如何从JSON文件Java中的特定索引获取数据

Question

JSONParser parses all json objects in a given file but i want to parse json objects starting from 100th index to end of the file.

I can do this later using subList but if I have 1 Million json objects in my json file I dont want to parse everything because efficiency will be reduced.

public static void readJsonFile() {

    JSONParser parser = new JSONParser();

    try {
        JSONArray a = (JSONArray) parser.parse(new FileReader("D:\\2018-4-21.json"));

        for (Object o : a.subList(100,a.size())) {
            JSONObject checkIn = (JSONObject) o;

            String userId = (String) checkIn.get("UserID");
            System.out.print(userId);

            String inout = (String) checkIn.get("INOUT");
            System.out.print("   " + inout);

            String swippedDateTime = (String) checkIn.get("SwippedDateTime");
            System.out.print("   " + swippedDateTime);

            System.out.println("");
        }
    } catch (FileNotFoundException e) {
        e.printStackTrace();
    } catch (IOException e) {
        e.printStackTrace();
    } catch (org.json.simple.parser.ParseException e) {
        e.printStackTrace();
    }
}

My Json File

[
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:25"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:36"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:36"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:36"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:38"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:38"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:38"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:39"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:39"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:39"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:42"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:42"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:42"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:42"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:42"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:42"
    },
    {
        "UserID": "2",
        "INOUT": null,
        "SwippedDateTime": "2018-4-23 22:49"
    }
]

Answer 1

The only way to locate index 100, is to parse everything up to index 100.

I think what you're really asking, is how to do that without creating unnecessary objects in memory.

The answer to that can actually also help you manage files with millions of records, without running out of memory:

Use a streaming parser.

With a streaming parser, you will get the data as it is parsed, so you can quickly skip the first X records, then begin processing records one at a time, so you never have to keep more than one record in memory.

That mean you can actually parse files of unlimited size, with a very small memory footprint.

Since you're using GSON, that means you need to use JsonReader instead of JsonParser .

Answer 2

If you have 1,000,000 records then memory usage is a concern.

The most efficient way to do this is to manually read past the first part of the file -- in the case you have shown, all your records are the same size, so you could simply use InputStream.skip() -- of course if your String fields like UserID can be different lengths then this won't work.

You could read the file character by character, counting (say) the commas to determine when you've skipped 100 records.

After you've skipped the first part of the file, you should use a streaming parser to read the rest. Gson will do that: https://sites.google.com/site/gson/streaming

You can also use a streaming parser to efficiently skip the first part of your file.