$ git clone git://github.com/rklancer/Smartgraphs.git
$ cd Smartgraphs
$ rvm update && rvm reload
$ rvm install 1.9.2
Create a Smartgraphs gemset and set RVM to use it if you (a) just cloned the repo or (b) updated your Ruby version:
$ rvm use 1.9.2
$ rvm gemset create Smartgraphs
$ rvm use 1.9.2@Smartgraphs
$ bundle install
$ echo "rvm use 1.9.2@Smartgraphs" > .rvmrc
Trust the .rvmrc file
$ mkdir -p frameworks
$ git submodule update --init --recursive
cp applets/config/config.sample.yml applets/config/config.yml
(This step is temporarily unnecessary.) See below for instructions for installing CouchDB on OS X using MacPorts or Homebrew.
(This step is temporarily unnecessary.) Try the following:
$ curl http://127.0.0.1:5984/
If you get the following response
{"couchdb":"Welcome","version":"1.0.1"}
then you're good to go. (Obviously, a version >= 1.0.1 is fine.)
You may want to set yourself up as an administrator on your local CouchDB instance. If you do not, everyone who can
access your machine via the network can administer your CouchDB instance Add a line to the file local.ini (located
in /opt/local/etc/couchdb/ if using MacPorts) as follows:
<username> = <password>
Where <username> and <password> are the username and password you want to use to access your local CouchDB instance.
Visit http://127.0.0.1:5984/_utils/ to see the web interface to CouchDB and to verify your username and password. Once
you do, your plaintext password in the local.ini file will be replaced by a hashed version.
Create the CouchDB database called smartgraphs as follows:
$ curl -X PUT http://<username>:<password>@127.0.0.1:5984/smartgraphs
The response should be:
{"ok":true}
Finally, replicate the Smartgraphs database into your local machine:
$ curl -i -H 'Content-Type: application/json' -X POST \
-d '{"source":"http://couchdb.cosmos.concord.org/smartgraphs","target":"http://<username>:<password>@127.0.0.1:5984/smartgraphs"}' http://127.0.0.1:5984/_replicate
The response should be something like:
{"ok":true,"session_id":"94c64a6984b88ff2dade30783df468b3","source_last_seq":15,
"history":[{"session_id":"94c64a6984b88ff2dade30783df468b3","start_time":"Wed, 03 Nov 2010 17:40:11 GMT",
"end_time":"Wed, 03 Nov 2010 17:40:12 GMT","start_last_seq":0,"end_last_seq":15,"recorded_seq":15,
"missing_checked":0,"missing_found":9,"docs_read":9,"docs_written":9,"doc_write_failures":0}]}
More information about replicating couchdb databases
If you will be using the sensor applet, start the applet server in another console window:
$ ruby applets/bin/local_server.rb
(in the root of the Smartgraphs project:)
$ sc-server -v
This will take over the Terminal window. Do subsequent work in a new Terminal window.
If you visit http://localhost:4020/ you should be greeted by the SproutCore Welcome app; if you visit http://localhost:4020/db/_utils/ you should be greeted by the CouchDB web administration app, Futon.
If these addresses work, visit http://localhost:4020/smartgraphs to see Smartgraphs in action.
To use TestRunner, open http://localhost:4020/sproutcore/tests The currently valid tests are written using Jasmine. Ignore the QUnit folder.
To visit all integration tests directly, open http://localhost:4020/static/smartgraphs/en/current/tests/jasmine/integration.html
To visit all unit tests directly, open http://localhost:4020/static/smartgraphs/en/current/tests/jasmine/unit.html
The following approach is not ideal but works until we improve it. Currently these files are at apps/tests/jasmine/integration/src and apps/tests/jasmine/unit/src When working with them it is most useful to have them be compiled to js files automatically.
- Install node
- Install jitter (note: you might want to learn more about npm so you know what this command is doing) sudo npm install --global jitter
- run jitter to watch that folder and compile to the parent jitter apps/smartgraphs/tests/jasmine/integration/src apps/smartgraphs/tests/jasmine/integration
How to install CouchDB on OS X using macports
$ sudo port install couchdb
$ sudo dscl localhost
> cd /Local/Default/Users
/Local/Default/Users > change couchdb dsAttrTypeNative:home /dev/null /opt/local/var/lib/couchdb
/Local/Default/Users > change couchdb dsAttrTypeNative:shell /dev/null /bin/bash
$ sudo chown -R couchdb:couchdb /opt/local/var/lib/couchdb
$ sudo chown -R couchdb:couchdb /opt/local/var/log/couchdb
$ sudo chown -R couchdb:couchdb /opt/local/etc/couchdb
$ sudo launchctl load -w /Library/LaunchDaemons/org.apache.couchdb.plist
How to install CouchDB on Mac OS X using homebrew
$ brew install couchdb
Follow the instructions displayed after a successful installation.
These instructions can also be displayed with the following command:
$ brew info couchdb
couchdb 1.0.1
http://couchdb.apache.org/
Depends on: spidermonkey, icu4c, erlang
/usr/local/Cellar/couchdb/1.0.1 (281 files, 2.4M)
If this is your first install, automatically load on login with:
cp /usr/local/Cellar/couchdb/1.0.1/Library/LaunchDaemons/org.apache.couchdb.plist ~/Library/LaunchAgents
launchctl load -w ~/Library/LaunchAgents/org.apache.couchdb.plist
If this is an upgrade and you already have the org.apache.couchdb.plist loaded:
launchctl unload -w ~/Library/LaunchAgents/org.apache.couchdb.plist
cp /usr/local/Cellar/couchdb/1.0.1/Library/LaunchDaemons/org.apache.couchdb.plist ~/Library/LaunchAgents
launchctl load -w ~/Library/LaunchAgents/org.apache.couchdb.plist
Or start manually with:
couchdb
http://github.com/mxcl/homebrew/commits/master/Library/Formula/couchdb.rb
$ curl -i -H 'Content-Type: application/json' -X POST \
-d '{"source":"http://<remote_host>/smartgraphs","target":"http://<user>:<password>@127.0.0.1:5984/smartgraphs"}' http://127.0.0.1:5984/_replicate
Do not include the <user>:<password>@ section in the target url unless it is required to write to your local database.
More information about replicating couchdb databases
On OS X, turn on Web Sharing via (Apple Menu) -> System Preferences -> Sharing -> Web Sharing
Now, make sure that virtual hosting is enabled by editing /private/etc/apache2/httpd.conf and uncommenting the
virtual hosting line (at about line 465 of the stock httpd.conf) as follows:
# Virtual hosts
Include /private/etc/apache2/extra/httpd-vhosts.conf
At the top of httpd-vhosts.conf enable name-based virtual hosts for port 80 on all interfaces:
NameVirtualHost *:80
Edit the virtual hosting configuration file /private/etc/apache2/extra/httpd-vhosts.conf to include the entry:
<VirtualHost *:80>
ServerName sc.local
ServerAdmin webmaster@localhost
DocumentRoot "/opt/local/www/dummy"
ProxyRequests Off
KeepAlive Off
<Proxy *>
Order deny,allow
Deny from all
Allow from 127.0.0.1
</Proxy>
ProxyPass /db/ http://127.0.0.1:5984/ nocanon retry=0
ProxyPassReverse /db/ http://127.0.0.1:5984/
ProxyPass / http://127.0.0.1:4020/ retry=0
ProxyPassReverse / http://127.0.0.1:4020/
</VirtualHost>
after making changes ...
- test the config:
apachectl configtest
(apachectl may complain that the directory /opt/local/www/dummy doesn't exist. You can create it if you like,
possibly at an alternate location of your choosing.)
- restart apache:
sudo apachectl restart
(For more instructions, set http://shapeshed.com/journal/setting_up_local_websites_on_snow_leopard/.)
And, finally, edit your /etc/hosts file to include the following line:
127.0.0.1 sc.local
Confirm that the new entry works:
$ dscacheutil -q host -a name sc.local
name: sc.local
ip_address: 127.0.0.1
It might be necessary to flush the local DNS cache:
$ sudo dscacheutil -flushcache
sc-build-number smartgraphs
If you have a large or 'speculative' set of commits to make, don't be afraid to create a new branch to hold your work as you go.
If you are making smaller or more routine commits, it's better not to create a new branch. However, if others are working on the master branch at the same time, this will create unnecessary merge commits as you reconcile their work with yours (or vice versa).
Therefore, please checkout a local branch, called for example 'tmp', before starting work:
(master) $ git checkout -b tmp
(tmp) $
When you are ready to push your work to the repo, check for the latest updates to master:
(tmp) $ git checkout master
(master) $ git pull
There should be no merges or conflicts, because you didn't make changes to master.
Then, rebase the changes from 'tmp' onto 'master':
(master)$ git checkout tmp
(tmp) $ git rebase master
First, rewinding head to replay your work on top of it...
...
(tmp) $ git checkout master
(master) $ git merge tmp
If there are rebase conflicts, you have to deal with them at this step.
Once you are sure the work is ready, push your changes:
(master) $ git push origin master
Ideally, you will have done this in a short window of time and git push will report no conflicts. If there are, you
can merge 'master' into 'tmp' again:
(master) $ git checkout tmp
(tmp) $ git merge master
(tmp) $
Then reset 'master' to the last common commit with origin/master, and repeat the pull-rebase cycle.
Once you have successfully pushed your changes, don't forget to update 'tmp' to point to 'master' again:
(master) $ git checkout tmp
(tmp) $ git merge master
(tmp) $