git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7100 2fb0f335-1f38-0410-981e-8018bf24f1b0

now it will fail if the fluid system tries to access any saturation or
a quantity for a phase which was not queried. also, the parameter
cache may no longer access quantities of phases which have not been
modified in the updatePhase*() methods.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7099 2fb0f335-1f38-0410-981e-8018bf24f1b0

dammed copy & paste!

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7098 2fb0f335-1f38-0410-981e-8018bf24f1b0

if new fluid systems become available, the test should be
extended. also, there the same test should be used for all fluid
systems in dumux-devel.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7097 2fb0f335-1f38-0410-981e-8018bf24f1b0

now we assume an ideal mixture of the gas phase again. we probably
want that, as the 2p2c model as currently implemented does not support
non-ideal mixtures, because the MiscibleMultiPhaseComposition
constraint solver does not support this yet. (On a different note, a
reference is missing for the approach taken for the fugacity
coefficients of the gas phase for the "gas as non-ideal mixture"
code.)

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7096 2fb0f335-1f38-0410-981e-8018bf24f1b0

if the mean molar mass is zero, then the sum of the mass fractions
will be zero anyway, which means that the result should be still correct.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7095 2fb0f335-1f38-0410-981e-8018bf24f1b0

and remove the "FluidSystem" postfix from the class names. e.g.
Dumux::H2ON2FluidSystem becomes Dumux::FluidSystems::H2ON2 . this
commit does not change anything beyond that...

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7094 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7093 2fb0f335-1f38-0410-981e-8018bf24f1b0

causes compilation to fail

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7092 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7091 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7090 2fb0f335-1f38-0410-981e-8018bf24f1b0

"newer" means opensuse 11.4 with GCC 4.6.1

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7089 2fb0f335-1f38-0410-981e-8018bf24f1b0

this means GCC 4.2.4, automake  1.10.1, autoconf 2.61, autoheader 2.61, libtool 1.5.26

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7088 2fb0f335-1f38-0410-981e-8018bf24f1b0

it determines pressures + saturations if there are compressible fluid
phases, if not it determines all saturations but leaves the pressures
alone...

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7087 2fb0f335-1f38-0410-981e-8018bf24f1b0

it returns -- guess what? -- whether a fluid phase is assumed to be
incompressible. this method is useful e.g. for flash calculations.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7086 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7085 2fb0f335-1f38-0410-981e-8018bf24f1b0

decoupled 2p2c: do not define the BoundaryMobility property for the DecoupledTwoPTwoC type tag twice

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7084 2fb0f335-1f38-0410-981e-8018bf24f1b0

these are no longer required in the new framework

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7083 2fb0f335-1f38-0410-981e-8018bf24f1b0

also make the densities in the water-nitrogen fluid system
consistent. this makes the results identical to to ones obtained by
the new fluid framework

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7082 2fb0f335-1f38-0410-981e-8018bf24f1b0

oops.

if "useSimpleRelations" in the H2O-N2 fluid system is set to true and
the density of the phases is not normalized by sum of the mole
fractions, the results are identical with the ones on trunk.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7081 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7080 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7079 2fb0f335-1f38-0410-981e-8018bf24f1b0

the fugacity coefficient must be independent of the fugacity since its
whole point is to, well, define the fugacity. if this worked before,
then probably only because the fugacity was undefined and smaller than
10^-3...

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7078 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7077 2fb0f335-1f38-0410-981e-8018bf24f1b0

where correct means "identical to the reference solution". so far,
only a reference solution for the test of the 2p box model is provided

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7076 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7075 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7074 2fb0f335-1f38-0410-981e-8018bf24f1b0

gets rid of the deprecation warnings for the box models

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7073 2fb0f335-1f38-0410-981e-8018bf24f1b0

defaults for properties come with quite a few issues, probably the
most galling one is that they can only be specified once in the whole
code base. This means so they are best avoided.

Further, the PTAG() in the GET_PROP* is not useful in 99% of all
cases. It was there to also allow full c++ types (e.g. template
arguments) as parameters. For the code that really needs this, there
now is GET_PROP*_ and PTAG_ (mind the underscores). To my knowledge
that only affects the run-time parameter system. The parameter system
has been adapted, everyone else can now write GET_PROP_TYPE(TypeTag,
Scalar) instead of GET_TYPE_PROP(TypeTag, PTAG(Scalar)). (the PTAG
macro is still around but just does nothing.)

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7072 2fb0f335-1f38-0410-981e-8018bf24f1b0

also rename the "fluidsystems" directory to "old_fluidsystems". Since
this is a separate SVN branch, nobody's feelings should be hurt by
this commit...

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7071 2fb0f335-1f38-0410-981e-8018bf24f1b0

the build broke because EquilibriumFluidState was renamed to
CompositionalFluidState

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7070 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7069 2fb0f335-1f38-0410-981e-8018bf24f1b0

also add a test

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7068 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7067 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7066 2fb0f335-1f38-0410-981e-8018bf24f1b0

Implemented completeFluidState for 1p, 1p2c, richards. Moved convenience function to the base box model.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7065 2fb0f335-1f38-0410-981e-8018bf24f1b0

Moved the actual implementation of completeFluidState to the volume variables. For the box models, this appears to be the natural place where it belongs to. Inserted a convenience function in the model.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7064 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7063 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7062 2fb0f335-1f38-0410-981e-8018bf24f1b0

All fluid systems which do not need a parameter cache now offer both the simple and the general interface.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7061 2fb0f335-1f38-0410-981e-8018bf24f1b0