You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I know this library has been around for some time, but I wanted to reach out from the Viper "team" to congratulate you on it. It keeps me on my toes to keep improving Viper.
I also wanted to reach out about an upcoming release (currently in alpha) that may render some of the information incorrect in the comparison:
v1.20.0 is going to drop a lot of dependencies from the core, so I expect this will significantly improve. Also, the comparison is a bit apples to oranges: Viper supports searching files, hence multiple parsers are included by default. I suspect after the update binary sizes will be somewhere on the same scale.
Tightly couples config parsing with file extensions.
v1.20.0 is going to introduce a new encoding layer. It still has a concept of "format", but that can be anything.
various parses are hardcoded in the core. There are no primitives that can be extended.
See the above comment: the encoding layer can be completely replaced now.
Pulls a large number of spf13/viper#707 into the core package. For instance, even if you do not use YAML or flags, the dependencies are still pulled as a result of the coupling.
Although there are still a number of third-party dependencies, this will also significantly improve with v1.20.0.
All of the other points remain valid I'm actively working towards v2 to improve the situation on those fronts as well.
The text was updated successfully, but these errors were encountered:
Hey there!
I know this library has been around for some time, but I wanted to reach out from the Viper "team" to congratulate you on it. It keeps me on my toes to keep improving Viper.
I also wanted to reach out about an upcoming release (currently in alpha) that may render some of the information incorrect in the comparison:
v1.20.0 is going to drop a lot of dependencies from the core, so I expect this will significantly improve. Also, the comparison is a bit apples to oranges: Viper supports searching files, hence multiple parsers are included by default. I suspect after the update binary sizes will be somewhere on the same scale.
v1.20.0 is going to introduce a new encoding layer. It still has a concept of "format", but that can be anything.
See the above comment: the encoding layer can be completely replaced now.
Although there are still a number of third-party dependencies, this will also significantly improve with v1.20.0.
All of the other points remain valid I'm actively working towards v2 to improve the situation on those fronts as well.
The text was updated successfully, but these errors were encountered: