What are the differences between json and simplejson Python modules

Question

I have seen many projects using simplejson module instead of json module from the Standard Library  Also  there are many different simplejson modules  Why would use these alternatives  instead of the one in the Standard Library

User · Answer

Some values are serialized differently between simplejson and json.

Notably, instances of collections.namedtuple are serialized as arrays by json but as objects by simplejson. You can override this behaviour by passing namedtuple_as_object=False to simplejson.dump, but by default the behaviours do not match.

>>> import collections, simplejson, json
>>> TupleClass = collections.namedtuple("TupleClass", ("a", "b"))
>>> value = TupleClass(1, 2)
>>> json.dumps(value)
'[1, 2]'
>>> simplejson.dumps(value)
'{"a": 1, "b": 2}'
>>> simplejson.dumps(value, namedtuple_as_object=False)
'[1, 2]'

User · Answer

json seems faster than simplejson in both cases of loads and dumps in latest version  Tested versions    python  3 6 8 json  2 0 9 simplejson  3 16 0   Results    gt  gt  gt  def test obj  call  data  times         s   datetime now         print  calling     call    in    obj       times    times          for   in range times           r   getattr obj  call  data        e   datetime now         print  total time     str e-s         return r   gt  gt  gt  test json   dumps   data  10000  calling   dumps  in   lt module  json  from  C   Users  jophine antony  AppData  Local  Programs  Python  Python36-32  lib  json    init   py  gt    10000  times total time   0 00 00 054857   gt  gt  gt  test simplejson   dumps   data  10000  calling   dumps  in   lt module  simplejson  from  C   Users  jophine antony  AppData  Local  Programs  Python  Python36-32  lib  site-packages  simplejson    init   py  gt    10000  times total time   0 00 00 419895    1   100   2    acs    3 5   3 5567   d    1   23     e     a    A       gt  gt  gt  test json   loads   strdata  1000  calling   loads  in   lt module  json  from  C   Users  jophine antony  AppData  Local  Programs  Python  Python36-32  lib  json    init   py  gt    1000  times total time   0 00 00 004985   1   100   2    acs    3 5   3 5567   d    1   23     e     a    A      gt  gt  gt  test simplejson   loads   strdata  1000  calling   loads  in   lt module  simplejson  from  C   Users  jophine antony  AppData  Local  Programs  Python  Python36-32  lib  site-packages  simplejson    init   py  gt    1000  times total time   0 00 00 040890   1   100   2    acs    3 5   3 5567   d    1   23     e     a    A      For versions    python  3 7 4 json  2 0 9 simplejson  3 17 0   json was faster than simplejson during dumps operation but both maintained the same speed during loads operations

User · Answer

I came across this question as I was looking to install simplejson for Python 2 6  I needed to use the  object pairs hook  of json load   in order to load a json file as an OrderedDict  Being familiar with more recent versions of Python I didn t realize that the json module for Python 2 6 doesn t include the  object pairs hook  so I had to install simplejson for this purpose  From personal experience this is why i use simplejson as opposed to the standard json module

User · Answer

All of these answers aren t very helpful because they are time sensitive  After doing some research of my own I found that simplejson is indeed faster than the builtin  if you keep it updated to the latest version  pip easy install wanted to install 2 3 2 on ubuntu 12 04  but after finding out the latest simplejson version is actually 3 3 0  so I updated it and reran the time tests   simplejson is about 3x faster than the builtin json at loads simplejson is about 30  faster than the builtin json at dumps  Disclaimer  The above statements are in python-2 7 3 and simplejson 3 3 0  with c speedups  And to make sure my answer also isn t time sensitive  you should run your own tests to check since it varies so much between versions  there s no easy answer that isn t time sensitive  How to tell if C speedups are enabled in simplejson  import simplejson   If this is True  then c speedups are enabled  print bool getattr simplejson    speedups   False    UPDATE  I recently came across a library called ujson that is performing  3x faster than simplejson with some basic tests

User · Answer

Here s  a now outdated  comparison of Python json libraries   Comparing JSON modules for Python  archive link   Regardless of the results in this comparison you should use the standard library json if you are on Python 2 6   And   might as well just use simplejson otherwise

User · Answer

The builtin json module got included in Python 2 6   Any projects that support versions of Python  lt  2 6 need to have a fallback   In many cases  that fallback is simplejson

User · Answer

An API incompatibility I found  with Python 2 7 vs simplejson 3 3 1 is in whether output produces str or unicode objects  e g    gt  gt  gt  from json import JSONDecoder  gt  gt  gt  jd   JSONDecoder    gt  gt  gt  jd decode       a   b         u a   u b     vs   gt  gt  gt  from simplejson import JSONDecoder  gt  gt  gt  jd   JSONDecoder    gt  gt  gt  jd decode       a   b          a    b     If the preference is to use simplejson  then this can be addressed by coercing the argument string to unicode  as in    gt  gt  gt  from simplejson import JSONDecoder  gt  gt  gt  jd   JSONDecoder    gt  gt  gt  jd decode unicode       a   b         utf-8     u a   u b     The coercion does require knowing the original charset  for example    gt  gt  gt  jd decode unicode       a              f e          Traceback  most recent call last     File   lt stdin gt    line 1  in  lt module gt  UnicodeDecodeError   ascii  codec can t decode byte 0xce in position 8  ordinal not in range 128    This is the won t fix issue 40

User · Answer

json is simplejson  added to the stdlib  But since json was added in 2 6  simplejson has the advantage of working on more Python versions  2 4      simplejson is also updated more frequently than Python  so if you need  or want  the latest version  it s best to use simplejson itself  if possible   A good practice  in my opinion  is to use one or the other as a fallback   try      import simplejson as json except ImportError      import json

User · Answer

I ve been benchmarking json  simplejson and cjson    cjson is fastest simplejson is almost on par with cjson json is about 10x slower than simplejson   http   pastie org 1507411     python test serialization speed py  --------------------    Encoding Tests -------------------- Encoding  100000 x   m    asdsasdqwqw    t   3         json  1 12385 seconds for 100000 runs  avg  0 011239ms  simplejson  0 44356 seconds for 100000 runs  avg  0 004436ms       cjson  0 09593 seconds for 100000 runs  avg  0 000959ms  Encoding  10000 x   m      0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     t   3         json  7 76628 seconds for 10000 runs  avg  0 776628ms  simplejson  0 51179 seconds for 10000 runs  avg  0 051179ms       cjson  0 44362 seconds for 10000 runs  avg  0 044362ms  --------------------    Decoding Tests -------------------- Decoding  100000 x   m    asdsasdqwqw    t   3         json  3 32861 seconds for 100000 runs  avg  0 033286ms  simplejson  0 37164 seconds for 100000 runs  avg  0 003716ms       cjson  0 03893 seconds for 100000 runs  avg  0 000389ms  Decoding  10000 x   m      0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19     t   3         json  37 26270 seconds for 10000 runs  avg  3 726270ms  simplejson  0 56643 seconds for 10000 runs  avg  0 056643ms       cjson  0 33007 seconds for 10000 runs  avg  0 033007ms

User · Answer

In python3  if you a string of b bytes    with json you have to  decode   the content before you can load it   simplejson takes care of this so you can just do simplejson loads byte string

User · Answer

I have to disagree with the other answers  the built in json library  in Python 2 7  is not necessarily slower than simplejson  It also doesn t have this annoying unicode bug   Here is a simple benchmark   import json import simplejson from timeit import repeat  NUMBER   100000 REPEAT   10  def compare json and simplejson data          Compare json and simplejson - dumps and loads        compare json and simplejson data   data     compare json and simplejson dump   json dumps data      assert json dumps data     simplejson dumps data      result   min repeat  json dumps compare json and simplejson data     from   main   import json  compare json and simplejson                     repeat   REPEAT  number   NUMBER       print        json dumps    seconds  format result      result   min repeat  simplejson dumps compare json and simplejson data     from   main   import simplejson  compare json and simplejson                     repeat   REPEAT  number   NUMBER       print  simplejson dumps    seconds  format result      assert json loads compare json and simplejson dump     data     result   min repeat  json loads compare json and simplejson dump     from   main   import json  compare json and simplejson                     repeat   REPEAT  number   NUMBER       print        json loads    seconds  format result      result   min repeat  simplejson loads compare json and simplejson dump     from   main   import simplejson  compare json and simplejson                     repeat   REPEAT  number   NUMBER       print  simplejson loads    seconds  format result    print  Complex real world data    COMPLEX DATA     status   1   timestamp   1362323499 23   site code    testing123    remote address    212 179 220 18    input text   u ny monday for less than  u20aa123    locale value    UK    eva version    v1 0 3286    message    Successful Parse    muuid1    11e2-8414-a5e9e0fd-95a6-12313913cc26    api reply     api reply     Money     Currency    ILS    Amount    123    Restriction    Less     ProcessedText    ny monday for less than   u20aa123    Locations      Index   0   Derived From    Default    Home    Default    Departure     Date    2013-03-04     Next   10     Arrival     Date    2013-03-04    Calculated   True    Index   10   All Airports Code    NYC    Airports    EWR JFK LGA PHL    Name    New York City  New York  United States  GID 5128581     Latitude   40 71427   Country    US    Type    City    Geoid   5128581   Longitude   -74 00597      compare json and simplejson COMPLEX DATA  print   nSimple data   SIMPLE DATA    1  2  3   asasd     a   b    compare json and simplejson SIMPLE DATA    And the results on my system  Python 2 7 4  Linux 64-bit       Complex real world data          json dumps 1 56666707993 seconds   simplejson dumps 2 25638604164 seconds         json loads 2 71256899834 seconds   simplejson loads 1 29233884811 seconds        Simple data          json dumps 0 370109081268 seconds   simplejson dumps 0 574181079865 seconds         json loads 0 422876119614 seconds   simplejson loads 0 270955085754 seconds     For dumping  json is faster than simplejson  For loading  simplejson is faster   Since I am currently building a web service  dumps   is more important   and using a standard library is always preferred   Also  cjson was not updated in the past 4 years  so I wouldn t touch it

User · Answer

Another reason projects use simplejson is that the builtin json did not originally include its C speedups  so the performance difference was noticeable

User · Answer

simplejson module is simply 1 5 times faster than json  On my computer  with simplejson 2 1 1 and Python 2 7 x86    If you want  you can try the benchmark  http   abral altervista org jsonpickle-bench zip On my PC simplejson is faster than cPickle  I would like to know also your benchmarks   Probably  as said Coady  the difference between simplejson and json is that simplejson includes  speedups c  So  why don t python developers use simplejson

[python] What are the differences between json and simplejson Python modules?

Examples related to python

Examples related to json

Examples related to simplejson