API

class gtable.Table(data={})

Table is a class for fast columnar storage using a bitmap index for sparse storage

add_column(k, v, dtype=None, index=None, align='top')

Column concatenation.

copy()

Returns a copy of the table

crop(key)

Purge the records where the column key is empty

del_column(k)

Column deletion

dropnan(clip=False)

Drop the NaNs and leave missing values instead

fill_column(key, fillvalue)

Fill N/A elements in the given columns with fillvalue

Parameters:

• key – String, list or tuple with the column names to be filled.
• fillvalue – Scalar to fill the N/A elements

Returns:

fillna_column(key, reverse=False, fillvalue=None)

Fillna on a column inplace

Parameters:

• key – string or list
• reverse –
• fillvalue –

Returns:

filter(predicate)

Filter table using a column specification or predicate

first_record(fill=False)

Returns the first record of the table

static from_chunks(chunks)

Create a table from table chunks

Parameters:chunks – Returns:

classmethod from_pandas(dataframe)

Create a table from a pandas dataframe

get(key, copy=False)

Gets a column or a table with columns

last_record(fill=False)

Returns the last record of the table

merge(table, column)

Merge two tables using two dense and sorted columns

records(fill=False)

Generator that returns a dictionary for each row of the table

reduce_by_key(column, check_sorted=False)

Reduce by key

Parameters:

• column –
• check_sorted –

Returns:

rename_column(old_name, new_name)

Rename a column of the table

Parameters:

• old_name –
• new_name –

Returns:

required_column(key, dtype)

Enforce the required column with a dtype

Parameters:

• key –
• dtype –

Returns:

required_columns(*args)

Enforce the required columns. Create empty columns if necessary.

Parameters:args – Returns:

sieve(idx)

Filter table using a one-dimensional array of boolean values

sort_by(column)

Sorts by values of a column

stack(table)

Vertical (Table) concatenation.

to_dict()

Translate the table to a dict {key -> array_of_values}

to_pandas(fill=False)

Translate the table to a pandas dataframe

class gtable.Column(values, index)

Indexed column view of the table

astype(dtype)

Changes the (numpy) datatype of the values

contains(item)

Returns a column with the value of the column present in item.

Parameters:item – Returns:

copy()

Return a copy of the column

date_range(fr='1970-01-01', to='2262-01-01', include_fr=True, include_to=True)

Filter a column by date range.

Parameters:

• fr –
• to –
• include_fr –
• include_to –

Returns:

dtype

Returns the datatype of the column

fill(fillvalue)

Fills the N/A values of the column with the fillvalue :param fillvalue: :return:

fillna(reverse=False, fillvalue=None)

Fills the non available value sequentially with the previous available position. Operates inplace.

is_empty()

True if the column is empty

Returns:

is_sorted()

True if the column is sorted :return:

mask(mask)

Apply mask on data to the data and the index out of place :param mask: :return:

reindex(index)

Reindex according to a global index

Parameters:index – Returns:

reorder(order)

Reorder the column inplace :param order: :return: