Apache Hop 1.1.0 released!
Hop 1.1.0 - Apache Hop continues to move fast!
Apache Hop 1.1.0, the first Hop release as an Apache Top-Level Project, was released late last week.
This Hop 1.1.0 release comes four months after the 1.0 release and about a month after the Top-Level graduation, and as always, is the result of a major effort by the entire Hop community.
What do you get in 1.1.0?
More robustness and stability, faster than ever
The quickly growing Hop community means more people test Hop, and find and report bugs. A growing number of contributors is eager to assist in bug hunting and fixing. As a result, a lot of the 235 JIRA tickets that have been worked on for Hop 1.1.0 were bug fixes or at least partial code cleanup and improvements.
The Hop development team started creating integration tests well before the 1.0 release. This has resulted in a library of close to 200 integration tests, that are run on a daily basis. These tests include regression testing for significant bugs, to ensure that all bugs need to be fixed once and only once. Regressions are detected and fixed as early as possible. A performance degradation that was detected has received the same treatment, which was the start of performance benchmark testing in Hop development.
All of this has resulted in a Hop 1.1.0 release that is lightning fast and as stable and robust as it gets.
New functionality and featuresHop Server
- Hop Server now supports projects and environments, asynchronous web services for long-running workflows and pipelines, and can be started with SSL.
- Hop 1.1.0 also comes with a new transform that lets you query the server status from within a pipeline.
Docker: The default Hop Docker container now supports optional support for projects and environments in both the long- and short-lived container. The container now contains additional support for logging variables and returns improved exit codes.Apache Beam: Apache Beam has been upgraded to 2.35.0, with support for Apache Spark 3.1.2 and Apache Flink 1.13.5. Service accounts on Google Cloud Dataflow are now supported, as is the DataflowServiceOptions, which pass various options through if they’re not available in the GUI
New and improved plugins:
- Hop Server Status: retrieves a lot of server and operating system metrics from one of your Hop Servers.
- Apache Tika: parses files in all sorts of formats and extracts the text content as well as available metadata it can extract.
- Metadata input: gives you access to the Hop Metadata items in the current project. It outputs all the metadata objects of all types unless one or more types are filtered.
- Neo4j Graph Output: now supports models with multiple relationships between two nodes, with multiple node labels, and with self-relationships for even more flexibility when loading data to Neo4j.
- Hop Gui has seen lots of tiny improvements that together make it a lot easier and more productive to develop or work with workflows and pipelines. These improvements include lazy file loading, reinstated git file status color coding, editable text files, and more.
- Log4j was completely removed
- Kafka consumer and producer improvements
- All the details are available in the official Apache Hop 1.1.0 release announcement.
The Hop community has grown significantly since the 1.0 release and Top-Level graduation.
The already large user bases in Europe, Latin America and North America continue to grow, but especially the explosion of interest out of Asia (mainly China) is remarkable.
With the growth in community, Hop also sees a strong growth in the number of contributors and contributions. The Hop PMC (Project Management Committee) has always been very clear: contributions can come in any shape or form, not just code. The Hop community seems to have read the memo: a diverse type of contributions have been added in code, documentation, samples, even discussions or informal feedback are considered valid contributions.
Apache Hop and know.bi
know.bi has been involved with Apache Hop since the very beginning as PMC members and contributors. We obviously will continue to do so, we'll even increase our efforts to make Hop the best data integration and data orchestration available, bar none.
As we are moving away from the "pure" consultancy and time-and-material services towards a more solution, platform, and product-based approach, we will increasingly use Apache Hop as part of our offering for robust and reliable data solutions that let you ignore the technical complexities.
Our history and background as Pentaho experts, combined with our engagement in and experience with Apache Hop since its earliest days, puts know.bi in pole position to support you in your new or existing Hop projects. We're also happy to help if you're looking for a way to upgrade your legacy Pentaho projects to Apache Hop. After all, Pentaho and Hop are two completely separate platforms, but Hop is the only logical way out for your Pentaho projects: true open-source, innovative, support for your entire project's life cycle, and overall a ton of fun!
Want to find out more? Download our free Hop fact sheet now!