Decorana

[jobdiogenes]

Some comments.

In readme say that exe binary are for Windows, so, decorana was made for Mainframes in Fortran.
and could be compiled to all platforms that support GCC Gfortran are supported.

I'm converting R metacom to Python, and found this issue. because use a decorana.
original decorana binary could be build in unix just using

gfortran --static decorana.f -o decorana

in windows, just add .exe at end.

decorana R source, its little different.
they use some original routines. as a external Fortran call, from a C program.
in that C program they adjust data to be processed to decorana routines.
its a two layers solution. R code, that cal a C code that call a Fortran code.
They have reason for that, but in python we could follow another approach.

• Calling

1. use decorana.F as external module with some adjusts in fortran code. using f2py.
2. use python lib fanalysis to do CA, in testing this now, and eig values are the same as come from R decorana. row,col projections I'm still checking
3. create a python interface that call decorana.F routines, as R decorana did, bit without de C layer.

I'm looking to all these.

I'm glad to read that data are in place to be run by decorana.
So, then could call FORTRAN easily.

I did a fork, and could do some tests to ecopy.

anyone could mail me to jobdiogenes - gmail.

I really like to help efforts to ecology with python .

[maurobio]

Hi. I found your English quite impossible to understand, but you may want to take a look at this: https://pypi.org/project/cornpy/ Good luck! Em sex., 2 de out. de 2020 às 16:20, Job Diógenes Ribeiro Borges < notifications@github.com> escreveu:

…

Some comments. In readme say that exe binary are for Windows, so, decorana was made for Mainframes in Fortran. and could be compiled to all platforms that support GCC Gfortran are supported. I'm converting R metacom to Python, and found this issue. because use a decorana. original decorana binary could be build in unix just using gfortran --static decorana.f -o decorana in windows, just add .exe at end. decorana R source, its little different. they use some original routines. as a external Fortran call, from a C program. in that C program they adjust data to be processed to decorana routines. its a two layers solution. R code, that cal a C code that call a Fortran code. They have reason for that, but in python we could follow another approach. - Calling 1. use decorana.F as external module with some adjusts in fortran code. using f2py. 2. use python lib fanalysis <https://github.com/OlivierGarciaDev/fanalysis> to do CA, in testing this now, and eig values are the same as come from R decorana. row,col projections I'm still checking 3. create a python interface that call decorana.F routines, as R decorana did, bit without de C layer. I'm looking to all these. I'm glad to read that data are in place to be run by decorana. So, then could call FORTRAN easily. I did a fork, and could do some tests to ecopy. anyone could mail me to jobdiogenes - gmail. I really like to help efforts to ecology with python . — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#28>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAPMACN6MHPJTY2IMVFAVZ3SIYRZDANCNFSM4SCCAMOQ> .

[jobdiogenes]

Dear @maurobio ,

I found you answer very unpleasant, but you may want to take a look at this: You repo cornpy before trying to advise someone that come here to contribute.
Good Luck.
Guide to build OpenSource communities:

[maurobio]

One cannot communicate with anyone else using hardly understandable language. And please notice that, besides, your unintelligible comments were quite misplaced, as "cornpy" is an independent project, without any direct connection with the "ecopy" package. Good luck. Em sex., 2 de out. de 2020 às 18:41, Job Diógenes Ribeiro Borges < notifications@github.com> escreveu:

…

Dear @maurobio <https://github.com/maurobio> , I found you answer very unpleasant, but you may want to take a look at this: You repo cornpy <https://github.com/maurobio/cornpy/network/members> before trying to advise someone that come here to contribute. Good Luck. Guide to build OpenSource communities: <http://opensource.guide/building-community/> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#28 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAPMACMALZDDEFSHNYY5G7TSIZCKLANCNFSM4SCCAMOQ> .

-- Dr. Mauro J. Cavalcanti E-mail: maurobio@gmail.com Web: http://sites.google.com/site/maurobio "Life is complex. It consists of real and imaginary parts."

[jobdiogenes]

@maurobio

Funny, ecopy has 6 members, but only you, a Brazilian like me, come here, in a very unpleasant way, to be unpolished to someone that come here to contribute, and answer to promote your repo as an answer to my write. Yes, your project are independent. But you appoint to be an answer to my comments. Check your repo, and you will know, that I did a fork, some days ago, I get the code and make some tests. Due this, I could write about "gfortran ---". Could be anyone else that come here to contribute, could be in a very bad English. Even native English writers, could not be expressive.
A good advise to anyone member of opensource project, is to be educated to others.

Cheers

[maurobio]

Because anyone else will not spend time trying to understand your unintelligible English, much less replying to a meaningless 'question' made by someone who seems unable to see that the compy project has absolutely no relationship to ecopy. Therefore, for ecopy developers, you question, besides unintelligible, is just white noise. By the way, I am not here to 'educate' anyone, and do not feel obliged to be 'educated' (you did mean 'polite', in your very broken English) to morons who still take 'open source' as a synonym of FREE SOFTWARE. Last, but not least: I'm not 'Brazilian' - I'm stateless.