Skip to content

chore(release): prepare v1.0.2#299

Merged
ebfull merged 2 commits into
mainfrom
bump-v1.0.2
Jul 20, 2026
Merged

chore(release): prepare v1.0.2#299
ebfull merged 2 commits into
mainfrom
bump-v1.0.2

Conversation

@ebfull

@ebfull ebfull commented Jul 20, 2026

Copy link
Copy Markdown
Contributor

Motivation

Prepare main for the stable v1.0.2 release, graduating the v1.0.2-rc0/rc1
release candidates. Everything user-visible in this release has already
shipped on main, almost all of it exercised in the two rcs; this PR moves
the crate graph and release metadata to their stable versions.

Solution

Crate graph — graduate all 9 crates changed since their last stable
publish (verified against the crates.io index and
check-crate-version-bumps.sh with BASE_TAG=v1.0.1):

Crate Last stable This release
zakura 1.0.1 1.0.2
zakura-chain 1.1.0 1.2.0
zakura-consensus 2.0.0 3.0.0
zakura-network 2.0.0 3.0.0
zakura-rpc 2.0.0 3.0.0
zakura-state 2.0.0 3.0.0
zakura-node-services 1.0.0 1.0.1
zakura-script 1.0.0 1.0.1
zakura-utils 1.0.1 1.0.2

zakura-test, zakura-tower-batch-control, zakura-tower-fallback, and
zakura-jsonl-trace have zero diff since v1.0.0 and are not bumped or
published. Post-rc1 changes are covered by the graduation: #281 touched
zakura-chain and zakura-script, #296 touched zakura-network, and #300
touched zakura-state — all four crates were already in the bump set.

Changelogs — the rc0/rc1 sections in the root and crate changelogs are
consolidated into single stable sections (entries verbatim, rc-version
cross-references reworded to stable), matching the changelog policy that rcs
are pre-releases of a version and don't keep standalone entries.
zakura-chain's rc1 test-only entry was dropped as not consumer-visible.
#300's root and zakura-state entries (added on main under
[Unreleased]) are folded into the stable [1.0.2] and [3.0.0] sections.

Release metadata

  • Stored config fixture zakurad/tests/common/configs/v1.0.2.toml generated
    from this branch via zakurad generate (byte-identical to the rc1 fixture —
    no config surface changed since rc1).
  • ESTIMATED_RELEASE_HEIGHT stays at the rc1 estimate of 3,419,500 (tip was
    3,418,970 at 2026-07-20 13:54 UTC). v1.0.2 spans Ironwood activation and
    halts ~2026-08-07 at height 3,440,236.
  • README install example tag → v1.0.2.
  • Installer metadata is left pinned to v1.0.2-rc1; the release workflow
    rewrites it and opens the follow-up PR after the release exists.

Testing

  • BASE_TAG=v1.0.1 ./scripts/check-crate-version-bumps.sh — 9 changed crates,
    all bumped, no warnings.
  • cargo semver-checks against last stable baselines: zakura-chain
    1.1.0→1.2.0 (minor), zakura-script and zakura-node-services
    1.0.0→1.0.1 (patch) — all pass; major bumps need no check.
  • cargo public-api diff 1.1.0 for zakura-chain: exactly one addition
    (NoteCommitmentTrees::update_sprout_tree), nothing removed or changed —
    the changelog matches the full API delta.
  • ./scripts/check-release-version.sh v1.0.2 — tag matches package version.
  • cargo metadata --locked, cargo fmt --check, bash -n scripts/install-zakura.sh, markdownlint (CI config) on all changed
    Markdown — clean.
  • stored_configs_parsed_correctly passes (includes the new fixture);
    the last_config_is_stored comparison was reproduced manually (the fixture
    is the checklist command's exact output). The full config_tests sequence
    cannot run on the prep machine — a long-running sync node holds the default
    state-DB lock and port 8233 — so node-starting config tests rely on draft
    CI here.
  • ./scripts/check-crate-packaging.sh --verify on the release commit — all
    13 crates package cleanly at their stable versions and the packaged
    zakura-1.0.2 archive rebuilds.
  • Clippy relies on draft CI (-D warnings); the only Rust change is two
    constants and a doc comment in end_of_support.rs.

Specifications & References

Follow-up Work

  • Checkpoints have not been refreshed since the fork (last checkpoint height
    3,358,006 vs tip ~3,418,970). v1.0.0/v1.0.1/rc0/rc1 all shipped this way —
    refresh or record an explicit waiver before dispatching the release.
  • After merge + explicit go-ahead: dispatch create-release.yml with
    v1.0.2, publish the 9 changed crates in dependency order, then promote to
    Latest and merge the installer-metadata follow-up PR.

🤖 Generated with Claude Code

@ebfull ebfull added A-release Release pull requests; enables the check-no-git-dependencies gate and release CI paths C-exclude-from-changelog Exclude this PR from the draft changelog P-Critical 🚑 Critical priority labels Jul 20, 2026
Comment thread zakurad/src/components/sync/end_of_support.rs Outdated
[network]
cache_dir = true
crawl_new_peer_interval = "1m 1s"
expose_peer_addresses = false

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Confirmed false default

@ebfull
ebfull marked this pull request as ready for review July 20, 2026 14:45
ebfull and others added 2 commits July 20, 2026 11:25
Graduate the v1.0.2-rc1 crate graph to stable: zakura 1.0.2,
zakura-chain 1.2.0, zakura-consensus/network/rpc/state 3.0.0,
zakura-node-services/script 1.0.1, zakura-utils 1.0.2. Consolidate the
rc0/rc1 changelog sections into the stable release sections, store the
v1.0.2 config fixture, and update the README install tag and estimated
release height.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@ebfull
ebfull merged commit 25909b0 into main Jul 20, 2026
52 checks passed
@ebfull
ebfull deleted the bump-v1.0.2 branch July 20, 2026 17:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

A-release Release pull requests; enables the check-no-git-dependencies gate and release CI paths C-exclude-from-changelog Exclude this PR from the draft changelog P-Critical 🚑 Critical priority

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants