淺談JsonPath
JsonPath,類似于XPath在XML中的作用。其提供了對Json格式數(shù)據(jù)的解析能力

操作符
$
查詢的根節(jié)點(diǎn),其中根節(jié)點(diǎn)可以是數(shù)組或?qū)ο?/p>

.或 ['name']
在JsonPath表達(dá)式可以使用點(diǎn)語法、括號語法來訪問子節(jié)點(diǎn)

..
可進(jìn)行遞歸搜索

['name'(, 'name')]
對于括號語法而言,其還支持訪問多個子節(jié)點(diǎn)

[(,)]
針對數(shù)組元素的索引操作符,其中0為起始索引。負(fù)數(shù)索引表示數(shù)組中倒數(shù)第幾個元素,例如,-1表示倒數(shù)第一個元素,-2表示倒數(shù)第2個元素

[start:end]
針對數(shù)組元素的切片操作符,其表示獲取索引在[start,end)區(qū)間范圍的元素。顯然這里是左閉右開區(qū)間

特別地,當(dāng)start省略時,默認(rèn)為0;當(dāng)end省略時,則可以獲取數(shù)組中剩余部分的全部元素。但二者不可同時省略

此外,在切片操作符中同樣支持負(fù)數(shù)索引

*
通配符,在任何需要名稱、數(shù)字的地方都可以使用

@
用于下文所述過濾器表達(dá)式當(dāng)中,用于指代過濾器當(dāng)前正在處理的節(jié)點(diǎn)對象。其效果類似于Java中的this關(guān)鍵字
[?()]
過濾器表達(dá)式,表達(dá)式結(jié)果必須是布爾值。下圖即是一個典型的使用過濾器對數(shù)組元素進(jìn)行過濾的示例

其中過濾器支持地操作符,常見地有:
「==」 :判斷是否相等 「!=」 :判斷是否不相等 「<」 :判斷是否小于 「<=」 :判斷是否小于等于 「>」 :判斷是否大于 「>=」 :判斷是否大于等于 「=~」 :判斷左側(cè) 是否 匹配右側(cè)的正則。例如:[? (@.age =~ /\d+/)] 「in」 :判斷左側(cè) 是否 存在于 右側(cè)的集合中。例如:[? (@.size in ['S','M','L'])] 「nin」 :判斷左側(cè) 是否 不存在于 右側(cè)的集合中。例如:[? (@.size nin ['S','M','L'])] 「subsetof」 :判斷左側(cè) 是否為 右側(cè)集合的子集。例如:[? (@.sizes subsetof ['S','M','L'])] 「anyof」 :判斷左側(cè) 是否與 右側(cè)集合 存在交集。例如:[? (@.sizes anyof ['S','M','L'])] 「noneof」 :判斷左側(cè) 是否與 右側(cè)集合 無交集。例如:[? (@.sizes noneof ['S','M','L'])] 「size」 :判斷左側(cè) 數(shù)組長度 或 字符串長度 是否為 指定值。例如:[? (@.name size 3)]
此外對于

基于Java的實(shí)踐——Jayway JsonPath
Jayway JsonPath則提供了Java版本的實(shí)現(xiàn),方便開發(fā)者進(jìn)行集成使用。只需引入下述依賴即可
<!--Json Path-->
<dependency>
<groupId>com.jayway.jsonpath</groupId>
<artifactId>json-path</artifactId>
<version>2.7.0</version>
</dependency>
為了便于后續(xù)行文演示方便,這里準(zhǔn)備了一個較為復(fù)雜的Json數(shù)據(jù)
{
"store":{
"book":[
{
"category":"reference",
"author":"Nigel Rees",
"title":"Sayings of the Century",
"price":8.95
},
{
"category":"fiction",
"author":"Evelyn Waugh",
"title":"Sword of Honour",
"price":12.99
},
{
"category":"fiction",
"author":"Herman Melville",
"title":"Moby Dick",
"isbn":"0-553-21311-3",
"price":8.99
},
{
"category":"fiction",
"author":"J. R. R. Tolkien",
"title":"The Lord of the Rings",
"isbn":"0-395-19395-8",
"price":22.99
}
],
"bicycle":{
"color":"red",
"price":19.95
},
"clothes":[
{
"name":"牛仔褲",
"sizes":"S",
"price":94
},
{
"name":"背心",
"sizes":"M",
"price":48
},
{
"name":"裙子",
"sizes":["S", "M"],
"price":1.24
},
{
"name":"羊毛衫",
"sizes":["XS", "XL"],
"price":78.99
},
{
"name":"Polo衫",
"sizes":["XS", "XL", "M"],
"price":18.99
}
]
},
"expensive":10
}
快速入門
Jayway JsonPath 非常方便,開箱即用
public class Demo1 {
private String json = "{\"store\":{\"book\":[{\"category\":\"reference\",\"author\":\"Nigel Rees\",\"title\":\"Sayings of the Century\",\"price\":8.95},{\"category\":\"fiction\",\"author\":\"Evelyn Waugh\",\"title\":\"Sword of Honour\",\"price\":12.99},{\"category\":\"fiction\",\"author\":\"Herman Melville\",\"title\":\"Moby Dick\",\"isbn\":\"0-553-21311-3\",\"price\":8.99},{\"category\":\"fiction\",\"author\":\"J. R. R. Tolkien\",\"title\":\"The Lord of the Rings\",\"isbn\":\"0-395-19395-8\",\"price\":22.99}],\"bicycle\":{\"color\":\"red\",\"price\":19.95},\"clothes\":[{\"name\":\"牛仔褲\",\"sizes\":\"S\",\"price\":94},{\"name\":\"背心\",\"sizes\":\"M\",\"price\":48},{\"name\":\"裙子\",\"sizes\":[\"S\",\"M\"],\"price\":1.24},{\"name\":\"羊毛衫\",\"sizes\":[\"XS\",\"XL\"],\"price\":78.99},{\"name\":\"Polo衫\",\"sizes\":[\"XS\",\"M\",\"XL\"],\"price\":18.99}]},\"expensive\":10}\n";
/**
* 每次讀取時均會解析
*/
@Test
public void start1() {
List<String> authors = JsonPath.read(json, "$.store.book[*].author");
System.out.println("authors :" + authors);
}
/**
* 多次讀取路徑時, 避免重復(fù)解析
*/
@Test
public void start2() {
Object document = Configuration.defaultConfiguration()
.jsonProvider()
.parse(json);
List<String> prices = JsonPath.read(document, "$.store.book[*].price");
List<String> names = JsonPath.read(document, "$.store.clothes[*].name");
System.out.println("prices :" + prices);
System.out.println("names :" + names);
}
}
其中,start1的方式適用于僅僅需要讀取1次數(shù)據(jù);而start2則通過先解析后讀取的方式,適用于多次讀取的場景。避免start1方式重復(fù)解析帶來損耗

反序列化
在Jayway JsonPath中提供了多種JsonProvider,其中默認(rèn)的為JsonSmartJsonProvider。這里我們期望能夠直接對讀取的數(shù)據(jù)進(jìn)行反序列化,這里我們選用JacksonJsonProvider,此時要求jackson-databind依賴的版本至少為2.4.5。故這里我們先添加Jackson依賴
<!--Jackson-->
<dependency>
<groupId>com.fasterxml.jackson.core</groupId>
<artifactId>jackson-databind</artifactId>
<version>2.12.7</version>
</dependency>
Demo如下所示
import com.jayway.jsonpath.Configuration;
import com.jayway.jsonpath.JsonPath;
import com.jayway.jsonpath.ReadContext;
import com.jayway.jsonpath.TypeRef;
import com.jayway.jsonpath.spi.mapper.JacksonMappingProvider;
public class Demo1 {
private String json = "{\"store\":{\"book\":[{\"category\":\"reference\",\"author\":\"Nigel Rees\",\"title\":\"Sayings of the Century\",\"price\":8.95},{\"category\":\"fiction\",\"author\":\"Evelyn Waugh\",\"title\":\"Sword of Honour\",\"price\":12.99},{\"category\":\"fiction\",\"author\":\"Herman Melville\",\"title\":\"Moby Dick\",\"isbn\":\"0-553-21311-3\",\"price\":8.99},{\"category\":\"fiction\",\"author\":\"J. R. R. Tolkien\",\"title\":\"The Lord of the Rings\",\"isbn\":\"0-395-19395-8\",\"price\":22.99}],\"bicycle\":{\"color\":\"red\",\"price\":19.95},\"clothes\":[{\"name\":\"牛仔褲\",\"sizes\":\"S\",\"price\":94},{\"name\":\"背心\",\"sizes\":\"M\",\"price\":48},{\"name\":\"裙子\",\"sizes\":[\"S\",\"M\"],\"price\":1.24},{\"name\":\"羊毛衫\",\"sizes\":[\"XS\",\"XL\"],\"price\":78.99},{\"name\":\"Polo衫\",\"sizes\":[\"XS\",\"M\",\"XL\"],\"price\":18.99}]},\"expensive\":10}\n";
@Test
public void start3() {
// 使用 JacksonJsonProvider 實(shí)現(xiàn)反序列化
Configuration conf = Configuration
.builder()
.mappingProvider( new JacksonMappingProvider() )
.build();
ReadContext ctx = JsonPath.using( conf )
.parse(json);
TypeRef<List<Book>> typeRef = new TypeRef<List<Book>>() {};
List<Book> books = ctx.read("$.store.book[*]", typeRef);
books.forEach( System.out::println );
}
}
@AllArgsConstructor
@NoArgsConstructor
@Builder
@Data
class Book {
private String category;
private String title;
private String author;
private Double price;
private String isbn;
}
效果如下所示

過濾器謂詞
前面提到JsonPath中支持過濾器表達(dá)式,為此在Jayway JsonPath中提供了相應(yīng)的謂詞過濾器。具體地,我們可以使用內(nèi)聯(lián)謂詞、Filter謂詞、自定義謂詞3種方式進(jìn)行實(shí)踐。其中對于Filter謂詞、自定義謂詞而言,需要在jsonpath字符串中使用占位符?來代替所傳遞的過濾器謂詞。如果jsonpath字符串中使用多個占位符?,則應(yīng)按相應(yīng)順序傳入過濾器謂詞參數(shù)
import com.jayway.jsonpath.*;
import com.jayway.jsonpath.spi.mapper.JacksonMappingProvider;
import static com.jayway.jsonpath.Criteria.where;
import static com.jayway.jsonpath.Filter.filter;
public class Demo2 {
private String json = "{\"store\":{\"book\":[{\"category\":\"reference\",\"author\":\"Nigel Rees\",\"title\":\"Sayings of the Century\",\"price\":8.95},{\"category\":\"fiction\",\"author\":\"Evelyn Waugh\",\"title\":\"Sword of Honour\",\"price\":12.99},{\"category\":\"fiction\",\"author\":\"Herman Melville\",\"title\":\"Moby Dick\",\"isbn\":\"0-553-21311-3\",\"price\":8.99},{\"category\":\"fiction\",\"author\":\"J. R. R. Tolkien\",\"title\":\"The Lord of the Rings\",\"isbn\":\"0-395-19395-8\",\"price\":22.99}],\"bicycle\":{\"color\":\"red\",\"price\":19.95},\"clothes\":[{\"name\":\"牛仔褲\",\"sizes\":\"S\",\"price\":94},{\"name\":\"背心\",\"sizes\":\"M\",\"price\":48},{\"name\":\"裙子\",\"sizes\":[\"S\",\"M\"],\"price\":1.24},{\"name\":\"羊毛衫\",\"sizes\":[\"XS\",\"XL\"],\"price\":78.99},{\"name\":\"Polo衫\",\"sizes\":[\"XS\",\"M\",\"XL\"],\"price\":18.99}]},\"expensive\":10}\n";
/**
* Predicate Filter 謂詞過濾器
*/
@Test
public void usePredicateFilter() {
// 使用 JacksonJsonProvider 實(shí)現(xiàn)反序列化
Configuration conf = Configuration
.builder()
.mappingProvider( new JacksonMappingProvider() )
.build();
ReadContext ctx = JsonPath.using( conf )
.parse(json);
// 方式1 : 內(nèi)聯(lián)謂詞
TypeRef<List<Clothes>> typeRef = new TypeRef<List<Clothes>>() {};
List<Clothes> clothes1 = ctx.read("$.store.clothes[?( @.price>50 || @.sizes anyof ['M'] ) ]", typeRef);
System.out.println("-------------- clothes1 ---------------");
clothes1.forEach( System.out::println );
// 方式2 : Filter謂詞
Filter filter = filter( where("price").gt(50) )
.or( where("sizes").anyof( Arrays.asList("M") ) );
// 使用謂詞的占位符?
Clothes[] clothes2 = ctx.read("$.store.clothes[?]", Clothes[].class, filter);
System.out.println("-------------- clothes2 ---------------");
for (Clothes clothes : clothes2) {
System.out.println(clothes);
}
// 方式3 : 自定義謂詞
Predicate rule = ctx1 -> {
Map map = ctx1.item( Map.class );
boolean b1 = false;
Object priceObj = map.getOrDefault("price",null);
if( priceObj!=null ) {
String priceStr = priceObj.toString();
Double price = 0d;
try {
price = Double.parseDouble( priceStr );
} catch (Exception e) {
}
b1 = price > 50d;
}
boolean b2 = false;
Object sizes = map.getOrDefault("sizes", null);
if( sizes!=null && sizes instanceof List ) {
List<String> sizeList = (List<String>) sizes;
List<String> targetList = Arrays.asList("M");
for (String size : sizeList) {
if( targetList.contains(size) ) {
b2 = true;
break;
}
}
}
return b1 || b2;
};
// 使用謂詞的占位符?
Clothes[] clothes3 = ctx.read("$.store.clothes[?]", Clothes[].class, rule);
System.out.println("-------------- clothes3 ---------------");
for (Clothes clothes : clothes3) {
System.out.println(clothes);
}
}
}
@AllArgsConstructor
@NoArgsConstructor
@Builder
@Data
class Clothes {
private String name;
private Double price;
private Object sizes;
}
效果如下所示

