API Documentation

Table of Contents

• • h5 data model
  • h5 types
  • Read and write from/to HDF5
  • Serialize/Deserialize
  • Utilities

The h5 library implements various high-level C++ wrappers around their low-level C counterparts. It is not intended as a full replacement for the HDF5 C implementation. Instead it tries to help the user with the most common tasks like opening/closing files, creating groups/datasets and writing/reading data to/from HDF5 files.

The following provides a detailed reference documentation grouped into logical units.

If you are looking for a specific function, class, etc., try using the search bar in the top left corner.

h5 data model

The h5 data model forms the backbone of the h5 library by implementing some of the key concepts from the abstract HDF5 data model. It provides concrete types for HDF5 files, groups, datasets, dataspaces and so on.

The most fundamental type in h5 is the h5::object. It serves either as a base class for

• h5::file
• h5::group

or as a type alias for

• h5::dataset
• h5::datatype
• h5::attribute
• h5::proplist
• h5::attribute

For the average user, the two important types are the h5::file and h5::group. They simplify basic tasks like creating, opening and manipulating HDF5 files, groups, subgroups, datasets and links.

h5 types

h5 types offers various functionality related to HDF5 datatypes, e.g. mapping C++ types to their corresponding HDF5 datatypes (see h5::hdf5_type) or getting the type stored in an HDF5 dataset (see h5::get_hdf5_type).

Read and write from/to HDF5

Read and write from/to HDF5 contains all the tools to actually read or write HDF5 files.

At the highest level it provides a generic interface for reading/writing from/to HDF5. Internally, the generic functions call more specialized implementations depending on what types are being passed to the interface.

h5 has native support for various types:

• Arithmetic scalar types
• std::array
• std::map
• std::optional
• std::pair
• std::tuple
• std::variant
• std::vector

The array interface helps with loading and storing n-dimensional arrays.

Furthermore, the generic design of the read/write functionality makes it easily extendible to support custom user types as well. Example 2: Write/Read a custom C++ type shows how to make a user defined type HDF5 serializable.

Serialize/Deserialize

h5 can also be used to Serialize/Deserialize an object to/from a byte buffer using the two generic functions:

• h5::serialize: Store a given object in a byte buffer.
• h5::deserialize: Restore an object from a given byte buffer.

Utilities

Utilities is a collection of convenience functions, definitions and various other tools which do not fit into any other category above but are used throughout the h5 library.