End of Google Summer of Code
Thank You!
Yesterday was the last day of my Google Summer of Code project. I want to use this blog post to say thank you. Thank you Google for making this program possible and thank you Carol Smith for the great help and administrative work throughout the whole GSoC 2013. Thanks to my mentor, Florian Mayer, for helping me understand the existing VSO package code and for general guidance and tips. Thanks to Steven Christe for various ideas, especially the function display_entries and the ability to fetch a subset of all entries by using the slicing syntax.
Plans
The database package is not finished to a 100%; it has at least one bug that I know of (but I'm pretty there are more that will be found only if this package is used by real users), documentation can be improved and some features need to be added.
The bug that I know of is that the undo method of the Database class is broken for the method set_cache_size if it reduces the number of database entries. But I already know where the problem is, so it should be easy to fix: removing the database entries can be undone, but the original internal cache is not saved so it is not restored after the undo call.
One thing that I couldn't commit to my database branch was the feature of dumping and loading of database entries to a single text file. Use cases for this feature is the possibility to share this file with other people and also use it as a backup.
Another thing I want to implement is the support of adding database entries by an HEK query result. Thanks to the work of Michael Malocha (the other GSoC student working for SunPy) this should be quite simple because he wrote a module which can translate VSO queries to HEK queries.
Currently, only FITS files are supported in the download method of the Database class. But the files that will be downloaded via VSOClient.get may also be *.tar or *.tar.gz files, so I plan to support these kind of files as well.