Oooh, I’ve dreamed of similar before, especially one that would track multiple files at once.
So, just for the fun of it, I added multiple file tracking. It differentiates each file with a different color. If you pass more files than there are colors it’ll cycle over the color list. Colors can also be supplied via the command line as a comma separated list
Recently I’ve been working on improving the systemd integration for our single Elixir / Phoenix app that we’ve got running so far, as well as improving logging, fixing some data synchronization bugs, working on our deployment config a bit, etc.
All this is mostly to prepare for moving from a manual workflow, where operators have been entering data into forms in the web UI in order to handle particular types of requests, to an automated workflow where these requests will be entering the system via Kafka.
I’ll be exciting to start working a bit with Kafka going forward, so really looking forward to that!
I’m very new to Elixir (and still a beginner programmer).
I’m working through the Udemy course by Stephen Grinder and the Pragmatic Programmers course.
After I’ve learned a bit more about Elixir and Phoenix, I would like to create a simple CRUD application.
I find Elixir a friendly language. The syntax is quite nice (except for anonymous functions imho). Pattern-matching is awesome.
I finished the Elixir/Phoenix course by Stephen Grinder on Udemy. It was a nice overview but a bit shallow. I’m also not that happy that it uses Phoenix 1.2.
I will continue with the Pragmatic Programmers course which is much more indepth and then continue with the book “Programming Phoenix”.
I do lots of code reviews, and we have a standard where you always type a jira task number of the first part of a commit message. I had developed quite a few batch files for making this a little easier, but I always had to copy and paste the shas and files when i wanted to specific a subset of files (like non-compiled typescript files from node). I decided to write an elixir escript to make things even easier. now I just run gittool.bat and it’ll provide me a list like this
Now I have a list of all the files modified, who did it, when. Then I can just type the number of the commit I wish to inspect, and it’ll launch the default git difftool for each file in that commit. I haven’t made this one public yet, I still have some clean up and documentation to do. If you are interested in this let me know, if I get enough interest I’ll work on cleaning it up to make public.
As a motivating exercise, I’m pondering the scenario that @Cruz posted regarding calculating subtotals and seeing what I come up with. Along the way I’m discovering that pattern matching + function heads really simplifies things. Very cool stuff.
Meanwhile, I’m reading some books. I’ve made a first reading through the sequential-logic portion of Programming Elixir 1.6 (PragProg) and made it to the Ecto chapter of Programming Phoenix 1.4 (PragProg). I’m really enjoying these two books but I feel like I’m missing a little bit of detail and context, so I’m going to slow my pace just a touch.
I’ve also been catching up on some sample chapters of other books. I read the first chapter of Programming Erlang (PragProg) and instantly wanted more; Joe’s writing style really speaks to me (rest in peace, Joe), so I plan to take a break from the Elixir books and read this one to the end.
Writing a queueing service that reads large csv / other files to save these in the database. Only problem 1. was to control the resources (cores that are reading/building structs and parsing strings) and 2. the database insert statements (too long causes timeouts, transaction conlicts etc…)
Next up is cleaning up the application! Want to follow “the little elixir & otp guidebook” principles for cleaner code with supervisors, after that going to try distribution for the processing part. (what if 2 people upload 2 huge zips at the same time?)
Just published my first (written with my awesome coworker Bryan Hunter) library to Hex.pm! – the first open-source Elixir library to be released by HCA Healthcare.
I’m developing an academic information system while i’m learning Elixir/Phoenix
Actually i’m reading docs of every module I have to treat with to make things work well.
I’ve recenty migrated my project to Phoenix 1.4, that changed a lot of stuff like brunch -> webpack and Poison -> Jason
Finally I’ve been able to start digging into the Kafka integration for our app, and it’s been an adventure… For one thing, getting KafkaEx to interact conditionally with Kafka seems to be a challenge; eg. I want to have an on/off switch, and further I don’t want my entire app to die just because Kafka happens to go down. The Kafka integration, for us, is just one part of the app’s functionality, and it’s completely viable to live without for a few days at a time if needed.
Currently looking to see how much I can decouple our GenConsumer + the Kafka worker process from the supervision tree, so that we can handle reconnection attempts w/ backoff etc.
I had a similar experience trying to get kafka to work. In the end we decided to use redis streams and roll our own client for it which ended up taking less time than I spent trying to get kafka to work properly.
Hi AstonJ, I am just curious about Elixir, in particular the message passing, which seems quite powerful and simpler than Akka in Scala… if you know about any good resource to learn more about it, please do feel fee to point me out.