[Ndlb-developers] ndlb project is now set on source forge

[Andres M. <amo...@gm...>]

Hi guys,

We will use this e-mail for all development related communications so that a
log is kept.

Right now Pema, Sanjay, Tenzin, and me are on it, since those were the only
source forge accounts I knew. For now Pema and I are the project
administrators. Pema, I leave it up to you if you want to give administrator
privileges to others. Sanjay and Tenzin I put as developers.

Dechen, Uden, and Rinzin, please send either me or Pema your source forge
account, or to create one if you don't have ( http://sf.net ). Either Pema
or I can add you as developers. 

I have uploaded the Ruby on Rails code to an svn repository at source-forge
and I have added a hook to send the commit diffs and comments to ndlb-svn.

Only source code is there; no data at all (so when running the migrations
data has to be copied into the migration folders separately, such as the
dictionary files, the pictures, etc.). This takes care of the issues of
privacy. Since the project started from the beginning in the spirit of GPL,
I consider that it is ok for other people to look at it; regardless, I
consider very few people will since there is so much stuff out there and it
we aren't developing a general-use application, but something tailored for
our specific needs.

The repository is organized like this:

   portal
     |
    ror
   /   \
tags   trunk

All web applications should be under "portal". The off-line tools that we
develop would be in another folder. Under "ror" is the Ruby on Rails
application that will be only one (all new modules will be subsumed under
one big RoR application). Web applications developed under other platforms
would still be under portal but not under "ror". "trunk" has the edge code
where we commit the new code and "tags" has stable versions marking the
various phases that we will reach in time. We are currently at release 0.1!

The production server should check-out from:
https://ndlb.svn.sourceforge.net/svnroot/ndlb/portal/ror/tags/release-0.1

And with each new release a switch to it should be run on the server's
working copy.

The development machines should check-out from:
https://ndlb.svn.sourceforge.net/svnroot/ndlb/portal/ror/trunk

Please check out a fresh copy. In eclipse do file -> new -> project... ->
SVN -> Check Projects from SVN -> Create new repository location ->
https://ndlb.svn.sourceforge.net/svnroot/ndlb -> and then browse to
portal/ror/trunk -> Check out as project in workspace -> project name
"ndlb". 

the svn location you can write https://ndlb.svn.sourceforge.net/svnroot/ndlb
and then

A lot of files have been revised, including all migrations. Remember to fix
the config/database.yml with the appropriate socket, user, and password (for
linux it is probably much faster to connect via socket instead of host).

If you don't have the mysql databases you should create them
(ndlb_development, _test, and _production) with utf8 as default character
set (create database ndlb_development default character set utf8;) as well a
the user to connect to it (grant all on ndlb_%.* to rubyuser@'localhost'
identified by 'rubypassword';). For the new database you should run:

% rake globalize:setup

And for the other two environments as well (rake globalize:setup
RAILS_ENV=test, rake globalize:setup RAILS_ENV=production)

If you already have a database, try "rake db:migrate VERSION=0" to drop all
tables (except the globalize tables which where not created by the
migration, but by the above statement; so in the server we won't loose the
translated messages).

Then after commenting the stuff under "Code for globalization" in
config/environment.rb and adding the extra files (for the data importation,
see below) run:

% rake db:migrate

(also for the other two environments as well: rake db:migrate
RAILS_ENV=test, and rake db:migrate RAILS_ENV=production).

And then uncomment those lines again. Copy into the new project any files
that you have modified and revise them running right-click -> compare with
-> latest from repository and do the necessary changes. Then commit them.
After all of us are on the same page and all code that we have done is there
we can assess were we are and make a plan on how to move forward.

Pema, I will write a separate e-mail to keep you up to date on what we
discussed on the various meetings at the University of Virginia in my latest
visit.

Sanjay, I will send you the CSV with the location names that worked for the
migration

Dechen, I will send you the three sample dictionary files that worked with
the migration, so that you can add all letters to the English to Dzongkha
dictionary.

Again all data I will send separately so that it is not in source forge (nor
in the mailing lists).

For the images, under db\migrate\images create folders with the names of
places (exactly matching what was imported for location names) and on those
folders put the images that should be associated to that place. DO NOT
create a tree structure there, only the name of the folder for the district,
village, or block where the picture should be at the first level.

Rinzin, the xml and xslt files we will not put under portal/ror. Let me
figure out were we will put them and then let you know.

Take care and let me know when you have updated the svn server with the
latest code for me to revise and then we can move forward,

	Andres