I have been working on a problem statement where we have a huge JSON response coming in and when we were parsing it using conventional gson parsing technique, it used to give OutOfMemoryException as this method stores the data in memory before processing it, so as a solution to this i have worked on streaming the JSON response where it won't put everything in memory, so it worked fine till somewhere around 1.6 million records and after that even that broke. So this is the exception we are getting.
Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
This is the entire code i'm using for this:
// Getting reponse into InputStream and casting it to JsonReader object for parsing
InputStream liInStream = luURLConn.getInputStream();
lCycleTimeReader = new JsonReader(new InputStreamReader(liInStream, "UTF-8"));
Our JSON looks like this:
{
"Report_Entry": [
{
"key1": "value",
"key2": "value",
"key3": "value",
"key4": "value",
"key5": "value"
},
{
"key1": "value",
"key2": "value",
"key3": "value",
"key4": "value",
"key5": "value"
}
]}
Using this object into our parsing method:
public HashMap<String, HashMap<String, String>> getcycleTimeMap(JsonReader poJSONReaderObj,
CycleTimeConstant cycleTimeConstant, int processId) {
Integer counter = 0;
HashMap<String, HashMap<String, String>> cycleTimeMap = new HashMap<String, HashMap<String, String>>();
HashMap<String, HashMap<String, String>> finalcycleTimeMap = new HashMap<String, HashMap<String, String>>();
try {
CycleTime cycleTime = new CycleTime();
poJSONReaderObj.beginObject();
while (poJSONReaderObj.hasNext()) {
String name = poJSONReaderObj.nextName();
if (name.equals("Report_Entry")) {
poJSONReaderObj.beginArray();
while (poJSONReaderObj.hasNext()) {
JsonToken nextToken2 = poJSONReaderObj.peek();
if (JsonToken.BEGIN_OBJECT.equals(nextToken2)) {
poJSONReaderObj.beginObject();
} else if (JsonToken.END_OBJECT.equals(nextToken2)) {
poJSONReaderObj.endObject();
} else {
String nextString = "";
if (JsonToken.STRING.equals(nextToken2)) {
nextString = poJSONReaderObj.nextString();
} else if (JsonToken.NAME.equals(nextToken2)) {
nextString = poJSONReaderObj.nextName();
}
switch (nextString) {
case "key1":
cycleTime.setKey1(poJSONReaderObj.nextString());
break;
case "key2":
cycleTime.setKey2(poJSONReaderObj.nextString());
break;
case "key3":
cycleTime.setKey3(poJSONReaderObj.nextString());
break;
case "key4":
cycleTime.setKey4(poJSONReaderObj.nextString());
break;
case "key5":
cycleTime.setKey5(poJSONReaderObj.nextString());
break;
}
}
poJSONReaderObj.endObject();
System.out
.println("Value of Map is : " + new Gson().toJson(cycleTime) + "counter : " + counter);
counter++;
System.out.println("Counter : " + counter);
cycleTimeMap = (HashMap<String, HashMap<String, String>>) cycleTimeBpProcessIterator(
cycleTime, cycleTimeConstant, counter, processId);
}
finalcycleTimeMap.putAll(cycleTimeMap);
}
}
JsonToken nextToken = poJSONReaderObj.peek();
if (JsonToken.END_OBJECT.equals(nextToken)) {
poJSONReaderObj.endObject();
} else if (JsonToken.END_ARRAY.equals(nextToken)) {
poJSONReaderObj.endArray();
}
} catch (IOException ioException) {
ioException.printStackTrace();
}
System.out.println("FINAL MAP TO BE LOADED : " + new Gson().toJson(finalcycleTimeMap));
return finalcycleTimeMap;
}
POJO class for handling response:
public class CycleTime {
private String key1 = "";
private String key2 = "";
private String key3 = "";
private String key4 = "";
private String key5 = "";
public String getKey1() {
return key1;
}
public void setKey1(String key1) {
this.key1 = key1;
}
public String getKey2() {
return key2;
}
public void setKey2(String key2) {
this.key2 = key2;
}
public String getKey3() {
return key3;
}
public void setKey3(String key3) {
this.key3 = key3;
}
public String getKey4() {
return key4;
}
public void setKey4(String key4) {
this.key4 = key4;
}
public String getKey5() {
return key5;
}
public void setKey5(String key5) {
this.key5 = key5;
}
}
I'm not sure what might be a culprit here but seems it is giving the same error, i'm wondering what should be the next approach to avoid this OutOfMemoryException.
Reading the entire document into a single object does not mean that the streamed reading would help you.
Moreover, Gson uses streaming under the hood because it is just an optional way of reading and writing.
Your approach, however, is very far from being good:
Gson things:
The main thing is: use Gson properly in full and let it do its job. I couldn't run your code for the JSON document you provided: it works neither for the root JSON object, nor for the only top object entry (so your deserializer is broken due to improper used of the hasNext and the beginObject/endObject pair).
Common Java things:
don't catch exceptions in middle returning a partially composed object (is it correct?);
don't use Throwable.printStackTrace (use proper logging facilities);
if you don't want using loggers, then print it to System.err (this is just a proper standard stream for such purposes);
Integer as a counter is a bad idea because of creating many boxed values, especially for huge documents (use int -- it is just fine);
enum values can (and should be) checked for equality with == (it is safe since they are singletons);
then, you can use switch to for enums too (both shorted, more compile-time safe);
don't create Gson instances in a loop especially that has that many iterations (Gson instances are known to be immutable as thread-safe, but not that cheap at constructing its objects);
don't use maps where you can have statically typed plain objects (for good);
what's the purposes of returning an always-one-key-value-pair map; (return the value);
Common design things:
use as common types as possible for declarations: not HashMap, but Map (what if someday you need another map with ordered keys? or what if you don't need a map after all?);
inverse dependencies (what if you don't need CycleTime with five keys?);
Streaming things:
if it runs in an OOM error, then what's the point of collecting a huge map that obviously cannot fit your app RAM? (use callbacks or promises (pushing approach) to process a single element, iterators or streams (pulling approach), reactive streams (pushing approach), whatever);
collect the result only for a small memory foot-print or use aggregation (otherwise you're at risk of having OOM).
This is how you can reduce the memory foot-print by using a pushing approach via callbacks:
#UtilityClass
public final class StreamSupport {
public static void acceptArrayElements(#WillNotClose final JsonReader jsonReader, final Consumer<? super JsonReader> acceptElement)
throws IOException {
jsonReader.beginArray();
while ( jsonReader.hasNext() ) {
acceptElement.accept(jsonReader);
}
jsonReader.endArray();
}
}
#UtilityClass
public final class CycleDeserializer {
public static void readCycles(final JsonReader jsonReader, final Consumer<? super JsonReader> acceptJsonReader)
throws IOException {
jsonReader.beginObject();
while ( jsonReader.hasNext() ) {
switch ( jsonReader.nextName() ) {
case "Report_Entry":
StreamSupport.acceptArrayElements(jsonReader, acceptJsonReader);
break;
default:
jsonReader.skipValue();
break;
}
}
jsonReader.endObject();
}
}
private static final Gson gson = new GsonBuilder()
.disableHtmlEscaping()
.disableInnerClassSerialization()
.create();
#Test
public void test()
throws IOException {
try ( final JsonReader jsonReader = openTheHugeDocument() ) {
CycleDeserializer.readCycles(jsonReader, jr -> {
final CycleTime cycleTime = gson.fromJson(jr, CycleTime.class);
System.out.println(cycleTime);
});
}
// do the simplest aggregation operation: `COUNT`
try ( final JsonReader jsonReader = openTheHugeDocument() ) {
final AtomicInteger count = new AtomicInteger();
CycleDeserializer.readCycles(jsonReader, jr -> {
try {
jr.skipValue();
count.incrementAndGet();
} catch ( final IOException ex ) {
throw new RuntimeException(ex);
}
});
System.out.println("Count = " + count);
}
// this will probably fail when the document is huge because it is collected into a single collection
// (you need to let your JVM use as much RAM as possible if it is a must for you)
try ( final JsonReader jsonReader = openTheHugeDocument() ) {
final Collection<CycleTime> cycleTimes = new ArrayList<>();
CycleDeserializer.readCycles(jsonReader, jr -> {
final CycleTime cycleTime = gson.fromJson(jr, CycleTime.class);
cycleTimes.add(cycleTime);
});
System.out.println("Count in list = " + cycleTimes.size());
}
}
As you can see, in the runner above you can choose the way you prefer to process your entries: either a dumb logging, or a simple count, or a simple collect-to operation.
For the pull approach via Stream approach please see: https://stackoverflow.com/a/69282822/12232870
I am trying to write a lambda function that gets employee location preference and have the code sample below.
But for my lambda function I get a compilation error at flatMap(this::buildEmployeeGeolocation)
saying Bad return type in method reference: cannot convert com.abc.EmployeeGeolocation to java.util.Optional<U>.
What am I missing here?
public Optional<EmployeeGeolocation> getEmployee(final SessionId sessionId) {
return Optional.ofNullable(employeePreferencesStore.getEmployeeAccountPreferences(sessionId))
.map(preferences -> preferences.getPreference(PreferenceKey.Location))
.filter(StringUtils::isNotBlank)
.map(this::readEmployeelocation)
.flatMap(this::buildEmployeeGeolocation);
}
private Optional<EncryptedGeolocation> readEmployeeLocation(#NonNull final String encryptedGeolocation) {
try {
return Optional.ofNullable(objectMapper.readValue(encryptedGeolocation, EmployeeGeolocation.class));
} catch (final IOException e) {
log.error("Error while reading the encrypted geolocation");
throw new RuntimeException(e);
}
}
private EmployeeGeolocation buildEmployeeGeolocation(#NonNull final EncryptedGeolocation unditheredEncryptedGeolocation) {
return EmployeeGeolocation.builder()
.latitude(10.0)
.longitude(10.0)
.accuracy(1.0)
.locationType(ADDRESS)
.build();
}
It seems like what you really need to do is swap the map and flatMap. Change the code
.map(this::readEmployeeLocation)
.flatMap(this::buildEmployeeGeolocation);
to
.flatMap(this::readEmployeeLocation) // since you already have an Optional<String>
.map(this::buildEmployeeGeolocation); // above results in Optional<EncryptedGeolocation>
Important: Inferred from the code Optional.ofNullable(...).map(...).filter(StringUtils::isNotBlank), that it would result in an Optional<String> until this operation.
I am having some trouble with converting a Set of POJO's to a JSON string so I can store it in a database column. The conversion to JSON works as expected but when the conversion from JSON to a Set< Qualification > happens it always returns a LinkedHashSet<LinkedHashMap> which is causing issues.
The weird thing is that inside my converter the JSON string is converted successfully to a Set<Qualification>. When I debug in my IDE and step through the execution I can see that after it calls a deepCopy method in the MutableMutabilityPlan abstract class. At this point the data is of type LinkedHashSet<LinkedHashMap> and not Set<Qualification> from the conversion.
Here is my converter.
public class SetToStringConverter implements AttributeConverter<Set<Qualification>, String> {
private final ObjectMapper mapper = new ObjectMapper();
private final String errorMessage = "converter.invalid";
#Override
public String convertToDatabaseColumn(final Set<Qualification> items) {
try {
return mapper.writeValueAsString(items);
} catch (JsonProcessingException e) {
throw new ConverterFailureException(errorMessage);
}
}
#Override
public Set<Qualification> convertToEntityAttribute(final String data) {
try {
if (data != null) {
final Set<Qualification> s = mapper.readValue(data, new TypeReference<Set<Qualification>>() {});
return s; // Debugging here I have the correct type
}
return new HashSet<>();
} catch (IOException e) {
throw new ConverterFailureException(errorMessage);
}
}
}
I have done some research and have tried various approaches but the result is always the same.
Has anyone run into this issue before or can see anything wrong with my converter. Let me know if anything isn't clear so I can provide more information.
Thanks very much for the help.
I have the following code:
public JSONObject addProductToOrder(int orderId, int productId, String name, double price) {
JSONObject json = new JSONObject();
try {
json.put("orderId", orderId);
json.put("productId", productId);
json.put("name", name);
json.put("price", price);
} catch (JSONException e) {
Debugger.out(e);
}
return this.dataSource.write("/orderItems/addproductToOrder", json);
}
I want to test so using Mockito and I have done the following:
public void getOrder() throws JSONException {
JSONObject json = new JSONObject("{result: 'OK', error: 0}");
doReturn(json)
.when(this.clientSpy) // this.clientSpy = Mockito.spy(client);
.post("/orderItems/addProductToOrder", new JSONObject("{productId: 1, orderId: 1, price: 15.99, name: 'Product 1'}"));
OrderItem orderItem = new OrderItem(this.api);
assertEquals("OK", orderItem.addProductToOrder(1, 1, "Product 1", 15.99).get("result"));
}
The way I understand things, my doReturn() is not triggered because new JSONObject() !== new JSONObject().
Is there a way to make it not compare the objects, but instead just the contents?
What happens here is that Mockito is calling equals() to compare the object that you provided to the specification to the one that is used for the actual method call.
Basically, there are three options:
if you just provide an object, equals() is used to compare
then you have a variety of other ArgumentMatchers, such as any() or isNull(). Please note: when using such matchers, all arguments must be matched, you can't do: foo(any(), 5) for example.
if that doesn't do: you can also use an ArgumentCaptor. Meaning: instead of having Mockito compare objects, you simply record the object that was passed to the mock. And then you add another step in your test case where you check the captured argument as required.
This argThat function can help you.
private static String eqJson(String expectedJson) {
return argThat(argument -> {
try {
JSONAssert.assertEquals(expectedJson, argument, true);
return true;
} catch (JSONException e) {
return false;
}
});
}
Then in your test
// import
import org.skyscreamer.jsonassert.JSONAssert;
import static org.mockito.ArgumentMatchers.argThat;
// then
then(mock).should(times(1)).method(eqJson(expectedJson));
Inshort : I am trying to find some api that could just change the value by taking first parameter as jsonString , second parameter as JSONPath and third will be new value of that parameter. But, all I found is this..
https://code.google.com/p/json-path/
This api allows me to find any value in JSON String. But, I am not finding easy way to update the value of any key. For example, Here is a book.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,
"isbn":"0-553-21311-3"
}
],
"bicycle":{
"color":"red",
"price":19.95
}
}
}
I can access color of bicycle by doing this.
String bicycleColor = JsonPath.read(json, "$.store.bicycle.color");
But I am looking for a method in JsonPath or other api some thing like this
JsonPath.changeNodeValue(json, "$.store.bicycle.color", "green");
String bicycleColor = JsonPath.read(json, "$.store.bicycle.color");
System.out.println(bicycleColor); // This should print "green" now.
I am excluding these options,
Create a new JSON String.
Create a JSON Object to deal with changing value and convert it back to jsonstring
Reason: I have about 500 different requests for different types of service which return different json structure. So, I do not want to manually create new JSON string always. Because, IDs are dynamic in json structure.
Any idea or direction is much appreciated.
Updating this question with following answer.
Copy MutableJson.java.
copy this little snippet and modify as per you need.
private static void updateJsonValue() {
JSONParser parser = new JSONParser();
JSONObject jsonObject = new JSONObject();
FileReader reader = null;
try {
File jsonFile = new File("path to book.json");
reader = new FileReader(jsonFile);
jsonObject = (JSONObject) parser.parse(reader);
} catch (Exception ex) {
System.out.println(ex.getLocalizedMessage());
}
Map<String, Object> userData = null;
try {
userData = new ObjectMapper().readValue(jsonObject.toJSONString(), Map.class);
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
MutableJson json = new MutableJson(userData);
System.out.println("Before:\t" + json.map());
json.update("$.store.book[0].author", "jigish");
json.update("$.store.book[1].category", "action");
System.out.println("After:\t" + json.map().toString());
}
Use these libraries.
import org.json.simple.JSONObject;
import org.json.simple.parser.JSONParser;
import org.codehaus.jackson.map.ObjectMapper;
The thing is that the functionality you want is already an undocumented feature of JsonPath. Example using your json structure:
String json = "{ \"store\":{ \"book\":[ { \"category\":\"reference\", \"author\":\"Nigel Rees\", \"title\":\"Sayings of the Century\", \"price\":8.95 }, { \"category\":\"fiction\", \"author\":\"Evelyn Waugh\", \"title\":\"Sword of Honour\", \"price\":12.99, \"isbn\":\"0-553-21311-3\" } ], \"bicycle\":{ \"color\":\"red\", \"price\":19.95 } } }";
DocumentContext doc = JsonPath.parse(json).
set("$.store.bicycle.color", "green").
set("$.store.book[0].price", 9.5);
String newJson = new Gson().toJson(doc.read("$"));
Assuming that parsed JSON can be represented in memory as a Map, you can build an API similar to JsonPath that looks like:
void update(Map<String, Object> json, String path, Object newValue);
I've quickly done a gist of a dirty implementation for simple specific paths (no support for conditions and wildcards) that can traverse json tree, E.g. $.store.name, $.store.books[0].isbn. Here it is: MutableJson.java. It definitely needs improvement, but can give a good start.
Usage example:
import java.util.*;
public class MutableJson {
public static void main(String[] args) {
MutableJson json = new MutableJson(
new HashMap<String, Object>() {{
put("store", new HashMap<String, Object>() {{
put("name", "Some Store");
put("books", Arrays.asList(
new HashMap<String, Object>() {{
put("isbn", "111");
}},
new HashMap<String, Object>() {{
put("isbn", "222");
}}
));
}});
}}
);
System.out.println("Before:\t" + json.map());
json.update("$.store.name", "Book Store");
json.update("$.store.books[0].isbn", "444");
json.update("$.store.books[1].isbn", "555");
System.out.println("After:\t" + json.map());
}
private final Map<String, Object> json;
public MutableJson(Map<String, Object> json) {
this.json = json;
}
public Map<String, Object> map() {
return json;
}
public void update(String path, Object newValue) {
updateJson(this.json, Path.parse(path), newValue);
}
private void updateJson(Map<String, Object> data, Iterator<Token> path, Object newValue) {
Token token = path.next();
for (Map.Entry<String, Object> entry : data.entrySet()) {
if (!token.accept(entry.getKey(), entry.getValue())) {
continue;
}
if (path.hasNext()) {
Object value = token.value(entry.getValue());
if (value instanceof Map) {
updateJson((Map<String, Object>) value, path, newValue);
}
} else {
token.update(entry, newValue);
}
}
}
}
class Path {
public static Iterator<Token> parse(String path) {
if (path.isEmpty()) {
return Collections.<Token>emptyList().iterator();
}
if (path.startsWith("$.")) {
path = path.substring(2);
}
List<Token> tokens = new ArrayList<>();
for (String part : path.split("\\.")) {
if (part.matches("\\w+\\[\\d+\\]")) {
String fieldName = part.substring(0, part.indexOf('['));
int index = Integer.parseInt(part.substring(part.indexOf('[')+1, part.indexOf(']')));
tokens.add(new ArrayToken(fieldName, index));
} else {
tokens.add(new FieldToken(part));
}
};
return tokens.iterator();
}
}
abstract class Token {
protected final String fieldName;
Token(String fieldName) {
this.fieldName = fieldName;
}
public abstract Object value(Object value);
public abstract boolean accept(String key, Object value);
public abstract void update(Map.Entry<String, Object> entry, Object newValue);
}
class FieldToken extends Token {
FieldToken(String fieldName) {
super(fieldName);
}
#Override
public Object value(Object value) {
return value;
}
#Override
public boolean accept(String key, Object value) {
return fieldName.equals(key);
}
#Override
public void update(Map.Entry<String, Object> entry, Object newValue) {
entry.setValue(newValue);
}
}
class ArrayToken extends Token {
private final int index;
ArrayToken(String fieldName, int index) {
super(fieldName);
this.index = index;
}
#Override
public Object value(Object value) {
return ((List) value).get(index);
}
#Override
public boolean accept(String key, Object value) {
return fieldName.equals(key) && value instanceof List && ((List) value).size() > index;
}
#Override
public void update(Map.Entry<String, Object> entry, Object newValue) {
List list = (List) entry.getValue();
list.set(index, newValue);
}
}
A JSON string can be easily parsed into a Map using Jackson:
Map<String,Object> userData = new ObjectMapper().readValue("{ \"store\": ... }", Map.class);
Just answering for folks landing on this page in future for reference.
You could consider using a Java implementation of jsonpatch. RFC can be found here
JSON Patch is a format for describing changes to a JSON document. It can be used to avoid sending a whole document when only a part has changed. When used in combination with the HTTP PATCH method it allows partial updates for HTTP APIs in a standards compliant way.
You can specify the operation that needs to be performed (replace, add....), json path at which it has to be performed, and the value which should be used.
Again, taking example from the RFC :
[
{ "op": "test", "path": "/a/b/c", "value": "foo" },
{ "op": "remove", "path": "/a/b/c" },
{ "op": "add", "path": "/a/b/c", "value": [ "foo", "bar" ] },
{ "op": "replace", "path": "/a/b/c", "value": 42 },
{ "op": "move", "from": "/a/b/c", "path": "/a/b/d" },
{ "op": "copy", "from": "/a/b/d", "path": "/a/b/e" }
]
For Java implementation, I have not used it myself, but you can give a try to https://github.com/fge/json-patch
So in order to change a value within a JSon string, there are two steps:
Parse the JSon
Modify the appropriate field
You are trying to optimize step 2, but understand that you are not going to be able to avoid step 1. Looking at the Json-path source code (which, really, is just a wrapper around Jackson), note that it does do a full parse of the Json string before being able to spit out the read value. It does this parse every time you call read(), e.g. it is not cached.
I think this task is specific enough that you're going to have to write it yourself. Here is what I would do:
Create an object that represents the data in the parsed Json string.
Make sure this object has, as part of it's fields, the Json String pieces that you do not expect to change often.
Create a custom Deserializer in the Json framework of your choice that will populate the fields correctly.
Create a custom Serializer that uses the cached String pieces, plus the data that you expect to change
I think the exact scope of your problem is unusual enough that it is unlikely a library already exists for this. When a program receives a Json String, most of the time what it wants is the fully deserialized object - it is unusual that it needs to FORWARD this object on to somewhere else.