json_filter/tool/gabs.go

259 lines
6.7 KiB
Go
Raw Normal View History

2023-02-12 18:26:27 +08:00
// Package json_tool ...
//
// Description : json_tool ...
//
// Author : go_developer@163.com<张德满>
//
// Date : 2022/01/22 9:19 PM
package json_tool
import (
"fmt"
"io"
"log"
"os"
"strings"
2023-02-12 18:26:27 +08:00
"github.com/pkg/errors"
"github.com/tidwall/gjson"
2023-09-01 16:26:17 +08:00
"github.com/tidwall/sjson"
2023-02-12 18:26:27 +08:00
)
2023-09-01 18:17:20 +08:00
// FilterOption 过滤选项
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 17:59 2023/9/1
type FilterOption struct {
DebugModel bool // 调试模式
LogInstance io.Writer // 日志实例
2023-09-01 18:17:20 +08:00
}
2023-02-12 18:26:27 +08:00
// FilterDataRule 参数过滤规则
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 2022/1/22 9:44 PM
type FilterDataRule struct {
SourceKey string // 原始数据路径
MapKey string // 提取后映射到的数据路径
DefaultValue interface{} // 原始数据路径不存在时的默认值
WithDefault bool // 是否使用默认值
}
// NewDataFilter 获取数据过滤方法实例
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 2022/1/22 9:50 PM
2023-09-01 18:17:20 +08:00
func NewDataFilter(source string, filterRule []*FilterDataRule, filterOption *FilterOption) *DataFilter {
if nil == filterOption {
filterOption = &FilterOption{}
}
2023-09-01 19:18:39 +08:00
if filterOption.DebugModel && nil == filterOption.LogInstance {
filterOption.LogInstance = os.Stdout
log.SetOutput(filterOption.LogInstance)
2023-09-01 19:18:39 +08:00
}
2023-09-02 12:35:03 +08:00
// 去除末尾的 .[]
for _, item := range filterRule {
item.MapKey = strings.TrimRight(item.MapKey, ".[]")
item.SourceKey = strings.TrimRight(item.SourceKey, ".[]")
}
2023-02-12 18:26:27 +08:00
return &DataFilter{
2023-09-01 16:26:17 +08:00
source: source,
filterRule: filterRule,
rewriteResult: "{}",
2023-09-01 18:17:20 +08:00
filterOption: filterOption,
2023-02-12 18:26:27 +08:00
}
}
// DataFilter 数据过滤
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 2022/1/22 9:20 PM
type DataFilter struct {
2023-09-01 16:26:17 +08:00
source string
filterRule []*FilterDataRule
2023-09-01 18:17:20 +08:00
rewriteResult string // json数据重写结果
filterOption *FilterOption // 过滤选项
2023-02-12 18:26:27 +08:00
}
// Filter 数据过滤
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 2022/1/22 9:36 PM
func (df *DataFilter) Filter() (string, error) {
df.logPrint(logLevelDebug, "source_data => ", df.source)
2023-02-12 18:26:27 +08:00
var (
2023-09-01 16:26:17 +08:00
err error
2023-02-12 18:26:27 +08:00
)
2023-09-01 16:26:17 +08:00
for _, itemRule := range df.filterRule {
sourceIsArr := df.isArrPath(itemRule.SourceKey)
mapIsArr := df.isArrPath(itemRule.MapKey)
if !sourceIsArr && !mapIsArr {
2023-09-01 16:26:17 +08:00
// 输入输出均不是数组, 最简单的场景
if err = df.setKV(itemRule); nil != err {
2023-02-12 18:26:27 +08:00
return "", err
}
2023-09-02 12:35:03 +08:00
continue
2023-02-12 18:26:27 +08:00
}
sourcePathArr := df.getArrPathList(itemRule.SourceKey)
mapPathArr := df.getArrPathList(itemRule.MapKey)
if len(mapPathArr) > len(sourcePathArr) {
df.logPrint(logLevelFatal, "映射的层级深度大于数据源深度", "source_path => "+itemRule.SourceKey, "map_path => "+itemRule.MapKey)
return "", fmt.Errorf("映射的层级深度大于数据源深度, source_path => %v map_path => %v", itemRule.SourceKey, itemRule.MapKey)
}
2023-09-02 12:35:03 +08:00
// 映射至非数组字段
if !mapIsArr {
2023-09-02 11:38:15 +08:00
if err = df.setValue(itemRule.MapKey, df.getDataAsSlice(df.source, df.getArrPathList(itemRule.SourceKey))); nil != err {
df.logPrint(logLevelFatal, "映射非数组, 数据源为数组, 设置失败", "source_path => "+itemRule.SourceKey, "map_path => "+itemRule.MapKey, " err => "+err.Error())
return "", fmt.Errorf("映射的层级深度大于数据源深度, source_path => %v map_path => %v", itemRule.SourceKey, itemRule.MapKey)
}
2023-09-02 12:35:03 +08:00
continue
}
if len(mapPathArr) == len(sourcePathArr) {
// 数组深度一致
continue
}
2023-08-31 21:06:35 +08:00
}
2023-09-01 16:26:17 +08:00
return df.rewriteResult, nil
2023-02-12 18:26:27 +08:00
}
2023-09-01 16:26:17 +08:00
// isArrPath 是否为数组路径
2023-02-12 18:26:27 +08:00
//
// Author : go_developer@163.com<白茶清欢>
//
2023-09-01 16:26:17 +08:00
// Date : 16:00 2023/9/1
func (df *DataFilter) isArrPath(path string) bool {
2023-02-12 18:26:27 +08:00
return strings.Contains(path, "[]")
}
2023-09-02 12:35:03 +08:00
// setSameDeepArr 设置同深度的数组
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 12:19 2023/9/2
func (df *DataFilter) setSameDeepArr(sourceVal string, sourcePath string, mapPath string) error {
sourcePathArr := df.getArrPathList(sourcePath)
mapPathArr := df.getArrPathList(mapPath)
for idx, itemSourcePath := range sourcePathArr {
sourceValueArr := gjson.Get(sourceVal, sourcePath).Array()
}
return nil
}
// a.[].b.[].c.[].d
// e.[].f.[].g
2023-09-02 12:35:03 +08:00
// getDataAsSlice 抽取制定深度生成list
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 12:17 2023/9/2
func (df *DataFilter) getDataAsSlice(sourceData string, pathList []string) []interface{} {
//fmt.Println(sourceData, pathList)
result := make([]interface{}, 0)
if len(pathList) == 0 {
return result
}
if len(pathList) == 1 {
return []interface{}{gjson.Get(sourceData, pathList[0]).Value()}
}
for idx, itemPath := range pathList {
if len(pathList)-1 == idx {
val := gjson.Get(sourceData, itemPath).Value()
if nil == val {
return result
}
result = append(result, val)
return result
}
currentPathVal := gjson.Get(sourceData, itemPath).Array()
for _, sonItem := range currentPathVal {
result = append(result, df.getDataAsSlice(sonItem.String(), pathList[idx:])...)
}
}
return result
}
// setValue 设置值
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 22:04 2023/9/1
func (df *DataFilter) setValue(path string, value interface{}) error {
var (
err error
)
df.rewriteResult, err = sjson.Set(df.rewriteResult, path, value)
return err
}
2023-09-01 16:26:17 +08:00
// setKV 设置相关值
2023-02-12 18:26:27 +08:00
//
// Author : go_developer@163.com<白茶清欢>
//
2023-09-01 16:26:17 +08:00
// Date : 16:03 2023/9/1
func (df *DataFilter) setKV(rule *FilterDataRule) error {
var (
err error
)
sourceValue := gjson.Get(df.source, rule.SourceKey)
if sourceValue.Exists() {
// 原始数据存在对应路径
df.rewriteResult, err = sjson.Set(df.rewriteResult, rule.MapKey, sourceValue.Value())
return err
}
if !rule.WithDefault {
// 路径不存在, 且禁用默认值
return errors.New(rule.SourceKey + " : source path not found, and default value is forbidden")
}
// 使用默认值填充
df.rewriteResult, err = sjson.Set(df.rewriteResult, rule.MapKey, rule.DefaultValue)
return err
2023-02-12 18:26:27 +08:00
}
2023-09-01 18:17:20 +08:00
// getArrPathList 获取路径列表
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 21:32 2023/9/1
func (df *DataFilter) getArrPathList(inputPath string) []string {
pathArr := strings.Split(inputPath, "[]")
arr := make([]string, 0)
for _, item := range pathArr {
arr = append(arr, strings.Trim(item, "."))
}
return arr
}
2023-09-01 18:17:20 +08:00
// logPrint 打印日志
//
// Author : go_developer@163.com<白茶清欢>
//
// Date : 18:00 2023/9/1
2023-09-01 19:18:39 +08:00
func (df *DataFilter) logPrint(level string, msg string, logAttr ...interface{}) {
2023-09-01 18:17:20 +08:00
if !df.filterOption.DebugModel {
// 未开启调试模式
return
}
logData := append([]interface{}{level, msg}, logAttr...)
log.Println(logData...)
2023-09-01 18:17:20 +08:00
}
// 日志等级定义
const (
logLevelPanic = "PANIC"
2023-09-01 18:17:20 +08:00
logLevelFatal = "FATAL"
logLevelWarn = "WARN"
logLevelInfo = "INFO"
logLevelDebug = "DEBUG"
)