epsi-builds/msspec_python3/pipeline/head This commit looks goodDetails
In the top Makefile, the Python path and version are first checked
before going any further. This is now done in a more cleaner way
by looking at the "sys.version_info variable"
epsi-builds/msspec_python3/pipeline/head This commit looks goodDetails
It is impossible to compute the crosssection if LMAX > 30.
This is due to a lack of precision in the djmn.f subroutine which
is written in simple precision. This commit is a first step to promote
this subroutine and all its dependensices to doucle precision.
epsi-builds/msspec_python3/pipeline/head This commit looks goodDetails
Just began to rewrite the topmost Makefile which will provide
targets for:
* runing the tests
* creating the Python venv
* building and installing the code
* building the manual and the website
Makefiles in subfolder were also upgraded to fix some issues in
target filenames.
epsi-builds/msspec_python3/pipeline/head This commit looks goodDetails
Makefiles are easier to maintain, I eventually prefer using GNU Make
instead of scons. SConstruct files will be removed when makefiles will
be ready in a future commit.
epsi-builds/msspec_python3/pipeline/head There was a failure building this commitDetails
The automatic resizing is not working very well in WXPython along with
matplotlib with the WxAgg backend. It is an attempt to improve this but
does not change anything unfortunately. This is a problem of WX + mpl.
When setting calc.spec_parameters.iprint=3, Spec prints out a huge
amount of information about path taken into account using the series
expansion algorithm.
Some files were not closed in spec at the end of the program. This
caused some troubles in runing the phagen-spec cycle several times.
A function (CLOSE_ALL_FILES) was added in misc.f to force opened
units in the range [7,200] to be closed before exiting the main
subroutine.
The Curve comparison is possible through R-Factor analysis.
Further comparison like shape analysis, similarity index... will
be included later.
Data export is now possible through the gui menu or as a method of
the Data object.
X, Y, Z axes were not in the right-hand orientation and the
Z-stack for drawing axes was not updated resulting in strange
graphical output (axis could "jump" over another instead of being
drawn behind)
A module (cli.py) is used to provide an entry point for the msspec
package. It is here to be PEX-compatible (https://github.com/pantsbuild/pex)
that may be used to provide a zero-install msspec distribution.
F2PY can generate inconsistent suffix for *.so files due to a
different version of numpy if not the same as the one used
in the virtualenv. The workaround is to remove the
"cpython-36m-x86_64-linux-gnu"-like part of the name. The import
mechanism works well without. This is a tiny hack though.
The f2py *.so suffix is not consistent with python virtualenv version.
By using the source dir it is easier to fix manually since the *.so are not
installed
The sys module should be imported first to decalre msspec
path as soon as possible. It enables to load the version module
necessary to avoid hardcoded version number in the setup.py file.
Please note that this import order is not pep8 compliant.
Sylvain Tricot
sylvain tricot
sylvain tricot