ben/coze-studio
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

chore: replace all cn comments to en version by volc api (#313)

Browse Source
This commit is contained in:
tecvan
2025-07-31 15:18:11 +08:00
committed by GitHub
parent 91d6cdb430
commit 5abc63fba6
254 changed files with 5899 additions and 5844 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
backend/domain/knowledge/entity/strategy.go
View File

@@ -25,18 +25,18 @@ type RetrievalStrategy = knowledge.RetrievalStrategy
// ParsingStrategy for document parse before indexing
type ParsingStrategy struct {
ParsingType ParsingType `json:"parsing_type"` // 解析类型
ParsingType ParsingType `json:"parsing_type"` // parse type
// Doc
ExtractImage bool `json:"extract_image"` // 提取图片元素
ExtractTable bool `json:"extract_table"` // 提取表格元素
ImageOCR bool `json:"image_ocr"` // 图片 ocr
FilterPages []int `json:"filter_pages"` // 过滤页数
ExtractImage bool `json:"extract_image"` // Extract image elements
ExtractTable bool `json:"extract_table"` // Extract table elements
ImageOCR bool `json:"image_ocr"` // Image ocr
FilterPages []int `json:"filter_pages"` // filter pages
// Sheet
SheetID int64 `json:"sheet_id"` // xlsx sheet id
HeaderLine int `json:"header_line"` // 表头行
DataStartLine int `json:"data_start_line"` // 数据起始行
RowsCount int `json:"rows_count"` // 读取数据行数
HeaderLine int `json:"header_line"` // header row
DataStartLine int `json:"data_start_line"` // Data start row
RowsCount int `json:"rows_count"` // number of rows read
// Image
CaptionType *parser.ImageAnnotationType `json:"caption_type"`
@@ -52,13 +52,13 @@ const (
type ChunkingStrategy struct {
ChunkType parser.ChunkType `json:"chunk_type"`
// custom chunk config
ChunkSize int64 `json:"chunk_size"` // 分段最大长度
Separator string `json:"separator"` // 分段标识符
Overlap int64 `json:"overlap"` // 分段重叠
ChunkSize int64 `json:"chunk_size"` // maximum segmentation length
Separator string `json:"separator"` // segmentation identifier
Overlap int64 `json:"overlap"` // segmented overlap
TrimSpace bool `json:"trim_space"`
TrimURLAndEmail bool `json:"trim_url_and_email"`
// 按层级分段
MaxDepth int64 `json:"max_depth"` // 按层级分段时的最大层级
SaveTitle bool `json:"save_title"` // 保留层级标题
// segmentation by hierarchy
MaxDepth int64 `json:"max_depth"` // Maximum level when segmented by level
SaveTitle bool `json:"save_title"` // Preserve Hierarchical Titles
}