Table Classes

import pyarrow as pa

data = {
    'n_legs': [2, 4, 5, 100],
    'animals': ['Flamingo', 'Horse', 'Brittle stars', 'Centipede']
}
table = pa.table(data)
print(table)

pa.table(data, schema=None, metadata=None)

table = pa.table({'a': [1, 2], 'b': [3, 4]})
print(table.schema)
# a: int64
# b: int64

table = pa.table({'a': [1, 2, 3]})
print(table.num_rows)  # 3

table = pa.table({'a': [1, 2], 'b': [3, 4]})
print(table.num_columns)  # 2

table = pa.table({'a': [1, 2], 'b': [3, 4]})
print(table.column_names)  # ['a', 'b']

table = pa.table({'a': [1, 2], 'b': [3, 4]})
for col in table.columns:
    print(col)

table.column(i)

table.select(columns)

table.slice(offset=0, length=None)

table.filter(mask, null_selection_behavior='drop')

table.take(indices)

table.sort_by(sorting, **kwargs)

table.group_by(keys)

table = pa.table({'a': [1, 2], 'b': [3, 4]})
py_dict = table.to_pydict()
print(py_dict)  # {'a': [1, 2], 'b': [3, 4]}

table.to_pandas(self_destruct=False, **kwargs)

table.add_column(i, field_, column)

table.remove_column(i)

table.rename_columns(names)

import pyarrow as pa

data = [
    pa.array([1, 2, 3]),
    pa.array(['a', 'b', 'c'])
]
schema = pa.schema([('nums', pa.int64()), ('letters', pa.string())])
batch = pa.record_batch(data, schema=schema)
print(batch)

pa.record_batch(data, schema=None, metadata=None)

pa.concat_tables(tables, promote=False, memory_pool=None)

pa.concat_batches(batches, memory_pool=None)

import pyarrow as pa

# Create a reader from batches
batches = [pa.record_batch([[1, 2], [3, 4]], names=['a', 'b'])]
reader = pa.RecordBatchReader.from_batches(batches[0].schema, batches)

for batch in reader:
    print(batch)

RecordBatchReader.from_batches(schema, batches)

reader.read_all()

reader.read_next_batch()