I apologize ahead for this long ranty post, and thank you for bearing with me 
by spending your precious time.

It is unfortunate for Nepal that joda-time does not already have support for 
Vikram Samvat let alone any other hindu calendar system. But then it may be the 
case that Nepal perhaps needs to adopt a more algorithmically straightforward 
calendar system for official purposes.

All this maybe out of the scope of the initiative :( ?

I have been desperately researching on how I can perhaps add support and I have 
found some promising leads(at the moment these are way over my head) that might 
lead to an answer:

Books/Research Paper:
        *  Calendrical Calculations 
 (Provides Common Lisp Code for Saka Samvat) 
        *  Astronomical Algorithms 
        * Indian Calendrical Calculations 
 (Common Lisp, same authors as #1) 
There are also a bunch of github repositories on the subject with different 
        *  https://github.com/espinielli/pycalcal 
(https://github.com/espinielli/pycalcal) - Python Port of "Calendrical 
Calculations" Common Lisp Code. has Saka Samvat calculations. 
        *  https://github.com/dzucconi/calendrical 
(https://github.com/dzucconi/calendrical) - Javascript (Saka Samvat) 
        *  https://github.com/kbwood/calendars 
(https://github.com/kbwood/calendars) - jQuery - has Nepali Calendar based on 
data of precalculated dates (I'd say that's a inelegant but practical hack that 
may pontentially bite the users in the *** like Y2K). This approach is what has 
been used in local software, perhaps even in proprietary Government(Tax, 
Internal Revenue) software. 

Note: My preliminary knowledge indicates that Saka Samvat is similar to Vikram 
Samvat with maybe only few differences. Vikram Samvat is more usually spelled 
as Bikram Sambat(BS) in Nepal.

My thinking is that it may suffice to have application/UI level support for 
Bikram Sambat. What I mean by that is that we keep the storage and manipulation 
of dates in AD as it is. Date/Time is stored as Unix Time isn't it anyway? We 
display and do all UI level manipulation in BS based on Calendar code we derive 
from above resources. Actually there exists in github, a Java and an AngularJS 
library too for jQuery lib above(#3 - hack) kind of approach .

There might only be a need to do a little bit of architectural change required 
in the core in the definition of months and fiscal year. If definition of month 
and fiscal year could be customized(if it is not already customizable - I know 
fiscal year is customizable), that maybe all we need, besides the UI widgets. 
If that is the case then Aliens in another galaxy could theoretically use 
MifosX? :D


I am also thinking of joining or igniting a movement for government to base 
official work be done in terms of AD like most of the world does including our 
neighbor India. As far as I have gone in finding the solution to perceived 
problem, I still do not know the government's actual position on AD usage so 
I'll be doing that soon first. It certainly is the case though that doing 
official work in BS makes things easier with government in Nepal.

Thanks Much!


September 20, 2016 4:03 PM, "Ed Cable"  wrote:

 Hi Yalu,
The Fineract developer is the right place so thanks for starting this 
discussion. Keith Woodlock, are you able to respond more about the calendar 
library that was chosen - it think it was Joda right? It looks like that only 
supports 8 calendars - http://www.joda.org/joda-time/cal.html 
(http://www.joda.org/joda-time/cal.html) - but Vikram Samvat isn't one of them. 
On Mon, Sep 19, 2016 at 8:54 PM, Yalu  wrote:

 Dear Dev Community,

This is my first post to this list and I apologize if the content of this 
message is not appropriate.

I am from Nepal and here the official Calendar System in widespread use is 
still exclusively a Hindu calendar system called Vikram Samvat. I am trying to 
get Mifos to be usable in Nepal and this is one of the major hurdle that is 
impeding adoption.

I am trying to determine what kind of Calendar Date/Time architecture/module 
Mifos X has so as to localize this aspect of the software faithfully and 
seamlessly; Whether the module is pluggable so that we could just plug in 
Vikram Samvat based calendar system(if it exists) or any other Calendar system, 
and we would be good to go.

Determining or making the architecture pluggable and having a slot to 
accommodate different kinds of Calendar/Date/Time system may only be one of a 
few related problems in this problem set. On the other side, I am not even sure 
if Nepal has a reliable computational Calendar library to plugin for Vikram 

I know there exists an approach that some local developers have been using to 
address Calendar solutions for Vikram Samvat. From what I know these systems 
are made not with pure algorithmic approach(thus scalable into the future) but 
by using pre-calculated data set of dates. These set of dates range between 
arbitrary(?) past and future dates. The authority that publishes the pre 
calculated dates and the legitimacy of their calculations may also be 
questionable. I have done some research on this and I will share more of my 
findings soon.

I would really appreciate any help localizing Mifos X in this manner. It would 
be a huge win for Nepali adoption of Mifos X. I will assist in any ways I can, 
so please ask questions, share your insights and discuss.

Thank You immensely for all your hard work. You all are all simply amazing.


 Ed Cable
Director of Community Programs, Mifos Initiative 
 edca...@mifos.org (mailto:edca...@mifos.org) | Skype: edcable | Mobile: 
Collectively Creating a World of 3 Billion Maries | http://mifos.org 
(http://mifos.org)   (http://facebook.com/mifos)

Reply via email to