New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Prepare for 1.10.2 release #8135
Conversation
… and thus caused desyncs. Use DC_AUTOREPLACE for actions that shall be reversibe, in this case: - Do not rearrange free wagons in test-run. - Do not discard OrderBackups. The latter was not triggered by actual auto-replace, since it does not set a 'user'.
… Windows version (OpenTTD#8114) * Fix OpenTTD#8102: Update Windows installer strings to reference versions newer than 7 * Fix 2c53390: Version strings incorrectly capitalised
…agonal rail next to a dock end (OpenTTD#8124)
Variable-length values would write threshold values with a byte too many. System Exclusive messages would cause write to fail since the end byte was treated as part of next message.
NB waiting to see if @ldpl gets any further with desync debugging |
Dismiss so it doesn't get merged by accident
The presence of sdl2-config is used go determine whether to look for sdl2 first, or just sdl1. However, when sdl2-config is *not* present, `which` returns an empty string. Due to lack of quoting, this produces `[ -x ]`, rather than `[ -x "" ]` and it turns out the former actually has a succesful exit status for some reason. This was not a problem when just running configure, because it would then just fail to detect sdl2 and fall back to sdl1. However, when passing `--with-sdl` (without specifying a version), this would only attempt to detect sdl2, even when sdl2-config was not present, but sdl1 is. Adding quotes makes the check work as intended.
I really want a fix for #8066 into 1.10.2, we keep getting reports of it and it feels bad to tell people to play 1.10.0-RC1. |
I agree, but I'd rather not wait any longer. The amount of desyncs that @ldpl keeps finding isn't good... |
If we have fixes for bugs in prod, we should ship those fixes early and often :) "Releases are cheap"? :) |
88b1ead
to
99f2b27
Compare
…o editor was doubled
Previously the internal content list was invalidated and sorted for every new item added. Now the sorting is delayed until the GUI is drawn, which means we only sort once per GUI tick. Since the amount of incoming items per GUI tick is not controlled by the GUI but rather by network speed, we were previously doing a lot of duplicate work per tick, causing the mouse cursor to lag while the list was initialized.
Initialization code for GRFFile::roadtype_map was copied from railtype_map. But while RailType is a byte-sized enum and could thus be initialized via memset, RoadType doesn't have a defined size.
Quite a few changes in the end, but don't think we need an RC?