idaholab / stork

DEPRECATED

THIS REPOSITORY IS DEPRECATED.

To create new MOOSE-based applications, use the stork.sh script in the main MOOSE repository. For more information, see http://mooseframework.org/create-an-app/

stork's Issues

Need new revision logic

idaholab/moose#4027 Contains new revision logic. It needs to be incorporated into Stork.

doxygen fails from fresh cloned stork

warning: Tag `SHOW_DIRECTORIES' at line 519 of file doc/doxygen/Doxyfile has become obsolete.
To avoid this warning please remove this line from your configuration file or upgrade it using "doxygen -u"
warning: Tag `HTML_ALIGN_MEMBERS' at line 902 of file doc/doxygen/Doxyfile has become obsolete.
To avoid this warning please remove this line from your configuration file or upgrade it using "doxygen -u"
warning: Tag `USE_INLINE_TREES' at line 1089 of file doc/doxygen/Doxyfile has become obsolete.
To avoid this warning please remove this line from your configuration file or upgrade it using "doxygen -u"
warning: tag INPUT: input source `../elk' does not exist
warning: source ../elk is not a readable file or directory... skipping.

make_new_module.py creates template that uses old constructors

The <my_module>App.C file created by make_new_module.py got created with old-style constructors. This issue may also apply to other scripts in Stork -- I just ran into this issue when I tried using that particular script.

"module" prototype file not getting removed

the app.C.module file is not getting removed when you run ./make_new_application

make_new_application.py is not python3 compatible

If a user tries to run ./make_new_application.py with python3.x, the script will fail.

Add the <App>Revision.h file to .gitignore

Also, make sure that the create application script finds and updates the .gitignore file once a user creates a named application.

InterfaceOrientationMaterial not accepting input parameters

Hello,

When trying to define parameters in this material, I'm running into the issue of the material not recognizing the input parameters listed in the source code. If I only define "block" and "c", the input file will run witht the default parameters. However, when I define an input parameter like I do in this input file, I get the error "input parameter not used". Furthermore, most of these input parameters do not appear when I open the material in Peacock.

Atom seems to recognize the input parameters. I've recompiled MOOSE and it tested fine. I'm using Terminal on a Mac desktop running El Capitan. Someone mentioned that there were Windows Newlines in my input file which may be the issue, but I was unable to find or remove them.

Any thoughts on what the issue with the material or my input file might be? Forgive me if I'm misusing Github or not following the MO for working through these types of things.

Many thanks,
Patrick Shower

newly created module unit Makefile slightly incorrect

MOOSE_DIR ?= $(shell dirname pwd)/../moose

should possibly be (?)

MOOSE_DIR ?= $(shell dirname pwd)/../../../moose

Look into removing internal stork and using this version for new applications

Remove reference to Herd in Makefile.module

Really minor, but I noticed when modifying the porous_flow makefile that the comments reference the HERD_TRUNK_DIR. This should be changed to MODULE_DIR.

Enable the ability to have src/include for tests in separate location

I would really like Pika to have tests/src and tests/include or something similar. So, when a pure user is getting started with input files and Peacock they do not see a whole bunch of objects that are truly not indented for use, but are there for testing purposes. For example, we have a set of Kernels for MMS forcing functions, an end-user should be using or seeing these.

EOF in simple_diffusion tests file from make_new_module

At the moment it doesn't end in a line-feed but it needs to to conform with our coding standards

App names with dashes (or any C++ operators) generate invalid class names

MyAPP-Extreme::registerObjects(Factory & factory) will not compile...

new module arguments for App constructor

the make_new_module script should produce, for example:

PorousFlowApp::PorousFlowApp(const InputParameters &parameters) :
MooseApp(parameters)

and

public:
PorousFlowApp(const InputParameters &parameters);

Currently it still has the old moose version that includes the "name" argument.

Needs update for new output syntax changes

Stork needs to be updated for the new output syntax changes.

Currently there is no solver output at all (Is that a bug??)

Add "lib" directory to .gitignore

On linux you end up with a directory named lib that has your built libraries in it. This should go into the .gitignore

Warn people when MOOSE submodule is out of date

This is what we have in RELAP-7's Makefile:

CURRENT_DIR        := $(shell pwd)
APP_DIR            ?= $(CURRENT_DIR)

# moose submodule status
moose_status := $(shell git -C $(APP_DIR) submodule status 2>/dev/null | grep moose | cut -c1)
ifneq (,$(findstring +,$(moose_status)))
  ifneq ($(origin APP_DIR),environment)
    moose_status_msg = "WARNING: Your MOOSE submodule is out of date.\n"
  endif
endif

all:: moose_submodule_status

moose_submodule_status:
    @if [ x$(moose_status_msg) != "x" ]; then echo $(moose_status_msg); fi

When people's MOOSE submodule is out of date, the first message they see is the WARNING line.

This seemed to be working for us for some time, so should we add this into stork? There might be some corner cases with this that I do not see....