wehub-resource-sync
a06f331eb8
CI / benchmark (push) Has been skipped
install-script / posix-syntax (push) Successful in 6m1s
CI / build-onnx (push) Failing after 6m43s
init-smoke / dry-run (push) Failing after 15m57s
security / govulncheck (push) Has been cancelled
security / trivy-fs (push) Has been cancelled
CI / test (1.26, ubuntu-latest) (push) Has been cancelled
Scorecard supply-chain security / Scorecard analysis (push) Has been cancelled
CI / test (1.26, macos-latest) (push) Has been cancelled
CI / build-windows (push) Has been cancelled
CI / lint (push) Has been cancelled
install-script / powershell-syntax (push) Has been cancelled
install-script / install (macos-14) (push) Has been cancelled
install-script / install (ubuntu-latest) (push) Has been cancelled
69 lines
2.2 KiB
Go
69 lines
2.2 KiB
Go
package lsp
|
|
|
|
import (
|
|
"encoding/json"
|
|
"sync"
|
|
"sync/atomic"
|
|
"testing"
|
|
"time"
|
|
|
|
"github.com/stretchr/testify/assert"
|
|
"github.com/stretchr/testify/require"
|
|
)
|
|
|
|
// TestLSP_Enrich_CrashLoopGuardAbandonsPass pins the crash-loop guard: a server
|
|
// that connects cleanly but keeps exiting mid-request (the clangd/clang-tidy
|
|
// failure shape) must not pin the pass in an endless crash -> reconnect -> crash
|
|
// loop. After the per-pass cap the provider's enrichment is abandoned with an
|
|
// error and a bounded reconnect count, and whatever landed earlier stays.
|
|
func TestLSP_Enrich_CrashLoopGuardAbandonsPass(t *testing.T) {
|
|
t.Setenv("GORTEX_LSP_SWEEP", "full") // exercise the full per-file sweep, not the demand-gated default
|
|
repoRoot, g := seedRepo(t, 60)
|
|
|
|
server1 := newInstrumentedServer()
|
|
p, cleanup := providerWithInstrumentedServer(t, server1, []string{"go"}, 4)
|
|
defer cleanup()
|
|
// Keep backoff tiny so the crash loop churns quickly.
|
|
p.dialBackoffStart = 1 * time.Millisecond
|
|
p.maxDialBackoff = 2 * time.Millisecond
|
|
|
|
kill := func(c *Client) {
|
|
c.mu.Lock()
|
|
if !c.closed {
|
|
c.closed = true
|
|
close(c.done)
|
|
}
|
|
c.mu.Unlock()
|
|
}
|
|
// dyingHover kills its own client on the first hover, so the initial server
|
|
// and every reconnect's replacement die mid-request in turn.
|
|
dyingHover := func(c *Client) func(json.RawMessage) (any, *jsonRPCError) {
|
|
var once sync.Once
|
|
return func(_ json.RawMessage) (any, *jsonRPCError) {
|
|
once.Do(func() { kill(c) })
|
|
return nil, &jsonRPCError{Code: -32603, Message: "server exited"}
|
|
}
|
|
}
|
|
server1.handle("textDocument/hover", dyingHover(p.client))
|
|
|
|
var reconnects atomic.Int64
|
|
p.connectOnce = func(_ string) error {
|
|
reconnects.Add(1)
|
|
srv := newInstrumentedServer()
|
|
c, in, out, cl := newPipedClient(t)
|
|
go srv.run(in, out)
|
|
t.Cleanup(cl)
|
|
srv.handle("textDocument/hover", dyingHover(c))
|
|
p.client = c
|
|
return nil
|
|
}
|
|
|
|
err := runEnrich(t, p, g, repoRoot, 20*time.Second)
|
|
require.Error(t, err, "a server that keeps crashing must abandon the pass, not loop forever")
|
|
|
|
n := int(reconnects.Load())
|
|
assert.Greater(t, n, 0, "should have attempted at least one reconnect")
|
|
assert.LessOrEqual(t, n, maxEnrichReconnectCycles+1,
|
|
"reconnects must be bounded by the crash-loop cap, got %d", n)
|
|
}
|