z3c.schema2xml 1.0

简介

这个包可以将zope 3模式描述的对象转换为简单的 XML结构。它还可以将这个xml转换回对象。 导出和导入过程完全由模式驱动；任何 此系统在上看不到架构中未描述的属性 全部。

此系统可用于为Zope 3创建导出和导入系统 应用。它还可以用来提供 用于其他目的的对象，例如xslt转换，甚至只是 以获取用于索引目的的全文表示。

包位于lxml上，用于序列化为XML。

序列化

我们首先定义一个简单的zope 3模式：

>>> from zope import interface, schema
>>> class IName(interface.Interface):
...     first_name = schema.TextLine(title=u'First name')
...     last_name = schema.TextLine(title=u'Last name')

现在让我们创建一个实现此架构的类：

>>> from zope.interface import implements
>>> class Name(object):
...     implements(IName)
...     def __init__(self, first_name, last_name):
...         self.first_name = first_name
...         self.last_name = last_name

让我们制作一个类的实例：

>>> name = Name('Karel', 'Titulaer')

现在我们将其序列化为xml:

>>> from z3c.schema2xml import serialize
>>> print serialize('container', IName, name)
<container>
  <first_name>Karel</first_name>
  <last_name>Titulaer</last_name>
</container>

这也适用于其他类型的字段：

>>> from zope import interface, schema
>>> class IAddress(interface.Interface):
...     street_name = schema.TextLine(title=u'Street name')
...     number = schema.Int(title=u'House number')
>>> class Address(object):
...     implements(IAddress)
...     def __init__(self, street_name, number):
...         self.street_name = street_name
...         self.number = number
>>> address = Address('Hofplein', 42)
>>> print serialize('container', IAddress, address)
<container>
  <street_name>Hofplein</street_name>
  <number>42</number>
</container>

如果未填写字段，则序列化将导致空 元素：

>>> address2 = Address(None, None)
>>> print serialize('container', IAddress, address2)
<container>
  <street_name/>
  <number/>
</container>

如果架构用自己的架构定义了一个对象字段，则序列化 也可以处理此问题：

>>> class IPerson(interface.Interface):
...     name = schema.Object(title=u"Name", schema=IName)
...     address = schema.Object(title=u"Address", schema=IAddress)

>>> class Person(object):
...     implements(IPerson)
...     def __init__(self, name, address):
...         self.name = name
...         self.address = address

>>> person = Person(name, address)
>>> print serialize('person', IPerson, person)
<person>
  <name>
    <first_name>Karel</first_name>
    <last_name>Titulaer</last_name>
  </name>
  <address>
    <street_name>Hofplein</street_name>
    <number>42</number>
  </address>
</person>

模式还可以定义一个列表字段，其中包含自己的元素 架构。让我们创建一个对象并序列化它：

>>> class ICommission(interface.Interface):
...     members = schema.List(
...         title=u"Commission",
...         value_type=schema.Object(__name__='person',
...         schema=IPerson))

注意，我们必须为 用于此处的值类型，否则我们没有要序列化到的名称 XML与。

>>> class Commission(object):
...     implements(ICommission)
...     def __init__(self, members):
...         self.members = members

>>> commission = Commission(
...     [person, Person(Name('Chriet', 'Titulaer'), Address('Ruimteweg', 3))])
>>> print serialize('commission', ICommission, commission)
<commission>
  <members>
    <person>
      <name>
        <first_name>Karel</first_name>
        <last_name>Titulaer</last_name>
      </name>
      <address>
        <street_name>Hofplein</street_name>
        <number>42</number>
      </address>
    </person>
    <person>
      <name>
        <first_name>Chriet</first_name>
        <last_name>Titulaer</last_name>
      </name>
      <address>
        <street_name>Ruimteweg</street_name>
        <number>3</number>
      </address>
    </person>
  </members>
</commission>

每当我们尝试序列化 没有序列化程序：

>>> class IWithNonSerializableField(interface.Interface):
...     field = schema.Field(title=u"Commission")
>>> class NotSerializable(object):
...     implements(IWithNonSerializableField)
...     def __init__(self, value):
...         self.field = value
>>> not_serializable = NotSerializable(None)
>>> serialize('noway', IWithNonSerializableField, not_serializable)
Traceback (most recent call last):
 ...
TypeError: ('Could not adapt', <zope.schema._bootstrapfields.Field object at ...>, <InterfaceClass z3c.schema2xml._schema2xml.IXMLGenerator>)

反序列化

现在我们要根据一个模式将xml反序列化为 提供此架构。

>>> from z3c.schema2xml import deserialize
>>> xml = '''
...  <container>
...    <first_name>Karel</first_name>
...    <last_name>Titulaer</last_name>
...  </container>
...  '''
>>> name = Name('', '')
>>> deserialize(xml, IName, name)
>>> name.first_name
u'Karel'
>>> name.last_name
u'Titulaer'

XML中字段的顺序无关紧要：

>>> xml = '''
...  <container>
...    <last_name>Titulaer</last_name>
...    <first_name>Karel</first_name>
...  </container>
...  '''
>>> name = Name('', '')
>>> deserialize(xml, IName, name)
>>> name.first_name
u'Karel'
>>> name.last_name
u'Titulaer'

反序列化后，对象还提供架构接口：

>>> IName.providedBy(name)
True

这也适用于其他类型的字段：

>>> xml = '''
...  <container>
...    <street_name>Hofplein</street_name>
...    <number>42</number>
...  </container>
...  '''
>>> address = Address('', 0)
>>> deserialize(xml, IAddress, address)
>>> address.street_name
u'Hofplein'
>>> address.number
42

如果架构用自己的架构定义了一个对象字段，则序列化 也可以处理此问题：

>>> xml = '''
...  <person>
...    <name>
...      <first_name>Karel</first_name>
...      <last_name>Titulaer</last_name>
...    </name>
...    <address>
...      <street_name>Hofplein</street_name>
...      <number>42</number>
...    </address>
...  </person>
...  '''
>>> person = Person(Name('', ''), Address('', 0))
>>> deserialize(xml, IPerson, person)
>>> person.name.first_name
u'Karel'
>>> person.name.last_name
u'Titulaer'
>>> person.address.street_name
u'Hofplein'
>>> person.address.number
42
>>> IPerson.providedBy(person)
True
>>> IName.providedBy(person.name)
True
>>> IAddress.providedBy(person.address)
True

同样，字段以xml形式出现的顺序也不重要：

>>> xml = '''
...  <person>
...    <address>
...      <number>42</number>
...      <street_name>Hofplein</street_name>
...    </address>
...    <name>
...      <last_name>Titulaer</last_name>
...      <first_name>Karel</first_name>
...    </name>
...  </person>
...  '''
>>> person = Person(Name('', ''), Address('', 0))
>>> deserialize(xml, IPerson, person)
>>> person.name.first_name
u'Karel'
>>> person.name.last_name
u'Titulaer'
>>> person.address.street_name
u'Hofplein'
>>> person.address.number
42
>>> IPerson.providedBy(person)
True
>>> IName.providedBy(person.name)
True
>>> IAddress.providedBy(person.address)
True

>>> xml = '''
... <commission>
...  <members>
...    <person>
...      <name>
...        <first_name>Karel</first_name>
...        <last_name>Titulaer</last_name>
...      </name>
...      <address>
...        <street_name>Hofplein</street_name>
...        <number>42</number>
...      </address>
...    </person>
...    <person>
...      <name>
...        <first_name>Chriet</first_name>
...        <last_name>Titulaer</last_name>
...     </name>
...      <address>
...        <street_name>Ruimteweg</street_name>
...        <number>3</number>
...      </address>
...    </person>
...  </members>
... </commission>
... '''

>>> commission = Commission([])
>>> deserialize(xml, ICommission, commission)
>>> len(commission.members)
2
>>> member = commission.members[0]
>>> member.name.first_name
u'Karel'
>>> member.address.street_name
u'Hofplein'
>>> member = commission.members[1]
>>> member.name.first_name
u'Chriet'
>>> member.address.street_name
u'Ruimteweg'

当xml元素为空时，结果值应为none:

>>> from z3c.schema2xml import deserialize
>>> xml = '''
...  <container>
...    <first_name></first_name>
...    <last_name/>
...  </container>
...  '''
>>> name = Name('', '')
>>> deserialize(xml, IName, name)
>>> name.first_name is None
True
>>> name.last_name is None
True

对于所有类型的字段，如字符串和整数…：

>>> xml = '''
...  <container>
...    <street_name/>
...    <number/>
...  </container>
...  '''
>>> address = Address('', 0)
>>> deserialize(xml, IAddress, address)
>>> address.street_name is None
True
>>> address.number is None
True

…以及子对象的字段（但不是子对象本身！）：

>>> xml = '''
...  <person>
...    <name>
...      <first_name/>
...      <last_name/>
...    </name>
...    <address>
...      <street_name/>
...      <number/>
...    </address>
...  </person>
...  '''
>>> person = Person(Name('', ''), Address('', 0))
>>> deserialize(xml, IPerson, person)
>>> person.name.first_name is None
True
>>> person.name.last_name is None
True
>>> IPerson.providedBy(person)
True
>>> IName.providedBy(person.name)
True
>>> person.address is None
False
>>> person.address.street_name is None
True
>>> person.address.number is None
True
>>> IAddress.providedBy(person.address)
True

类似地，如果需要序列，则该值应为空序列：

>>> xml = '''
... <commission>
...   <members/>
... </commission>
... '''
>>> commission = Commission([])
>>> deserialize(xml, ICommission, commission)
>>> len(commission.members)
0

textline、int、object和list都刚刚测试过。现在跟着测试 对于其他具有序列化程序的字段类型。

日期时间

日期时间对象：

>>> from datetime import datetime
>>> class IWithDatetime(interface.Interface):
...     datetime = schema.Datetime(title=u'Date and time')
>>> class WithDatetime(object):
...     implements(IWithDatetime)
...     def __init__(self, datetime):
...         self.datetime = datetime
>>> with_datetime = WithDatetime(datetime(2006, 12, 31))
>>> xml = serialize('container', IWithDatetime, with_datetime)
>>> print xml
<container>
  <datetime>2006-12-31T00:00:00</datetime>
</container>
>>> new_datetime = WithDatetime(None)
>>> deserialize(xml, IWithDatetime, new_datetime)
>>> new_datetime.datetime.year
2006
>>> new_datetime.datetime.month
12
>>> new_datetime.datetime.day
31

让我们在未填写字段的情况下进行尝试：

>>> with_datetime = WithDatetime(None)
>>> xml = serialize('container', IWithDatetime, with_datetime)
>>> print xml
<container>
  <datetime/>
</container>
>>> new_datetime= WithDatetime(None)
>>> deserialize(xml, IWithDatetime, new_datetime)
>>> new_datetime.datetime is None
True

选择

选择字段。目前，我们只处理具有 文本值：

>>> from zc.sourcefactory.basic import BasicSourceFactory
>>> class ChoiceSource(BasicSourceFactory):
...     def getValues(self):
...         return [u'alpha', u'beta']
>>> class IWithChoice(interface.Interface):
...     choice = schema.Choice(title=u'Choice', required=False,
...                            source=ChoiceSource())
>>> class WithChoice(object):
...     implements(IWithChoice)
...     def __init__(self, choice):
...         self.choice = choice
>>> with_choice = WithChoice('alpha')
>>> xml = serialize('container', IWithChoice, with_choice)
>>> print xml
<container>
  <choice>alpha</choice>
</container>
>>> new_choice = WithChoice(None)
>>> deserialize(xml, IWithChoice, new_choice)
>>> new_choice.choice
'alpha'
>>> with_choice = WithChoice(None)
>>> xml = serialize('container', IWithChoice, with_choice)
>>> print xml
<container>
  <choice/>
</container>
>>> deserialize(xml, IWithChoice, new_choice)
>>> new_choice.choice is None
True

设置

集合字段与列表字段非常相似：

>>> class IWithSet(interface.Interface):
...     set = schema.Set(title=u'Set', required=False,
...                      value_type=schema.Choice(__name__='choice',
...                                               source=ChoiceSource()))
>>> class WithSet(object):
...     implements(IWithSet)
...     def __init__(self, set):
...         self.set = set
>>> with_set = WithSet(set(['alpha']))
>>> xml = serialize('container', IWithSet, with_set)
>>> print xml
<container>
  <set>
    <choice>alpha</choice>
  </set>
</container>
>>> with_set = WithSet(set(['alpha', 'beta']))
>>> xml = serialize('container', IWithSet, with_set)
>>> print xml
<container>
  <set>
    <choice>alpha</choice>
    <choice>beta</choice>
  </set>
</container>
>>> new_set = WithSet(None)
>>> deserialize(xml, IWithSet, new_set)
>>> new_set.set
set(['alpha', 'beta'])

1.0（2008-12-05）

• 更改了对grokcore.component的依赖关系，因此 在直zope 3应用程序中也很有用。
• 对lxml 2.0.9运行测试。

0.10（2008-03-10）

• 首先登录svn.zope.org。