I know… It’s been a long time since the first post, but I’ve been kinda busy with all the Haskell mind-blowing assignments (oh my favourite “boobs”), Crazy affine spaces and other beautiful things that brings the University into the life.
The ESA SOCIS 2013 coding period started a bit late due to the some technical difficulties but it does not matter because we are back on the track and rolling like never before!
I and Cadair (aka Stuart Mumford) were thinking about the right design for the HyperMap project for some time. Well, I have to say that we have some really good design sorted out and now working really hard on the implementation.
The whole HyperMap design is trying to be as much modular as possible. The heavy data processing and header parsing will be separated from the client API (that’s the plan), so it will be easier to maintain the whole thing and add support for new instruments. We are currently working on the header parsing and moving forward to the other parts of the module.
The progress on the HyperMap looks good and I have to say that we have solid foundations.
I will try to write another blog post as we move on.
Happy hacking!
I’ve been accepted for the ESA Summer of Code in Space 2013 AKA SOCIS 2013.
ESA Summer of Code in Space 2013 (SOCIS 2013) is a program run by the European Space Agency. It aims at offering student developers stipends to write code for various space-related open source software projects.
I will work on the SunPy, which is cool open-source software library for solar physics using the Python programming language.
Although, classes at the University are starting sometime in September, I will also work on the SunPy. My job will be to design and implement new data type for
the SunPy, let’s call it HyperMap (you can find more info about HyperMap in the doc here).
At the moment there are three core data types in the SunPy:
- LightCurve, which is for 1D time series
- Spectra, 1D wavelength and 2D wavelength vs time data
- Map, 2 spatial dimensions
Apart from this, there is a requirement for more general spatial data type, one that can handle time/wavelength series of images.
There are data products like from a lots of different instruments that would provide data that should this new data type handle.
Something like:
- 1 spatial dimension, wavelength and time
- 2 spatial dimensions, 1 wavelength and 1 time
Check this for more info.
This is rather non-trivial stuff so I will dive more deeply into the SunPy doc and code for the following week or so before I will start thinking about HyberMap design and implementation.
THE PLAN
I tried to put some rough plan for the following week or so, here it is:
- check SunPy doc, get used to the library from the user perspective
- dive into spectra, map & lightcurve code
- check mapfactory code
- look at AstroPy’s NDData
- profit
Happy hacking!
examon's blog 