lp:~simon-funke/fluidity/fluidity_tsunami

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

3465. By Simon Funke on 2011-05-13

a first draft of the run up forest problem described in the thuy paper.

The test already has the correct domain, but the trees are not simulated yet.

3464. By Simon Funke on 2011-05-13

diamond option which specifies the exponent order of the depth term in the manning strickler drag.

3463. By Stephan Kramer on 2011-05-03

Fix for SP build.

3462. By Patrick Farrell on 2011-05-01

In the previous change to the build process, I remarked that deleting libfluidity.a
by itself was a bad idea. Well, it turns out we had it configured so that if any
file failed to build, it deleted libfluidity.a, thus making the next build fail
confusingly. This commit turns off that deletion of libfluidity.a, so people can
make mistakes to their heart's content.

3461. By Stephan Kramer on 2011-04-29

Fix flredecomp for gfortran 4.5. Pass input and output basename strings from c to fortran using bind(c).

3460. By Patrick Farrell on 2011-04-28

An orthogonal change to the build process.

As discussed in yesterday's dev meeting, I've replaced

default: $(OBJS)
  ar ...

with

default: $(LIB)
$(LIB): $(OBJS)
  ar ..

which does /not/ execute the ar if nothing has been built. This makes
compiles from mostly-built sources much faster: on my laptop, the time
taken for make in a fully-built checkout goes from ~12s to ~4s. It would be
even faster if we could properly state the dependencies of check_options
and register_diagnostics; that would probably get it down to a second or so.
It would be even faster if we then properly stated the dependencies of
bin/fluidity, so we didn't have to recompile that unnecessarily. Then it would
take 0s!

Admittedly, this is still a little experimental. There is one known bug
that this introduces into the build system: if someone deletes lib/libfluidity.a
without deleting the */*.o, the build will fail. So don't do that (use make clean
instead, of course).

If this causes problems for anyone, feel free to revert it (or better yet, improve it).

3459. By Stephan Kramer on 2011-04-28

Finally getting rid of new_options logical.

  !------------------------------------------------------------------------
  ! Global flag for whether we are running off the new xml options file.
  ! Legacy variable - to be removed in the future
  !------------------------------------------------------------------------
  logical, parameter :: new_options = .true.

Today is the future!

3458. By Simon Mouradian on 2011-04-27

Small bug fix.

(This is the final commit, henceforth merge requests will be made)

3457. By Jon Saunders on 2011-04-27

Adding structures to the multiphase schema to enable us
to read in eos, rel perm and capillary pressure options
as required by the prototype code

3456. By Patrick Farrell on 2011-04-27

Add some documentation to the python diagnostic algorithms, mentioning
that 'time' and 'dt' are also set.

Sort the diagnostic algorithms so that the python diagnostic (the most common)
is first, and then the rest are sorted alphabetically.