Mid-SoK Blog

Wednesday, 4 March 2026  |  Navya Sai Sadu

#SoK   #mentorship  

I'm half-way through the Season of KDE 2026 and wanted to share the journey so far.

I had subscribed to the kde-soc mailing list after I returned from IndiaFOSS'25, where I met KDE contributors who really encouraged me to join the community and told me that one can always learn while building. The first step, always, is to start. Then I got carried away with life until I saw "call to action" in my mailbox in January. It was about SoK'26. I had then recently set up Kubuntu and was in awe about what people can build out of passion and by collaborating with others. I felt mentorship was the way to get started. I explored the projects and found Task-3 under Lokalize as something that I can contribute to while learning new skills- programming in CPP, debugging, and exploring an old repo.

Here's everything i did before the season officialy began.

Before Proposal

  • I installed ==Lokalize== and reproduced the bug.
  • Cloned the repo. Used grep to find “Approve and go next” in codebase. Turned out EditorTab::gotoNextFuzzyUntr() is the origin of bug. I thought about "Approve and go prev" and tried that. It was buggy too!
  • Wrote proposal

After Proposal

  • kde-builder failed due to missing libraries. So after lot of fixes, i did ubuntu update to 25.10(has qt6). *Created a test branch and added debug log in forementioned method.

Before 23rd

  • Got myself acquainted with Kate.
  • Fidgeted around files to understand architecture.

Got to know about

  • qDebug()
  • why use .h and .cpp files
  • LSP

The weekly status reports contain links to supoorting documents, commits and MRs.

Week 1

  • Spent more time understanding the repo, the files,classes and methods which were involved in the bug using debug logs, grep and manual look-over.

  • Wrote a few comments to document better.

  • Had an introductory call, was nice to know the contributors.

Week 2

  • Researched past commit to understand the intent of navigational shortcuts.

  • Added SPDX license headers( did not know it was being actively developed in GSOC too). Aided Kumud by sending an email to the KDE i18n mailing list regarding the cyclic traversal on entries for keyboard shortcuts in the Translation Units View. Typically, KDE applications don't use that, as followed from the replies and was not needed.

  • Had a group call for understanding rebase but i still mess up sometimes.

Week 3

While Kumud's implementation addressed gotoNextFuzzyUntr(), analyzed their approach to generalise it to make it work for gotoPrevFuzzyyUntr() and added a parameter to move in either direction for which a new Enum was used instead of magic numbers(they have a name for it T-T). I could see magic numbers used previously in the code and do get it why it isn't preferred.

Week 4

Continued with the week-3's work

Week 5

Worked on fixing other navigational shortcuts in editortab.cpp which implied generalising the function furthermore to take another parameter for entry state. StackOverflow is a saviour(Of course, after my mentor). Throughtout this time, it felt as if things were unravelling themselves to me. New bugs were found and worked upon. Looking back, I mostly improvised upon Kumud's solution but still got to learn a lot on the way.

Week 6

Will work on other bugs identified on the way or scout for new ones! See if I make past this which was primarily the task for SoK.