mirror of
https://github.com/sipeed/picoclaw.git
synced 2026-06-12 18:08:54 +00:00
Zhaoyikaiii
1f7cbd9164
Avoid rebuilding the entire system prompt on every BuildMessages() call by caching the static portion (identity, bootstrap, skills summary, memory) and only recomputing it when workspace source files change. Key changes: - ContextBuilder caches the static prompt behind an RWMutex with double-checked locking. Source file changes are detected via cheap os.Stat mtime checks so no explicit invalidation is needed. - Track file existence at cache time (existedAtCache map) so that newly created or deleted bootstrap/memory files also trigger a rebuild — the old modifiedSince() silently returned false on os.IsNotExist. - Walk the skills directory recursively with filepath.WalkDir to catch content-only edits at any nesting depth; directory mtime alone misses in-place file modifications on most filesystems. - ToolRegistry.sortedToolNames() sorts tool names before iteration, ensuring deterministic tool definition order across calls — a prerequisite for LLM-side prefix/KV cache reuse. - Merge all context (static + dynamic + summary) into a single system message for provider compatibility: the Anthropic adapter extracts messages[0] as the top-level system parameter, and Codex reads only the first system message as instructions. - Fix a data race in BuildMessages() where cachedSystemPrompt was read without holding the lock in a debug log statement. - Add tests: single system message invariant, mtime auto-invalidation, new-file creation detection, skill file content change, explicit InvalidateCache, cache stability, concurrent access (20 goroutines x 50 iterations, passes go test -race), and a benchmark.
82 lines
2.2 KiB
Go
82 lines
2.2 KiB
Go
package providers
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
|
|
"github.com/sipeed/picoclaw/pkg/providers/protocoltypes"
|
|
)
|
|
|
|
type (
|
|
ToolCall = protocoltypes.ToolCall
|
|
FunctionCall = protocoltypes.FunctionCall
|
|
LLMResponse = protocoltypes.LLMResponse
|
|
UsageInfo = protocoltypes.UsageInfo
|
|
Message = protocoltypes.Message
|
|
ToolDefinition = protocoltypes.ToolDefinition
|
|
ToolFunctionDefinition = protocoltypes.ToolFunctionDefinition
|
|
ExtraContent = protocoltypes.ExtraContent
|
|
GoogleExtra = protocoltypes.GoogleExtra
|
|
ContentBlock = protocoltypes.ContentBlock
|
|
CacheControl = protocoltypes.CacheControl
|
|
)
|
|
|
|
type LLMProvider interface {
|
|
Chat(
|
|
ctx context.Context,
|
|
messages []Message,
|
|
tools []ToolDefinition,
|
|
model string,
|
|
options map[string]any,
|
|
) (*LLMResponse, error)
|
|
GetDefaultModel() string
|
|
}
|
|
|
|
type StatefulProvider interface {
|
|
LLMProvider
|
|
Close()
|
|
}
|
|
|
|
// FailoverReason classifies why an LLM request failed for fallback decisions.
|
|
type FailoverReason string
|
|
|
|
const (
|
|
FailoverAuth FailoverReason = "auth"
|
|
FailoverRateLimit FailoverReason = "rate_limit"
|
|
FailoverBilling FailoverReason = "billing"
|
|
FailoverTimeout FailoverReason = "timeout"
|
|
FailoverFormat FailoverReason = "format"
|
|
FailoverOverloaded FailoverReason = "overloaded"
|
|
FailoverUnknown FailoverReason = "unknown"
|
|
)
|
|
|
|
// FailoverError wraps an LLM provider error with classification metadata.
|
|
type FailoverError struct {
|
|
Reason FailoverReason
|
|
Provider string
|
|
Model string
|
|
Status int
|
|
Wrapped error
|
|
}
|
|
|
|
func (e *FailoverError) Error() string {
|
|
return fmt.Sprintf("failover(%s): provider=%s model=%s status=%d: %v",
|
|
e.Reason, e.Provider, e.Model, e.Status, e.Wrapped)
|
|
}
|
|
|
|
func (e *FailoverError) Unwrap() error {
|
|
return e.Wrapped
|
|
}
|
|
|
|
// IsRetriable returns true if this error should trigger fallback to next candidate.
|
|
// Non-retriable: Format errors (bad request structure, image dimension/size).
|
|
func (e *FailoverError) IsRetriable() bool {
|
|
return e.Reason != FailoverFormat
|
|
}
|
|
|
|
// ModelConfig holds primary model and fallback list.
|
|
type ModelConfig struct {
|
|
Primary string
|
|
Fallbacks []string
|
|
}
|