Searching from 2023-02-18 00:00:00 to 2023-02-18 23:59:59.999999.
Query completed in 0.40 seconds
[2023-02-18 08:01:35] --> monr0e has joined the channel
[2023-02-18 08:01:39] --> delinquent has joined the channel
[2023-02-18 08:33:39] <-- monr0e has quit (Ping timeout: 255 seconds)
[2023-02-18 08:37:19] <-- delinquent has quit (Ping timeout: 480 seconds)
[2023-02-18 11:31:35] <Z-Man> Lucifer_arma: I feel like I'm done with cerritos adaptions: https://github.com/davefancella/cerritos/pull/5
[2023-02-18 11:31:35] <Z-Man> It's a bit of a mess. CMake docs are all like "if you want a, do this, if you want b, do that, and so on" but the reality is IF YOU DO ANYTHING OTHER THAN d things are going to fail on some system.
[2023-02-18 11:31:35] <Z-Man> For me, building now works out of the box on macOS and Linux, the examples mostly run as intended. I don't have the font you are testing anywhere, so dunno about that.
[2023-02-18 11:31:35] <Z-Man> On macOS and Windows, the dadsteroids sample does not compile, both because destructors are called explicitly in a fishy way. Maybe you can get them reproduced if you switch to clang from GCC.
[2023-02-18 11:31:37] <Z-Man> On Windows, the Conan step is required. The main CMakeList.txt is set up so it uses the output of conan if it is there, otherewise it is ignored. What conan does is that it fetches the dependencies, then puts appropriate FindXXX.cmake files into the build directory, which CMake then is instructed to use first. What builds/runs without Conan also builds/runs with conan.
[2023-02-18 11:31:38] <Z-Man> Exception: dadsteroids on Linux with Conan does not find its data files. I can help sort that out once it compiles everywhere.
[2023-02-18 11:31:40] <Z-Man> Oh, right, probably important: spriteTest compiles everywhere, but crashes everywhere. Here is a backtrace: https://pastebin.com/T5k1FJpv
[2023-02-18 11:31:42] <Z-Man> Also important: You seem to have noticed that filesystem::path has different internal string representations on Windows and everything else. Your solution was to define strings differently depending on the system... that is probably a bad idea, evidenced by the torrent of compiler errors I got :) Since std::wstring naturally has no easy conversion to/from char * strings, that means every user has to be aware of that difference in almost
[2023-02-18 11:31:44] <Z-Man> eveny string operation...
[2023-02-18 11:31:46] <Z-Man> What I did in the last commit, easy to just leave out if you don't want it, is to define std::string as the string to use every time, and use explicit conversion from path to utf8 strings where that was required. I'd personally probably just keep conversions to a minimum; just have functions that want a Dirpath ultimately accept only a Dirpath, and functions that return a Dirpath from storage return a Dirpath. Urgh. It's all filesystem::
[2023-02-18 11:31:49] <Z-Man> path's fault.
[2023-02-18 11:52:29] --> monr0e has joined the channel
[2023-02-18 11:52:34] --> delinquent has joined the channel
[2023-02-18 17:48:01] <-- monr0e has quit (Ping timeout: 252 seconds)
[2023-02-18 17:51:44] <-- delinquent has quit (Ping timeout: 480 seconds)
View entire month
DISCLAIMER: These logs of public chat may contain some content which may not be appropriate for all audiences. Use at your own risk.
Logs from 2006-2009 pulled from wrtlprnft
Format changes at: 2015-08-25, 2017-02-20, and 2020-03-23. Times (2015 and later) should be Eastern.