Hi Sven
some of that is covered here:
http://flyingsheeponsailfish.blogspot.ch/2013/11/deploying-additional-packages-to.html
You should not need to map any additional drives for Qt Creator to
find headers, the SDK VM already has shared folders that access you
home drive (within which you have to install the Sailfish SDK).
One point additional point discussed in many mails over the past few
weeks is what / where can you install shared libraries to make your
app compliant with the rules of the Jolla Harbour app store. My
(shaky) understanding is that while technically you can use the Yaml
/ spec file to automatically install additional packages, for the app
store you will have to bundle these with your app (i.e. not shared).
mfg
Chris
Zitat von "Putze Sven" <sailfish...@hardcodes.de>:
Hi there,
For some time now I am wondering what the best practice would be for
using third party components. First off all I want to sketch my
understanding so far:
# I can use any RPM package that is available via "zypper" on the
emulator AKA "The SailfishOS Emulator". SSH in that thingy and
install the files. That way the package is there, where it is needed
and used in the end.
# The same package plus header files are installed in the "Mer build
engine for cross compilation", so the compiler and linker can
find/use them during compilation/linking time (and packaging time
later on)
Development is done on the "QtCreator integrated development
environment" plus some plugins for the communication with the
virtual machines (and later on the physical device). This is the
only platform dependent piece of software since the rest is done in
VMs. So far so good, now the washy part.
# For developing convenience it would be nice if the editor in
QtCreator knew about the header files that are part of the earlier
mentioned packages.
What's the best practice here? Do you guys mount a drive from the
"Mer build engine for cross compilation" to simply re-use those
header files for the QtCreator? Do you copy them over to a local
path? Am I on the wrong path? ;-)
# Finally there must be a way to tell the build system that there is
a dependency to an existing RPM package.
I assume that is somehow done in the .YAML file in the "Requires"
section. Do I simply add the name of the package here?
If that has been asked and answered before just point me to the
appropriate resource, that I have googled it wrong.
Best regards
Sven
_______________________________________________
SailfishOS.org Devel mailing list
_______________________________________________
SailfishOS.org Devel mailing list
