def from_json(cls, path, keys=None, **kwargs):
"""
Create a new :class:`TableSet` from a directory of JSON files or a
single JSON object with key value (Table key and list of row objects)
pairs for each :class:`Table`.
See :meth:`.Table.from_json` for additional details.
:param path:
Path to a directory containing JSON files or filepath/file-like
object of nested JSON file.
:param keys:
A list of keys of the top-level dictionaries for each file. If
specified, length must be equal to number of JSON files in path.
"""
if isinstance(path, six.string_types) and not os.path.isdir(
path) and not os.path.isfile(path):
raise IOError('Specified path doesn\'t exist.')
tables = OrderedDict()
if isinstance(path, six.string_types) and os.path.isdir(path):
filepaths = glob(os.path.join(path, '*.json'))
if keys is not None and len(keys) != len(filepaths):
raise ValueError(
'If specified, keys must have length equal to number of JSON files'
)
for i, filepath in enumerate(filepaths):
name = os.path.split(filepath)[1].strip('.json')
if keys is not None:
tables[name] = Table.from_json(filepath, keys[i], **kwargs)
else:
tables[name] = Table.from_json(filepath, **kwargs)
else:
if hasattr(path, 'read'):
js = json.load(path,
object_pairs_hook=OrderedDict,
parse_float=Decimal,
**kwargs)
else:
with open(path, 'r') as f:
js = json.load(f,
object_pairs_hook=OrderedDict,
parse_float=Decimal,
**kwargs)
for key, value in js.items():
output = StringIO(json.dumps(value))
tables[key] = Table.from_json(output)
return TableSet(tables.values(), tables.keys())
def test_monkeypatch_shadow(self):
before_table = Table([['blah'], ], ['foo'], [Text()])
Table.monkeypatch(TryPatchShadow)
after_table = Table([['blah'], ], ['foo'], [Text()])
self.assertIsInstance(before_table.columns, MappedSequence)
self.assertIsInstance(after_table.columns, MappedSequence)
with self.assertRaises(AttributeError):
after_table.foo == 'foo'
def test_monkeypatch_shadow(self):
before_table = Table([['blah'], ], [('foo', Text())])
Table.monkeypatch(TryPatchShadow)
after_table = Table([['blah'], ], [('foo', Text())])
self.assertIsInstance(before_table.columns, MappedSequence)
self.assertIsInstance(after_table.columns, MappedSequence)
with self.assertRaises(AttributeError):
after_table.foo == 'foo'
def from_json(cls, path, column_names=None, column_types=None, keys=None, **kwargs):
"""
Create a new :class:`TableSet` from a directory of JSON files or a
single JSON object with key value (Table key and list of row objects)
pairs for each :class:`Table`.
See :meth:`.Table.from_json` for additional details.
:param path:
Path to a directory containing JSON files or filepath/file-like
object of nested JSON file.
:param keys:
A list of keys of the top-level dictionaries for each file. If
specified, length must be equal to number of JSON files in path.
:param column_types:
See :meth:`Table.__init__`.
"""
from agate.tableset import TableSet
if isinstance(path, six.string_types) and not os.path.isdir(path) and not os.path.isfile(path):
raise IOError('Specified path doesn\'t exist.')
tables = OrderedDict()
if isinstance(path, six.string_types) and os.path.isdir(path):
filepaths = glob(os.path.join(path, '*.json'))
if keys is not None and len(keys) != len(filepaths):
raise ValueError('If specified, keys must have length equal to number of JSON files')
for i, filepath in enumerate(filepaths):
name = os.path.split(filepath)[1].strip('.json')
if keys is not None:
tables[name] = Table.from_json(filepath, keys[i], column_types=column_types, **kwargs)
else:
tables[name] = Table.from_json(filepath, column_types=column_types, **kwargs)
else:
if hasattr(path, 'read'):
js = json.load(path, object_pairs_hook=OrderedDict, parse_float=Decimal, **kwargs)
else:
with open(path, 'r') as f:
js = json.load(f, object_pairs_hook=OrderedDict, parse_float=Decimal, **kwargs)
for key, value in js.items():
tables[key] = Table.from_object(value, column_types=column_types, **kwargs)
return TableSet(tables.values(), tables.keys())
def from_csv(cls, dir_path, column_names=None, column_types=None, row_names=None, header=True, **kwargs):
"""
Create a new :class:`TableSet` from a directory of CSVs.
See :meth:`.Table.from_csv` for additional details.
:param dir_path:
Path to a directory full of CSV files. All CSV files in this
directory will be loaded.
:param column_names:
See :meth:`Table.__init__`.
:param column_types:
See :meth:`Table.__init__`.
:param row_names:
See :meth:`Table.__init__`.
:param header:
See :meth:`Table.from_csv`.
"""
if not os.path.isdir(dir_path):
raise IOError('Specified path doesn\'t exist or isn\'t a directory.')
tables = OrderedDict()
for path in glob(os.path.join(dir_path, '*.csv')):
name = os.path.split(path)[1].strip('.csv')
tables[name] = Table.from_csv(path, column_names, column_types, row_names=row_names, header=header, **kwargs)
return TableSet(tables.values(), tables.keys())
def from_csv(cls, dir_path, column_info, header=True, **kwargs):
"""
Create a new :class:`TableSet` from a directory of CSVs. This method
will use csvkit if it is available, otherwise it will use Python's
builtin csv module.
``kwargs`` will be passed through to :meth:`csv.reader`.
If you are using Python 2 and not using csvkit, this method is not
unicode-safe.
:param dir_path: Path to a directory full of CSV files. All CSV files
in this directory will be loaded.
:param column_info: See :class:`.Table` constructor.
:param header: If `True`, the first row of the CSV is assumed to contains
headers and will be skipped.
"""
from agate.table import Table
if not os.path.isdir(dir_path):
raise IOError(
'Specified path doesn\'t exist or isn\'t a directory.')
tables = OrderedDict()
for path in glob(os.path.join(dir_path, '*.csv')):
name = os.path.split(path)[1].strip('.csv')
table = Table.from_csv(path, column_info, header=header, **kwargs)
tables[name] = table
return TableSet(tables)
def from_csv(cls, dir_path, column_names=None, column_types=None, row_names=None, header=True, **kwargs):
"""
Create a new :class:`TableSet` from a directory of CSVs.
See :meth:`.Table.from_csv` for additional details.
:param dir_path:
Path to a directory full of CSV files. All CSV files in this
directory will be loaded.
:param column_names:
See :meth:`Table.__init__`.
:param column_types:
See :meth:`Table.__init__`.
:param row_names:
See :meth:`Table.__init__`.
:param header:
See :meth:`Table.from_csv`.
"""
if not os.path.isdir(dir_path):
raise IOError('Specified path doesn\'t exist or isn\'t a directory.')
tables = OrderedDict()
for path in glob(os.path.join(dir_path, '*.csv')):
name = os.path.split(path)[1].strip('.csv')
tables[name] = Table.from_csv(path, column_names, column_types, row_names=row_names, header=header, **kwargs)
return TableSet(tables.values(), tables.keys())
def aggregate(self, aggregations):
"""
Aggregate data from the tables in this set by performing some
set of column operations on the groups and coalescing the results into
a new :class:`.Table`.
:code:`aggregations` must be a sequence of tuples, where each has two
parts: a :code:`new_column_name` and a :class:`.Aggregation` instance.
The resulting table will have the keys from this :class:`TableSet` (and
any nested TableSets) set as its :code:`row_names`. See
:meth:`.Table.__init__` for more details.
:param aggregations:
A list of tuples in the format
:code:`(new_column_name, aggregation)`.
:returns:
A new :class:`.Table`.
"""
column_names, column_types, output, row_name_columns = _aggregate(
self, aggregations)
if len(row_name_columns) == 1:
row_names = row_name_columns[0]
else:
def row_names(r):
return tuple(r[n] for n in row_name_columns)
return Table(output, column_names, column_types, row_names=row_names)
def from_csv(cls,
dir_path,
column_info,
row_names=None,
header=True,
**kwargs):
"""
Create a new :class:`TableSet` from a directory of CSVs. This method
will use csvkit if it is available, otherwise it will use Python's
builtin csv module.
``kwargs`` will be passed through to :meth:`csv.reader`.
If you are using Python 2 and not using csvkit, this method is not
unicode-safe.
:param dir_path: Path to a directory full of CSV files. All CSV files
in this directory will be loaded.
:param column_info: A sequence of pairs of column names and types. The latter
must be instances of :class:`.DataType`. Or, an instance of
:class:`.TypeTester` to infer types.
:param row_names: See :meth:`Table.__init__`.
:param header: If `True`, the first row of the CSV is assumed to contains
headers and will be skipped.
"""
use_inference = isinstance(column_info, TypeTester)
if use_inference and not header:
raise ValueError(
'Can not apply TypeTester to a CSV without headers.')
if not os.path.isdir(dir_path):
raise IOError(
'Specified path doesn\'t exist or isn\'t a directory.')
tables = OrderedDict()
if use_inference:
has_inferred_columns = False
for path in glob(os.path.join(dir_path, '*.csv')):
name = os.path.split(path)[1].strip('.csv')
table = Table.from_csv(path,
column_info,
row_names=row_names,
header=header,
**kwargs)
if use_inference and not has_inferred_columns:
column_info = tuple(zip(table.column_names,
table.column_types))
has_inferred_columns = True
tables[name] = table
return TableSet(tables.values(), tables.keys())
def merge(cls, tables, row_names=None, column_names=None):
"""
Create a new table from a sequence of similar tables.
This method will not carry over row names from the merged tables, but new
row names can be specified with the :code:`row_names` argument.
It is possible to limit the columns included in the new :class:`.Table`
with :code:`column_names` argument. For example, to only include columns
from a specific table, set :code:`column_names` equal to
:code:`table.column_names`.
:param tables:
An sequence of :class:`.Table` instances.
:param row_names:
See :class:`.Table` for the usage of this parameter.
:param column_names:
A sequence of column names to include in the new :class:`.Table`. If
not specified, all distinct column names from `tables` are included.
:returns:
A new :class:`.Table`.
"""
from agate.table import Table
new_columns = OrderedDict()
for table in tables:
for i in range(0, len(table.columns)):
if column_names is None or table.column_names[i] in column_names:
column_name = table.column_names[i]
column_type = table.column_types[i]
if column_name in new_columns:
if not isinstance(column_type, type(new_columns[column_name])):
raise DataTypeError('Tables contain columns with the same names, but different types.')
else:
new_columns[column_name] = column_type
column_keys = new_columns.keys()
column_types = new_columns.values()
rows = []
for table in tables:
# Performance optimization for identical table structures
if table.column_names == column_keys and table.column_types == column_types:
rows.extend(table.rows)
else:
for row in table.rows:
data = []
for column_key in column_keys:
data.append(row.get(column_key, None))
rows.append(Row(data, column_keys))
return Table(rows, column_keys, column_types, row_names=row_names, _is_fork=True)
def test_table_from_csv(self):
import csvkit
from agate import table
table.csv = csvkit
if six.PY2:
table = Table.from_csv('examples/test.csv', self.tester, encoding='utf8')
else:
table = Table.from_csv('examples/test.csv', self.tester)
self.assertSequenceEqual(table.get_column_names(), ['one', 'two', 'three'])
self.assertSequenceEqual(tuple(map(type, table.get_column_types())), [Number, Number, Text])
self.assertEqual(len(table.columns), 3)
self.assertSequenceEqual(table.rows[0], [1, 4, 'a'])
self.assertSequenceEqual(table.rows[1], [2, 3, 'b'])
self.assertSequenceEqual(table.rows[2], [None, 2, u'👍'])
def print_structure(self, output=sys.stdout, max_rows=None):
"""
Print this table's column names and types as a plain-text table.
:param output:
The output to print to.
"""
from agate.table import Table
name_column = [n for n in self._column_names]
type_column = [t.__class__.__name__ for t in self._column_types]
rows = zip(name_column, type_column)
column_names = ['column', 'data_type']
text = Text()
column_types = [text, text]
table = Table(rows, column_names, column_types)
return table.print_table(output=output, max_column_width=None, max_rows=max_rows)
def print_structure(self, output=sys.stdout):
"""
Print this table's column names and types as a plain-text table.
:param output:
The output to print to.
"""
from agate.table import Table
name_column = [n for n in self.column_names]
type_column = [t.__class__.__name__ for t in self.column_types]
rows = zip(name_column, type_column)
column_names = ['column', 'data_type']
text = Text()
column_types = [text, text]
table = Table(rows, column_names, column_types)
return table.print_table(output=output, max_column_width=None)
def test_table_from_csv(self):
table = Table.from_csv('examples/test.csv', self.tester)
self.assertSequenceEqual(table.column_names, ['one', 'two', 'three'])
self.assertSequenceEqual(tuple(map(type, table.column_types)), [Number, Number, Text])
self.assertEqual(len(table.columns), 3)
self.assertSequenceEqual(table.rows[0], [1, 4, 'a'])
self.assertSequenceEqual(table.rows[1], [2, 3, 'b'])
self.assertSequenceEqual(table.rows[2], [None, 2, u'👍'])
def test_table_from_csv(self):
table = Table.from_csv('examples/test.csv', column_types=self.tester)
self.assertSequenceEqual(table.column_names, ['one', 'two', 'three'])
self.assertSequenceEqual(tuple(map(type, table.column_types)), [Number, Number, Text])
self.assertEqual(len(table.columns), 3)
self.assertSequenceEqual(table.rows[0], [1, 4, 'a'])
self.assertSequenceEqual(table.rows[1], [2, 3, 'b'])
self.assertSequenceEqual(table.rows[2], [None, 2, u'👍'])
def from_json(cls, path, row_names=None, key=None, newline=False, column_types=None, **kwargs):
"""
Create a new table from a JSON file.
Once the JSON has been deseralized, the resulting Python object is
passed to :meth:`.Table.from_object`.
If the file contains a top-level dictionary you may specify what
property contains the row list using the :code:`key` parameter.
:code:`kwargs` will be passed through to :meth:`json.load`.
:param path:
Filepath or file-like object from which to read JSON data.
:param row_names:
See the :meth:`.Table.__init__`.
:param key:
The key of the top-level dictionary that contains a list of row
arrays.
:param newline:
If `True` then the file will be parsed as "newline-delimited JSON".
:param column_types:
See :meth:`.Table.__init__`.
"""
from agate.table import Table
if key is not None and newline:
raise ValueError('key and newline may not be specified together.')
if newline:
js = []
if hasattr(path, 'read'):
for line in path:
js.append(json.loads(line, object_pairs_hook=OrderedDict, parse_float=Decimal, **kwargs))
else:
with open(path, 'r') as f:
for line in f:
js.append(json.loads(line, object_pairs_hook=OrderedDict, parse_float=Decimal, **kwargs))
else:
if hasattr(path, 'read'):
js = json.load(path, object_pairs_hook=OrderedDict, parse_float=Decimal, **kwargs)
else:
with open(path, 'r') as f:
js = json.load(f, object_pairs_hook=OrderedDict, parse_float=Decimal, **kwargs)
if isinstance(js, dict):
if not key:
raise TypeError('When converting a JSON document with a top-level dictionary element, a key must be specified.')
js = js[key]
return Table.from_object(js, row_names=row_names, column_types=column_types)
def test_table_from_csv(self):
import csvkit
from agate import table
table.csv = csvkit
if six.PY2:
table = Table.from_csv('examples/test.csv',
self.tester,
encoding='utf8')
else:
table = Table.from_csv('examples/test.csv', self.tester)
self.assertSequenceEqual(table.column_names, ['one', 'two', 'three'])
self.assertSequenceEqual(tuple(map(type, table.column_types)),
[Number, Number, Text])
self.assertEqual(len(table.columns), 3)
self.assertSequenceEqual(table.rows[0], [1, 4, 'a'])
self.assertSequenceEqual(table.rows[1], [2, 3, 'b'])
self.assertSequenceEqual(table.rows[2], [None, 2, u'👍'])
def rename(self, column_names=None, row_names=None, slug_columns=False, slug_rows=False, **kwargs):
"""
Create a copy of this table with different column names or row names.
By enabling :code:`slug_columns` or :code:`slug_rows` and not specifying
new names you may slugify the table's existing names.
:code:`kwargs` will be passed through to
`awesome-slugify's <https://github.com/dimka665/awesome-slugify>`_
:code:`UniqueSlugify` class.
:param column_names:
New column names for the renamed table. May be either an array or
a dictionary mapping existing column names to new names. If not
specified, will use this table's existing column names.
:param row_names:
New row names for the renamed table. May be either an array or
a dictionary mapping existing row names to new names. If not
specified, will use this table's existing row names.
:param slug_columns:
If True, column names will be converted to slugs and duplicate names
will have unique identifiers appended.
:param slug_rows:
If True, row names will be converted to slugs and dupicate names will
have unique identifiers appended.
"""
from agate.table import Table
if isinstance(column_names, dict):
column_names = [column_names[name] if name in column_names else name for name in self._column_names]
if isinstance(row_names, dict):
row_names = [row_names[name] if name in row_names else name for name in self._row_names]
if slug_columns:
column_names = column_names or self._column_names
if column_names is not None:
column_names = utils.slugify(column_names, ensure_unique=True, **kwargs)
if slug_rows:
row_names = row_names or self.row_names
if row_names is not None:
row_names = utils.slugify(row_names, ensure_unique=True, **kwargs)
if column_names is not None and column_names != self._column_names:
if row_names is None:
row_names = self._row_names
return Table(self._rows, column_names, self._column_types, row_names=row_names, _is_fork=False)
else:
return self._fork(self._rows, column_names, self._column_types, row_names=row_names)
def test_monkeypatch(self):
before_table = Table([], ['foo'], [Text()])
Table.monkeypatch(TryPatch)
after_table = Table([], ['foo'], [Text()])
self.assertSequenceEqual(Table.__bases__, [Patchable, TryPatch])
self.assertIsNotNone(getattr(before_table, 'test'))
self.assertIsNotNone(getattr(before_table, 'testcls'))
self.assertIsNotNone(getattr(after_table, 'test'))
self.assertIsNotNone(getattr(after_table, 'testcls'))
self.assertEqual(before_table.test(5), 5)
self.assertEqual(after_table.test(5), 5)
self.assertEqual(Table.testcls(5), 5)
def print_structure(self, max_rows=20, output=sys.stdout):
"""
Print the keys and row counts of each table in the tableset.
:param max_rows:
The maximum number of rows to display before truncating the data.
Defaults to 20.
:param output:
The output used to print the structure of the :class:`Table`.
:returns:
None
"""
max_length = min(len(self.items()), max_rows)
name_column = self.keys()[0:max_length]
type_column = [str(len(table.rows)) for key, table in self.items()[0:max_length]]
rows = zip(name_column, type_column)
column_names = ['table', 'rows']
text = Text()
column_types = [text, text]
table = Table(rows, column_names, column_types)
return table.print_table(output=output, max_column_width=None)
def from_csv(cls, dir_path, column_info, header=True, **kwargs):
"""
Create a new :class:`TableSet` from a directory of CSVs. This method
will use csvkit if it is available, otherwise it will use Python's
builtin csv module.
``kwargs`` will be passed through to :meth:`csv.reader`.
If you are using Python 2 and not using csvkit, this method is not
unicode-safe.
:param dir_path: Path to a directory full of CSV files. All CSV files
in this directory will be loaded.
:param column_info: A sequence of pairs of column names and types. The latter
must be instances of :class:`.DataType`. Or, an instance of
:class:`.TypeTester` to infer types.
:param header: If `True`, the first row of the CSV is assumed to contains
headers and will be skipped.
"""
from agate.table import Table
use_inference = isinstance(column_info, TypeTester)
if use_inference and not header:
raise ValueError('Can not apply TypeTester to a CSV without headers.')
if not os.path.isdir(dir_path):
raise IOError('Specified path doesn\'t exist or isn\'t a directory.')
tables = OrderedDict()
if use_inference:
has_inferred_columns = False
for path in glob(os.path.join(dir_path, '*.csv')):
name = os.path.split(path)[1].strip('.csv')
table = Table.from_csv(path, column_info, header=header, **kwargs)
if use_inference and not has_inferred_columns:
column_info = tuple(zip(table.get_column_names(), table.get_column_types()))
has_inferred_columns = True
tables[name] = table
return TableSet(tables)
def merge(self, groups=None, group_name=None, group_type=None):
"""
Convert this TableSet into a single table. This is the inverse of
:meth:`.Table.group_by`.
Any `row_names` set on the merged tables will be lost in this
process.
:param groups:
A list of grouping factors to add to merged rows in a new column.
If specified, it should have exactly one element per :class:`Table`
in the :class:`TableSet`. If not specified or None, the grouping
factor will be the name of the :class:`Row`'s original Table.
:param group_name:
This will be the column name of the grouping factors. If None,
defaults to the :attr:`TableSet.key_name`.
:param group_type:
This will be the column type of the grouping factors. If None,
defaults to the :attr:`TableSet.key_type`.
:returns:
A new :class:`Table`.
"""
if type(groups) is not list and groups is not None:
raise ValueError('Groups must be None or a list.')
if type(groups) is list and len(groups) != len(self):
raise ValueError('Groups length must be equal to TableSet length.')
column_names = list(self.column_names)
column_types = list(self.column_types)
column_names.insert(0, group_name if group_name else self.key_name)
column_types.insert(0, group_type if group_type else self.key_type)
rows = []
for index, (key, table) in enumerate(self.items()):
for row in table.rows:
if groups is None:
rows.append(Row((key, ) + tuple(row), column_names))
else:
rows.append(
Row((groups[index], ) + tuple(row), column_names))
return Table(rows, column_names, column_types)
def from_csv(cls, dir_path, column_info, row_names=None, header=True, **kwargs):
"""
Create a new :class:`TableSet` from a directory of CSVs.
See :meth:`.Table.from_csv` for additional details.
:param dir_path:
Path to a directory full of CSV files. All CSV files in this
directory will be loaded.
:param column_info:
A sequence of pairs of column names and types. The latter must be
instances of :class:`.DataType`. Or, an instance of
:class:`.TypeTester` to infer types.
:param row_names:
See :meth:`Table.__init__`.
:param header:
If `True`, the first row of the CSV is assumed to contains headers
and will be skipped.
"""
use_inference = isinstance(column_info, TypeTester)
if use_inference and not header:
raise ValueError('Can not apply TypeTester to a CSV without headers.')
if not os.path.isdir(dir_path):
raise IOError('Specified path doesn\'t exist or isn\'t a directory.')
tables = OrderedDict()
if use_inference:
has_inferred_columns = False
for path in glob(os.path.join(dir_path, '*.csv')):
name = os.path.split(path)[1].strip('.csv')
table = Table.from_csv(path, column_info, row_names=row_names, header=header, **kwargs)
if use_inference and not has_inferred_columns:
column_info = tuple(zip(table.column_names, table.column_types))
has_inferred_columns = True
tables[name] = table
return TableSet(tables.values(), tables.keys())
def test_monkeypatch(self):
before_table = Table([], [('foo', Text())])
Table.monkeypatch(TryPatch)
after_table = Table([], [('foo', Text())])
self.assertSequenceEqual(Table.__bases__, [Patchable, TryPatch])
self.assertIsNotNone(getattr(before_table, 'test'))
self.assertIsNotNone(getattr(before_table, 'testcls'))
self.assertIsNotNone(getattr(after_table, 'test'))
self.assertIsNotNone(getattr(after_table, 'testcls'))
self.assertEqual(before_table.test(5), 5)
self.assertEqual(after_table.test(5), 5)
self.assertEqual(Table.testcls(5), 5)
def merge(self):
"""
Convert this TableSet into a single table. This is the inverse of
:meth:`.Table.group_by`.
Any :code:`row_names` set on the merged tables will be lost in this
process.
:returns: A new :class:`Table`.
"""
column_names = list(self.column_names)
column_types = list(self.column_types)
column_names.insert(0, self.key_name)
column_types.insert(0, self.key_type)
rows = []
for key, table in self.items():
for row in table.rows:
rows.append(Row((key,) + tuple(row), column_names))
return Table(rows, column_names, column_types)
def from_object(cls, obj, row_names=None, column_types=None):
"""
Create a new table from a Python object.
The object should be a list containing a dictionary for each "row".
Nested objects or lists will also be parsed. For example, this object:
.. code-block:: python
{
'one': {
'a': 1,
'b': 2,
'c': 3
},
'two': [4, 5, 6],
'three': 'd'
}
Would generate these columns and values:
.. code-block:: python
{
'one/a': 1,
'one/b': 2,
'one/c': 3,
'two.0': 4,
'two.1': 5,
'two.2': 6,
'three': 'd'
}
Column names and types will be inferred from the data.
Not all rows are required to have the same keys. Missing elements will
be filled in with null values.
:param obj:
Filepath or file-like object from which to read JSON data.
:param row_names:
See :meth:`.Table.__init__`.
:param column_types:
See :meth:`.Table.__init__`.
"""
from agate.table import Table
column_names = []
row_objects = []
for sub in obj:
parsed = utils.parse_object(sub)
for key in parsed.keys():
if key not in column_names:
column_names.append(key)
row_objects.append(parsed)
rows = []
for sub in row_objects:
r = []
for name in column_names:
r.append(sub.get(name, None))
rows.append(r)
return Table(rows, column_names, row_names=row_names, column_types=column_types)
def test_monkeypatch_deprecated(self):
with warnings.catch_warnings():
warnings.simplefilter('error')
with self.assertRaises(DeprecationWarning):
Table.monkeypatch(TryPatch)
def test_monkeypatch_double(self):
Table.monkeypatch(TryPatch)
Table.monkeypatch(TryPatch)
Table.monkeypatch(TryPatch)
self.assertSequenceEqual(Table.__bases__, [Patchable, TryPatch])
def denormalize(self,
key=None,
property_column='property',
value_column='value',
default_value=utils.default,
column_types=None):
"""
Create a new table with row values converted into columns.
For example:
+---------+-----------+---------+
| name | property | value |
+=========+===========+=========+
| Jane | gender | female |
+---------+-----------+---------+
| Jane | race | black |
+---------+-----------+---------+
| Jane | age | 24 |
+---------+-----------+---------+
| ... | ... | ... |
+---------+-----------+---------+
Can be denormalized so that each unique value in `field` becomes a
column with `value` used for its values.
+---------+----------+--------+-------+
| name | gender | race | age |
+=========+==========+========+=======+
| Jane | female | black | 24 |
+---------+----------+--------+-------+
| Jack | male | white | 35 |
+---------+----------+--------+-------+
| Joe | male | black | 28 |
+---------+----------+--------+-------+
If one or more keys are specified then the resulting table will
automatically have :code:`row_names` set to those keys.
This is the opposite of :meth:`.Table.normalize`.
:param key:
A column name or a sequence of column names that should be
maintained as they are in the normalized table. Typically these
are the tables unique identifiers and any metadata about them. Or,
:code:`None` if there are no key columns.
:param field_column:
The column whose values should become column names in the new table.
:param property_column:
The column whose values should become the values of the property
columns in the new table.
:param default_value:
Value to be used for missing values in the pivot table. If not
specified :code:`Decimal(0)` will be used for aggregations that
return :class:`.Number` data and :code:`None` will be used for
all others.
:param column_types:
A sequence of column types with length equal to number of unique
values in field_column or an instance of :class:`.TypeTester`.
Defaults to a generic :class:`.TypeTester`.
:returns:
A new :class:`.Table`.
"""
from agate.table import Table
if key is None:
key = []
elif not utils.issequence(key):
key = [key]
field_names = []
row_data = OrderedDict()
for row in self.rows:
row_key = tuple(row[k] for k in key)
if row_key not in row_data:
row_data[row_key] = OrderedDict()
f = six.text_type(row[property_column])
v = row[value_column]
if f not in field_names:
field_names.append(f)
row_data[row_key][f] = v
if default_value == utils.default:
if isinstance(self.columns[value_column].data_type, Number):
default_value = Decimal(0)
else:
default_value = None
new_column_names = key + field_names
new_rows = []
row_names = []
for k, v in row_data.items():
row = list(k)
if len(k) == 1:
row_names.append(k[0])
else:
row_names.append(k)
for f in field_names:
if f in v:
row.append(v[f])
else:
row.append(default_value)
new_rows.append(Row(row, new_column_names))
key_column_types = [
self.column_types[self.column_names.index(name)] for name in key
]
if column_types is None or isinstance(column_types, TypeTester):
tester = TypeTester() if column_types is None else column_types
force_update = dict(zip(key, key_column_types))
force_update.update(tester._force)
tester._force = force_update
new_column_types = tester.run(new_rows, new_column_names)
else:
new_column_types = key_column_types + list(column_types)
return Table(new_rows,
new_column_names,
new_column_types,
row_names=row_names)
def test_monkeypatch_double(self):
Table.monkeypatch(TryPatch)
Table.monkeypatch(TryPatch)
Table.monkeypatch(TryPatch)
self.assertSequenceEqual(Table.__bases__, [Patchable, TryPatch])
def normalize(self,
key,
properties,
property_column='property',
value_column='value',
column_types=None):
"""
Create a new table with columns converted into rows values.
For example:
+---------+----------+--------+-------+
| name | gender | race | age |
+=========+==========+========+=======+
| Jane | female | black | 24 |
+---------+----------+--------+-------+
| Jack | male | white | 35 |
+---------+----------+--------+-------+
| Joe | male | black | 28 |
+---------+----------+--------+-------+
can be normalized on columns 'gender', 'race' and 'age':
+---------+-----------+---------+
| name | property | value |
+=========+===========+=========+
| Jane | gender | female |
+---------+-----------+---------+
| Jane | race | black |
+---------+-----------+---------+
| Jane | age | 24 |
+---------+-----------+---------+
| ... | ... | ... |
+---------+-----------+---------+
This is the opposite of :meth:`.Table.denormalize`.
:param key:
A column name or a sequence of column names that should be
maintained as they are in the normalized self. Typically these
are the tables unique identifiers and any metadata about them.
:param properties:
A column name or a sequence of column names that should be
converted to properties in the new self.
:param property_column:
The name to use for the column containing the property names.
:param value_column:
The name to use for the column containing the property values.
:param column_types:
A sequence of two column types for the property and value column in
that order or an instance of :class:`.TypeTester`. Defaults to a
generic :class:`.TypeTester`.
:returns:
A new :class:`.Table`.
"""
from agate.table import Table
new_rows = []
if not utils.issequence(key):
key = [key]
if not utils.issequence(properties):
properties = [properties]
new_column_names = key + [property_column, value_column]
row_names = []
for row in self.rows:
k = tuple(row[n] for n in key)
left_row = list(k)
if len(k) == 1:
row_names.append(k[0])
else:
row_names.append(k)
for f in properties:
new_rows.append(
Row(tuple(left_row + [f, row[f]]), new_column_names))
key_column_types = [
self.column_types[self.column_names.index(name)] for name in key
]
if column_types is None or isinstance(column_types, TypeTester):
tester = TypeTester() if column_types is None else column_types
force_update = dict(zip(key, key_column_types))
force_update.update(tester._force)
tester._force = force_update
new_column_types = tester.run(new_rows, new_column_names)
else:
new_column_types = key_column_types + list(column_types)
return Table(new_rows,
new_column_names,
new_column_types,
row_names=row_names)
def from_json(cls,
path,
row_names=None,
key=None,
newline=False,
column_types=None,
encoding='utf-8',
**kwargs):
"""
Create a new table from a JSON file.
Once the JSON has been deseralized, the resulting Python object is
passed to :meth:`.Table.from_object`.
If the file contains a top-level dictionary you may specify what
property contains the row list using the :code:`key` parameter.
:code:`kwargs` will be passed through to :meth:`json.load`.
:param path:
Filepath or file-like object from which to read JSON data.
:param row_names:
See the :meth:`.Table.__init__`.
:param key:
The key of the top-level dictionary that contains a list of row
arrays.
:param newline:
If `True` then the file will be parsed as "newline-delimited JSON".
:param column_types:
See :meth:`.Table.__init__`.
:param encoding:
According to RFC4627, JSON text shall be encoded in Unicode; the default encoding is
UTF-8. You can override this by using any encoding supported by your Python's open() function
if :code:`path` is a filepath. If passing in a file handle, it is assumed you have already opened it with the
correct encoding specified.
"""
from agate.table import Table
if key is not None and newline:
raise ValueError('key and newline may not be specified together.')
close = False
try:
if newline:
js = []
if hasattr(path, 'read'):
for line in path:
js.append(
json.loads(line,
object_pairs_hook=OrderedDict,
parse_float=Decimal,
**kwargs))
else:
f = io.open(path, encoding=encoding)
close = True
for line in f:
js.append(
json.loads(line,
object_pairs_hook=OrderedDict,
parse_float=Decimal,
**kwargs))
else:
if hasattr(path, 'read'):
js = json.load(path,
object_pairs_hook=OrderedDict,
parse_float=Decimal,
**kwargs)
else:
f = io.open(path, encoding=encoding)
close = True
js = json.load(f,
object_pairs_hook=OrderedDict,
parse_float=Decimal,
**kwargs)
if isinstance(js, dict):
if not key:
raise TypeError(
'When converting a JSON document with a top-level dictionary element, a key must be specified.'
)
js = js[key]
finally:
if close:
f.close()
return Table.from_object(js,
row_names=row_names,
column_types=column_types)
def from_fixed(cls,
path,
schema_path,
column_names=utils.default,
column_types=None,
row_names=None,
encoding='utf-8',
schema_encoding='utf-8'):
"""
Create a new table from a fixed-width file and a CSV schema.
Schemas must be in the "ffs" format. There is a repository of such schemas
maintained at `wireservice/ffs <https://github.com/wireservice/ffs>`_.
:param path:
File path or file-like object from which to read fixed-width data.
:param schema_path:
File path or file-like object from which to read schema (CSV) data.
:param column_names:
By default, these will be parsed from the schema. For alternatives, see
:meth:`.Table.__init__`.
:param column_types:
See :meth:`.Table.__init__`.
:param row_names:
See :meth:`.Table.__init__`.
:param encoding:
Character encoding of the fixed-width file. Note: if passing in a file
handle it is assumed you have already opened it with the correct
encoding specified.
:param schema_encoding:
Character encoding of the schema file. Note: if passing in a file
handle it is assumed you have already opened it with the correct
encoding specified.
"""
from agate.table import Table
close_f = False
if not hasattr(path, 'read'):
f = io.open(path, encoding=encoding)
close_f = True
else:
f = path
close_schema_f = False
if not hasattr(schema_path, 'read'):
schema_f = io.open(schema_path, encoding=schema_encoding)
close_schema_f = True
else:
schema_f = path
reader = fixed.reader(f, schema_f)
rows = list(reader)
if close_f:
f.close()
if close_schema_f:
schema_f.close()
if column_names == utils.default:
column_names = reader.fieldnames
return Table(rows, column_names, column_types, row_names=row_names)
def from_csv(cls, path, column_names=None, column_types=None, row_names=None, skip_lines=0, header=True, sniff_limit=0,
encoding='utf-8', row_limit=None, **kwargs):
"""
Create a new table from a CSV.
This method uses agate's builtin CSV reader, which supplies encoding
support for both Python 2 and Python 3.
:code:`kwargs` will be passed through to the CSV reader.
:param path:
Filepath or file-like object from which to read CSV data. If a file-like
object is specified, it must be seekable. If using Python 2, the file
should be opened in binary mode (`rb`).
:param column_names:
See :meth:`.Table.__init__`.
:param column_types:
See :meth:`.Table.__init__`.
:param row_names:
See :meth:`.Table.__init__`.
:param skip_lines:
The number of lines to skip from the top of the file.
:param header:
If :code:`True`, the first row of the CSV is assumed to contain column
names. If :code:`header` and :code:`column_names` are both specified
then a row will be skipped, but :code:`column_names` will be used.
:param sniff_limit:
Limit CSV dialect sniffing to the specified number of bytes. Set to
None to sniff the entire file. Defaults to 0 (no sniffing).
:param encoding:
Character encoding of the CSV file. Note: if passing in a file
handle it is assumed you have already opened it with the correct
encoding specified.
:param row_limit:
Limit how many rows of data will be read.
"""
from agate import csv
from agate.table import Table
close = False
try:
if hasattr(path, 'read'):
f = path
else:
if six.PY2:
f = open(path, 'Urb')
else:
f = io.open(path, encoding=encoding)
close = True
if isinstance(skip_lines, int):
while skip_lines > 0:
f.readline()
skip_lines -= 1
else:
raise ValueError('skip_lines argument must be an int')
contents = six.StringIO(f.read())
if sniff_limit is None:
kwargs['dialect'] = csv.Sniffer().sniff(contents.getvalue())
elif sniff_limit > 0:
kwargs['dialect'] = csv.Sniffer().sniff(contents.getvalue()[:sniff_limit])
if six.PY2:
kwargs['encoding'] = encoding
reader = csv.reader(contents, header=header, **kwargs)
if header:
if column_names is None:
column_names = next(reader)
else:
next(reader)
if row_limit is None:
rows = tuple(reader)
else:
rows = tuple(itertools.islice(reader, row_limit))
finally:
if close:
f.close()
return Table(rows, column_names, column_types, row_names=row_names)
