Hi! I can spend a part of my tokens and do this at least for some reviews. I'll do this for qtbase other modules if Marc is doing the core
br, Jani ________________________________ From: Volker Hilsheimer <[email protected]> Sent: Friday, June 12, 2026 9:50 AM To: Marc Mutz <[email protected]> Cc: Jani Heikkinen <[email protected]>; [email protected] <[email protected]> Subject: Re: [Development] HEADS-UP: Qt 6.12 API change review Thanks Marc! Very much agree that this is an excellent use case for AI to support us with; not just for the pre-release header review, but also to support any reviews of changes to public headers (if not for any reviews). The AI reviews of integration failures are already very valuable. It is however work (and token budget…) to make it so. Volker On 11 Jun 2026, at 16:40, Marc Mutz via Development <[email protected]> wrote: The first two are up on Gerrit. I guess I'll do the other QtCore ones, too, now. But then I'd really like someone else to take over. Adjusted prompt I'll use for further such changes: Please review attached Qt 6.11..6.12 header diff for violations of https://wiki.qt.io/Things_To_Look_Out_For_In_Reviews and https://community.kde.org/Policies/Binary_Compatibility_Issues_With_C%2B%2B. The primary goal is to find BC and SC problems that would be hard or impossible to fix after the release. The secondary goal is to find style violations for rules listed in TTLOFIR (see link above), incl. the canons linked from there. Output as a lJSON file compatible with gerrit review --json. Use the same base name for the json file as the input .diff file. If something isn't clearly BC or SC, output a question asking for clarification. This is part of the process described in https://contribute.qt-project.org/quips/10. The Qt project distinguishes between acceptable and unacceptable SiC, this is defined in https://contribute.qt-project.org/quips/6. Assume readers of the comments see such a diff all the time, so be very brief like "mising trailing comma" instead of writing several lines of text. For TTLOFIR, specifically, use a deep link of the form"https://wiki.qt.io/Things_To_Look_Out_For_In_Reviews#Includes Item 2.1"<https://wiki.qt.io/Things_To_Look_Out_For_In_Reviews#IncludesItem2.1>. Don't comment on acceptable changes, unless the comment helps other reviewers understand this is ok even though it may not look like it at first sight. Again, assume that the readers of the comments know everything in TTLOFIR and the KDE BC page, no need to explain. Finally, this is an open source project, the diff is public. On 11.06.26 15:53, Marc Mutz via Development wrote: I've let Opus 4.8 lose on the two XL patches, so consider those covered. But I don't want to burn though all my tokens for this, so I would prefer if someone else could do it for the others. I used this prompt, but I can't say, yet, whether it was a good choice: Please review attached Qt 6.11..6.12 header diff for violations of https://wiki.qt.io/Things_To_Look_Out_For_In_Reviews and https://community.kde.org/Policies/Binary_Compatibility_Issues_With_C%2B%2B. The primary goal is to find BC and SC problems that would be hard or impossible to fix after the release. The secondary goal is to find style violations for rules listed in TTLOFIR (see link above), incl. the canons linked from there. Output as a list of comments with <file>:<line> anchors. If something isn't clearly BC or SC, output a question asking for clarification. This is part of the process described in https://contribute.qt-project.org/quips/10. The Qt project distinguishes between acceptable and unacceptable SiC, this is defined in https://contribute.qt-project.org/quips/6. Thanks, Marc On 11.06.26 15:38, Marc Mutz via Development wrote: Seeing as the very first one I looked at revealed a threefold human fail in the 6.11 review¹, can we please point Claude at these, first, and tell it to review the patches according to TTLOFIR and KDE's BC page? And post the result as comments? ¹ https://codereview.qt-project.org/c/qt/qtbase/+/743936/comment/47772d51_cbd88b4c/ On 11.06.26 12:26, Jani Heikkinen via Development wrote: Confidential Hardly, seeing as this was posted on the public ML ;) Thanks, Marc -- Marc Mutz <[email protected]><mailto:[email protected]> (he/his) Principal Software Engineer The Qt Company Erich-Thilo-Str. 10 12489 Berlin, Germany www.qt.io<http://www.qt.io/> Geschäftsführer: Mika Pälsi, Juha Varelius, Juha Puputti Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B Public -- Marc Mutz <[email protected]><mailto:[email protected]> (he/his) Principal Software Engineer The Qt Company Erich-Thilo-Str. 10 12489 Berlin, Germany www.qt.io<http://www.qt.io/> Geschäftsführer: Mika Pälsi, Juha Varelius, Juha Puputti Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B Public -- Marc Mutz <[email protected]><mailto:[email protected]> (he/his) Principal Software Engineer The Qt Company Erich-Thilo-Str. 10 12489 Berlin, Germany www.qt.io<http://www.qt.io/> Geschäftsführer: Mika Pälsi, Juha Varelius, Juha Puputti Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B Public -- Development mailing list [email protected]<mailto:[email protected]> https://lists.qt-project.org/listinfo/development Confidential
-- Development mailing list [email protected] https://lists.qt-project.org/listinfo/development
