从Gson更改为Jackson以进行JSON解析

Question

我目前正在使用GSON使用inputstream/reader解析一个相当大的JSONfile。在我的android设备上解析大约需要35秒，我从一些基准测试中了解到Jackson的性能要好得多。但是我不知道如何使用jackson解析我的JSON文件。有人能帮我吗？

我的JSON如下所示：

[

    {
        "venue": { … }
    },
    {
        "venue": {
            "venue_seasons": [ … ],
            "address": "the address",
            "city": "the city",
            "name": "the name",
            "created_at": "2011-05-31T07:55:33Z",
            "latitude": 00.000000,
            "country": "the country",
            "internal_link_en": null,
            "internal_link_nl": null,
            "updated_at": "2011-09-15T14:46:09Z",
            "zipcode": "the zipcode",
            "foursquare_link": "foursquare url",
            "url": null,
            "id": 3,
            "tip": "some tip",
            "uid": "4ab5e205f964a520317620e3",
            "phone": "phonenr",
            "recommended": null,
            "website": "someurl",
            "venue_photos": [ … ],  //array containing objects with urls of images
            "description": null,
            "longitude": 00.000000,
            "thumbnail_location": null,
            "subcategories": [ … ],
            "opening_en": null,
            "opening_nl": null,
            "hidden": false,
            "twitter": "thetwitteraccount",
            "themes": [ … ]
        }
    }, //more venues
]

我的GSON代码如下所示，它可以工作：

AssetManager assetManager = getAssets();
InputStream inputStream = null;
try {
    inputStream = assetManager.open("filename.json");
} catch (IOException e) {
    Log.e("tag", e.getMessage());
}

Reader reader = new InputStreamReader(inputStream);

Gson gson = new Gson();
List<JsonResponse> venueList = gson.fromJson(reader, new TypeToken<List<JsonResponse>>() {}.getType());
JsonResponse naam = venueList.get(12);
String denaam = naam.venue.getName;
Log.i("nr12",denaam);
Log.i("timetracker","stop" ); 

// just some logging to test if the parser works 
for (JsonResponse venue : venueList) {
    String tijdel = String.valueOf(venue.venue.id);
    Log.i(venuetag,"name of venue"+ tijdel+ " is: " + venue.venue.getName);
}

...

class JsonResponse
{
    Venues venue;
}

class Venues
{
    public List<VenueSeasons> venue_seasons;

    public List<VenuePhotos> venue_photos;

    public List<SubCategories> subcategories;

    public List<Themes> themes;

    @SerializedName("address")
    public String getAdress;

    @SerializedName("city")
    public String getCity;

    @SerializedName("country")
    public String getCountry;

    @SerializedName("name")
    public String getName;

    @SerializedName("created_at")
    public Date getCreatedAt;

    @SerializedName("updated_at")
    public Date getUpdatedAt;

    @SerializedName("internal_link_nl")
    public String getInternalLinkNl;

    @SerializedName("internal_link_en")
    public String getInternalLinkEN;

    @SerializedName("latitude")
    public Double getLatitude;

    @SerializedName("longitude")
    public Double getLongitude;

    @SerializedName("foursquare_link")
    public String getFoursquareLink;

    @SerializedName("url")
    public String getURL;

    @SerializedName("phone")
    public String getPhone;

    @SerializedName("zipcode")
    public String getZipCode;

    public String tip;

    public String tip_en;

    public String uid;

    public int id;

    @SerializedName("website")
    public String getWebsite;

    @SerializedName("recommended")
    public Boolean getRecommended;

    @SerializedName("description")
    public String getDescription;

    @SerializedName("hidden")
    public Boolean getHidden;

    @SerializedName("opening_en")
    public String getOpeningEN;

    @SerializedName("opening_nl")
    public String getOpeningNL;

    @SerializedName("twitter")
    public String getTwitter;

    @SerializedName("thumbnail_location")
    public String getThumbnailLocation;
}


public class VenuePhotos 
{
    @SerializedName("large")
    public String getLargePhotoURL;

    @SerializedName("medium")
    public String getMediumPhotoURL;

    @SerializedName("small")
    public String getSmallPhotoURL; 

    @SerializedName("original")
    public String getOriginalPhotoURL;

    public String uid;

    public int id;

    public int venue_id;

    public boolean selected;

    @SerializedName("created_at")
    public Date getCreatedAt;

    @SerializedName("updated_at")
    public Date getUpdatedAt;
}

现在这是可行的。我对数据做了一些处理，解析后效果非常好，但我认为启动我的应用程序需要这么长时间，这太糟糕了。

我的杰克逊代码（失败的代码）是：

AssetManager assetManager = getAssets();
InputStream inputStream = null;
try {
    inputStream = assetManager.open("originalDelftJson.json");
} catch (IOException e) {
    Log.e("tag", e.getMessage());
}
Reader reader = new InputStreamReader(inputStream);

ObjectMapper objectMapper = new ObjectMapper();
JsonFactory jsonFactory = new JsonFactory();
JJsonResponse response = null;
try {
    JsonParser jp = jsonFactory.createJsonParser(reader);
    response = objectMapper.readValue(jp, JJsonResponse.class);
    String test = String.valueOf(response.venue.size());

课程：

public class JJsonResponse
{
     public List<JVenue> venue;
}

class Venues
{
    public List<VenueSeasons> venue_seasons;

    public List<VenuePhotos> venue_photos;

    public List<SubCategories> subcategories;

    public List<Themes> themes;

    @SerializedName("address")
    public String getAdress;

    @SerializedName("city")
    public String getCity;

    @SerializedName("country")
    public String getCountry;

    @SerializedName("name")
    public String getName;

    @SerializedName("created_at")
    public Date getCreatedAt;

    @SerializedName("updated_at")
    public Date getUpdatedAt;

    @SerializedName("internal_link_nl")
    public String getInternalLinkNl;

    @SerializedName("internal_link_en")
    public String getInternalLinkEN;

    @SerializedName("latitude")
    public Double getLatitude;

    @SerializedName("longitude")
    public Double getLongitude;

    @SerializedName("foursquare_link")
    public String getFoursquareLink;

    @SerializedName("url")
    public String getURL;

    @SerializedName("phone")
    public String getPhone;

    @SerializedName("zipcode")
    public String getZipCode;

    public String tip;

    public String tip_en;

    public String uid;

    public int id;

    @SerializedName("website")
    public String getWebsite;

    @SerializedName("recommended")
    public Boolean getRecommended;

    @SerializedName("description")
    public String getDescription;

    @SerializedName("hidden")
    public Boolean getHidden;

    @SerializedName("opening_en")
    public String getOpeningEN;

    @SerializedName("opening_nl")
    public String getOpeningNL;

    @SerializedName("twitter")
    public String getTwitter;

    @SerializedName("thumbnail_location")
    public String getThumbnailLocation;
}

public class JVenue
{
    public String name;
    public int id;
    public String city;
    public String address;
    public String country;
    public String internal_link_nl;
    public String internal_link_en;
    public String zipcode;
    public String foursquare_link;
    public String tip_en;
    public String url;
    public Date created_at;
    public Date updated_at;
    public float latitude;
    public float longitude;
    public String tip;
    public String uid;
    public String phone;
    public String recommended;
    public String website;
    public String description;
    public String thumbnail_location;
    public boolean hidden;
    public String twitter;
    public String opening_en;
    public String opening_nl;
}

我想我很接近，但我做了一些错误的事情，因为我得到的错误：org.codehaus.jackson.map.JsonMappingExc0019：不能反序列化实例com.jacksonrecipes.testapp.model.JJsonRestor-outSTART_ARRAY令牌

我真的不明白杰克逊是如何工作的，以及如何实施它。有人知道如何在我的Android代码中更改我的Jackson实现，这样它就可以工作，我可以访问数据吗？

编辑：找到我的解决方案了吗

借助MH的回答。我能找到它。我现在使用：

List<JJsonResponse> venueCounter = objectMapper.readValue(inputStream, new TypeReference<List<JJsonResponse>>() { });

Answer 1

看起来Json内容只是一个对象列表。这意味着，在我看来，您的代码需要如下所示：

List<JVenue> venues = mapper.readValue(inputStream, new TypeReference<List<JVenue>>() { });

不需要将InputStream包装到Reader中，也不需要创建自己的JsonFactory和JsonParser实例。

Answer 2

需要记住的一件非常重要的事情是，您总是希望重用ObjectMapper实例——每个请求创建一个实例会大大降低速度，可能会降低一个数量级。通常，要么只创建一个静态实例（在创建之后立即配置），要么通过DI框架（如Guice）注入一个静态实例。

杰克逊的其他最佳性能实践可以从以下网站找到：http://wiki.fasterxml.com/JacksonBestPracticesPerformance

Answer 3

以下是使用Jackson的工作示例，在原始问题中使用JSON（更正为有效并匹配Java数据结构），使用原始Gson示例中的Java数据结构（更正为使用Jackson的@JsonProperty而不是Gson的@SeriazedName）。

import static com.fasterxml.jackson.annotation.JsonAutoDetect.Visibility.ANY;
import static com.fasterxml.jackson.annotation.PropertyAccessor.FIELD;

import java.io.File;
import java.util.Date;
import java.util.List;

import com.fasterxml.jackson.annotation.JsonProperty;
import com.fasterxml.jackson.core.type.TypeReference;
import com.fasterxml.jackson.databind.ObjectMapper;

public class JacksonFoo
{
  public static void main(String[] args) throws Exception
  {
    ObjectMapper mapper = new ObjectMapper().setVisibility(FIELD, ANY);
    List<JsonResponse> responses = mapper.readValue(new File("input.json"), new TypeReference<List<JsonResponse>>() {});
    System.out.println(responses);
    System.out.println(mapper.writeValueAsString(responses));
  }
}

class JsonResponse
{
  Venues venue;

  @Override
  public String toString()
  {
    return venue.toString();
  }
}

class Venues
{
  List<VenueSeasons> venue_seasons;
  @JsonProperty("address") String getAdress;
  @JsonProperty("city") String getCity;
  @JsonProperty("name") String getName;
  @JsonProperty("created_at") Date getCreatedAt;
  @JsonProperty("latitude") Double getLatitude;
  @JsonProperty("country") String getCountry;
  @JsonProperty("internal_link_en") String getInternalLinkEN;
  @JsonProperty("internal_link_nl") String getInternalLinkNl;
  @JsonProperty("updated_at") Date getUpdatedAt;
  @JsonProperty("zipcode") String getZipCode;
  @JsonProperty("foursquare_link") String getFoursquareLink;
  @JsonProperty("url") String getURL;
  int id;
  String tip;
  String uid;
  @JsonProperty("phone") String getPhone;
  @JsonProperty("recommended") Boolean getRecommended;
  @JsonProperty("website") String getWebsite;
  List<VenuePhotos> venue_photos;
  @JsonProperty("description") String getDescription;
  @JsonProperty("longitude") Double getLongitude;
  @JsonProperty("thumbnail_location") String getThumbnailLocation;
  List<SubCategories> subcategories;
  @JsonProperty("opening_en") String getOpeningEN;
  @JsonProperty("opening_nl") String getOpeningNL;
  @JsonProperty("hidden") Boolean getHidden;
  @JsonProperty("twitter") String getTwitter;
  List<Themes> themes;
  String tip_en; // not in example JSON

  @Override
  public String toString()
  {
    return String.format("Venues: id=%d", id);
  }
}

class VenuePhotos
{
  @JsonProperty("large") String getLargePhotoURL;
  @JsonProperty("medium") String getMediumPhotoURL;
  @JsonProperty("small") String getSmallPhotoURL;
  @JsonProperty("original") String getOriginalPhotoURL;
  String uid;
  int id;
  int venue_id;
  boolean selected;
  @JsonProperty("created_at") Date getCreatedAt;
  @JsonProperty("updated_at") Date getUpdatedAt;
}

enum VenueSeasons
{
  Spring, Summer, Fall, Winter
}

enum SubCategories
{
  SubCat1, SubCat2, SubCat3, SubCat4 
}

enum Themes
{
  Theme1, Theme2, Theme3, Theme4
}