Like-api/relay/model/message.go

package model

import (
	"github.com/songquanpeng/one-api/common/logger"
)

type Message struct {
	Role             string  `json:"role,omitempty"`
	Content          any     `json:"content,omitempty"`
	ReasoningContent any     `json:"reasoning_content,omitempty"`
	ReasoningEncryptedContent string `json:"reasoning_encrypted_content,omitempty"`
	Name             *string `json:"name,omitempty"`
	ToolCalls        []Tool  `json:"tool_calls,omitempty"`
	ToolCallId       string  `json:"tool_call_id,omitempty"`
}

func (m Message) IsStringContent() bool {
	_, ok := m.Content.(string)
	return ok
}

func (m Message) StringContent() string {
	content, ok := m.Content.(string)
	if ok {
		return content
	}
	contentList, ok := m.Content.([]any)
	if ok {
		var contentStr string
		for _, contentItem := range contentList {
			contentMap, ok := contentItem.(map[string]any)
			if !ok {
				continue
			}
			if contentMap["type"] == ContentTypeText {
				if subStr, ok := contentMap["text"].(string); ok {
					contentStr += subStr
				}
			}
		}
		return contentStr
	}
	return ""
}

func (m Message) ParseContent() []MessageContent {
	var contentList []MessageContent
	content, ok := m.Content.(string)
	if ok {
		contentList = append(contentList, MessageContent{
			Type: ContentTypeText,
			Text: content,
		})
		return contentList
	}
	anyList, ok := m.Content.([]any)
	if ok {
		logger.Debugf(nil, "=== ParseContent: Found array content with %d items ===", len(anyList))
		for i, contentItem := range anyList {
			contentMap, ok := contentItem.(map[string]any)
			if !ok {
				logger.Debugf(nil, "=== ParseContent: Item %d is not a map, type=%T ===", i, contentItem)
				continue
			}
			logger.Debugf(nil, "=== ParseContent: Item %d has type=%v ===", i, contentMap["type"])
			switch contentMap["type"] {
			case ContentTypeText:
				if subStr, ok := contentMap["text"].(string); ok {
					contentList = append(contentList, MessageContent{
						Type: ContentTypeText,
						Text: subStr,
					})
				}
			case ContentTypeImageURL:
				logger.Debugf(nil, "=== ParseContent: Processing image_url item ===")
				if subObj, ok := contentMap["image_url"].(map[string]any); ok {
					if url, ok := subObj["url"].(string); ok {
						logger.Debugf(nil, "=== ParseContent: Found image URL: %s ===", truncateURL(url))
						contentList = append(contentList, MessageContent{
							Type: ContentTypeImageURL,
							ImageURL: &ImageURL{
								Url: url,
							},
						})
					} else {
						logger.Errorf(nil, "=== ParseContent: image_url.url is not a string, type=%T ===", subObj["url"])
					}
				} else {
					logger.Errorf(nil, "=== ParseContent: image_url is not a map, type=%T ===", contentMap["image_url"])
				}
			case ContentTypeVideoURL:
				if subObj, ok := contentMap["video_url"].(map[string]any); ok {
					if url, ok := subObj["url"].(string); ok {
						contentList = append(contentList, MessageContent{
							Type:     ContentTypeVideoURL,
							VideoURL: &VideoURL{Url: url},
						})
					}
				}
			case ContentTypeInputAudio:
				if subObj, ok := contentMap["input_audio"].(map[string]any); ok {
					data, _ := subObj["data"].(string)
					format, _ := subObj["format"].(string)
					if data != "" {
						contentList = append(contentList, MessageContent{
							Type:       ContentTypeInputAudio,
							InputAudio: &InputAudio{Data: data, Format: format},
						})
					}
				}
			}
		}
		imageCount := 0
		for _, c := range contentList {
			if c.Type == ContentTypeImageURL {
				imageCount++
			}
		}
		logger.Debugf(nil, "=== ParseContent: Parsed %d content items, image_count=%d ===", len(contentList), imageCount)
		return contentList
	}
	logger.Debugf(nil, "=== ParseContent: Content is neither string nor array, type=%T ===", m.Content)
	return nil
}

func truncateURL(url string) string {
	if len(url) > 100 {
		return url[:100] + "..."
	}
	return url
}

type ImageURL struct {
	Url    string `json:"url,omitempty"`
	Detail string `json:"detail,omitempty"`
}

type VideoURL struct {
	Url string `json:"url,omitempty"`
}

type InputAudio struct {
	Data   string `json:"data,omitempty"`   // base64-encoded audio (no data: prefix)
	Format string `json:"format,omitempty"` // e.g. "mp3", "wav", "webm", "ogg"
}

type MessageContent struct {
	Type       string      `json:"type,omitempty"`
	Text       string      `json:"text"`
	ImageURL   *ImageURL   `json:"image_url,omitempty"`
	VideoURL   *VideoURL   `json:"video_url,omitempty"`
	InputAudio *InputAudio `json:"input_audio,omitempty"`
}