Skip to main content

Help composing objects to build JSON.

Project description

Help composing objects to emit JSON.

emitjson.repository() creates a repository of functions to convert various type of objects to JSON-friendly objects. The repository is a single-dispatch generic function that returns a converted object if the converter for the type is registered.

The repository contains some default converters for types that are not supported by json module such as set() or datetime.datetime(). Also, default repository returns (shallow) copies of standard container types such as collections.abc.Mapping or collection.abc.Sequence. Items in the container object are also converted recursively.

>>> import datetime
>>> import emitjson
>>> my_repo = emitjson.repository()  # create emitjson repository
>>> src = [1, 2, datetime.datetime.now(), {4, 5, 6}]
>>> new = my_repo(src)
>>> print(new)
(1, 2, '2016-04-20T22:09:18.731157', (4, 5, 6))

You can add your own converters as singledispatch function. Following example add a converter to generate Base64 strings from bytes objects.

>>> import emitjson
>>> import base64
>>> my_repo = emitjson.repository()  # create emitjson repository
>>> @my_repo.register(bytes)
>>> def conv_bytes(obj):
...    # encode bytes object in Base64 format
...    return base64.b64encode(obj).decode('ascii')
...
>>> print(my_repo([b'abcd']))
('YWJjZA==', )

You can also define mapper class to convert objects to dictionary.

from emitjson repository, attr, ObjConverter

class Class1:
    def __init__(self):
        self.prop1 = 'spam'
        self.prop2 = 'ham'

class Class2:
    def __init__(self):
        self.prop3 = [Class1()]


my_repo = emitjson.repository()  # create emitjson repository

@my_repo.register(Class1)
class Class1Converter(ObjConverter):
    prop1 = attr    # get obj.prop1
    prop2 = attr    # get obj.prop2

@my_repo.register(Class2)
class Class2Converter(ObjConverter):
    prop_x = attr('prop3')  # get obj.prop3 but store as 'prop_x'

Requirements

  • Python 3.4 or later

Functions

@repository()

Create a repository of object converters. The repository is an instance of the single-dispatch generic function.

The repository overrides following types by default.

  • collections.abc.Sequence objects are converted to tuple. Elements in the sequence are converted recursively.

  • Keys and values of the collections.abc.Mapping are converted recursively.

  • datetime.date and datetime.datetime objects are converted to isoformat() string.

ObjConverter class

ObjConverter class create a converter from an arbitrary object to a dictionary. Values obtained from objects are also converted recursively.

Attribute names to be converted are defined as attr class members of ObjConverter class.

If an attribute of the ObjConverter is an instance of attr, attrname arguments specifies attribute name to be converted.

Resulting dict objects are also converted recursively.

ObjConverter.on_convert(obj, values)

on_convert method is called to obtain attribute of obj and store the value to the values dictionary. You can override on_convert() method to customize conversion.

class Class3:
    prop = 1

@myrepo.register(Class3)
class Class3Converter(ObjConverter):
    def on_convert(self, obj, values):
        super().on_convert(obj, values)
        values['dynamic_prop'] = obj.prop

print(myrepo(Class3())  # prints {'dynamic_prop': 1}

attr(attrname=None, map=None) class

Converts an attribute of object.

attrname overrides the attribute name if not omitted.

map argument is a functon to convert a target object to arbitrary objects.

SAModelConverter class

SAModelConverter class create a converter from a SQLAlchemy model. Columns of the model are converted unless specified in IGNORES member of the class.

class Test(Base):
    __tablename__ = 'test'
    a = Column(Integer, primary_key=True)
    b = Column(Integer)
    c = Column(Integer)
    d = Column(Integer)

@myrepo.register(Test)
class TestModelConverter(SAModelConverter):
    IGNORES = ('c', 'd')    # ommit Test.c and Test.d

repository.fromSQLAlchemyModel(model, attrs=None, ignores=None)

Another way to register a converter of SQLAlchemy model. Columns of the model are converted unless the name of column is listed in ignore argument. attrs is a dictionary of key name and attr object.

# Converts Test model
myrepo.fromSQLAlchemyModel(Test,
    attrs={
        'fld1': attr,           # Emits Test.fld1 as value of 'fld1'
        'X_VALUE':attr('fld_X') # Emits Test.fld_X as value of 'X_VALUE'
    },
    ignores=('fld3', 'fld4')) # ignore Test.fld3 and Test.fld2

DjangoModelConverter class

DjangoModelConverter class create a converter from a Django model. Columns of the model are converted unless specified in IGNORES member of the class.

class DjangoModel(models.Model):
    charattr = models.CharField(max_length=10)
    intattr = models.IntegerField()

@myrepo.register(DjangoModel)
class TestModelConverter(DjangoModelConverter):
    pass

repository.fromSQLAlchemyModel(model, attrs=None, ignores=None)

Another way to register a converter of Django model. Columns of the model are converted unless the name of column is listed in ignore argument. attrs is a dictionary of key name and attr object.

# Converts Test model
myrepo.fromDjangoModel(Test,
    attrs={
        'fld1': attr,           # Emits Test.fld1 as value of 'fld1'
        'X_VALUE':attr('fld_X') # Emits Test.fld_X as value of 'X_VALUE'
    },
    ignores=('fld3', 'fld4')) # ignore Test.fld3 and Test.fld2

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

emitjson-0.0.5.tar.gz (5.2 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page