GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles
  • Enterprises

This view

  • Original page
  • New address

GeezerVison is showing github.com — Land the rest of the state hints stack (#412, #413, #414) by CatarinaGamboa · Pull Request #417 · liquid-java/liquidjava

Conversation

@CatarinaGamboa

@CatarinaGamboa CatarinaGamboa commented Oct 24, 2026

Copy link
Copy Markdown
Collaborator

Description

#412, #413 and #414 were merged into their stacked base branches instead of main: once #411 was merged, their bases were never retargeted. Their code is only on feat/state-hints-unknown.

This PR applies the three reviewed commits on top of main, unchanged:

  • Explain dead ends in failing state transitions (Explain dead ends in failing state transitions #412)
  • List the calls that reach an allowed state when the state is unknown (List the calls that reach an allowed state when the state is unknown #413)
  • Build each class's state machine once per checker (Build each class's state machine once per checker #414)

The hint and state machine files are identical to feat/state-hints-unknown. Rebase-merging keeps the three commits separate.

Checklist

  • mvn test passes locally on top of the current main (Join an if on the instances its condition read #408, Check each branch of a conditional expression under its condition #409)

🤖 Generated with Claude Code

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

All reactions
CatarinaGamboa and others added 3 commits October 24, 2026 12:13
@CatarinaGamboa @claude
Explain dead ends in failing state transitions
0b5259e
When no sequence of calls leads to a state that allows the failing
call, the error says so and suggests the constructors that create the
object in such a state, or, if the object can be in several states
depending on the path taken, asks for the same state on every path.
When no call is allowed in the object's state, the error says it is a
final state.

These claims are only made when the state machine models every
state-changing method of the class and of its super and subclasses.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
@CatarinaGamboa @claude
List the calls that reach an allowed state when the state is unknown
c1f53c5
When nothing is known about the receiver's state, the error lists the
constructors and the calls, with the states they are made from, that
lead to a state allowing the failing call.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
@CatarinaGamboa @claude
Build each class's state machine once per checker (#414)
dc4ef43
Stacked on #413.

## Description
Building a state machine takes about 25 SMT checks for a small class,
and each check creates its own Z3 context. The machine only depends on
specifications, which are all registered in the first pass, so
`TypeChecker` now builds it on the first state error and reuses it for
every later error in the same checker. Finding the receiver's current
states still runs at each error.

Other options considered:
- **Eagerly, after the first pass, for every class with `@StateSet`:**
correct programs would pay the SMT cost too. Only worth it once
something needs machines without an error, e.g. the fsm diagram.
- **Once per run, cached in `Context`:** `Context` would depend on
`state_machine`, which depends on `context` (a package cycle). The other
way is a static cache that tracks when specifications change. Neither
seemed worth it for the extra hits across classes.

## Cost
Extra time of `TestExamples` on the 269 cases that exist on `main`
(summed per case, one run each; `main` takes 80.3 s):
- #413, without the cache: +29.5 s
- with this PR: +10.6 s, spread over the 54 files with state errors

## Type of change
- [x] Code refactoring

## Checklist
- [x] `mvn test` passes locally

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <[email protected]>
@CatarinaGamboa
CatarinaGamboa merged commit 69895fa into main Oct 24, 2026
1 check passed

Uh oh!

There was an error while loading. Please reload this page.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Uh oh!

There was an error while loading. Please reload this page.

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Uh oh!

There was an error while loading. Please reload this page.

1 participant

@CatarinaGamboa
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.
You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.