Problems building Subsurface for Android

Anton Lundin glance at acc.umu.se
Thu Feb 2 12:10:36 PST 2017


On 02 February, 2017 - Joakim Bygdell wrote:

> On 2 February 2017 at 17:01, Dirk Hohndel <dirk at hohndel.org> wrote:
> 
> > >
> > > originating at the line:
> > >
> > > "$SUBSURFACE_SOURCE"/../libdivecomputer/configure --host=${BUILDCHAIN}
> > --prefix="$PREFIX" --enable-static --disable-shared --enable-examples=no
> > >
> > > Does any one have an idea?
> > >
> > > Of course there is not a libdivecomputer directory in src because I
> > started with a clean src directory.
> > >
> > > Do I need a manual install of libdivecomputer? As fa ar I am aware this
> > should not be necessary.
> >
> > It looks like the Android build.sh doesn't do that, yet.
> >
> No, but the android-build.sh sets up the current src directory correctly.
> The only time this will be an issue is when people runs the build script in
> packagin/android directly without running any of the other scripts first.
> 
> 
> > Goes to show you that whatever we do, we always make assumptions.

Yea, everything is based on lots of assumptions. I wrote that script to
make it possible for others than me to build the apk's , and at least 
both Dirk and Jocke have managed to build apk's so it can't be that
broken.


All of our build scripts makes lots of assumptions, and some of them are
silly, and others just plain broken. Is the right solution to glue even
more shell around everything?


And for everyone who tries to write a shellscript, you might think you
know what you're doing, and you most probably don't. Please use
https://www.shellcheck.net/ before you end up doing rm -rf / directory.


//Anton - Who sees the horror of shell every day


-- 
Anton Lundin	+46702-161604


More information about the subsurface mailing list