Hi Rommaso, see my related complaints at https://musescore.org/en/comment/768356
Am Freitag, 18. August 2017 15:59 CEST, Tommaso Cucinotta <[email protected]> schrieb: > Hi Joachim, > > thanks for your reply. I aimed at 2.0.x because it's the latest official on > Ubuntu 17.04, and I've read about the requirement of Qt 5.8 which I don't > have (5.7 on my laptop), and I could also exploit "apt-get build-dep > musescore" which is normally the smoothest way for recompiling a pkg on > Ubuntu & Debian-based distros. > > Also, I forgot to mention: the final mscore executable (2.0.2 or .3), when > launched, used to launch a "Loading..." dialog which wouldn't go away, until > I killed it via Alt+F4 -- apart from that, the software seemed to work > (except that, if I didn't kill that dialog,. > > I'm compiling origin/2.2 right now: Where did you get your sources from and how exactly did you try to build. > 1) Makefile is still overwritten by 'make' -- what's the rationale for doing > that ? Hmm - that allmost sounds like you did run cmake in the toplevel source directory. This, as you have seen doesn't work since cmake does (at least with the default settings) create a Makefile (after all, that's the whole purpose of cmake) which would overwrite the checked-in Makefile. > 2) first 'make' gives same errors > > make[1]: Entering directory '/home/tommaso/mscore/build.release' > make[1]: *** No rule to make target 'manpages'. Stop. > make[1]: Leaving directory '/home/tommaso/mscore/build.release' > make[1]: Entering directory '/home/tommaso/mscore/build.release' > make[1]: *** No targets specified and no makefile found. Stop. > make[1]: Leaving directory '/home/tommaso/mscore/build.release' > Makefile:38: recipe for target 'release' failed > make: *** [release] Error 2 > > 3) I have tons of > > cc1plus: warning: /home/tommaso/mscore/all.h.gch: not used because > `__STRICT_ANSI__' not defined [-Winvalid-After that pretty much everything is > lost since pch] > (gcc version 6.3.0 20170406 (Ubuntu 6.3.0-12ubuntu2) ) See my forum post. I think these are "jus t" collateral damages from non-propagating compile options and ill-defined dependencies. > 4) plus, I've just got a problem with linking, so trying to clean again the > sources (apart from 'make clean', is there another way to clean-up more?), > but 'make' would refuse again, until I made a 'git checkout Makefile' I think the most obvious problem id that MS seem to be a project with a custom, handtailored makefile that (ab?)uses cmake internally. Once you run cmake in the source root directory you've destroyed the central makefile. Cmake strongly fafours an out-of-source build (mkdir mybuild & cd mybuild & cmake ../ % make) As I wrote in my forum post, that alone doesn't work thank's to some "magic" sprinkled over the build process. > > About 3) above, 'make' seems really wasting plenty of time on these .gch not > used, seems not making progress, stuck at [1%], but now it claims I guess that as soon as pchs do work you'll see an incredible speedup during compilation ;-) What I don't understand: why not use one of the "standard" cmake solutions for precompiled headers (like, i.e. cotire)? Cheers, Ralf Mattes ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Mscore-developer mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mscore-developer
