Prefix binding #415
Draft
Prefix binding #415
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rocketraman
force-pushed
the
rg/prefix-binding
branch
from
f7c7456
to
450fc6a
Compare
The purpose of the path normalizer is to normalize all inbound paths by making them lower case, and removing any "-" characters. Normalizing paths means sources with different idiomatic approaches to defining key values will all map correctly to the defined config classes. The only downside to this is multiple config attributes in the same class that differ only by case can no longer be disambiguated. This should be a rare case and the advantages are more than worth losing this "feature". We also add a LowercaseParameterMapper by default which can handle the normalized paths.
With path normalization (and even before path normalization), it was possible for the reported path to not match the input value. For example, with environment variables, the environment variable may have been `FOO__BAR`, but the report showed `FOO.BAR`, which makes it harder for users to trace the report back to the original source. We now add a `sourceKey` attribute to `Node` and report on it, to show user's the key value as it was originally present in the source.
rocketraman
force-pushed
the
rg/prefix-binding
branch
2 times, most recently
from
b151ea7
to
16024e5
Compare
Remove both the `SnakeCaseParamMapper` and `KebabCaseParamMapper` by default, add the `PathNormalizer` by default. Add removal of `_` to path normalizer. Fix some issues with the `HikariDataSourceDecoder` when enabling path normalization by default -- that decoder requires the original key case as its props are case-sensitive. Create an abstract `UnnormalizedKeysDecoder` which has the ability to restore the case of keys via the `sourceKey`. Fix breaking explicit sealed types with normalization because the discriminator field defaults to `_type` which normalizes to `type`. Disable normalization if the field name matches the discriminator field name, and the node is a `MapNode`. Fix reporting for strict mode to use the `sourceKey` value, so that reporting matches the source value, not the normalized value. Update Preprocessor implementations to correctly copy the source key when new Map and Array nodes are created.
rocketraman
force-pushed
the
rg/prefix-binding
branch
from
16024e5
to
d196ce4
Compare
With this approach, if there are multiple prefixes and config classes to be bound, we parse the configuration only a single time, and decode multiple times.
rocketraman
force-pushed
the
rg/prefix-binding
branch
from
d196ce4
to
8974098
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Modify
ConfigLoaderandConfigParserso that the config is parsed only a single time.ConfigLoader.configBinder()can now be called to obtain the instance of a new classConfigBinder, which can in turn be used to bind the already parsed config to multiple data classes, given a prefix.This follows the general approach as suggested by @cloudshiftchris.
This PR is stacked on top of #414 and will be rebased and moved out of draft once that PR is merged.