Commit 60416260 authored by Marc Tamlyn's avatar Marc Tamlyn
Browse files

Added array field support for PostgreSQL.

The first part of django.contrib.postgres, including model and two form
fields for arrays of other data types.

This commit is formed of the following work:

    Add shell of postgres app and test handling.

    First draft of array fields.

    Use recursive deconstruction.

    Stop creating classes at lookup time.

    Add validation and size parameter.

    Add contained_by lookup.

    Add SimpleArrayField for forms.

    Add SplitArrayField (mainly for admin).

    Fix prepare_value for SimpleArrayField.

    Stop using MultiValueField and MultiWidget.

    They don't play nice with flexible sizes.

    Add basics of admin integration.

    Missing:
    - Tests
    - Fully working js

    Add reference document for django.contrib.postgres.fields.ArrayField.

    Various performance and style tweaks.

    Fix internal docs link, formalise code snippets.

    Remove the admin code for now.

    It needs a better way of handing JS widgets in the admin as a whole
    before it is easy to write. In particular there are serious issues
    involving DateTimePicker when used in an array.

    Add a test for nested array fields with different delimiters.

    This will be a documented pattern so having a test for it is useful.

    Add docs for SimpleArrayField.

    Add docs for SplitArrayField.

    Remove admin related code for now.

    definition -> description

    Fix typo.

    Py3 errors.

    Avoid using regexes where they're not needed.

    Allow passing tuples by the programmer.

    Add some more tests for multidimensional arrays.

    Also fix slicing as much as it can be fixed.

    Simplify SplitArrayWidget's data loading.

    If we aren't including the variable size one, we don't need to search
    like this.
parent d9d92425
Loading
Loading
Loading
Loading
+0 −0

Empty file added.

+1 −0
Original line number Diff line number Diff line
from .array import *  # NOQA
+254 −0
Original line number Diff line number Diff line
import json

from django.contrib.postgres.forms import SimpleArrayField
from django.contrib.postgres.validators import ArrayMaxLengthValidator
from django.core import checks, exceptions
from django.db.models import Field, Lookup, Transform, IntegerField
from django.utils import six
from django.utils.translation import string_concat, ugettext_lazy as _


__all__ = ['ArrayField']


class AttributeSetter(object):
    def __init__(self, name, value):
        setattr(self, name, value)


class ArrayField(Field):
    empty_strings_allowed = False
    default_error_messages = {
        'item_invalid': _('Item %(nth)s in the array did not validate: '),
        'nested_array_mismatch': _('Nested arrays must have the same length.'),
    }

    def __init__(self, base_field, size=None, **kwargs):
        self.base_field = base_field
        self.size = size
        if self.size:
            self.default_validators = self.default_validators[:]
            self.default_validators.append(ArrayMaxLengthValidator(self.size))
        super(ArrayField, self).__init__(**kwargs)

    def check(self, **kwargs):
        errors = super(ArrayField, self).check(**kwargs)
        if self.base_field.rel:
            errors.append(
                checks.Error(
                    'Base field for array cannot be a related field.',
                    hint=None,
                    obj=self,
                    id='postgres.E002'
                )
            )
        else:
            # Remove the field name checks as they are not needed here.
            base_errors = self.base_field.check()
            if base_errors:
                messages = '\n    '.join('%s (%s)' % (error.msg, error.id) for error in base_errors)
                errors.append(
                    checks.Error(
                        'Base field for array has errors:\n    %s' % messages,
                        hint=None,
                        obj=self,
                        id='postgres.E001'
                    )
                )
        return errors

    def set_attributes_from_name(self, name):
        super(ArrayField, self).set_attributes_from_name(name)
        self.base_field.set_attributes_from_name(name)

    @property
    def description(self):
        return 'Array of %s' % self.base_field.description

    def db_type(self, connection):
        size = self.size or ''
        return '%s[%s]' % (self.base_field.db_type(connection), size)

    def get_prep_value(self, value):
        if isinstance(value, list) or isinstance(value, tuple):
            return [self.base_field.get_prep_value(i) for i in value]
        return value

    def get_db_prep_lookup(self, lookup_type, value, connection, prepared=False):
        if lookup_type == 'contains':
            return [self.get_prep_value(value)]
        return super(ArrayField, self).get_db_prep_lookup(lookup_type, value,
                connection, prepared=False)

    def deconstruct(self):
        name, path, args, kwargs = super(ArrayField, self).deconstruct()
        path = 'django.contrib.postgres.fields.ArrayField'
        args.insert(0, self.base_field)
        kwargs['size'] = self.size
        return name, path, args, kwargs

    def to_python(self, value):
        if isinstance(value, six.string_types):
            # Assume we're deserializing
            vals = json.loads(value)
            value = [self.base_field.to_python(val) for val in vals]
        return value

    def value_to_string(self, obj):
        values = []
        vals = self._get_val_from_obj(obj)
        base_field = self.base_field

        for val in vals:
            obj = AttributeSetter(base_field.attname, val)
            values.append(base_field.value_to_string(obj))
        return json.dumps(values)

    def get_transform(self, name):
        transform = super(ArrayField, self).get_transform(name)
        if transform:
            return transform
        try:
            index = int(name)
        except ValueError:
            pass
        else:
            index += 1  # postgres uses 1-indexing
            return IndexTransformFactory(index, self.base_field)
        try:
            start, end = name.split('_')
            start = int(start) + 1
            end = int(end)  # don't add one here because postgres slices are weird
        except ValueError:
            pass
        else:
            return SliceTransformFactory(start, end)

    def validate(self, value, model_instance):
        super(ArrayField, self).validate(value, model_instance)
        for i, part in enumerate(value):
            try:
                self.base_field.validate(part, model_instance)
            except exceptions.ValidationError as e:
                raise exceptions.ValidationError(
                    string_concat(self.error_messages['item_invalid'], e.message),
                    code='item_invalid',
                    params={'nth': i},
                )
        if isinstance(self.base_field, ArrayField):
            if len({len(i) for i in value}) > 1:
                raise exceptions.ValidationError(
                    self.error_messages['nested_array_mismatch'],
                    code='nested_array_mismatch',
                )

    def formfield(self, **kwargs):
        defaults = {
            'form_class': SimpleArrayField,
            'base_field': self.base_field.formfield(),
            'max_length': self.size,
        }
        defaults.update(kwargs)
        return super(ArrayField, self).formfield(**defaults)


class ArrayContainsLookup(Lookup):
    lookup_name = 'contains'

    def as_sql(self, qn, connection):
        lhs, lhs_params = self.process_lhs(qn, connection)
        rhs, rhs_params = self.process_rhs(qn, connection)
        params = lhs_params + rhs_params
        return '%s @> %s' % (lhs, rhs), params


ArrayField.register_lookup(ArrayContainsLookup)


class ArrayContainedByLookup(Lookup):
    lookup_name = 'contained_by'

    def as_sql(self, qn, connection):
        lhs, lhs_params = self.process_lhs(qn, connection)
        rhs, rhs_params = self.process_rhs(qn, connection)
        params = lhs_params + rhs_params
        return '%s <@ %s' % (lhs, rhs), params


ArrayField.register_lookup(ArrayContainedByLookup)


class ArrayOverlapLookup(Lookup):
    lookup_name = 'overlap'

    def as_sql(self, qn, connection):
        lhs, lhs_params = self.process_lhs(qn, connection)
        rhs, rhs_params = self.process_rhs(qn, connection)
        params = lhs_params + rhs_params
        return '%s && %s' % (lhs, rhs), params


ArrayField.register_lookup(ArrayOverlapLookup)


class ArrayLenTransform(Transform):
    lookup_name = 'len'

    @property
    def output_type(self):
        return IntegerField()

    def as_sql(self, qn, connection):
        lhs, params = qn.compile(self.lhs)
        return 'array_length(%s, 1)' % lhs, params


ArrayField.register_lookup(ArrayLenTransform)


class IndexTransform(Transform):

    def __init__(self, index, base_field, *args, **kwargs):
        super(IndexTransform, self).__init__(*args, **kwargs)
        self.index = index
        self.base_field = base_field

    def as_sql(self, qn, connection):
        lhs, params = qn.compile(self.lhs)
        return '%s[%s]' % (lhs, self.index), params

    @property
    def output_type(self):
        return self.base_field


class IndexTransformFactory(object):

    def __init__(self, index, base_field):
        self.index = index
        self.base_field = base_field

    def __call__(self, *args, **kwargs):
        return IndexTransform(self.index, self.base_field, *args, **kwargs)


class SliceTransform(Transform):

    def __init__(self, start, end, *args, **kwargs):
        super(SliceTransform, self).__init__(*args, **kwargs)
        self.start = start
        self.end = end

    def as_sql(self, qn, connection):
        lhs, params = qn.compile(self.lhs)
        return '%s[%s:%s]' % (lhs, self.start, self.end), params


class SliceTransformFactory(object):

    def __init__(self, start, end):
        self.start = start
        self.end = end

    def __call__(self, *args, **kwargs):
        return SliceTransform(self.start, self.end, *args, **kwargs)
+1 −0
Original line number Diff line number Diff line
from .array import *  # NOQA
+185 −0
Original line number Diff line number Diff line
import copy

from django.contrib.postgres.validators import ArrayMinLengthValidator, ArrayMaxLengthValidator
from django.core.exceptions import ValidationError
from django import forms
from django.utils.safestring import mark_safe
from django.utils import six
from django.utils.translation import string_concat, ugettext_lazy as _


class SimpleArrayField(forms.CharField):
    default_error_messages = {
        'item_invalid': _('Item %(nth)s in the array did not validate: '),
    }

    def __init__(self, base_field, delimiter=',', max_length=None, min_length=None, *args, **kwargs):
        self.base_field = base_field
        self.delimiter = delimiter
        super(SimpleArrayField, self).__init__(*args, **kwargs)
        if min_length is not None:
            self.min_length = min_length
            self.validators.append(ArrayMinLengthValidator(int(min_length)))
        if max_length is not None:
            self.max_length = max_length
            self.validators.append(ArrayMaxLengthValidator(int(max_length)))

    def prepare_value(self, value):
        if isinstance(value, list):
            return self.delimiter.join([six.text_type(self.base_field.prepare_value(v)) for v in value])
        return value

    def to_python(self, value):
        if value:
            items = value.split(self.delimiter)
        else:
            items = []
        errors = []
        values = []
        for i, item in enumerate(items):
            try:
                values.append(self.base_field.to_python(item))
            except ValidationError as e:
                for error in e.error_list:
                    errors.append(ValidationError(
                        string_concat(self.error_messages['item_invalid'], error.message),
                        code='item_invalid',
                        params={'nth': i},
                    ))
        if errors:
            raise ValidationError(errors)
        return values

    def validate(self, value):
        super(SimpleArrayField, self).validate(value)
        errors = []
        for i, item in enumerate(value):
            try:
                self.base_field.validate(item)
            except ValidationError as e:
                for error in e.error_list:
                    errors.append(ValidationError(
                        string_concat(self.error_messages['item_invalid'], error.message),
                        code='item_invalid',
                        params={'nth': i},
                    ))
        if errors:
            raise ValidationError(errors)

    def run_validators(self, value):
        super(SimpleArrayField, self).run_validators(value)
        errors = []
        for i, item in enumerate(value):
            try:
                self.base_field.run_validators(item)
            except ValidationError as e:
                for error in e.error_list:
                    errors.append(ValidationError(
                        string_concat(self.error_messages['item_invalid'], error.message),
                        code='item_invalid',
                        params={'nth': i},
                    ))
        if errors:
            raise ValidationError(errors)


class SplitArrayWidget(forms.Widget):

    def __init__(self, widget, size, **kwargs):
        self.widget = widget() if isinstance(widget, type) else widget
        self.size = size
        super(SplitArrayWidget, self).__init__(**kwargs)

    @property
    def is_hidden(self):
        return self.widget.is_hidden

    def value_from_datadict(self, data, files, name):
        return [self.widget.value_from_datadict(data, files, '%s_%s' % (name, index))
                for index in range(self.size)]

    def id_for_label(self, id_):
        # See the comment for RadioSelect.id_for_label()
        if id_:
            id_ += '_0'
        return id_

    def render(self, name, value, attrs=None):
        if self.is_localized:
            self.widget.is_localized = self.is_localized
        value = value or []
        output = []
        final_attrs = self.build_attrs(attrs)
        id_ = final_attrs.get('id', None)
        for i in range(max(len(value), self.size)):
            try:
                widget_value = value[i]
            except IndexError:
                widget_value = None
            if id_:
                final_attrs = dict(final_attrs, id='%s_%s' % (id_, i))
            output.append(self.widget.render(name + '_%s' % i, widget_value, final_attrs))
        return mark_safe(self.format_output(output))

    def format_output(self, rendered_widgets):
        return ''.join(rendered_widgets)

    @property
    def media(self):
        return self.widget.media

    def __deepcopy__(self, memo):
        obj = super(SplitArrayWidget, self).__deepcopy__(memo)
        obj.widget = copy.deepcopy(self.widget)
        return obj

    @property
    def needs_multipart_form(self):
        return self.widget.needs_multipart_form


class SplitArrayField(forms.Field):
    default_error_messages = {
        'item_invalid': _('Item %(nth)s in the array did not validate: '),
    }

    def __init__(self, base_field, size, remove_trailing_nulls=False, **kwargs):
        self.base_field = base_field
        self.size = size
        self.remove_trailing_nulls = remove_trailing_nulls
        widget = SplitArrayWidget(widget=base_field.widget, size=size)
        kwargs.setdefault('widget', widget)
        super(SplitArrayField, self).__init__(**kwargs)

    def clean(self, value):
        cleaned_data = []
        errors = []
        if not any(value) and self.required:
            raise ValidationError(self.error_messages['required'])
        max_size = max(self.size, len(value))
        for i in range(max_size):
            item = value[i]
            try:
                cleaned_data.append(self.base_field.clean(item))
                errors.append(None)
            except ValidationError as error:
                errors.append(ValidationError(
                    string_concat(self.error_messages['item_invalid'], error.message),
                    code='item_invalid',
                    params={'nth': i},
                ))
                cleaned_data.append(None)
        if self.remove_trailing_nulls:
            null_index = None
            for i, value in reversed(list(enumerate(cleaned_data))):
                if value in self.base_field.empty_values:
                    null_index = i
                else:
                    break
            if null_index:
                cleaned_data = cleaned_data[:null_index]
                errors = errors[:null_index]
        errors = list(filter(None, errors))
        if errors:
            raise ValidationError(errors)
        return cleaned_data
Loading