#144HoursOfCode D34

Submitted almost 3 years ago

(8:09pm - 9:11pm, 9:40pm - 10:40pm , plus some extra here & there)

* Drafted first conference talk proposal and requested feedback on it
* Updated Zendesk API credentials
* Got stuck on an error produced by Pi Piper gem (I'm pretty sure that's the source)
* Started researching how to write tests for files in Pegacorn Project

I was camping over the weekend, and it's been the first weekend in a while that I didn't spend some time coding. But we'd planned this trip a while in advance, so I feel ok about that.

The conference proposal I drafted up was for RubyConf, and I've submitted it to some folks at work from a speaker-related Slack channel to request feedback. Honestly, I'm proud of myself just for getting this far since I'm terrified of public speaking and have never given a talk before. I'm interested to see what kind of feedback I receive from my colleagues at work and especially from the RubyConf reviewers.

I've noticed lately that I'm getting generic responses from my Pegacorn Project files and that goals are never showing as having been reached lately. Since I don't have tests in place yet, I wanted to refresh the credentials I'm using to access the Zendesk API. I started on this during D33 but ran into issues, so I continued my efforts today and came out successful. I continued getting similar results, but once I changed the limit parameter in the count_good_ratings.rb file I noticed that I was getting changes in results. It seems that setting this limit parameter to 1 was indeed limiting the count of good ratings returned from the Zendesk API (not shocking, I realize this) to 1. Based on previous success and my comment left in the file, I had been under the impression that I would still get an accurate count of good ratings with the limit set to 1, but would just need to change the limit to inspect the contents of those ratings. That doesn't seem to be the case now, so either the Zendesk API has changed how it handles that limit parameter or some other factor was causing it to work successfully in the past. Regardless, now that I was getting a correct count of good ratings it trigger the pegacorn to light.

When the block of code that triggers the pegacorn to light was invoked, I got an error message that included the following:

/usr/local/lib/ruby/gems/2.5.0/gems/ffi-1.9.23/lib/ffi/library.rb:147:in `block in ffi_lib': Could not open library '/usr/local/lib/ruby/gems/2.5.0/gems/pi_piper-2.0.0/lib/pi_piper/libbcm2835.so': dlopen(/usr/local/lib/ruby/gems/2.5.0/gems/pi_piper-2.0.0/lib/pi_piper/libbcm2835.so, 5): no suitable image found. Did find: (LoadError)
/usr/local/lib/ruby/gems/2.5.0/gems/pi_piper-2.0.0/lib/pi_piper/libbcm2835.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x01 0x01 0x01 0x00
/usr/local/lib/ruby/gems/2.5.0/gems/pi_piper-2.0.0/lib/pi_piper/libbcm2835.so: stat() failed with errno=25

This makes me think there's an issue between Pi Piper and the libbcm 2835 library it's depending on.

I didn't come across this error when googling variations on this error message, so my 2 next courses of action to tackle it are:
* run my updated code on my RPi connected to the light to confirm that not being connected to the light isn't the issue (I don't recall that being a requirement to invoke this section of code in the past)
* if the above generates the same results, submit an issue on the Pi Piper gem's Github page

From there, I've moved on to studying some materials related to RSpec. Even though I prefer the syntax of minitest based on my limited usage of the tools and research on them so far, it seems like RSpec is the more widely used and recognized tool and it'd be more helpful to be familiar with it. Here are a couple of tools I'm using for this RSpec research:
* https://www.youtube.com/watch?v=CHTep2zQVAc
* https://semaphoreci.com/community/tutorials/getting-started-with-rspec
* https://www.youtube.com/watch?v=K6RPMhcRICE

Here's what we've got for today:


Add a comment: