Rename the rdata modules to use names instead of numbers. #189
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
partim
added a commit
that referenced
this pull request
May 12, 2023
Breaking Changes * The minimal required Rust version is now 1.65. ([#160]) * The `random` feature has been dropped in favour of using `rand`. ([#204]) * The generic octets foundation has been moved to a new crate *[octseq]* and completely revamped with Generic Associated Types stabilized in Rust 1.65. This required changes all over the code but, hopefully, should result in relatively few changes when using the crate. ([#160]) * The range, slice, and split methods on the domain name types have changed. They have been merge into a single method taking ranges – except for those on `Dname` that require type changes. The split methods now take references and don’t change `self` anymore. ([#160]) * The `Parse`, `Compose`, and `Scan` traits have been demoted to mere extension traits for foreign types (primarily the built-in integers, so that you can do things like `u16::parse`). All other types now simply have methods matching the patterns. Where generics are necessary, dedicated traits have been added. E.g., there now are `ParseRecordData` and `ComposeRecordData` traits that are implemented by all record data types. ([#160]) * The `Deref` and `DerefMut` impls have been removed for most types that had them to follow guidance that they are exclusively for use by pointer types – which none of them are. `len` and `is_empty` methods have been added where appropriate, additional methods may be added. ([#205]) * Various functions and methods of the `tsig` module now expect the current time as an argument to allow use of the module in a no-std environment. ([#152]) * Parsing of representation format and zonefiles has been completely re-written. ([#142], based on work in [#109] by [Martin Fischer]) * All types that wrap an octets sequence only now allow unsized octets sequence types. They all have an associated function `from_slice` to create a reference to a value wrapping an (unsized) octets slice and method `for_slice` that converts a `&self` into such a reference. Where the latter already existed but returned a value wrapping a `&[u8]` (e.g., `Dname<_>` and `Message<_>`, the return type has changed accordingly. ([#168]) * Removed `CharStr::from_bytes`. Use `CharStr::from_octets` instead. ([#168]) * `Message::from_octets` now returns a new error type `ShortMessage`. ([#168]) * Dropped `Deref` impls for `Dname<_>`, `RelativeDname<_>`. ([#168]) * Renamed `opt::KeyTag::new` to `opt::KeyTag::from_octets`. ([#168]) * Renamed `rdata::Txt::try_from_slice` to `build_from_slice`. ([#168]) * The `new` method of the following record data types now check whether the wire format representation of the record data is too long and thus returns a result: `Tsig<_, _>`, `Dnskey<_>`, `Rrsig<_, _>`, `Ds<_>`, `Cdnskey<_>`, `Cds<_>`. ([#169]) * The `new` function for `rdata::Null<_>` has been replaced with a `from_octets` and `from_slice` pair. The `Deref` impl was removed. ([#169]) * The `rdata::svcb` module has been refactored to work in the same way as other type-length-value constructs. The names of types, methods, and functions have changed both to match the usual nomenclature as well as to match the terms used in the SVCB draft. ([#176]) * The `base::iana::SvcbParamKey` type has been renamed to `SvcParamKey` to match the terms used in the SVCB draft. ([#176]) * The `TcpKeepalive` option has been changed to use an `Option<u16>` as its data and allow for an empty option in accordance with the RFC. ([#185]) * Renamed the sub-modules of `rdata` that contain record data types to use a name derived from their content rather than their RFC number – with the exception of `rdata::rfc1035`. ([#189]) * Renamed the sub-modules of `base::opt` that contain option data types to use short-hand names rather than their RFC number. ([#190]) * TTL values are now using a newtype `base::record::Ttl` that wraps the raw `u32` and improves conversions. ([#202] by [@CrabNejonas]) * Changes all option data types to ensure their wire format is at most 65,535 octets long. This requires changing the signatures of some creator functions. Their naming scheme and signatures are also changed to follow the pattern established with record data. ([#193]) * Renamed `UnknownOptData::from_octets` to `new` and return a result. ([#193]) * Completely redesigns DNS cookie options, adding support for standard server cookies introduced in RFC 9018. ([#193]) * Change the type of `ExtendedError`’s text to `Str<Octs>` and change the return type of `set_text` to `()`. ([#193]) * Changed the type `TcpKeepalive`’s content to a newtype `IdleTimeout` to make it easier to convert to and from durations. ([#193]) * Changes Padding to just contain the padding octets and drop `PaddingMode`. Instead, the methods on `OptBuilder` should be used to add padding. ([#193]) New * `Display` impls are now available for all EDNS0 options. ([#157]) * Adds a `FromStr` implementation and related functions to `RelativeDname`. ([#177]) * Add a `Debug` impl to `base::message::Message` so it can be unwrapped etc. ([#199]) * New methods `make_canonical` on `Dname` and `RelativeDname` that convert the name into its canonical, i.e., lowercase form. Similarly, new methods `ToDname::to_canonical_dname` and `ToRelativeDname::to_canonical_relative_dname` that produce new canonical names. ([#200]) * Added a `MAX_LEN` constant to various types that wrap length-limited octets sequences. ([#201] by [@CrabNejonas])
Philip-NLnetLabs
pushed a commit
that referenced
this pull request
Jun 30, 2023
Breaking Changes * The minimal required Rust version is now 1.65. ([#160]) * The `random` feature has been dropped in favour of using `rand`. ([#204]) * The generic octets foundation has been moved to a new crate *[octseq]* and completely revamped with Generic Associated Types stabilized in Rust 1.65. This required changes all over the code but, hopefully, should result in relatively few changes when using the crate. ([#160]) * The range, slice, and split methods on the domain name types have changed. They have been merge into a single method taking ranges – except for those on `Dname` that require type changes. The split methods now take references and don’t change `self` anymore. ([#160]) * The `Parse`, `Compose`, and `Scan` traits have been demoted to mere extension traits for foreign types (primarily the built-in integers, so that you can do things like `u16::parse`). All other types now simply have methods matching the patterns. Where generics are necessary, dedicated traits have been added. E.g., there now are `ParseRecordData` and `ComposeRecordData` traits that are implemented by all record data types. ([#160]) * The `Deref` and `DerefMut` impls have been removed for most types that had them to follow guidance that they are exclusively for use by pointer types – which none of them are. `len` and `is_empty` methods have been added where appropriate, additional methods may be added. ([#205]) * Various functions and methods of the `tsig` module now expect the current time as an argument to allow use of the module in a no-std environment. ([#152]) * Parsing of representation format and zonefiles has been completely re-written. ([#142], based on work in [#109] by [Martin Fischer]) * All types that wrap an octets sequence only now allow unsized octets sequence types. They all have an associated function `from_slice` to create a reference to a value wrapping an (unsized) octets slice and method `for_slice` that converts a `&self` into such a reference. Where the latter already existed but returned a value wrapping a `&[u8]` (e.g., `Dname<_>` and `Message<_>`, the return type has changed accordingly. ([#168]) * Removed `CharStr::from_bytes`. Use `CharStr::from_octets` instead. ([#168]) * `Message::from_octets` now returns a new error type `ShortMessage`. ([#168]) * Dropped `Deref` impls for `Dname<_>`, `RelativeDname<_>`. ([#168]) * Renamed `opt::KeyTag::new` to `opt::KeyTag::from_octets`. ([#168]) * Renamed `rdata::Txt::try_from_slice` to `build_from_slice`. ([#168]) * The `new` method of the following record data types now check whether the wire format representation of the record data is too long and thus returns a result: `Tsig<_, _>`, `Dnskey<_>`, `Rrsig<_, _>`, `Ds<_>`, `Cdnskey<_>`, `Cds<_>`. ([#169]) * The `new` function for `rdata::Null<_>` has been replaced with a `from_octets` and `from_slice` pair. The `Deref` impl was removed. ([#169]) * The `rdata::svcb` module has been refactored to work in the same way as other type-length-value constructs. The names of types, methods, and functions have changed both to match the usual nomenclature as well as to match the terms used in the SVCB draft. ([#176]) * The `base::iana::SvcbParamKey` type has been renamed to `SvcParamKey` to match the terms used in the SVCB draft. ([#176]) * The `TcpKeepalive` option has been changed to use an `Option<u16>` as its data and allow for an empty option in accordance with the RFC. ([#185]) * Renamed the sub-modules of `rdata` that contain record data types to use a name derived from their content rather than their RFC number – with the exception of `rdata::rfc1035`. ([#189]) * Renamed the sub-modules of `base::opt` that contain option data types to use short-hand names rather than their RFC number. ([#190]) * TTL values are now using a newtype `base::record::Ttl` that wraps the raw `u32` and improves conversions. ([#202] by [@CrabNejonas]) * Changes all option data types to ensure their wire format is at most 65,535 octets long. This requires changing the signatures of some creator functions. Their naming scheme and signatures are also changed to follow the pattern established with record data. ([#193]) * Renamed `UnknownOptData::from_octets` to `new` and return a result. ([#193]) * Completely redesigns DNS cookie options, adding support for standard server cookies introduced in RFC 9018. ([#193]) * Change the type of `ExtendedError`’s text to `Str<Octs>` and change the return type of `set_text` to `()`. ([#193]) * Changed the type `TcpKeepalive`’s content to a newtype `IdleTimeout` to make it easier to convert to and from durations. ([#193]) * Changes Padding to just contain the padding octets and drop `PaddingMode`. Instead, the methods on `OptBuilder` should be used to add padding. ([#193]) New * `Display` impls are now available for all EDNS0 options. ([#157]) * Adds a `FromStr` implementation and related functions to `RelativeDname`. ([#177]) * Add a `Debug` impl to `base::message::Message` so it can be unwrapped etc. ([#199]) * New methods `make_canonical` on `Dname` and `RelativeDname` that convert the name into its canonical, i.e., lowercase form. Similarly, new methods `ToDname::to_canonical_dname` and `ToRelativeDname::to_canonical_relative_dname` that produce new canonical names. ([#200]) * Added a `MAX_LEN` constant to various types that wrap length-limited octets sequences. ([#201] by [@CrabNejonas])
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.
This PR renames the sub-modules of
rdatathat contain record data types to use a name derived from their content rather than their RFC number – with the exception ofrdata::rfc1035for which there isn’t really a good name.Using the RFC numbers was a bad idea to begin with – there is no association with the content so finding your record type is difficult. Hopefully, the impact of renaming for existing code will be minimal. All the record data types have been re-exported at in
rdata, so only the import paths of helper types need adjusting. But we feel this is for the better.