This project is meant for easy sharing of our version of LabRAD tools that are useful to the entire lab.

### What this Is ###
"LabRAD" refers to a variety of software packages based on this: http://en.wikipedia.org/wiki/LabRAD
The most detailed documentation on its motivation and design is section 7 of this thesis: http://web.physics.ucsb.edu/~martinisgroup/theses/Ansmann2009ds.pdf
The original software no longer has a tutorial but can be downloaded here: http://sourceforge.net/projects/labrad/
In most cases however, you will want to git pull the software from our page: https://github.com/HaeffnerLab/Haeffner-Lab-LabRAD-Tools

### How to Use ###
Instructions for using LabRAD with AMO labs are being written here: https://github.com/AMOLabRAD/AMOLabRAD/wiki
Follow those installation steps and basic tutorial, being aware that pylabrad is a python library that is stored separately from LabRAD.
Also note that the guide is still under construction, so it will not have everything you need.
You must also install paramiko or similar: http://www.paramiko.org/
Run labrad.exe and click on the button in the top left corner, which should list "manager" and "registry" in the center.
Open this in the command line to display "data vault": https://github.com/HaeffnerLab/Haeffner-Lab-LabRAD-Tools/blob/master/dataservers/dvascii.py
You will also need to put data in the data_vault, so put the relevant csv files here: https://github.com/HaeffnerLab/Haeffner-Lab-LabRAD-Tools/tree/master/dataservers
Note that you will need to create a folder named "__data__" and store your data in their own folders inside it.
You may need to rename the folder to "data" (or similar) and then edit the path defined in the registry (which will be created after starting LabRAD).
Finally, you need to allow your computer access by enabling it from another computer running LabRAD, as its IP may be blocked.

### Graphing ###
The most complicated section of this code is the pygrapherlive module: https://github.com/HaeffnerLab/Haeffner-Lab-LabRAD-Tools/tree/master/clients/pygrapherlive
For testing, you can use it directly with grapher, which uses the other scripts, described below:
canvas         - displays the plots on a Qt canvas
connections    - connects the displayed windows to the data_vault
dataset        - prepares incoming data for plotting
datavault      - refers to the data stored in the data_vault 
grapherwindow  - defines the objects (windows) that surround the canvas; right click options were added later
qt4reactor     - uses twisted to interface with the Qt GUI software, which LabRAD is based on
Other files are less important but some are notable, specifically ones that were added later and make the software more complex:
analysis       - added for processing data 
analysiswindow - allows for curve fitting
fitcurve       - what the other fit scripts are based on; in order to acquire the data, it must nonmodularly refer back 3 parent levels to the data_vault through grapher and connection

### Editing Grapher ###
The data from the data_vault is referenced separately for plotting (canvas) and curve fitting (analysiswindow). 
Note that the buttons directly over the plot (e.g. save) are matplotlib widgets and not part of the grapher software here.
Most interaction occurs through the checkboxes, which are controlled by methods of the class GreapherWindow in grapherwindow.
ParameterWindow and AnalysisWindow are classes that are opened by the "parameters" and "fit" options on right click, and both reference the data_vault.

### Publishing ###
The "publish" option (which allows the user to publish data directly to the webblog) has been added.
It includes changes to grapherwindow and a new script names publish_parameters.
Be sure to fill out publish_parameters properly; instudtions are in its comments.
It opens a parameter window much like the "parameters" and accesses the data in the same way that "fit" does, but without referencing analysiswindow or fitcurve.
It then saves a plot of the data without displaying it and copies the parameters to the clipboard inside the ParameterWindow methods.
You cna also add comments and select specific parameters to publish.
Next, it uploads the plot to the server with sftp methods from paramiko so that it can be referred to inline in blog posts.
Finally, it uses built in xmlrpclib to create a new post on the blog. 

### SSH Details ###
The public key is stored in authorized_keys in the .ssh folder on the server, one step higher than the web files.
We temporarily have one private key, which is stored in the group share.
Changes to the webpage are done by the old synch, which now copies over changes without deleting everything on the server first.
You can also make changes to the site manually through sftp, but this is not recommended because we prefer centralization.