Patryk Obara

Uploaded patch series 4. It consists of all the commits from series 2, 3 and all new changes since then (in total: 78 patches); content can be also browsed on GitHub: https://github.com/dreamer/dosbox-staging/compare/317861791151%5E...b86c178e67d0 dist-clean topic: I squashed 3 relevant patches into 1 splash vector: My questions were not answered, so I had no actionable way to address concerns New patches in the series fix several small bugs, resource leaks, do general cleanup enabled by C++11 usage...

trunk is missing NEWS and ChangeLog about 0.74-3

For some of the patches, I am really not sure what to do with them. We can go in details and iterate on these patches until they are ok for inclusion, or just deemed unacceptable. dist-clean patch I believe this will explain: https://github.com/dreamer/dosbox-staging/issues/14 Personally, I use git clean -fdx for this purpose, but users receiving the tarball with the code (e.g. from their OS distribution or as source code distributed with the game) won't have such option. The file listing was later...

None of patches from patch series 2 was applied to SVN yet, but in the meantime, we had a lot of work done in dosbox-staging. Attaching patch series 3. These are the commits: https://github.com/dreamer/dosbox-staging/compare/a183237d0707%5E...ee54da6ee765 They depend on changes introduced in patch series 2. Short summary of changes: ALL of krcroft's work regarding sound handling, including the latest improvements merged in today Lots of warning fixes (or silencing, if it can be trivially accomplished)...

I think we can close this issue? Continued in #284

Prerequisites for SDL_sound replacement and SDL2, plus some other changes

Thanks :) These changes pushed number of GCC9 warnings down from 382 to 380 and Clang static analyzer issues from 109 to 100 - you can check details/download static analysis report here: https://github.com/dreamer/dosbox-staging/runs/247088678 Can I get a clarification about the newest allowed C++ standard in DOSBox code? Is C++03 ok? What about patches 1 and 2?

@qbix79 I think we posted at the same time :) Static analyzer reported the memory leak for ther "else" branch - most likely it couldn't correlate that else branch is always taken when fstype == "none" (that would require dynamic analysis; I hope value of that string is not taken literally from user input) - I will look at it again tomorrow. There's no rush with merging these patches to the trunk, take your time to vet them - I can work on my changes on my own Git branch and rebase on top of newer...