Skip to content

fix: Prevent undefined result in reasoning budget assertions #523

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
FL4TLiN3 merged 2 commits into from
Feb 15, 2026
Merged

fix: Prevent undefined result in reasoning budget assertions #523

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f14a29d
fix: Prevent undefined result in reasoning budget assertions
FL4TLiN3 Feb 15, 2026
9591fec
style: Fix formatting in reasoning-budget test
FL4TLiN3 Feb 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 8 additions & 13 deletions e2e/perstack-cli/reasoning-budget.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,8 +109,7 @@ describe("Reasoning Budget", () => {
// Extended thinking should produce reasoning tokens or thinking text
// Note: AI SDK currently doesn't report reasoningTokens for Anthropic,
// but we can verify thinking content is generated
const hasThinking =
result.reasoningTokens > 0 || (result.thinking && result.thinking.length > 0)
const hasThinking = result.reasoningTokens > 0 || (result.thinking?.length ?? 0) > 0
expect(hasThinking).toBe(true)
},
LLM_TIMEOUT,
Expand Down Expand Up @@ -166,10 +165,9 @@ describe("Reasoning Budget", () => {
// Both should produce reasoning tokens or thinking text
// Note: AI SDK currently doesn't report reasoningTokens for Anthropic
const minimalHasThinking =
minimalResult.reasoningTokens > 0 ||
(minimalResult.thinking && minimalResult.thinking.length > 0)
minimalResult.reasoningTokens > 0 || (minimalResult.thinking?.length ?? 0) > 0
const highHasThinking =
highResult.reasoningTokens > 0 || (highResult.thinking && highResult.thinking.length > 0)
highResult.reasoningTokens > 0 || (highResult.thinking?.length ?? 0) > 0
expect(minimalHasThinking).toBe(true)
expect(highHasThinking).toBe(true)
},
Expand Down Expand Up @@ -202,8 +200,7 @@ describe("Reasoning Budget", () => {
expect(result.success).toBe(true)
// OpenAI reasoning models may not always surface reasoning token counts,
// so verify either tokens or thinking text is present
const hasReasoning =
result.reasoningTokens > 0 || (result.thinking && result.thinking.length > 0)
const hasReasoning = result.reasoningTokens > 0 || (result.thinking?.length ?? 0) > 0
expect(hasReasoning).toBe(true)
},
LLM_TIMEOUT,
Expand All @@ -221,7 +218,7 @@ describe("Reasoning Budget", () => {

// At least one budget level should produce reasoning tokens or thinking text
const highHasReasoning =
highResult.reasoningTokens > 0 || (highResult.thinking && highResult.thinking.length > 0)
highResult.reasoningTokens > 0 || (highResult.thinking?.length ?? 0) > 0
expect(highHasReasoning).toBe(true)
},
LLM_TIMEOUT * 2, // Two API calls
Expand Down Expand Up @@ -252,8 +249,7 @@ describe("Reasoning Budget", () => {

expect(result.success).toBe(true)
// Flash thinking should produce reasoning tokens or thinking text
const hasThinking =
result.reasoningTokens > 0 || (result.thinking && result.thinking.length > 0)
const hasThinking = result.reasoningTokens > 0 || (result.thinking?.length ?? 0) > 0
expect(hasThinking).toBe(true)
},
LLM_TIMEOUT,
Expand All @@ -271,10 +267,9 @@ describe("Reasoning Budget", () => {

// Both should produce reasoning tokens or thinking text
const minimalHasThinking =
minimalResult.reasoningTokens > 0 ||
(minimalResult.thinking && minimalResult.thinking.length > 0)
minimalResult.reasoningTokens > 0 || (minimalResult.thinking?.length ?? 0) > 0
const highHasThinking =
highResult.reasoningTokens > 0 || (highResult.thinking && highResult.thinking.length > 0)
highResult.reasoningTokens > 0 || (highResult.thinking?.length ?? 0) > 0
expect(minimalHasThinking).toBe(true)
expect(highHasThinking).toBe(true)
},
Expand Down