* Lock files
* Add flag to disable cover traffic
* Add flag to disable per hop delays
* Add flag to enable mixed size packets
* Add meta flag to set medium speed
* Special case zero averge hop delay to be exactly zero
* Extract out generate_hop_delays function
* added cargo config file to explicitly specify build target
* wip
* Config option to disable topology refreshing
* extracted common parsing code
* helper trait for working on wasm topology
* wasm topology parsing
* restored (slightly modified) old js-example
* wip
* Moved message preparation into a trait
* wip
* long-winded way of sending test packet
* standalone NymNodeTester
* finishing the test upon receiving all packets even if timeout wasnt reached
* initial round of cleanup
* sending multiple test packets in normal NymClient
* javascript-side cleanup
* starting mixnode test on btn click
* Improved NymNodeTester constructors
* improved error handling and constructors
* tester utils error handling
* further cleanup + using BTreeMap for NymTopology mixnodes
* handling missed errors
* splitting up 'test_node'
* split up and cleaned up generation of test result
* clippy + fixed example
* post rebase fixes
* another broken test
* prevent running multiple parallel tests
* cargo fmt
* Added nym- prefix to node tester utils
* Generalise MessageReceiver
* Generics all the way
* Generalise MessageReceiver
* Generics all the way
* Fix Cargo.lock
---------
Co-authored-by: benedettadavico <benedetta.davico@gmail.com>
* oak-2
* oak-8
* oak-13
* oak-15
* oak-18
* Minor clippy nit
* 2023-01-13-OAK-6
* 2023-01-13-OAK-3
* 2023-01-13-OAK-13
Implemented via direct dependency on cw2 and calling the appropriate code on migration
* Removed few instances of password being unecessarily copied
* 2023-01-13-OAK-10
* 2023-01-13-OAK-12
* 2021-09-13-JP-S-NYM-02
* 2021-09-13-JP-S-NYM-03
* Removed further instances of needlessly copying the mnemonic
* 2021-09-13-JP-O-PROT-03
* 2021-09-13-JP-S-NYM-01*
*: we still have one vulnerability on 'time' pulled from chrono via sqlx. However, apparently its usage is fine... Having said that, I'd still recommend removing all dependencies on chrono, but this will require some database migrations...
* 2023-01-13-OAK-11 (#3009)
* wip
* Introducing the concept of starting epoch transition in `nym-api`
* split epoch operations into multiple files
* epoch operation failure recovery
* sending rewarding transactions in correct order
* tests and fixes due to epoch state progression
* lint
* missed rebasing import changes
* Setting cw2 contract version during first migration run
* calling 'reconcile_epoch_events' at least once
* Made message to BeginEpochTransition more consistent with other variants
* Merge layer assignment updates
---------
Co-authored-by: Jędrzej Stuczyński <jedrzej.stuczynski@gmail.com>
* added additional workspace-wide dependencies
* Added conditional serialization on 'BinaryBuildInformationOwned'
* initial framework for service provider messages
* updated request/response tags to account for existing variants of Socks5Message
* handling legacy deserialization
* another serialization revamp to account for legacy version
* legacy client working with versioned network requester
* socks5 client deserializing responses into updated structures
* using new structures for sending in socks5 client
* SendRequest wrapper for Request::Send variant
* created named fields for all variants of 'ControllerCommand'
* Versioning socks5 requests + moving to proper Socks5Request struct
* Updated backwards compatible Socks5Response
* unused imports
* poc ServiceProvider trait
* wip
* implemented 'ServiceProvider' trait on the Network Requester
* Socks5RequestError
* added properly serialized ErrorResponse to ControlResponse
* fixed version serialization + feature selection
* handling of version control requests
* improved SocksProxyError by providing more concrete variants and removing generic case
* got rid of ServiceProviderClient trait and wrote simple example showing control requests
* tests for serialisation backwards compatibility
* post-merge fixes due to method renaming
* enum boxing to make clippy happier
* making sure to not drop buffer channel when starting `DirectClient`
* Using nym-sdk in the example
* Replaced printing version to stdout with proper log call
* fixed all uses of deprecated clap methods
* updated all uses of clap to 4.0
* unified obtaining build information
* moved around the imports
* Moved all nym-api arguments to ApiArgs and simplified parsing
* Using common shutdown signal code
* Using clap for parsing Vec<Url>
* stricter validation of socks5-client arguments
* ibid for the native client
* ibid for the gateway
* ibid for the mixnode
* clippy
* Preserve information about original transmission lanes when buffering reply packets
* Attempting to send partial data in 'handle_send_reply' if we don't have enough surbs immediately
* Display logging of reply surb request target
* promoted reply_controller to a directory
* moved channels and messages to separate file
* simplifications due to rust 1.66
* Using a shoarthand for obtaining connection_id
* made TransmissionBuffer generic
* Moved transmissaion buffer to a higher level directory + defined wasm helpers
* Using transmission buffer in reply controller
* Using the pending replies size in lane lenghts queries
* fixed an out of bounds use of fragments
* Fixed dropped channel in getting lane queue length
* Fixed an edge case failure for reply retransmissions
* measuring (and logging) time it takes to obtain lane lenghts
* decreased logging level
* Removed non-wasm lock on total_size
* cleaned up MixProcessingError
* Added Error impl to (hopefully) all error enums in the codebase
* Replaced all occurences of error("{0}") with error(transparent)
* Changelog entry
Jędrzej Stuczyński