Well, the point of Travis is to see whether nothing broke. If the build is totally fake, how can we trust the results? And no, the previous bundle wasn't the one for the last release, but it was for older ones, and yes, I did test the builds. I used the very same script on my machine to cross-compile several times, and tested the resulting binaries. What I don't want is the Travis builds to break in two weeks not because we did something wrong but because it wasn't a reliable test but some fragile thing that happened to pass build. That's why I didn't fancy my original idea of overwriting some part of the bundle with a newer one in a way I didn't even test produced a working executable. I don't fancy either editing the bundle to strip out some flags just because it doesn't build with them, in a way that wasn't tested either.
--- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/geany/geany/pull/1072#issuecomment-226023360
