It seems that systemd-nspawn, together with machinectl can use qcow2 images directly. This means I could use the images genereted by kiwi to start containers (for development and for running the tests).

Sometimes, you want a flexible way to handle certain situations involving C++ exceptions, particularly those for which alternative actions are either easily encapsulated, do not substantially interrupt the program flow, or must cross a foreign-function interface or thread boundary. A useful idiom is to rewrite attempts to perform an action with such forms of exception handling as instead returning a monadic value, like with C++17's std::optional<>:

Eliminate the overhead that usually comes with managing access details and credentials for many hosts in many locations by providing tools to install an 'anchor' service on one publicly accessible server, and a reverse tunnel and authentication synchronization service on each of the hosts one needs access to.

Python offers some useful libraries for this topic, I already know numpy and openCV, I want to try for example scikit-image. As a concrete goal, I want to try deconvolution with variable PSF - create lens profile and use this profile to improve sharpness and chromatic aberration.

There is a long-standing problem that mirrorbrain on download.opensuse.org does not do reasonable redirection when a user comes by with an ipv6 address... In my case for example, mirrorbrain claim I'm coming from an unknown country. It's a small country, ok, but by far not unknown (Switzerland)

Ever left a restaurant wanting to write a review, but thinking it wasn't worth the trouble to tap out all those words on your phone -- you just want to give the place your n stars and provide a few words of praise or condemnation? If only you could press a button to generate a plausible review. If this project happens, you will. We'll use the Yelp API to grab as many reviews of certain types of restaurants as the terms of service allow (I assume "Use any robot, spider, site search/retrieval application, or other automated device, process or means to access, retrieve, scrape, or index any portion of the Site or any Site Content;" doesn't apply to API users -- otherwise it wouldn't be much of an API).

RabbitMQ is, in this hacker's opinion, hard to run, hard to scale, hard to debug, and difficult to run in a HA situation. ZeroMQ takes a different approach from the centralized broker model, and instead runs a daemon on every machine that needs to send or receive message over the bus, and communicates directly between machines. * Package up zeromq.

Playing around with kvm, libvirt and related tools to setup an easy to use and quickly available test environment on my workstation. Find out which other tools or environments are available on SUSE side, which can be used instead of a local/private solution.

The goals: - Learn enough Java fundamentals to understand OpenDaylight code

GCC's IPA-SRA pass is run as a regular pass, not as an IPA pass. While this has simplified its implementation quite a bit, it's been creating pass-ordering issues for years now. So, let me try again to make it a real IPA pass, possibly dropping the capability to turn by-reference parameters to by-value ones, but definitely giving it the ability to work on strongly connected components of the call graph. And ending the pass-ordering issues.

There have some kernel mechanisms that they keep symmetric key or password in memory. Those password or key may leak through /dev/mem, kdump, hibernation, bpf print to userland. I want to think a design to:

kCFI is a tool that enables the compilation of commodity Operating Systems with Control-Flow Integrity protection. kCFI first prototype was developed/implemented during a PhD program held in the University of Campinas, in Brazil. Although fully functional, the tool remains in a very experimental shape, needing to be refactored prior to being released. As kCFI consists in multiple tools, which include compiler plugins, kernel patches and binary analyzers, its deployment is considerably complex. Once code refactoring is finished, the next step consists in setting up proper repositories for the tool and for each submodule, along with scripts and documentation to enable easy configuration and compilation on new environments.

In general plan for YaST is to use ruby only in future. So goal of this project is to move it forward and replace more parts with ruby. Current ideas is to replace binary y2base with ruby script, which allows easier integration with ruby tools and more important it solve some issues with embedding ruby, as it will be then first citizen.

I plan to learn openQA during this hack week. Below are the details: - Basic concepts

libmarpa is an parsing library based on Earley's algorithm, by Jeffrey Kegler.

I want to get started with developing apps for phones and tablets. It is a very general idea, rather open ended. To have tangible results, I want to make a trivial app and publish it in the official Google Play store: Just Roll One Die.

Take the source rpm of a package (e.g. systemd), find the upstream tests, run them, store the results to a database that a web ui can access and display them. Now, do this automatically, for every qam-sle update.

Nightwatch.js is a Node.js based End-to-End (E2E) testing solution for browser based apps. It's based on Webdrive, meaning it's a very reliable way for me to test HAWK UI! It's a great opportunity to learn Webpack, PhantomJS nightwatch and to set up a TDD approach with any new features written with vue.js, the framework which we will be using in HAWK from now on.

Implement a basic file search for a given OBS project. An example of basic functionality can already be found for Ubuntu or Debian. The goal is to implement,

Features

In the last hackweeks and my free time, I've been developing several tools to various degrees of completion. The purpose of this hackweek is to finish three of them, publish them on github and create openSUSE packages. * finddupmusic

1. currently when you navigate to Test Results page - it will load everything before show you the page. I plan to change this model into "load on demand" approach
2. Test Details page currently show you thumbnails for all modules. Want to add expand/collapse functionality for thumbnails - by default you will not see them for passed modules. But you will have ability to expand them. Also I will add "Expand All"/"Collapse All"

Just an excuse to learn more about antlr and parsing in general. Update Implemented a very simple interpreter https://github.com/mateiw/littlebasic