mirror of
https://github.com/garrytan/gstack.git
synced 2026-10-04 18:36:54 +02:00
test: retire the finding-count cluster and trim its helpers (C)
- C0/C1: the five never-green evals (skill-e2e-autoplan-chain and
skill-e2e-plan-{ceo,eng,design,devex}-finding-count) failed on harness and
budget, never on skill behavior; delete them, their touchfile/tier ids,
AUTOPLAN_CHAIN_BUDGET and the dedicated eighth periodic slice (--slices 7).
- C2: delete the helper groups whose only paid consumers were those files
(11 modules), trim claude-pty-runner and eng-seeded-coverage to the paid
closure, and delete the free replay tests whose assertions exercised only
that dead code (89 files, 135 orphaned fixtures). Blocks that used dead code
only as input for a live subject keep their assertions: the multiSelect
default moved to plan-review-decisions, runner PTY tests use inline caller
policies, and the timer-safe budget checks moved to eng-finding-retry-budget.
- The eight production-touching files stay except ceo-current-decision-record
(its template read only feeds the retired counter).
- CARVE_GUARDS.autoplan is behavioral 'none'; TODOS records the lost chain
and per-finding cadence coverage with their re-entry tests.
This commit is contained in:
1 parent
53e7f3212f
commit
6415690a18
317 files changed
+315
-54111
No files matched your search
@@ -134,7 +134,6 @@ import * as fs from 'node:fs'; import * as path from 'node:path'; import * as os
|
||||
import {launchClaudePty,resolveClaudeBinary} from ${JSON.stringify(helper('claude-pty-runner.ts'))};
|
||||
import {readPlanCountTranscript} from ${JSON.stringify(helper('plan-count-transcript.ts'))};
|
||||
import {readPendingQuestion} from ${JSON.stringify(helper('plan-count-pending-question.ts'))};
|
||||
import {autoplanSetupDecision} from ${JSON.stringify(helper('autoplan-setup-question.ts'))};
|
||||
if(resolveClaudeBinary()!==${JSON.stringify(fake)})throw Error('fake CLI binding failed');
|
||||
const root=${JSON.stringify(dir)},results=[];
|
||||
for(const enabled of [false,true]){
|
||||
@@ -147,18 +146,13 @@ try{await session.waitFor('Enter to select',{timeoutMs:5000,pollMs:20});
|
||||
const transcript=()=>readPlanCountTranscript(session.hermeticConfigDir,cwd);
|
||||
const pending=()=>readPendingQuestion(owned,cwd,session.hermeticConfigDir,startedAt,transcript());
|
||||
if(transcript().calls.length)throw Error('unpublished packet manufactured JSONL coverage');
|
||||
const seen=new Set(),inputs=[],first=await session.currentScreen();
|
||||
if(autoplanSetupDecision(first,new Set()).kind!=='waiting')throw Error('partial panel unexpectedly authorized input');
|
||||
const inputs=[];
|
||||
if(enabled){
|
||||
if(pending()?.source!=='pre_tool_use'||pending()?.answered)throw Error('pending source was lost or counted as answer');
|
||||
for(let step=0;step<3;step++){
|
||||
const current=await session.currentScreen(),call=pending();
|
||||
for(const [step,input] of ['1','1','\\r'].entries()){
|
||||
const call=pending();
|
||||
if(!call||transcript().calls.length)throw Error('pending identity lost or premature coverage');
|
||||
const action=autoplanSetupDecision(current,seen,call);
|
||||
if(action.kind!=='input')throw Error('full native packet not navigable: '+JSON.stringify({action,current,call}));
|
||||
for(const key of action.signatures)seen.add(key);
|
||||
if(autoplanSetupDecision(current,seen,call).kind!=='waiting')throw Error('redraw repeated input');
|
||||
inputs.push(action.input);session.send(action.input);
|
||||
inputs.push(input);session.send(input);
|
||||
await session.waitFor(step===0?'☒ '+call.questions[0].header:step===1?'Ready to submit your answers?':'NATIVE_PACKET_COMPLETE',{timeoutMs:3000,pollMs:20});
|
||||
}
|
||||
if(pending())throw Error('completed replay reopened pending identity');
|
||||
|
||||
Reference in new issue
Block a user