fix: handle async 202 job polling for analyze#16
Merged
greynewell merged 1 commit intosupermodeltools:mainfrom Apr 2, 2026
Merged
fix: handle async 202 job polling for analyze#16greynewell merged 1 commit intosupermodeltools:mainfrom
greynewell merged 1 commit intosupermodeltools:mainfrom
Conversation
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
WalkthroughThe API client was refactored to an asynchronous job flow: Changes
Sequence DiagramsequenceDiagram
participant Client
participant Analyze
participant postZip
participant Server
Client->>Analyze: Analyze(ctx, zip)
loop poll until done or error
Analyze->>postZip: postZip(ctx, zip)
postZip->>Server: POST /.../analyze (multipart ZIP)
Server-->>postZip: JobResponse(status, jobId, retryAfter, error?, result?)
postZip-->>Analyze: return JobResponse
alt status == "completed"
Analyze->>Analyze: unmarshal JobResponse.Result -> jobResult.Graph
Analyze-->>Client: return *Graph
else status == "pending" or "processing"
Analyze->>Analyze: wait retryAfter (fallback 5s) or ctx.Done()
else error set or unexpected status
Analyze-->>Client: return error
end
end
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@internal/api/client.go`:
- Around line 72-76: Check for null/empty payloads before attempting to
unmarshal the job result: in the code that uses json.Unmarshal into jobResult
(the jobResult type and the result variable), first verify job.Result is not nil
and not an empty/whitespace/"null" payload (e.g., len==0 or trimmed equals
"null") and return a descriptive error if it is; after unmarshalling, also
validate that result.Graph is present/non-empty and return an error if the graph
is missing so a completed job with result:null cannot silently succeed.
- Around line 47-63: The polling loop in internal/api/client.go that checks
job.Status and calls c.postZip can hang indefinitely; add a hard stop by
introducing a max poll guard (e.g., maxPollDuration or maxPollAttempts) measured
from a start time (time.Now()) or a counter, and after the limit is exceeded
return a clear error (or context.DeadlineExceeded) instead of looping forever;
keep existing handling of job.RetryAfter and the ctx.Done() case, but before
each sleep/iteration check the elapsed time or attempts and bail out with an
error referencing the job/idempotency context so callers can handle it.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 67f8d546-e468-4ded-8491-b68a54874a37
📒 Files selected for processing (2)
internal/api/client.gointernal/api/types.go
Comment on lines
+47
to
+63
| // Poll until the job completes. | ||
| for job.Status == "pending" || job.Status == "processing" { | ||
| wait := time.Duration(job.RetryAfter) * time.Second | ||
| if wait <= 0 { | ||
| wait = 5 * time.Second | ||
| } | ||
| select { | ||
| case <-ctx.Done(): | ||
| return nil, ctx.Err() | ||
| case <-time.After(wait): | ||
| } | ||
|
|
||
| job, err = c.postZip(ctx, zipPath, idempotencyKey) | ||
| if err != nil { | ||
| return nil, err | ||
| } | ||
| } |
There was a problem hiding this comment.
Add a hard stop for polling to avoid indefinite hangs.
Right now, Line 48 can loop forever if the API never reaches a terminal state. At call sites (internal/analyze/handler.go Line 57, internal/find/handler.go Line 161), ctx is passed through without a timeout wrapper, so this can block indefinitely.
Suggested guardrail
func (c *Client) Analyze(ctx context.Context, zipPath, idempotencyKey string) (*Graph, error) {
+ if _, hasDeadline := ctx.Deadline(); !hasDeadline {
+ var cancel context.CancelFunc
+ ctx, cancel = context.WithTimeout(ctx, defaultTimeout)
+ defer cancel()
+ }
+
job, err := c.postZip(ctx, zipPath, idempotencyKey)
if err != nil {
return nil, err
}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@internal/api/client.go` around lines 47 - 63, The polling loop in
internal/api/client.go that checks job.Status and calls c.postZip can hang
indefinitely; add a hard stop by introducing a max poll guard (e.g.,
maxPollDuration or maxPollAttempts) measured from a start time (time.Now()) or a
counter, and after the limit is exceeded return a clear error (or
context.DeadlineExceeded) instead of looping forever; keep existing handling of
job.RetryAfter and the ctx.Done() case, but before each sleep/iteration check
the elapsed time or attempts and bail out with an error referencing the
job/idempotency context so callers can handle it.
Comment on lines
+72
to
+76
| var result jobResult | ||
| if err := json.Unmarshal(job.Result, &result); err != nil { | ||
| return nil, fmt.Errorf("decode graph result: %w", err) | ||
| } | ||
| return &result.Graph, nil |
There was a problem hiding this comment.
Guard against completed responses with null/empty result payload.
A completed job with result: null would currently decode to an empty graph and look like success. Fail fast here so bad server payloads don’t silently pass.
Suggested validation
var result jobResult
+ if trimmed := bytes.TrimSpace(job.Result); len(trimmed) == 0 || bytes.Equal(trimmed, []byte("null")) {
+ return nil, fmt.Errorf("analysis completed without graph result")
+ }
if err := json.Unmarshal(job.Result, &result); err != nil {
return nil, fmt.Errorf("decode graph result: %w", err)
}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@internal/api/client.go` around lines 72 - 76, Check for null/empty payloads
before attempting to unmarshal the job result: in the code that uses
json.Unmarshal into jobResult (the jobResult type and the result variable),
first verify job.Result is not nil and not an empty/whitespace/"null" payload
(e.g., len==0 or trimmed equals "null") and return a descriptive error if it is;
after unmarshalling, also validate that result.Graph is present/non-empty and
return an error if the graph is missing so a completed job with result:null
cannot silently succeed.
greynewell
force-pushed
the
fix/async-job-polling
branch
from
cb38984 to
1a81429
Compare
The /v1/graphs/supermodel endpoint returns HTTP 202 with an async job envelope (status, jobId, retryAfter, result). The client was treating this as a completed response and unmarshalling directly into Graph, resulting in 0 nodes/relationships. Changes: - Add JobResponse and jobResult types for the async envelope - Poll with the same idempotency key until status is "completed" - Extract graph from the nested result.graph path - Extract postZip helper to avoid duplicating multipart construction Tested live against api.supermodeltools.com — correctly returns 362 files, 1527 functions, 7741 relationships for supermodel-public-api.
greynewell
force-pushed
the
fix/async-job-polling
branch
from
1a81429 to
ae6a423
Compare
This was referenced Apr 2, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary
{"status":"pending","jobId":"...","retryAfter":10,"result":null}) but the client was deserializing it directly intoGraph, producing 0 nodes/relationshipsstatus: "completed"result.graphpathChanges
internal/api/types.go— addJobResponseandjobResulttypesinternal/api/client.go— rewriteAnalyze()with poll loop, extractpostZiphelperTest results
Note
Depends on #15 for the correct endpoint path (
/v1/graphs/supermodel). This PR keeps the endpoint as-is to avoid conflicts — once #15 merges, theanalyzeEndpointconst just needs updating.Test plan
supermodel analyzeon a real repo returns non-zero node/relationship counts--forcebypasses cache and re-runs analysisSummary by CodeRabbit