I want to get an openSUSE based image working on Samsung DEX: https://www.linuxondex.com/

Debian

• https://wiki.debian.org/Apache

Storing the devices in libstorage as a graph looks like a natural approach. The boost graph library (BGL) could be used here instead of the selfmade iterator over iterator (not a graph) concept currently used. Together with a redesign of the current objects several features would be simple, e.g. using disks for filesystems, renaming LVM volume groups and logical volumes, switching to and from partitioned MD RAID.

Retty is a tool for changing stdin,stdout,stderr of a running process (http://pasky.or.cz/dev/retty/) which may be handy if you want to attach to a process running e.g. from a died ssh session, changing running process to run from under screen(1) etc.. Currently it doesn't work for 64-bit systems and it doesn't handle changing of a controlling tty of a process (which may be possible using setsid() and opening a new tty). The aim of this project it to fix the above deficiencies and package the tool in OBS.

Since I'm already working on a work-related project (https://hackweek.suse.com/14/projects/1388), I decided to also spend some time improving some of my toolbox. This includes configurations, scripts, pet projects and so on. I don't expect anyone to collaborate on this, but if someone is interested to know which scripts, programs and such I'm going to touch, feel free to ping me.

Using python's lib requests to write a small(PoC) Web spider. https://gitlab.suse.de/shukui/web-spider

On a smaller internet connection or offline the development experience can be severely degraded. This has a big impact on hackathons or other events where multiple people want to get the same content. Workarounds over USB sticks are less usable than an automated way can be. A content addressable transport that can work on a local network without internet would be better. There is a downside of local content addressable transports compared to protocols like HTTPS that AFAIK has no finished solution yet: Participants in the network can see what you request. If one wants to use such a transport by default this needs to be solved, otherwise it will remain an option that will only be used on networks where this downside is deemed acceptable. Find, ask around for more and summarize possible solutions.

Take the oportunity to learn Kubernetes, to play with AWS and compare performance and availability of openQA and OBS instances and openSUSE Mirrors running on AWS. - https://aws.amazon.com/

Play with kiwi and liveCd capabilities to create a Small liveCD to get a rescue and multimedia system in less than 700MB. Will be a liveCD in Full french.

Project Description

Uyuni recently made the switch from Javascript to Typescript. Alas, the team has a very mixed bag of experience with the technology and we could use a way to get everyone up to speed quickly.

Project Description

Many years back I created a simple python based CLI package that wrapped the NWS API to get weather forecasts, discussions and current conditions. Meanwhile I have not had time to keep it up-to-date so many pieces are broken or using deprecated features of the REST API. The package is useful to get weather information much quicker from CLI than clicking through the NWS website.

Project Description

We know that Kubernetes clusters need control planes but running them in dedicated VMs might not be always efficient, instead, we can run them as pods within the management cluster.

Project Description

Try container based openQA, including openQA webui, worker, postgres and etc. Setup complete openQA environment using various containers. I am expecting a more convenient way to deploy openQA service and, at the same time, migrate the service to another server.

We from the public cloud dev team regularly upload generic and product images to be available in public cloud systems like Amazon EC2, Google Compute Engine or Microsoft Azure. The task of uploading stuff there is not only about having the tools it's also about having the right environment which provides account credentials as well as access to the image when they are built in the buildservice. Also the location from where the upload happens can make a difference in performance. Therefore a framework to combine all this into a more clean process should be developed. The basic parts of the project consists of:

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.

I'd like to see a web-based desktop RSS reader with a simple 3-pane interface like SharpReader's but it runs in Firefox and uses Firefox shared storage to store the user's usage data on their local machines, not on a server. Technology:

learn both and be awesome

ABF is a buildsystem developed by ROSA (rosalab.com) and OpenMandriva (https://www.openmandriva.org/) currently being used for their distribution. The build system is FOSS and is available to be deployed anywhere. (https://github.com/rosa-abf) (https://github.com/OpenMandrivaSoftware/docker-builder)