❭ API reference ❭ data_base ❭ IO ❭ LoaderDumper ❭

meta

Read and write dataframe meta.

The meta of a dataframe is an empty dataframe containing only the columns, column dtypes, index, index name, and index dtype. These meta objects are used by dask and parquet to infer these properties during reading.

Functions

save_object_meta(obj, savedir)

Save an object’s meta in the file format configured in py:mod:config.

get_meta_filename(savedir, raise_)

Get the filename of the meta file in the savedir.

read_object_meta(savedir, raise_)

Read the meta of a dask/parquet object.

set_object_meta(obj, meta)

Set/reset the meta of a dataframe.