test with the MCD6.1alpha, compilation with python 3.6

[antoine-bierjon]

compile_fmcd.sh fails at his first call to f2py (which thus doesn't create a fmcd.pyf file).
The log for this first f2py call (/!\ this log is erased in default version by the 2nd call to f2py in compile_fmcd.sh, see issue #45 ) ends with these lines :

Traceback (most recent call last):
File "/opt/python3-anaconda50/bin/f2py", line 28, in
main()
File "/opt/python3-anaconda50/lib/python3.6/site-packages/numpy/f2py/f2py2e.py", line 650, in main
run_main(sys.argv[1:])
File "/opt/python3-anaconda50/lib/python3.6/site-packages/numpy/f2py/f2py2e.py", line 408, in run_main
postlist = callcrackfortran(files, options)
File "/opt/python3-anaconda50/lib/python3.6/site-packages/numpy/f2py/f2py2e.py", line 329, in callcrackfortran
postlist = crackfortran.crackfortran(files)
File "/opt/python3-anaconda50/lib/python3.6/site-packages/numpy/f2py/crackfortran.py", line 3246, in crackfortran
readfortrancode(files, crackline)
File "/opt/python3-anaconda50/lib/python3.6/site-packages/numpy/f2py/crackfortran.py", line 363, in readfortrancode
l = fin.readline()
File "/opt/python3-anaconda50/lib/python3.6/fileinput.py", line 299, in readline
line = self._readline()
File "/opt/python3-anaconda50/lib/python3.6/codecs.py", line 321, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xd7 in position 7369: invalid continuation byte

[antoine-bierjon]

By comparing with log files from compiling versions (MCD5.3/py2.7, MCD5.3/py3.6, MCD6.1/py2.7), the error seems to arise between the usual messages :
" Line #256 in constants_mcd.inc:" real ma3d(nbvar3d) /-1.0, -1.0, -1.0, -1.0, -1.0, 18.0, -1.0, 44.0, 28.0, 16.0, 32.0, 48.0, 1.0, 2.0, 28.0, 40.0, -1.0, 4.0, -1.0, -1.0, -1.0/"
updatevars: no name pattern found for entity='-1.0/'. Skipping. "

AND

" Post-processing... "

[antoine-bierjon]

Actually, since this message :
" Line #256 in constants_mcd.inc:" real ma3d(nbvar3d) /-1.0, -1.0, -1.0, -1.0, -1.0, 18.0, -1.0, 44.0, 28.0, 16.0, 32.0, 48.0, 1.0, 2.0, 28.0, 40.0, -1.0, 4.0, -1.0, -1.0, -1.0/"
updatevars: no name pattern found for entity='-1.0/'. Skipping. "
appears multiple times in the log file, the crash can happen at an occurrence other than the last one.
Especially, when comparing with log from MCD6.1/py2.7, it may better correspond to the occurrence of this message placed at the line 4342/4343, hence occurring just before the message :
" Reading file 'constants_mcd.inc' (format:fix)" (which also appears multiple times)

[aymeric-spiga]

Sometimes problems like this arise when there is an accent in a comment, or a tab somewhere, or a weird character