Skip to content

Commit cb027c3

Browse files
committed
Getting ready for release 2025.6.26
1 parent 73415fc commit cb027c3

3 files changed

Lines changed: 51 additions & 12 deletions

File tree

ANNOUNCE.md

Lines changed: 7 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -1,18 +1,14 @@
1-
Announcing Caterva2 2025.05.02
2-
==============================
1+
Announcing Caterva2 2025.6.26
2+
=============================
33

44
Caterva2 is a high-performance storage and computation system for
55
Blosc2 data repositories.
66

7-
This is a minor release that includes several bug fixes and
8-
improvements in the web UI. In particular, the `Client.lazyexpr`
9-
method has made the `operands` parameter optional (it was required),
10-
and it gained a new `compute` parameter, which allows to compute the
11-
lazy expression immediately, if desired.
12-
13-
Important: this release includes a breaking change in the client server
14-
communication, so your will need to update the client package to
15-
version 2025.5.2 or later for a smoother experience.
7+
This is a major release that includes native support for HDF5 datasets.
8+
Also, the tomography viewer has been improved to support navigation
9+
across different axes, not just the first one. In addition, we have
10+
added new `concatenate` and `stack` API methods to mirror new Blosc2
11+
capabilities.
1612

1713
For more info, you can have a look at the release notes in:
1814

RELEASE_NOTES.md

Lines changed: 43 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,48 @@
11
# Release notes
22

3+
## Changes from 2025.5.2 to 2025.6.26
4+
5+
### New support for native HDF5 files
6+
7+
* The HDF5 files can be uploaded to the server and 'unfold'ed into
8+
component datasets, which are then available for browsing and
9+
downloading.
10+
11+
* The datasets are suffixed with `.b2nd` to indicate that
12+
they can be browsed and downloaded as Blosc2-compressed arrays, but
13+
they are actually proxies to the original HDF5 datasets. This allows
14+
for a more efficient use of space and faster access to the data.
15+
16+
* New tutorial on how to use the HDF5 support:
17+
https://ironarray.io/caterva2-doc/tutorials/hdf5.html
18+
19+
### Web frontend
20+
21+
* Improvements in the tomography viewer:
22+
* It now supports travelling in any direction (not just the first dimension).
23+
* The size by default is now 512x512 pixels, which is more suitable for
24+
a faster browsing experience.
25+
* Update image with JavaScript instead of htmx for reducing latencies.
26+
27+
* The Download and Delete buttons now are always visible, not just on the
28+
Meta (previously Main) tab as before.
29+
30+
### API additions
31+
32+
* New `Client.concat` and `Client.stack` methods to be able to concatenate or
33+
stack datasets in the server. This is useful for creating new datasets from
34+
existing ones without downloading them.
35+
36+
* These have been added to the REST API as well, so they can be used
37+
from the web client or any other client that supports the REST API.
38+
39+
### Bug fixes and improvements
40+
41+
* Add field-indexing to caterva2. Fixes #187.
42+
* Enabled field filtering for Proxy sources.
43+
* Full support for HDF5Proxy instances in expressions (including where()).
44+
* Support for HDF5 proxies in lazy operands.
45+
346
## Changes from 2025.04.09 to 2025.5.2
447

548
### API changes

caterva2/__init__.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
"""Caterva2 - On demand access to remote Blosc2 data repositories"""
1111
from .client import BasicAuth, Client, Dataset, File, Root, sub_urlbase_default
1212

13-
__version__ = "2025.5.2"
13+
__version__ = "2025.6.26"
1414
"""The version in use of the Caterva2 package."""
1515

1616
__all__ = [

0 commit comments

Comments
 (0)