It's almost time

Oh, there we go! Time for GSoC’s pre-final blog post.

Ah, so the last week was a bit draining for me. My college started last monday and I accidently messed up my Linux machine the next day! Never ever remove (apt purge) the dbus package (especially not when playing with python-dbus package). It will render your system utter useless, atleast that’s what happened to me. It destroyed my X server and network-manager (and god knows how many other packages). Tried to install the missing .deb packages via external USB but those packages complained about several other missing packages and this kept recursing until I gave up. Couldn’t get chroot from a live USB to install the missing packages either. Finally takes a backup of the little things I wouldn’t be able to live without and installed Linux Mint 19 :). Tara (codename) has many cool features and UI improvements. Though I won’t be mentioning them here and consume more of your time hearing my nasty adventures. You’d be better off reading more on Tara here.

Oh and during these unfortunate times, I tried to set-up PlasmaPy on my brother’s Windows machine. Everything went fine except I wasn’t able to run the our test suite, so I filed a bug. Eventually, I was able to trace it back to astropy and pushed a fix to upstream. :)

Otherwise, this period went pretty smooth. We decided to not work further on JHTDP (atleast not in this phase) as we’re not yet perfectly clear on how we are going to be using this database and if it would be worth the effort. Additonally, I looked up some examples on using the pyJHTDB API and IMO this probably wasn’t going to be completed within a single GSoC phase. Those examples still scare me.

The community is working to get PlasmaPy on conda-forge and has made quite a progress! We might be coming to conda soon!

I also made a PR (not yet merged) to accept non-unit type annotations in @quantity_input in astropy. This will allow us to use this useful decorator in PlasmaPy in many places we avoided earlier. Many a times, we end up annotating our function definitions to whatever data type is expected by the function parameter. That’s some cool stuff.

At the moment, I am manually playing with our API hoping to find bugs and ideas. I did find a bug where our online_help function did not work as expected.

We also added another attribute isotope_name on Particle class a few days back. Yey!

By the way, I haven’t been able to spend as much time on PlasmaPy and other projects under OpenAstronomy due to my college classes and travelling to & fro, which take most of my day time. I hope this wouldn’t be a problem as I completed many of my GSoC’s main project goals already. ;)

Anyway, I’m still trying to make time to contribute the littlest of things I can manage during this phase.

Ok, so we’ll end this post with another last thing - I am also considering to get myself a new laptop with a long lasting battery-life and lightweight enough to carry it to my college without having to move around with the usual (2.6 * bulk that’s always trying to come out from the bottom of my bag. :/

If anyone has any suggestions, please let me know!

(I just realized I need to enable the comment section on these posts, oh crud, I felt some static. This would be more work than I expected.)

Written on July 23, 2018