Convert Domain Names correctly from IDN to Punycode and vice-versa and offer TR46 processing.
npm install @randock/idna-uts46-hx
This module is a IDNA UTS46 connector library for javascript. In addition to the default functionality of tr46, we offer converting domain names to unicode / punycode considering the respective registry provider's behavior.
The JS Punycode converter library is
a great tool for handling Unicode domain names, but it only implements the
Punycode encoding of domain labels, not the full IDNA algorithm. In simple
cases, a mere conversion to lowercase text before input would seem sufficient,
but the real mapping for strings is far more complex. This library implements
the full mapping for these strings, as defined by
UTS #46.
- Documentation
- Release Notes
With v6 we migrated our library to npm package tr46 as software dependency. By that step we use a library that is actively maintained in direction of correctly supporting the TR46 standard and supporting the latest Version of the Unicode Standard. Reinventing the wheel isn't useful and something we have time or resources for. We were able to dramatically decrease the number of lines of code on our end.
- method toUnicode comes with auto-detection of transitionalProcessing setting based on the provided domain name input
- method toAscii comes with auto-detection of transitionalProcessing setting based on the provided domain name input
In general, we don't see a blocker for upgrading to v6. Still, consider the below changes.
#### Performance
Runtime performance of v6 compared to v5 has slightly improved. The compression for the underlying idna mapping table is superfluous, tr46 covers it well.
#### New Labels for Options
The below configuration options for the methods toUnicodeand toAscii must be renamed in case you're using them:
| Option, old | Option, new |
| --------------- | ---------------------- |
| transitional | transitionalProcessing |
| useStd3ASCII | useSTD3ASCIIRules |
| verifyDnsLength | verifyDNSLength |
#### Behavior
Earlier versions kept option transitional by default to false which is now automatically detected and results may therefore differ.
This affects the toAscii method.
The toUnicode function did not allow for a options parameter in earlier versions, now it follows the exemplary way of package tr46.
Thanks for the below former contributions:
- Initial work done by jcranmer.
- v5: Migration of the IDNA Mapping Table's Build Process from Python to NodeJS5 by dawsbot
- v5: Performance Improvements for the Browser Bundle's Page Load by dawsbot
See also the list of contributors who participated in this project.
MIT