Fixed #3026 -- newforms: Form class now suppresses validation and errors if no data (or None) is passed in. Validation still happens if you pass in an empty dictionary. Also updated unit tests. Thanks, SmileyChris
git-svn-id: http://code.djangoproject.com/svn/django/trunk@4117 bcc190cf-cafb-0310-a4f2-bffc1f526a37
This commit is contained in:
Adrian Holovaty
parent
682e435c5f
commit
126e0ec0c3
|
|
@ -36,6 +36,7 @@ class Form(object):
|
|||
__metaclass__ = DeclarativeFieldsMetaclass
|
||||
|
||||
def __init__(self, data=None, auto_id=False): # TODO: prefix stuff
|
||||
self.ignore_errors = data is None
|
||||
self.data = data or {}
|
||||
self.auto_id = auto_id
|
||||
self.clean_data = None # Stores the data after clean() has been called.
|
||||
|
|
@ -65,22 +66,13 @@ class Form(object):
|
|||
|
||||
def is_valid(self):
|
||||
"""
|
||||
Returns True if the form has no errors. Otherwise, False. This exists
|
||||
solely for convenience, so client code can use positive logic rather
|
||||
than confusing negative logic ("if not form.errors").
|
||||
Returns True if the form has no errors. Otherwise, False. If errors are
|
||||
being ignored, returns False.
|
||||
"""
|
||||
return not bool(self.errors)
|
||||
return not self.ignore_errors and not bool(self.errors)
|
||||
|
||||
def as_table(self):
|
||||
"Returns this form rendered as HTML <tr>s -- excluding the <table></table>."
|
||||
return u'\n'.join([u'<tr><td>%s:</td><td>%s</td></tr>' % (pretty_name(name), BoundField(self, field, name)) for name, field in self.fields.items()])
|
||||
|
||||
def as_ul(self):
|
||||
"Returns this form rendered as HTML <li>s -- excluding the <ul></ul>."
|
||||
return u'\n'.join([u'<li>%s: %s</li>' % (pretty_name(name), BoundField(self, field, name)) for name, field in self.fields.items()])
|
||||
|
||||
def as_table_with_errors(self):
|
||||
"Returns this form rendered as HTML <tr>s, with errors."
|
||||
output = []
|
||||
if self.errors.get(NON_FIELD_ERRORS):
|
||||
# Errors not corresponding to a particular field are displayed at the top.
|
||||
|
|
@ -92,8 +84,8 @@ class Form(object):
|
|||
output.append(u'<tr><td>%s:</td><td>%s</td></tr>' % (pretty_name(name), bf))
|
||||
return u'\n'.join(output)
|
||||
|
||||
def as_ul_with_errors(self):
|
||||
"Returns this form rendered as HTML <li>s, with errors."
|
||||
def as_ul(self):
|
||||
"Returns this form rendered as HTML <li>s -- excluding the <ul></ul>."
|
||||
output = []
|
||||
if self.errors.get(NON_FIELD_ERRORS):
|
||||
# Errors not corresponding to a particular field are displayed at the top.
|
||||
|
|
@ -113,6 +105,9 @@ class Form(object):
|
|||
"""
|
||||
self.clean_data = {}
|
||||
errors = ErrorDict()
|
||||
if self.ignore_errors: # Stop further processing.
|
||||
self.__errors = errors
|
||||
return
|
||||
for name, field in self.fields.items():
|
||||
value = self.data.get(name, None)
|
||||
try:
|
||||
|
|
|
|||
|
|
@ -1137,31 +1137,8 @@ u''
|
|||
... first_name = CharField()
|
||||
... last_name = CharField()
|
||||
... birthday = DateField()
|
||||
>>> p = Person()
|
||||
>>> print p
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_table()
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_ul()
|
||||
<li>First name: <input type="text" name="first_name" /></li>
|
||||
<li>Last name: <input type="text" name="last_name" /></li>
|
||||
<li>Birthday: <input type="text" name="birthday" /></li>
|
||||
>>> print p.as_table_with_errors()
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_ul_with_errors()
|
||||
<li><ul><li>This field is required.</li></ul>First name: <input type="text" name="first_name" /></li>
|
||||
<li><ul><li>This field is required.</li></ul>Last name: <input type="text" name="last_name" /></li>
|
||||
<li><ul><li>This field is required.</li></ul>Birthday: <input type="text" name="birthday" /></li>
|
||||
|
||||
Pass a dictionary to a Form's __init__().
|
||||
>>> p = Person({'first_name': u'John', 'last_name': u'Lennon', 'birthday': u'1940-10-9'})
|
||||
>>> p.errors
|
||||
{}
|
||||
|
|
@ -1189,12 +1166,58 @@ u''
|
|||
<tr><td>Last name:</td><td><input type="text" name="last_name" value="Lennon" /></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" value="1940-10-9" /></td></tr>
|
||||
|
||||
Empty dictionaries are valid, too.
|
||||
>>> p = Person({})
|
||||
>>> p.errors
|
||||
{'first_name': [u'This field is required.'], 'last_name': [u'This field is required.'], 'birthday': [u'This field is required.']}
|
||||
>>> p.is_valid()
|
||||
False
|
||||
>>> print p
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_table()
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_ul()
|
||||
<li><ul><li>This field is required.</li></ul>First name: <input type="text" name="first_name" /></li>
|
||||
<li><ul><li>This field is required.</li></ul>Last name: <input type="text" name="last_name" /></li>
|
||||
<li><ul><li>This field is required.</li></ul>Birthday: <input type="text" name="birthday" /></li>
|
||||
|
||||
If you don't pass any values to the Form's __init__(), or if you pass None,
|
||||
the Form won't do any validation. Form.errors will be an empty dictionary *but*
|
||||
Form.is_valid() will return False.
|
||||
>>> p = Person()
|
||||
>>> p.errors
|
||||
{}
|
||||
>>> p.is_valid()
|
||||
False
|
||||
>>> print p
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_table()
|
||||
<tr><td>First name:</td><td><input type="text" name="first_name" /></td></tr>
|
||||
<tr><td>Last name:</td><td><input type="text" name="last_name" /></td></tr>
|
||||
<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>
|
||||
>>> print p.as_ul()
|
||||
<li>First name: <input type="text" name="first_name" /></li>
|
||||
<li>Last name: <input type="text" name="last_name" /></li>
|
||||
<li>Birthday: <input type="text" name="birthday" /></li>
|
||||
|
||||
Unicode values are handled properly.
|
||||
>>> p = Person({'first_name': u'John', 'last_name': u'\u0160\u0110\u0106\u017d\u0107\u017e\u0161\u0111'})
|
||||
>>> p = Person({'first_name': u'John', 'last_name': u'\u0160\u0110\u0106\u017d\u0107\u017e\u0161\u0111', 'birthday': '1940-10-9'})
|
||||
>>> p.as_table()
|
||||
u'<tr><td>First name:</td><td><input type="text" name="first_name" value="John" /></td></tr>\n<tr><td>Last name:</td><td><input type="text" name="last_name" value="\u0160\u0110\u0106\u017d\u0107\u017e\u0161\u0111" /></td></tr>\n<tr><td>Birthday:</td><td><input type="text" name="birthday" /></td></tr>'
|
||||
u'<tr><td>First name:</td><td><input type="text" name="first_name" value="John" /></td></tr>\n<tr><td>Last name:</td><td><input type="text" name="last_name" value="\u0160\u0110\u0106\u017d\u0107\u017e\u0161\u0111" /></td></tr>\n<tr><td>Birthday:</td><td><input type="text" name="birthday" value="1940-10-9" /></td></tr>'
|
||||
>>> p.as_ul()
|
||||
u'<li>First name: <input type="text" name="first_name" value="John" /></li>\n<li>Last name: <input type="text" name="last_name" value="\u0160\u0110\u0106\u017d\u0107\u017e\u0161\u0111" /></li>\n<li>Birthday: <input type="text" name="birthday" /></li>'
|
||||
u'<li>First name: <input type="text" name="first_name" value="John" /></li>\n<li>Last name: <input type="text" name="last_name" value="\u0160\u0110\u0106\u017d\u0107\u017e\u0161\u0111" /></li>\n<li>Birthday: <input type="text" name="birthday" value="1940-10-9" /></li>'
|
||||
|
||||
>>> p = Person({'last_name': u'Lennon'})
|
||||
>>> p.errors
|
||||
|
|
@ -1375,7 +1398,10 @@ MultipleChoiceField is a special case, as its data is required to be a list:
|
|||
There are a couple of ways to do multiple-field validation. If you want the
|
||||
validation message to be associated with a particular field, implement the
|
||||
clean_XXX() method on the Form, where XXX is the field name. As in
|
||||
Field.clean(), the clean_XXX() method should return the cleaned value:
|
||||
Field.clean(), the clean_XXX() method should return the cleaned value. In the
|
||||
clean_XXX() method, you have access to self.clean_data, which is a dictionary
|
||||
of all the data that has been cleaned *so far*, in order by the fields,
|
||||
including the current field (e.g., the field XXX if you're in clean_XXX()).
|
||||
>>> class UserRegistration(Form):
|
||||
... username = CharField(max_length=10)
|
||||
... password1 = CharField(widget=PasswordInput)
|
||||
|
|
@ -1386,6 +1412,9 @@ Field.clean(), the clean_XXX() method should return the cleaned value:
|
|||
... return self.clean_data['password2']
|
||||
>>> f = UserRegistration()
|
||||
>>> f.errors
|
||||
{}
|
||||
>>> f = UserRegistration({})
|
||||
>>> f.errors
|
||||
{'username': [u'This field is required.'], 'password1': [u'This field is required.'], 'password2': [u'This field is required.']}
|
||||
>>> f = UserRegistration({'username': 'adrian', 'password1': 'foo', 'password2': 'bar'})
|
||||
>>> f.errors
|
||||
|
|
@ -1399,8 +1428,10 @@ Field.clean(), the clean_XXX() method should return the cleaned value:
|
|||
Another way of doing multiple-field validation is by implementing the
|
||||
Form's clean() method. If you do this, any ValidationError raised by that
|
||||
method will not be associated with a particular field; it will have a
|
||||
special-case association with the field named '__all__'. Note that
|
||||
Form.clean() still needs to return a dictionary of all clean data:
|
||||
special-case association with the field named '__all__'.
|
||||
Note that in Form.clean(), you have access to self.clean_data, a dictionary of
|
||||
all the fields/values that have *not* raised a ValidationError. Also note
|
||||
Form.clean() is required to return a dictionary of all clean data.
|
||||
>>> class UserRegistration(Form):
|
||||
... username = CharField(max_length=10)
|
||||
... password1 = CharField(widget=PasswordInput)
|
||||
|
|
@ -1410,9 +1441,15 @@ Form.clean() still needs to return a dictionary of all clean data:
|
|||
... raise ValidationError(u'Please make sure your passwords match.')
|
||||
... return self.clean_data
|
||||
>>> f = UserRegistration()
|
||||
>>> f.errors
|
||||
{}
|
||||
>>> f = UserRegistration({})
|
||||
>>> print f.as_table()
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Username:</td><td><input type="text" name="username" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Password1:</td><td><input type="password" name="password1" /></td></tr>
|
||||
<tr><td colspan="2"><ul><li>This field is required.</li></ul></td></tr>
|
||||
<tr><td>Password2:</td><td><input type="password" name="password2" /></td></tr>
|
||||
>>> f.errors
|
||||
{'username': [u'This field is required.'], 'password1': [u'This field is required.'], 'password2': [u'This field is required.']}
|
||||
|
|
@ -1420,15 +1457,11 @@ Form.clean() still needs to return a dictionary of all clean data:
|
|||
>>> f.errors
|
||||
{'__all__': [u'Please make sure your passwords match.']}
|
||||
>>> print f.as_table()
|
||||
<tr><td>Username:</td><td><input type="text" name="username" value="adrian" /></td></tr>
|
||||
<tr><td>Password1:</td><td><input type="password" name="password1" value="foo" /></td></tr>
|
||||
<tr><td>Password2:</td><td><input type="password" name="password2" value="bar" /></td></tr>
|
||||
>>> print f.as_table_with_errors()
|
||||
<tr><td colspan="2"><ul><li>Please make sure your passwords match.</li></ul></td></tr>
|
||||
<tr><td>Username:</td><td><input type="text" name="username" value="adrian" /></td></tr>
|
||||
<tr><td>Password1:</td><td><input type="password" name="password1" value="foo" /></td></tr>
|
||||
<tr><td>Password2:</td><td><input type="password" name="password2" value="bar" /></td></tr>
|
||||
>>> print f.as_ul_with_errors()
|
||||
>>> print f.as_ul()
|
||||
<li><ul><li>Please make sure your passwords match.</li></ul></li>
|
||||
<li>Username: <input type="text" name="username" value="adrian" /></li>
|
||||
<li>Password1: <input type="password" name="password1" value="foo" /></li>
|
||||
|
|
@ -1486,6 +1519,55 @@ A Form's fields are displayed in the same order in which they were defined.
|
|||
<tr><td>Field12:</td><td><input type="text" name="field12" /></td></tr>
|
||||
<tr><td>Field13:</td><td><input type="text" name="field13" /></td></tr>
|
||||
<tr><td>Field14:</td><td><input type="text" name="field14" /></td></tr>
|
||||
|
||||
# Sample form processing (as if in a view) ####################################
|
||||
|
||||
>>> from django.template import Template, Context
|
||||
>>> class UserRegistration(Form):
|
||||
... username = CharField(max_length=10)
|
||||
... password1 = CharField(widget=PasswordInput)
|
||||
... password2 = CharField(widget=PasswordInput)
|
||||
... def clean(self):
|
||||
... if self.clean_data.get('password1') and self.clean_data.get('password2') and self.clean_data['password1'] != self.clean_data['password2']:
|
||||
... raise ValidationError(u'Please make sure your passwords match.')
|
||||
... return self.clean_data
|
||||
>>> def my_function(method, post_data):
|
||||
... if method == 'POST':
|
||||
... form = UserRegistration(post_data)
|
||||
... else:
|
||||
... form = UserRegistration()
|
||||
... if form.is_valid():
|
||||
... return 'VALID'
|
||||
... t = Template('<form action="" method="post">\n<table>\n{{ form }}\n</table>\n<input type="submit" />\n</form>')
|
||||
... return t.render(Context({'form': form}))
|
||||
|
||||
Case 1: GET (an empty form, with no errors).
|
||||
>>> print my_function('GET', {})
|
||||
<form action="" method="post">
|
||||
<table>
|
||||
<tr><td>Username:</td><td><input type="text" name="username" /></td></tr>
|
||||
<tr><td>Password1:</td><td><input type="password" name="password1" /></td></tr>
|
||||
<tr><td>Password2:</td><td><input type="password" name="password2" /></td></tr>
|
||||
</table>
|
||||
<input type="submit" />
|
||||
</form>
|
||||
|
||||
Case 2: POST with erroneous data (a redisplayed form, with errors).
|
||||
>>> print my_function('POST', {'username': 'this-is-a-long-username', 'password1': 'foo', 'password2': 'bar'})
|
||||
<form action="" method="post">
|
||||
<table>
|
||||
<tr><td colspan="2"><ul><li>Please make sure your passwords match.</li></ul></td></tr>
|
||||
<tr><td colspan="2"><ul><li>Ensure this value has at most 10 characters.</li></ul></td></tr>
|
||||
<tr><td>Username:</td><td><input type="text" name="username" value="this-is-a-long-username" /></td></tr>
|
||||
<tr><td>Password1:</td><td><input type="password" name="password1" value="foo" /></td></tr>
|
||||
<tr><td>Password2:</td><td><input type="password" name="password2" value="bar" /></td></tr>
|
||||
</table>
|
||||
<input type="submit" />
|
||||
</form>
|
||||
|
||||
Case 3: POST with valid data (the success message).
|
||||
>>> print my_function('POST', {'username': 'adrian', 'password1': 'secret', 'password2': 'secret'})
|
||||
VALID
|
||||
"""
|
||||
|
||||
if __name__ == "__main__":
|
||||
|
|
|
|||
Loading…
Reference in New Issue