The doc comments don't discuss goroutine safety so we have to just guess
It pulls in way to many dependencies
The dependency issue is the dealbreaker here. Most of the deps it pulls in have to do with testing or prometheus. We already know the code works because these unit tests are presumably being run upstream, so these dependencies are just dead weight here. We don't need or want prometheus because having a web server packaged in every application is extremely silly, and without that prometheus is useless. It's even more dead weight that go has to churn through.
The solution would probably be to maintain a downstream "clean" version with an automatic procedure to run the tests (as a sanity check), delete all test files, delete everything related to prometheus, and finally run go mod tidy. It would be set up to pull when there is a release from upstream (because presumably all tests will pass if it is really a release), run the procedure, delete the old release, and create a new one.
This is not necessary for the functionality of this package so it can wait a little bit, but this should be done before a stable release.
Reasons why:
- The API is a bit unpleasant to use
- The doc comments don't discuss goroutine safety so we have to just guess
- It pulls in way to many dependencies
The dependency issue is the dealbreaker here. Most of the deps it pulls in have to do with testing or prometheus. We already know the code works because these unit tests are presumably being run upstream, so these dependencies are just dead weight here. We don't need or want prometheus because having a web server packaged in every application is extremely silly, and without that prometheus is useless. It's even more dead weight that go has to churn through.
The solution would probably be to maintain a downstream "clean" version with an automatic procedure to run the tests (as a sanity check), delete all test files, delete everything related to prometheus, and finally run `go mod tidy`. It would be set up to pull when there is a release from upstream (because presumably all tests will pass if it is really a release), run the procedure, delete the old release, and create a new one.
This is not necessary for the functionality of this package so it can wait a little bit, but this should be done before a stable release.
It outputs log messages if your kernel is configured a certain way, and this cannot be disabled
It outputs even more log messages if you set a certain environment variable, and this cannot be disabled
There does not seem to be a straightforward way to create a connection, having to dig through the source code to figure it out
More problems with the module:
- It outputs log messages if your kernel is configured a certain way, and this cannot be disabled
- It outputs even more log messages if you set a certain environment variable, and this cannot be disabled
- There does not seem to be a straightforward way to create a connection, having to dig through the source code to figure it out
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Reasons why:
The dependency issue is the dealbreaker here. Most of the deps it pulls in have to do with testing or prometheus. We already know the code works because these unit tests are presumably being run upstream, so these dependencies are just dead weight here. We don't need or want prometheus because having a web server packaged in every application is extremely silly, and without that prometheus is useless. It's even more dead weight that go has to churn through.
The solution would probably be to maintain a downstream "clean" version with an automatic procedure to run the tests (as a sanity check), delete all test files, delete everything related to prometheus, and finally run
go mod tidy. It would be set up to pull when there is a release from upstream (because presumably all tests will pass if it is really a release), run the procedure, delete the old release, and create a new one.This is not necessary for the functionality of this package so it can wait a little bit, but this should be done before a stable release.
More problems with the module: