volbx

Table of content

• Example view
• Description
• History
• Used tools and libs
• Compilation and installation
• Licensing
• Updater
• Setup update server
• Testing
• Potential further improvements
• Screenshots

Example view

Description

Volbx is a graphical tool used for data manipulation written in C++/Qt. User can:

• load data - opens XLSX and ODS spreadsheet files,
• filter data - filters panel can be used to define data range on each column,
• select data - user can select rows on main data table,
• visualize data - multiple types of built-in diagrams (histogram, grouping, linear regression, quantiles) which adjusts dynamically according to user actions,
• order data - each column can be ordered ascending or descending,
• export data - selected set of data, together with linked plots, can be exported to XLSX spreadsheet file or CSV.

History

The tool was originally created for real estate appraisal as it gives the possibility, for people working in that area, to estimate past, current and future value of analyzed properties. When project/start-up was dropped, I’ve decided to release a less specialized application as an open source project. Most advanced and dedicated for real estates functionalities were cut out, basic and intermediate ones stayed.

Used tools and libs

Tool	Windows	Lubuntu
OS version	10 22H2	22.04
GCC	11.2.0	11.3.0
CMake	3.25.0	3.25.0
Git	2.38.1	2.34.1
Qt	6.5.2	6.5.2
Qt Creator	10.0.2	10.0.2
Qwt	6.2	6.2
QuaZip	1.4	1.4
Zlib	1.2.13	1.2.13
Eible	1.1.0	1.1.0
QwtBle	1.1.0	1.1.0
Wble	1.1.0	1.1.0

Compilation and installation

Use directly CMake or Qt Creator. CMake should:

• configure everything automatically,
• download dependencies using Git (QuaZip, Zlib, Eible, Wble, Qwtble, Qwt),
• build dependencies,
• compile Volbx and create binaries.

This is of course a happy path…

TIPS: set CMAKE_PREFIX_PATH env variable (should point to Qt) and add qmake location to PATH variable (QWT building expects it).

Licensing

Volbx is published using LGPLv3 license.

The project uses the following software and licenses:

• Qt, Qwt, QuaZip, Eible, Wble and Qwtble libraries - LGPL licenses (more on qt-project.org, quazip.sourceforge.net, qwt.sourceforge.net and github.com/przemek83),
• Zlib - Zlib license (can be found on zlib.net).

Updater

1. Build target VersionCheck.
2. Run VersionCheck.
3. Following window should be shown:

Setup update server

1. Get a domain and hosting 😃
2. Create a simple .php file which contains something similar to this:

<?
echo("Volbx-Update-Info\n");

echo("1.10");

chdir("current/");
foreach (glob("*") as $filename) {
    echo "\n$filename;" . filesize($filename);
}
?>

3. Copy .php file to hosting.
4. Create folder current in directory where .php file is located.
5. Add new/other version of Volbx and/or updater in current directory.
6. Change code of Networking namespace in Volbx to point to proper address. Look for
   QNetworkRequest Networking::getCurrentVersionRequest()
   and
   QNetworkRequest Networking::getDownloadFileRequest(const QString& file)
7. Compile VersionChecker and run it.
8. Application should connect to given address and download content of current folder.

Remarks:

• Updater is doing correctness checks using file sizes. Checksums should be used instead.
• When I’ve created VersionChecker I’ve decided to use the simplest known by me solution. I do not know if it is safe and ‘proper’ enough, but it worked for me. I’m not web developer 😉
• Windows needs admin rights (as of ~2012) to execute files having in name strings like “update”, “install” or “setup”. I’ve picked name VersionChecker to work around that problem.
• On Windows OS, updater cannot overwrite file which are being used (as of ~2012). There is a special mechanism in place for replacing used/locked files:
  • close Volbx,
  • run updater,
  • download files,
  • change not used binaries/libs,
  • on next run of Volbx replace automatically used/blocked ones.

Testing

1. Compile project.
2. Run target named tests.
3. Check output. All tests should have status PASSED.

Tests are done using Qt test framework.

Potential further improvements

• Upgrade code to use C++20.
• Measure and increase test coverage.
• Setup static analysis (CppCheck, Clang Tidy, Clazy).

Screenshots