* Update CONTRIBUTING.adoc
Fix broken links to labels, update link to style guide. Add breaksapi and breaksconsensus and add a section on easy, medium and mentor ticket to get started. Slight rephrasing to clarify current process.
* Update CONTRIBUTING.adoc
* grandpa finality proof
* prove GrandpaApi::grandpa_authorities using parent block + some docs
* create justification when consensus data is changed
* generate justifications periodically
* test for ConsensusChanges
* Updates for CC's recent upgrade
- Version bump to 0.9.2
- Runtime version bump to v14
- Avoid misuse of error return type for system_health RPC
* Fix tests
* Updated README.adoc
Added a new section with instructions of how to join the Krumme Lanke testnet.
Closes Issue 1217
* Update README.adoc
Instructions for joining Charred Cherry + some minor typos
* Updated Runtime & API in README.adoc
Properties of these functions are changing quickly, updated this section to be more generic.
* Some documentation for ext_ functions in Substrate Runtime API
This would be useful for using non-Rust language to write Substrate-compatible modules and debugging modules at the WebAssembly level.
* Update without_std.rs
Update comment styles
* BlockAuthorityId convenience type
* Rename AuthorityId -> Ed25519AuthorityId to make it more precise
* Generalize AuthorityId up to substrate-client
* Fix in client-db
* rename: BlockAuthorityId -> AuthorityIdFor
* typo: should be digest item
* Fix test-runtime authorityId mismatch
One states that AuthorityId is u64 while the other states that it's Ed25519AuthorityId.
* Fix more u64 - Ed25519AuthorityId mismatch
* Fix compile of most of the srml modules
* Continue to pin aura and grandpa with ed25519 and fix compile
* Add MaybeHash trait
* Fix node-runtime compile
* Fix network tests
* Missing trait def on two calls.
Slight mcla refact on transfo.
* Allways provide GenesisConfig even when useless (for runtime module
export).
* Revert "Allways provide GenesisConfig even when useless (for runtime module"
This reverts commit 84a29bc592023d1ee107519c58693ec19dbde2d2.
* Fix Vec usage (from rstd).
* fix: Add assertions to prevent excess votes and computation, and to end council seat election when no empty seats
* Add assertion to prevent a vote from voters that provide a list of votes that exceeds the desired seats length, since otherise an attacker may be able to submit a very long list of `votes` that far exceeds the amount of candidates and waste more computation than a reasonable voting bond would cover. Added additional associated test that may be run with `cargo test -p srml-council`
* Add assertion so expired council seats are not up for election when desired seat count changes during the voting period such that there are no longer any empty seats
* Update comment to refer to `reporter` instead of `who` (target of inactivity), since the origin is the `reporter`
* Update commment to refer more specifically to how many vote indexes, since `InactiveGracePeriod` is measured in vote indexes
* Update comment for `ApprovalsOf` since previously the comment was a duplicate of the comment for `LastActiveOf`'s
* Create variable to refer to `retaining_seats` to improve readability
* Reference Notes: https://hackmd.io/nr6kPD2sR4urmljtvHs0CQ
* WIP - length votes should be less than length candidates. pushing changes for reference so can fix
* fix: Amount of candidate approval votes cannot exceed candidates amount. Candidates amount must be over zero.
* Fix so that amount of candidate approval votes cannot exceed amount of candidates (instead of desired seats)
* Add assertion to `set_approvals` such that amount of candidates to receive approval votes must be greater than zero. Add associated test
* fix: Remove assertion preventing votes when empty seats is 0
* review-fix: Replace with is_zero and add corresponding test
* Update seats.rs
* Update seats.rs
Documentation says the `default` macro expands to `Module, Call, Storage, Event<T>, Config`, but it actually uses the generic config: `Module, Call, Storage, Event<T>, Config<T>`
* core: fix bug on ondemand dispatch
after a request was dispatched to the last peer the dispatch would loop
forever on subsequent requests that aren't able to be fulfilled by any
of the peers, since the last peer wasn't updated.
* core: try to dispatch all pending ondemand requests
Fabian Raetz