Contribute to Open Source. Search issue labels to find the right project for you!



“Eine Persona (lat. Maske) ist ein Modell aus dem Bereich der Mensch-Computer-Interaktion (MCI).[1] Die Persona stellt einen Prototyp für eine Gruppe von Nutzern dar, mit konkret ausgeprägten Eigenschaften und einem konkreten Nutzungsverhalten.” - Wikipedia /

Ein solches Tool ist gerade in der Zielgruppenbestimmung sehr hilfreich. Auf Basis einschlägiger Literatur, gilt es Personas als Feature für Ennoia zu implementieren.

Updated 27/03/2017 12:32

Don't save partial results with same name


It can be tedious to go through all ‘saves’ and manually determine which have completed and which haven’t.

While the partial saves can be very close to a successful save and may therefore want to be included, automatically including all partials can break analysis scripts and accidentally incorporate invalid results into the analysis.

Perhaps ‘partial-abc123.xml’ or ‘_save-abc123.xml’ would be a sensible naming convention.

Updated 27/03/2017 12:29

Provide virtualenv installation process


Today the Readme has instructions like:

sudo pip3 install -r requirements.txt

The problem is that this command alters the sytem, whereas it exists tools to have a local python environment using virtualenv.

I’m not an expert at all in python, so I don’t know which are the correct steps to build silkaj using virtualenv.

Updated 27/03/2017 12:33

SORMAS Transaction History Data Model- Point-in-Time architecture


@MartinWahnschaffeSymeda @StefanSzczesny @gstephan30 @stoikkanen @Krisistis

I have been brainstorming about the change history model for SORMAS Project so far. And I have come to the following conclusions that needs to be implemented. Two major thoughts:

  1. There is no clear separation in the data model between master data (i.e. attributes of an entity that are static and conceptually won’t change over its lifetime) and transactional data (i.e. dynamic attributes of an entity). This has at least two implications: Firstly, It is an explicit decision (involving our Nigerian colleagues and the evaluation team) has been made that it isimportant to store the history of these attributes. In many cases (e.g. RumorStatus, CaseStatus, ContactgroupStatus, …) it might be sufficient for purely operational purposes to only store the last value, but in order to evaluate the system, the full history of values might be needed. Secondly, do you intend to create a new entity every time a transaction takes place or refer to an entity’s last update (or to another significant transaction changing one or more attributes)?

  2. I would like to draw your attention to a specific feature of the selected technology platform for SORMAS-N SAP HANA: history (column) tables which was used in the previous project. When editing entries in history tables, e.g. insert, update, delete, a full history of changes will be automatically stored. As a result, you can derive the complete history of changes and restore the full database table content at any point in time. An example can be found at [1]


See a typical example:

How do you query the change history, producing the following table?

Key Value Hue Commit_Id A 1 - 200611 A 1 GREEN 200721 A 4 - 201039 B 2 - 200611 B 2 BLUE 200721 (B 2 BLUE 201039) C 2 - 200611 C 2 BLUE 200721 C 5 - 201039

So we can closely look at how to implement an “AS OF” Statement in a relational database maybe as a separate table as shown in the example below:

pastedgraphic-1 schema-02 (3).pdf

We need to discuss the best way to do this as we cant run away from this function. Else it would be impossible to do any evaluation of the system. Martin suggested for the pilot to use a Data warehouse which is also an option but we need to activate this for the prototype as well.

Lets think through this. I believe there is way


Some helpful links

Updated 27/03/2017 12:27

Restrict promulgation types by message type


It should be possible to restrict promulgation types by message types. This includes:

  • Model and UI for assigning list of message types to promulgation types
  • Whenever the message type changes in the message editor or template execution editor, the list of promulgation types must be updated.
Updated 27/03/2017 12:24

Possibility for user to change data in Empties inouts


Is this a bug or feature request?


What is the current behavior?

When creating an Empties inout in material receipt schedule, the bpartner from the material receipt schedule is already set, as are the warehouse and the order id of the purchase order (=> the latter only in Leergut Ausgabe). The bpartner can be changed, but the rest is read-only.

Which are the steps to reproduce?

Open material receipt schedule, select a line, and open the Empties windows.

What is the expected or desired behavior?

User should be able to set change not only the bpartner for the empties inout, but also the rest of the data.

Updated 27/03/2017 12:24

[WIP] Add ssl verification to Hawkular provider


Support verifying datawarehouse/hawkular TLS certificates, including self-signed certificates.

This is available only through REST API. Creating Datawarehouse provider will set endpoint (security_protocol, verify_ssl, certificate_authority) for containers, with 3 modes:

(ssl-with-validation, 1, nil) - default (ssl-with-validation-custom-ca, 1, certificate_authority text) (ssl-without-validation, 0, nil)

Updated 27/03/2017 12:32 2 Comments

Max height for sprite preview in Inspector


As you can see in the image, Godot does not display Sprites with a big difference in width/height aspect ratio properly. At least very tall sprites messes up the inspector view. The image used here is 800x20 pixels. As you can see the Inspector is maxed out in size but the properties are not even visible below the preview.

Solution: Make a max height for sprites displayed in the preview window.


Godot version 2.1.2

Updated 27/03/2017 12:34

Create a sample project that would use Travis as a build with utplsql


To make it easier for others to start on CI/CD, we should build a sample project that woud: - use travis as a CI - create a fresh DB on each run - pull latest(or specific) utplsql release from repo and deploy it to DB (as a requirements/dependencies library) - install example application - install tests for example application - run the tests on the example application

The example project should be organized in a way that clearly separates libraries/dependencies from the project itself.

Updated 27/03/2017 12:17

Improve color variable naming scheme


For a long time we’ve been using the double variables system, mostly in order to preserve the ability to easily create a themable stylesheet.

For example all basic colors were stored in variables beginning with $color-{{name}}: #e43e33;. Then we created color variables that made the semantic link to the actual color, like this: $color-background-primary: $color-cinnabar;

As I mentioned in the beginning we did that for a long time. By now I’m wondering if that‘s still necessary because I think we can switch to a single naming/variable system and still keep our options open regarding themability. Plus we would lose the tight coupling/chaining of semantic color variables.

So I would propose we handle it like this:

  1. Only use variables for color codes and name them after the color
  2. Place frequently (≥3⨉) used color variables in meta/_colors.scss
  3. Use color manipulating functions to create nuances (See Controlling color with Sass color functions for inspiration)

Yeah, gimme code. I need code.

$color-black: #000000;
$color-japanese-laurel: #06a20d;
$color-cinnabar: #e43e33;

// Far far down the Sass road…

.my-error {
    border: 1px solid darken($color-cinnabar, 10%);
    color: $color-cinnabar;

// 👆 If that darker red color comes up more then 2 times, 
// we’ll store it in a separate variable.

@gridonic/frontend Can I get a 👍?

Updated 27/03/2017 12:35 2 Comments

Give feedback on disabled keypress


If you type (say) h to hold a video, but you aren’t allowed to, you get no feedback and it simply seems like the hotkey is broken. It would be nice to give some visual feedback saying that the server understood what you were trying to do, but is refusing to fulfil it.

Updated 27/03/2017 12:05

Let HTML code inside the APIDoc Descriptions


Please don’t delete this template or we’ll close your issue

I’m submitting a … - [ ] bug report - [x] feature request - [ ] question

Define your petition as much as possible I would like to use HTML code inside the text descriptions when I define a component using the ApiDoc.

What is the current behavior? HTML code is shown as plain text.

If the current behavior is a bug, please provide the steps to reproduce it

What is the expected behavior?

What is the motivation / use case for changing the behavior?

Please tell us about your environment:

  • Egeo version: 1.x.x
  • Browser: [all | Chrome XX | Firefox XX | IE XX | Edge XX | Safari XX ]

Other information (e.g. detailed explanation, stacktraces, related issues, suggestions how to fix, links for us to have context, eg. stackoverflow, gitter, etc)

Updated 27/03/2017 12:03

Simplify the test infrastructure.


Eclipse product is launched for a large number of the unit tests. But some of the mare not requiring to have a Eclipse product launched.

The eclipse product is launched by the Tycho test plugin.

We should go through the unit tests and simplify the testing infrastructure in order to decrease the time spent for running the unit tests (around 30 minutes on the CI servers).

<!— @huboard:{“order”:2.5997537698508282e-54,“milestone_order”:6.901120163212979e-59,“custom_state”:“”} –>

Updated 27/03/2017 11:52



We want to be able to test latest AppFlow code based on Continuous Integration via Jenkins. The basic idea is to run Jenkins inside a Docker container or a FreeBSD Jail and trigger the build/test process on each git commit to master or a specific testing branch via pipeline.

The stages of the pipeline process (thanks to @sneaker for suggestions on this), should run in parallel and perform this specific tasks:

We should trigger the LXC/VM creation via API, like pvesh on Proxmox and vm on FreeNAS Corral, we also could use curl REST API calls.

Updated 27/03/2017 11:55

Markdown provider: Pass path & file info


Would be handy to pass info about the file to the JSON for use in the front end. For example, path and filename.


```JSON { “original”: “—\ndate: 2016-04-17\ntitle: Testing\nhandle: pages\nauthor: The DADI Team\n—\n\nhow to create a new page…”, “attributes”: { “date”: “2016-04-17T00:00:00.000Z”, “title”: “Testing”, “author”: “The DADI Team” }, “path”: [ “posts”, “setup” ], “filename”: “testing”, “contentText”: “how to create a new page…”, “contentHtml”: “<p>how to create a new page…</p>\n” }

Updated 27/03/2017 11:39

Exporter - Manage DB query, String building, and SD write task in 3 separated threads


In order to speedup the exportation process, should be fine to split the 3 macro tasks of the exporter into 3 separated threads.

Thread 1 = Performs DB queries and populates the lists (a linkedblockingqueue) of Location items; Thread 2 = Reads Location items from the list, builds strings for output file(s) and puts them into one list for each file format; Thread 3 = Reads strings from the string list(s) and write file(s);

I tried to split DB queries from the rest of exportation process a couple of months ago, and I found a noticeable performance improvement. The downside was the memory usage: the exportation was more memory hungry.

Updated 27/03/2017 11:35

RGB & RGBW channel missing


RGB(W) channels are now added.

Tested with:

*.things: rgbwLight rgbwLight01 [ nodeId="199", childId="0" ]

*.items: Color rgbwLight01 "RGBW Light" <slider> { channel="mysensors:rgbwLight:gateway:rgbwLight01:rgbw" }

*.sitemap Colorpicker item=rgbwLight01

Please use this binding to test the additional channels: Binding

RGB should work just fine, but I’m unsure about the RGBW implementation:

Updated 27/03/2017 11:31

"Triage or per default private" - tag


Allow an admin to configure one or more tags to be made private between the author and any recipients. This allows for tags to hold support issues or game master discussions you rather not want to be shared with others.

So the admin area will have the option to select a tag from the dropdown, with behind it the Recipient Select input field. From then on any new discussion with that tag will automatically hold the recipients specified ánd the author ; perhaps we can make the author optional as setting.

Updated 27/03/2017 11:32

Look up competencies from a training resource


As an end user I want to be able to see which competencies (and competency profiles) relate to a given training resource So that I can see what a given training resource is helping me with\

A/C - [ ] Reverse look up based on some identifier (perhaps the URL of the training resource) - [ ] Some mechanism to embed this into EBI or external training course pages, e.g. JavaScript

Updated 27/03/2017 11:27

Move JSDOMNodeJSEnv out of the core


With jsDependencies gone from the core #2841, there will be no further dependency on JSDOMNodeJSEnv in the rest of the sbt plugin. Therefore, JSDOMNodeJSEnv can itself be moved outside of the core, just like all the other JS envs (except the basic Node.js one).

Updated 27/03/2017 11:24

Creating knowledge and competency frameworks outside of BioExcel


As an administrator (with technical assistance) I want to be able to create knowledge bases and competency profiles for schemes other than BioExcel So that I can use this method of organising training resources outside of BioExcel

A/C - [ ] Although populating the competency profile and knowledge base should be mostly editorial it would be acceptable for some technical setup to be required first as a one off setup. For example defining the structure of the competency profile

Updated 27/03/2017 11:24 1 Comments

Move the jsDependencies out of the core


jsDependencies is a feature that never reached the level of stability and well-designed-ness of the rest of core. Moreover, there now exists a better alternative with npmDependencies of scalajs-bundler. We can’t just drop support for jsDependencies, though, as too many people rely on it still, and it’s not broken enough to justify the drop.

We can however extract the entire jsDependencies in a separate sbt plugin. scalajs-bundler proves that it is possible.

Historically, jsDependencies was added in the core because we needed it to resolve Jasmine, back when our internal testing framework was Jasmine-based. Now that our testing is JUnit-based, and does not rely on any JS library, we can move jsDependencies out. There is still one place where we actually use jsDependencies in the core, though: javalib-ex depends on JSZip. However, javalib-ex should itself go out of the core #2840.

The disappearance of the internal setting scalaJSRequestsDOM from the core means that we won’t support anymore the fact that jsdom support is automatically added as part of resolvedJSEnv. But IMO that’s OK; it will even allow to move JSDOMNodeJSEnv outside of the core as well.

Updated 27/03/2017 11:22

Move javalib-ex out of the core (or just plain remove it?)


Historically, javalib-ex has contained parts of the JDK that were implemented on top of non-standard JS features, or even depending on some JS libraries. Today, though, the only remaining thing is ZipInputStream, and it’s not even complete.

The reason this thing exists in Core is that we need for the bootstrap test. Indeed, we need to unzip the .jars containing the .sjsir files in the bootstrapped linker, and we do that through ZipInputStream. However, for the purposes of the bootstrap tests, I think we should simply integrate zip reading in the bootstrap test classpath (which is not published), rather than depending on javalib-ex for that purpose.

This would then allow us to move javalib-ex out of the core. Since it’s so small and rather incomplete, maybe it should just be removed. Scaladex suggests that no one depends on it. That said, to keep our promise on backward source compatibility, we should probably still publish it somewhere in a separate repo (possibly even a personal repo rather than one under org.scala-js).

Action items:

  • [ ] Remove the dependency on javalib-ex in the bootstrap test
  • [ ] Move javalib-ex out of the core
Updated 27/03/2017 11:36

Estimate random / structured seperately


For the total uncertainty in the easy FCDR, separately evaluate the fully random and “everything else” uncertainties. Although this could be done from my intermediate files, those only contain the uncertainties in source coordinates. For diagnostics scripts (see #65) I will ultimately want to evaluate all uncertainty components in radiance / BT units, but this is not needed for the pre-β FCDR. Consider what the easiest way is to estimate the fully random vs. everything else uncertainties for now. The fully random uncertainty has a relatively straightforward sensitivity coefficient, can perhaps estimate that one and subtract it? Another idea: for the intermediate format, add all sensitivities as data layers… would explode storage but make diagnostics and processing much easier.

Updated 27/03/2017 11:17

Exception in Ng2DropdownModule after migrating project to Angular4


I’m submitting a … (check one with “x”) [x ] bug report => search github for a similar issue or PR before submitting [ ] feature request [ ] support request

Current behavior Uncaught exception in browser console: ERROR Error: Uncaught (in promise): Error: Found the synthetic property @fade. Please include either "BrowserAnimationsModule" or "NoopAnimationsModule" in your application. Error: Found the synthetic property @fade. Please include either "BrowserAnimationsModule" or "NoopAnimationsModule" in your application. at checkNoSyntheticProp (vendor.bundle.js:40984) [angular] at EmulatedEncapsulationDomRenderer2.DefaultDomRenderer2.setProperty (vendor.bundle.js:40952) [angular] at DebugRenderer2.setProperty (vendor.bundle.js:13666) [angular] at setElementProperty (vendor.bundle.js:9943) [angular] at checkAndUpdateElementValue (vendor.bundle.js:9862) [angular] at checkAndUpdateElementInline (vendor.bundle.js:9802) [angular] at checkAndUpdateNodeInline (vendor.bundle.js:12487) [angular] at checkAndUpdateNode (vendor.bundle.js:12462) [angular] at debugCheckAndUpdateNode (vendor.bundle.js:13091) [angular] at debugCheckRenderNodeFn (vendor.bundle.js:13070) [angular] at Object.eval [as updateRenderer] (ng:///Ng2DropdownModule/Ng2DropdownMenu.ngfactory.js:300) [angular] at Object.debugUpdateRenderer [as updateRenderer] (vendor.bundle.js:13055) [angular] at checkAndUpdateView (vendor.bundle.js:12434) [angular] at callViewAction (vendor.bundle.js:12744) [angular] at checkNoSyntheticProp (vendor.bundle.js:40984) [angular] at EmulatedEncapsulationDomRenderer2.DefaultDomRenderer2.setProperty (vendor.bundle.js:40952) [angular] at DebugRenderer2.setProperty (vendor.bundle.js:13666) [angular] at setElementProperty (vendor.bundle.js:9943) [angular] at checkAndUpdateElementValue (vendor.bundle.js:9862) [angular] at checkAndUpdateElementInline (vendor.bundle.js:9802) [angular] at checkAndUpdateNodeInline (vendor.bundle.js:12487) [angular] at checkAndUpdateNode (vendor.bundle.js:12462) [angular] at debugCheckAndUpdateNode (vendor.bundle.js:13091) [angular] at debugCheckRenderNodeFn (vendor.bundle.js:13070) [angular] at Object.eval [as updateRenderer] (ng:///Ng2DropdownModule/Ng2DropdownMenu.ngfactory.js:300) [angular] at Object.debugUpdateRenderer [as updateRenderer] (vendor.bundle.js:13055) [angular] at checkAndUpdateView (vendor.bundle.js:12434) [angular] at callViewAction (vendor.bundle.js:12744) [angular] at resolvePromise (polyfills.bundle.js:5250) [angular] at resolvePromise (polyfills.bundle.js:5221) [angular] at :4200/polyfills.bundle.js:5298:17 [angular] at Object.onInvokeTask (vendor.bundle.js:4493) [angular] at ZoneDelegate.webpackJsonp.991.ZoneDelegate.invokeTask (polyfills.bundle.js:4935) [angular] at Zone.webpackJsonp.991.Zone.runTask (polyfills.bundle.js:4703) [<root> => angular] at drainMicroTaskQueue (polyfills.bundle.js:5131) [<root>]

Expected behavior No exception in browser console.

Minimal reproduction of the problem with instructions I have migrated my project to Angular4.0.0 and I get this exception in browser console.

What do you use to build your app? (SystemJS, Webpack). Please specify the version Angular CLI 1.0.0

  • Angular version: 4.0.0

  • Browser: [ Chrome XX | Firefox XX | Safari XX | Opera XX ]

Updated 27/03/2017 11:23 1 Comments




  • [x] Adds --time-limit CLI option, to override time_limit setting in .zunit.yml. (#57)
  • [x] Adds allow_risky setting in .zunit.yml, which is overridden by --allow-risky option. (#55)
  • [x] Adds fail_fast setting in .zunit.yml, which is overridden by --fail-fast option. (#55)


  • [ ] Allow tests to expect crash exceptions. (#51)


  • [x] Fix completion script

  • [x] Update Documentation (molovo/zunit-www#2)
  • [x] Bump Version
Updated 27/03/2017 11:12

Hooks // Better Git Hooks

    lovepack hooks --init-git

Add more control to the git hooks command. Right now the command will create connections for ALL the git hooks setup in the default list, but doesn’t take into account what the user wants, so it creates a lot of clutter and will execute empty hooks all the time.

Create two options: - have the script check the hooks folder for git hooks and then only install those connections. - have the user determine which new hooks to create by command line arguments.

Updated 27/03/2017 11:14

Fork me on GitHub