simpletable 0.2

wrapper around pytables/hd5f to simplify using structured data

This module removes some of the boiler-plate code required to use the excellent pytables module to save and access structured data.

Example Usage:

>>> from simpletable import SimpleTable
>>> import tables

define a table as a subclass of simple table.

>>> class ATable(SimpleTable):
...     x = tables.Float32Col()
...     y = tables.Float32Col()
...     name = tables.StringCol(16)

instantiate with: args: filename, tablename

>>> tbl = ATable('test_docs.h5', 'atable1')

insert as with pytables:

>>> row = tbl.row
>>> for i in range(50):
...    row['x'], row['y'] = i, i * 10
...    row['name'] = "name_%i" % i
...    row.append()
>>> tbl.flush()

access the entire array via the numpy array interface

>>> import numpy as np
>>> np.asarray(tbl)

there is also insert_many() method with takes an iterable of dicts with keys matching the colunns (x, y, name) in this case.

query the data (query() alias of tables’ readWhere()

>>> tbl.query('(x > 4) & (y < 70)') #doctest: +NORMALIZE_WHITESPACE
array([('name_5', 5.0, 50.0), ('name_6', 6.0, 60.0)],
      dtype=[('name', '|S16'), ('x', '<f4'), ('y', '<f4')])