merdacz/deserializer

RestSharp JSON deserialization playground

Deserializer

Summary

Deserializes JS objects of given shape using RestSharp JsonDeserializer underneath. Given library limitations as described on stack overflow the process happens two-phase. First is deserialization into interim representation using JsonArray for the outer Entries list. Thereafter individual entries from that array are being individually re-processed, which includes serialization and final deserialization. The logic is wrapped into DocumentDeserializer class and its usage is demonstrated in unit test DocumentDeserializerTests#Original_structure_indirect_deserialization.

Load tests

Since above mentioned re-processing happens to achieve fully-typed result class it obviously has performance impact. LoadTests 'unit-tests' serve the purpose of meassuring that against two variables - number of entries and number of loops. OriginalLoadTest uses the two-phase parsing as described above and AlternativeLoadTest depends strictly on JsonDeserializer by using a bit modified input JSON structure. The results are as follows (sorted by ascending entries multiplied loops):

Version	# of entries	# of loops	Total time [ms]
Original	2	500	147
Alternative	2	500	119
Original	10	500	505 *
Alternative	10	500	897 *
Original	2	5000	1548
Alternative	2	5000	1192
Original	20	500	919
Alternative	20	500	718
Original	10	5000	4906
Alternative	10	5000	3846
Original	20	5000	9171
Alternative	20	5000	7348

Most of the cases alternative is superior, however for 10/500 case it is the opposite. Since the result persisted as such on subsequent 1.0.7 and 1.0.8 runs the guess is that GC collection may be influencing the result. This requires further investigation.

Original format

{
  "has_title": true,
  "title": "GoodLuck",
  "entries": [
    [
      "/gettingstarted.pdf",
      {
        "thumb_exists": false,
        "path": "/GettingStarted.pdf",
        "client_mtime": "Wed, 08 Jan 2014 18:00:54+0000",
        "bytes": 249159
      }
    ],
    [
      "/task.jpg",
      {
        "thumb_exists": true,
        "path": "/Task.jpg",
        "client_mtime": "Tue, 14 Jan 2014 05:53:57+0000",
        "bytes": 207696
      }
    ]
  ]
}

Alternative format

{
  "has_title": true,
  "title": "GoodLuck",
  "entries": {
    "/gettingstarted.pdf": {
      "thumb_exists": false,
      "path": "/GettingStarted.pdf",
      "client_mtime": "Wed, 08 Jan 2014 18:00:54+0000",
      "bytes": 249159
    },
    "/task.jpg": {
      "thumb_exists": true,
      "path": "/Task.jpg",
      "client_mtime": "Tue, 14 Jan 2014 05:53:57+0000",
      "bytes": 207696
    }
  }
}