Skip to main content

Pretty Good Hat

Tag: shiny

The last couple of weekends I’ve been slowly working on some improvements to Armorer, my Destiny 2 loadout finder build in Shiny.

Screenshot showing an application that allows a user to specify several parameters for a Destiny 2 loadout: Class, exotic armor, and several stat criteria including minimum stat total, power level, and primary ability stat values.

Notable enhancements for this release include:

  • Huge revision of the backend method for stat calculation, making it more flexible and accurate
  • Highlighting of masterwork items in display
  • Proper forwarding of selected ability fragments to DIM for loadout building (this was harder than I thought it would be)

There are some bugs to address and optimization to perform, of course. But, with these improvements, the tool is finally just about as functional as I have long wanted it to be! I think it’s really pretty good, and I’m pleased and proud to have made something sophisticated and useful. I hope some more Destiny 2 players will try it out and let me know how it works for you.

I had some really nice hours this weekend working on my Destiny 2 loadout finder project, Armorer. All my current work is on the backend, and will result eventually in a lot more flexibility in armor piece stat calculation, and – I hope – faster performance. This big Shiny application is complex enough that I can see the rest of the work to do right in front of me, and I know that completing all the revisions is still a pretty substantial piece of work!

An infographic summarizing a Peloton high intensity and hills ride. Instructor Ally Love stands at the right side in a pink top, hands on her hips.

Vacation days are great! Today I improved RideShare, my Shiny app that makes shareable ride cards from my Peloton workouts. I had to update to accommodate the new images that the API sends, and I added a clickable history to build cards for any recent ride. I don’t like the styling of these new images quite as much – they’re not as powerful-looking cycling-specific images as the old feed provided – but I’m still really pleased with what I can build with it!

Peloton instructor Cody Rigsby standing in a bicycle and smiling, in the background of text describing a ride. The ride is titled 2000s Ride and shows a graph of time in different heart rate zones.

This morning’s ride was a good start to a busy Sunday. I also got to test out my custom little Shiny app for building nice little shareable images with the Peloton API!

The launch of the new Destiny 2 expansion, Nightfall, adds a new subclass – Strand – and restructures the mod system that affects player stats gained from armor. I’m happy to report that it only took me a couple of hours to revise my Shiny tool to find optimal armor loadouts using the new mods and subclass fragments! I had to hunt a little through some old code, and next time it should be a simple and easy update due to having fixed how I work with the manifest.

A bright yellow background highlights bold text showing my favorite new-to-my artists this year: Big Thief, Low, Japanese Breakfast, Alvvays, and Courtney Marie Andrews. The hashtag at the bottom of the image reads #tuneR.

A heatmap showing calendar days through twelve months, with light yellow showing ‘light’ listening days shading to purple for the heavist days where I listened to the most music. I listened to a lot of music in Januar, July, October and November.

I dusted off the last.fm data visualizer that I started working on last year around this time. This year I built a couple of fun Spotify-wrapped-like visualizations; why should Spotify users have all the fun?

If you’re a last.fm user, you can try it, too! TuneR is a small Shiny app that you can provide your last.fm username and see your year in music, a comparison of this year to your all-time most-listened artists, and a fun heatmap of your listening activity.

Got nginx running on my MacBook today, as part of building some working-with-APIs-infrastructure for a small tutorial I want to write on working with oauth in Shiny. Good step!

Two windows from the RStudio profvis tool, showing the times of several procesess. The second window shows the same process being completed dramatically more quickly than the first.

I used the RStudio tool profvis this weekend to find speed improvement opportunities in Armorer. I suspected that I could rewrite a big operation that calculates the maximum of many columns across several thousand rows. Holy smokes: Using matrixStats::rowMaxs cuts processing time by an amazing amount!