mirror of
https://github.com/infiniflow/ragflow.git
synced 2026-08-01 21:37:33 +08:00
feat(go-models): unify OpenAI-compatible token usage extraction (#17634)
## Summary Relate to #17284 . - Add `usage_parser.go`: `extractOpenAIUsage` / `extractOpenAIStreamUsage` with multi-field fallback (`prompt_tokens`/`input_tokens`, `completion_tokens`/`output_tokens`). - Add `parser_config.go`: `ParserConfig` struct and `OpenAIParserConfig`. - Add `response_handler.go`: `HandleNonStreamingResponse` / `HandleStreamingResponse` as single entry points that extract usage, populate `chatConfig.UsageResult`, and emit a `StreamUsage` log. - Extend `base_model.go` with `doRequest` / `doStreamRequest`. - Migrate `deepseek` driver to shared handlers, cutting ~100 lines. - Add `usage_parser_test.go`. --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Co-authored-by: Haruko386 <tryeverypossible@163.com> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
This commit is contained in:
77
internal/entity/models/usage_parser.go
Normal file
77
internal/entity/models/usage_parser.go
Normal file
@@ -0,0 +1,77 @@
|
||||
//
|
||||
// Copyright 2026 The InfiniFlow Authors. All Rights Reserved.
|
||||
//
|
||||
// Licensed under the Apache License, Version 2.0 (the "License");
|
||||
// you may not use this file except in compliance with the License.
|
||||
// You may obtain a copy of the License at
|
||||
//
|
||||
// http://www.apache.org/licenses/LICENSE-2.0
|
||||
//
|
||||
// Unless required by applicable law or agreed to in writing, software
|
||||
// distributed under the License is distributed on an "AS IS" BASIS,
|
||||
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
//
|
||||
|
||||
package models
|
||||
|
||||
import "encoding/json"
|
||||
|
||||
// extractOpenAIUsage extracts token usage from an OpenAI-compatible
|
||||
// non-streaming response body. Returns (nil, false) when the response
|
||||
// carries no usable usage block.
|
||||
func extractOpenAIUsage(body map[string]any) (*TokenUsage, bool) {
|
||||
rawUsage, ok := body["usage"].(map[string]any)
|
||||
if !ok {
|
||||
return nil, false
|
||||
}
|
||||
|
||||
return tokenUsageFromRaw(rawUsage), true
|
||||
}
|
||||
|
||||
// extractOpenAIStreamUsage extracts token usage from one OpenAI-compatible
|
||||
// streaming event. A missing or null usage field is not an error.
|
||||
func extractOpenAIStreamUsage(event map[string]any) (*TokenUsage, bool) {
|
||||
rawUsage, ok := event["usage"].(map[string]any)
|
||||
if !ok || rawUsage == nil {
|
||||
return nil, false
|
||||
}
|
||||
|
||||
return tokenUsageFromRaw(rawUsage), true
|
||||
}
|
||||
|
||||
func tokenUsageFromRaw(rawUsage map[string]any) *TokenUsage {
|
||||
usage := &TokenUsage{}
|
||||
usage.PromptTokens = extractToken(rawUsage, "prompt_tokens", "input_tokens")
|
||||
usage.CompletionTokens = extractToken(rawUsage, "completion_tokens", "output_tokens")
|
||||
usage.TotalTokens = extractToken(rawUsage, "total_tokens")
|
||||
|
||||
if usage.TotalTokens == 0 {
|
||||
usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
|
||||
}
|
||||
|
||||
return usage
|
||||
}
|
||||
|
||||
// extractToken reads a numeric field from a map, trying each key in order.
|
||||
// Returns 0 when no key is present or the value is not a number.
|
||||
func extractToken(m map[string]any, keys ...string) int {
|
||||
for _, k := range keys {
|
||||
if v, ok := m[k]; ok {
|
||||
switch val := v.(type) {
|
||||
case float64:
|
||||
return int(val)
|
||||
case int:
|
||||
return val
|
||||
case int64:
|
||||
return int(val)
|
||||
case json.Number:
|
||||
if n, err := val.Int64(); err == nil {
|
||||
return int(n)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
return 0
|
||||
}
|
||||
Reference in New Issue
Block a user