While I am still a fan of what fastlane does at its core, I have experienced the same pain points that Maxwell has. From its lacking documentation, constant releases with regressions and general bloat caused by years of feature growth it is hard to trust fastlane anymore.
I have also had poor experiences when it comes to logging bugs or making feature requests. Below are three GitHub issues I created that were closed due to inactivity on the fastlane team's part. I believe I provided more than enough information to deserve a response but it seems that fastlane has reached a scale where they simply cannot handle the deluge of issues.
- [Feature Request] Add support for the --path parameter to the SwiftLint action
- Shell action no longer outputs to the console in real time
- Can the sh action output to the console in real time?
I know there a lot of developers who will read this and say "but Reid, fastlane is open source so you can fix all these issues yourself". To that my answer is that I do not know Ruby well enough to make any significant changes. If I was going to spend my time fixing my development tools I would much rather do it in a language that I have a deep understanding of like Swift.
At this point in time fastlane is just not a dependency I want to deal with anymore. I am going to try to start building out my own Swift scripts to automate my dev environment. If I find myself wishing for a piece of fastlane functionality I'll try to dig into their Ruby source code and reverse engineer what I need just like Maxwell did.