Code at the speed of thought – Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.
open_ai: Fix parsing stream usage chunks with null prompt cache tokens (#62983)
Enoch693140abac79
diff --git a/crates/open_ai/src/completion.rs b/crates/open_ai/src/completion.rsindex 893a14ca5ec..94028b976e7 100644--- a/crates/open_ai/src/completion.rs+++ b/crates/open_ai/src/completion.rs@@ -820,11 +820,13 @@ impl OpenAiEventMapper {let cache_creation_input_tokens = usage.prompt_tokens_details.as_ref()- .map_or(0, |details| details.cache_write_tokens);+ .and_then(|details| details.cache_write_tokens)+ .unwrap_or(0);let cache_read_input_tokens = usage.prompt_tokens_details.as_ref()- .map_or(0, |details| details.cached_tokens);+ .and_then(|details| details.cached_tokens)+ .unwrap_or(0);events.push(Ok(LanguageModelCompletionEvent::UsageUpdate(TokenUsage {input_tokens: prompt_tokens.saturating_sub(cache_creation_input_tokens)diff --git a/crates/open_ai/src/open_ai.rs b/crates/open_ai/src/open_ai.rsindex f9a66db8d38..2b79a5c823d 100644--- a/crates/open_ai/src/open_ai.rs+++ b/crates/open_ai/src/open_ai.rs@@ -515,6 +515,25 @@ mod tests {expected_efforts.as_slice());}++ #[test]+ fn stream_usage_with_null_prompt_cache_tokens_is_not_an_error() {+ let chunk = r#"{"choices":[],"usage":{"prompt_tokens":5,"completion_tokens":3,"total_tokens":8,"prompt_tokens_details":{"cached_tokens":0,"cache_write_tokens":null}}}"#;++ let super::ResponseStreamResult::Ok(event) =+ serde_json::from_str::<super::ResponseStreamResult>(chunk).unwrap()+ else {+ panic!("usage-only chunk with null cache token fields must not fail to parse");+ };++ let details = event+ .usage+ .unwrap()+ .prompt_tokens_details+ .expect("prompt_tokens_details should be present");+ assert_eq!(details.cached_tokens, Some(0));+ assert_eq!(details.cache_write_tokens, None);+ }}#[derive(Debug, Serialize, Deserialize)]@@ -754,11 +773,9 @@ pub struct FunctionChunk {#[derive(Clone, Serialize, Deserialize, Debug, Default)]pub struct PromptTokensDetails {/// Tokens read from a prompt cache.- #[serde(default)]- pub cached_tokens: u64,+ pub cached_tokens: Option<u64>,/// Tokens written to a prompt cache.- #[serde(default)]- pub cache_write_tokens: u64,+ pub cache_write_tokens: Option<u64>,}#[derive(Clone, Serialize, Deserialize, Debug)]