Files
pezkuwi-subxt/scripts/ci/changelog
Branislav Kontur 445f9277ab Added new empty shell runtime for BridgeHubPolkadot (without bridging stuff) for onboarding parachain on Polkadot (#2137)
* Added new empty shell runtime for `BridgeHubPolkadot` (without bridging stuff) for onboarding parachain on Polkadot

* Fix benchmarks/try-runtime

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs frame_system

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_balances

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_multisig

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_xcm

* Fixes

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_session

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_utility

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_timestamp

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs pallet_collator_selection

* ".git/.scripts/commands/bench/bench.sh" pallet bridge-hub-polkadot bridge-hubs cumulus_pallet_xcmp_queue

* PR fixes

* Rebase fix

* Fix new runtime api

* ".git/.scripts/commands/bench/bench.sh" xcm bridge-hub-polkadot bridge-hubs pallet_xcm_benchmarks::fungible

* ".git/.scripts/commands/bench/bench.sh" xcm bridge-hub-polkadot bridge-hubs pallet_xcm_benchmarks::generic

* Updated aura/bootNodes in `create_bridge_hub_polkadot_spec.sh`

---------

Co-authored-by: command-bot <>
2023-02-02 10:35:11 +00:00
..
2022-03-28 10:00:07 +02:00

Changelog

Currently, the changelog is built locally. It will be moved to CI once labels stabilize.

For now, a bit of preparation is required before you can run the script:

  • fetch the srtool digests
  • store them under the digests folder as <chain>-srtool-digest.json
  • ensure the .env file is up to date with correct information

The content of the release notes is generated from the template files under the scripts/ci/changelog/templates folder. For readability and maintenance, the template is split into several small snippets.

Run:

./bin/changelog <ref_since> [<ref_until>=HEAD]

For instance:

./bin/changelog parachains-v7.0.0-rc8

A file called release-notes.md will be generated and can be used for the release.

ENV

You may use the following ENV for testing:

RUSTC_STABLE="rustc 1.56.1 (59eed8a2a 2021-11-01)"
RUSTC_NIGHTLY="rustc 1.57.0-nightly (51e514c0f 2021-09-12)"
PRE_RELEASE=true
HIDE_SRTOOL_ROCOCO=true
HIDE_SRTOOL_SHELL=true
REF1=statemine-v5.0.0
REF2=HEAD
DEBUG=1
NO_CACHE=1

By default, the template will include all the information, including the runtime data. For clients releases, we don't need those and they can be skipped by setting the following env:

RELEASE_TYPE=client

Considered labels

The following list will likely evolve over time and it will be hard to keep it in sync. In any case, if you want to find all the labels that are used, search for meta in the templates. Currently, the considered labels are:

  • Priority: C labels
  • Audit: D labels
  • E4 => new host function
  • B0 => silent, not showing up
  • B1-releasenotes (misc unless other labels)
  • B5-client (client changes)
  • B7-runtimenoteworthy (runtime changes)
  • T6-XCM

Note that labels with the same letter are mutually exclusive. A PR should not have both B0 and B5, or both C1 and C9. In case of conflicts, the template will decide which label will be considered.

Dev and debuggin

Hot Reload

The following command allows Hot Reload:

fswatch templates -e ".*\.md$" | xargs -n1 -I{} ./bin/changelog statemine-v5.0.0

Caching

By default, if the changelog data from Github is already present, the calls to the Github API will be skipped and the local version of the data will be used. This is much faster. If you know that some labels have changed in Github, you probably want to refresh the data. You can then either delete manually the cumulus.json file or export NO_CACHE=1 to force refreshing the data.