are left are written to UnknownErrors.txt. The script then returns ErrorArray.

In order to view the known and unknown errors more clearly, you can open up the Error Reporting Tool in the command line by typing:

pythonw ErrorReportingToolGUI.py

or if your Python 2.7 update is located in a different folder:

C:\Python27\python.exe ErrorReportingToolGUI.py

The GUI displays the known errors in a text box on the left hand side of the pop-up window. Note that the errors cannot be selected or manipulated in any way. The unknown errors are displayed in a listbox on the right side of the window. The unknown errors can be selected from the box.

There is a 'Report to JIRA' button that, if pressed, opens up the JIRA homepage in a browser. This is intended for users of the tool to report any unknown errors that are ship killers to a product or causing a BAT test case to fail to be reported to JIRA.