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

"norelease" branches


IMO we need some mechanism to show that the PR we merge should not create a release.

For example, a refactor PR, which should be tested as any other PR, but should not be released cause it provides no value for user.

Updated 16/10/2017 20:56

determining which nodeshift to use


This idea comes from the ember-cli. like nodeshift, ember-cli is usually installed globally

To use nodeshift, you usually install globally, npm install -g nodeshift so then to use you could use nodeshift --version and this would give you the nodeshift version installed globally.

the same thing happens with ember, but if you are in an ember project(created with the ember-cli) and you run the ember command, you get the version that is installed within that project

i feel like nodeshift should behave the same way. one way to do that now, is the use a npm script, like npm run openshift where the openshift script is nodeshift [args]. since npm uses your locally installed modules, it will use the version of nodeshift installed from your package.json

To summarize, assuming you have nodeshift installed globally, if you run nodeshift outside of a project that has nodeshift installed, you use the globally installed version

but if you run nodeshift inside a project that has it installed, you will get that locally installed version

Updated 16/10/2017 18:36

Template/form for rules page


What’s Wrong? Currently no way for admins to alter rules on the client side.

What do I Expect to Happen? Create a forms page for rules that admins can use to edit/submit as a form to modify the rules page. This will be stored in SQL as template categories.

Suggested Fix Create a template for admins to use to create their rules page. Create a form for admins to submit their rules page. Connect the rules page to the server.

Supporting Information

Updated 15/10/2017 18:33 1 Comments

Support . _ and $ in user-defined variables


Please answer these questions before submitting your issue. Thanks!

User variables are written as @var_name, where the variable name var_name consists of alphanumeric characters, ., _, and $. A user variable name can contain other characters if you quote it as a string or identifier (for example, @‘my-var’, @“my-var”, or @my-var).

  1. What did you do? If possible, provide a recipe for reproducing the error.
mysql> set @'my-var' = 42;
ERROR 1105 (HY000): line 1 column 5 near "'my-var' = 42" (total length 18)

mysql> set @`my-var` = 42;
ERROR 1105 (HY000): line 1 column 5 near "`my-var` = 42" (total length 18)

mysql> set @my.var = 42;
ERROR 1105 (HY000): line 1 column 8 near "var = 42" (total length 16)
  1. What did you expect to see?

Those should be valid user-defined variables.

Updated 16/10/2017 04:36

Make proprietary firmware and userspace optional


The built-in wifi and baseband will not work without proprietary firmware on practically all phones. However, we could allow the user to avoid installing these blobs, just like libre distributions do it.

Implementation idea

  1. Change the device package recipes to generate the following subpackages:
  2. device-vendor-name-nonfree-userland (we don’t have proprietary userland right now, so this is not needed)
  3. device-vendor-name-nonfree-firmware (depends on the firmware package(s))

(This is really easy in APKBUILDs and can be done with very few lines.)

  1. Ask in pmbootstrap init: (Explanation text) (pkgdesc from the nonfree packages, so they can tell which hardware components they enable) Use non-free components (none, firmware, userland, all)? [firmware]:

We could directly skip the question, in case there is no nonfree subpackage for the device, and only show the firmware/userland choice when such a package exists.

  1. Depending on what the user chose, the nonfree packages and its dependencies get built and installed in pmbootstrap install.


  • Make people aware, that there are proprietary components and what they are for
  • People who prefer a full libre stack can disable the firmware files which are enabled by default right now (they could use an external modem via usb etc.)
  • It would allow packaging proprietary userspace drivers with libhybris (#678) in theory, while keeping it cleanly separated for users who do not wish to install them.
  • Once we can run proprietary components in pmOS, it might be a lot easier to reverse engineer them.


Updated 13/10/2017 16:36 2 Comments

Obfuscation failed for code with `super`


Can be reproduced on online demo: ``` class BaseClass { constructor(foo) { this._foo = foo; }

  get foo() {
    return this._foo;


class BarClass extends BaseClass { constructor(foo, bar) { super(foo); this._bar = bar; }

  get bar() {
    return this._bar;


const a = new BarClass(‘foo’, ‘bar’); console.log(; console.log(; ```

Attempt to obfuscate this code snippet with everything off and Mangle Variable Names on returns: Error: Unknown node type Super.

Updated 11/10/2017 19:56 4 Comments

General event types


When coding the throttle and roll user event types I thought about having more general kinds of events, mainly: * action group calling events, * delegate events.

The first kind would allow easy automation of tasks like solar panel deployment.
The second would be a powerful tool for advanced users to allow PEGAS execute arbitrary code via function delegates.

Labelling this as a future enhancement until a more clear vision of those features emerges. Comments?

Updated 08/10/2017 17:20

Exercise Keys for Future KPIs and


What’s Wrong? Users may want to re-define a certain exercise, but we would want to compare apples to apples.

What do I Expect to Happen? We should be able to collect data that folks can use to track their progress over time. We shouldn’t have unlike exercises feeding into one another though.

Suggested Fix In case we have different leagues with the same named exercise, we should someday create a key for these exercises. If “running” allows for commute-walking in one league and not in another league, that would be bad. We want folks to be able to name their exercises “running” but differentiate them with a key. The key could connect exercises across leagues, but if someone wanted to redefine “Running”’s definition, their new exercise would get a new key.

Maybe we would have a “Running” exercise that’s standard. They could create a custom “Running” exercise, but when they are creating a new league they can only use the “Running” exercise we’ve defined or their custom exercise. This prevents them from being inundated with 7 kinds of “Running” exercises that other folks have created, but it also allows us to have a running exercise we can someday treat as a KPI that will compare apples-to-apples across leagues.

Supporting Information

Updated 09/10/2017 02:07

Sound visualisations


It’s been suggested that Phanary include some way of visualizing some kind of waveform to easily see where in a loop specifically you currently are. I can see this as being useful especially with some of the suspenseful or dramatic ambiences and songs to allow DM’s to time and pace their expositions and plot twists. I do think it might be kind of complex and resource-intensive to implement, so I’m tagging it as ‘future’ for now, but any input as to how it might be more easily achieved would be much appreciated!

Updated 04/10/2017 19:04

Cleanup HOGDescriptor


Before this pull request I opened #9224 and no-one could find what these parameters should do in two months. Remove unused parameters since several years <code>derivAperture</code> and <code>histogramNormType</code> (it was not used in “atomic bomb” commit).

Updated 04/10/2017 11:17 6 Comments

react-native-web compatibility


Eventually it would probably be a good idea to test out the library in react-native-web and see if it works. If there are small compatibility issues fix them. And add web as a test target if things work out.

There’s already a library directly for the web and this isn’t a parity feature for rnmd or rnmk so this will be something after 1.0.0 and some of the other bonus features are implemented.

Updated 02/10/2017 02:44

Add features suggestion


Great project I must say, however, let’s add new features. Here are some that should be implemented in future:

  1. Date Lock
  2. OS lock
  3. Browser lock
  4. Character to Ternary Operator
  5. String Splitting
  6. Extended Predicates 6.5. Property Keys reordering
  7. RSA instead of RC4 for string encryption for people who don’t care about performance
  8. RC4 or RSA key length setting
  9. Divide JS 2 different categories. One for node and other for the browser as there are differences on how the code is written.
  10. Function Outlining
Updated 02/10/2017 12:44 8 Comments

Create a LeagueType and MatchType Controller


… to automatically create the views (create, details, …) for the MatchReport.

The admin then can create new LeagueType and select the MatchTypes (that can be created also) he wants to have in the new league type.

So maybe he can create a FooLeague with 7 games, 2 mixed double, two “over 50yrs double”, …

Updated 30/09/2017 21:03

Display manager


Right now we configure the tty0 to automatically log in as user (Hildon, plasma?) or root (weston) on boot (by editing the inittab in the postmarketos-ui-* package’s postinstall script). Then we start the X11 server (for Hildon) or Wayland compositor (kwin for plasma-mobile, Weston) in an /etc/profile.d/ script, that checks if the current tty is the one where the autologin happened.

While that works for installing one device, it is not possible to cleanly switch out user interfaces after the installation (because replacing the entry in the inittab does probably not work when switching), and it is not possible to install multiple user interfaces in parallel and switch dynamically.

Our requirements for a display manager, as I see it: * Needs to be able to execute X11 UIs (desktop environments) as well as Wayland compositors (!) * It should set up the login session (consolekit?) and udev stuff (as @MartijnBraam suggested) * It should be lightweight, so we don’t blow up the image with something that could be done in a few MB (like the osk-sdl) * A config file, that selects the current UI would be nice as first step * Super fancy version with UI (future!) * if only one UI is installed, it should boot that * user can select the UI with touch buttons * display date, time and battery * reboot and power-off buttons

In my opinion, we don’t need to allow multiple users for now, or typing in a password (as we focus on security and have full disk encryption anyway, and getting multi-user secured is hard).

It might be possible to use an existing display manager for that. However, the big difference is, that for our use case we don’t need focus on user selection, but on selecting the session. On the other hand, we have a nice stack with osk-sdl already, with which touchscreens work across all these strange drivers, so it would be relatively easy to re-use that.

Updated 01/10/2017 17:14 3 Comments

Idea: UI for educators to confirm Service end dates



  • Now that we’ve landed #1105, we’re asking educators about future end dates for services. Suggestion from @kevinrobinson: a UI that shows educators Services (for example, services that have been active the longest) and asks them to confirm if they’re still active, discontinued, etc. so that we have some UI prompting to keep the data fresh and accurate
Updated 29/09/2017 18:41

Rewrite Hitoban


Rewrite Hitoban to use C++17, more class, make it using OOP (struct are C style 😂). And use shared_ptr / unique_ptr instead of raw pointers.

Updated 28/09/2017 19:58

PyQt4 doesn't find QtWebKit with *.deb distribution



I’m looking to make a Debian package to facilitate deployment (repo) and I discovered that PyQt4 doesn’t find QtWebKit when trying to use a recent version of Ubuntu (16.10) or Debian. It seems to be a known problem with Debian and Ubuntu.

  File "/opt/venvs/ndrive/local/lib/python2.7/site-packages/nxdrive/wui/", line 11, in <module>
    from PyQt4 import QtCore, QtGui, QtNetwork, QtWebKit


Updated 11/10/2017 09:13 2 Comments

Tech Skills Courses


I propose creating some courses on technical skills. These could even utilize interactive twitch sessions with developers who can run students through exercises in real time and stream the lessons.Other options are a slack / skype conference that allows for multiple people to work together to learn.

Traditional activities for coursework will be added to this class as well.

Updated 13/10/2017 03:05

ELMS:LN Open Ulmus Course


We need an intro course for ELMS:LN to use as a POC for other future Open Ulmus courses. I’m actively developing this currently, and will update with progress on this issue as it progressed. Below will be a checklist for the courses – as well as any links to streams about the building of these courses.

Updated 13/10/2017 03:05 1 Comments

Temp fix for remark


Revert once is resolved.

Updated 21/09/2017 04:26

Apply different ciphers to string literal encryption.


A great update would be support for difference ciphers rather than the weak RC4 algorithm currently implemented. Such ciphers may be aes, camellia, tripledes, seed, idea etc. I’d also suggest caching decrypted strings – possibly in Base64 for faster retrieval – to avoid lagging the program.

At the very least, you really should implement this:

Great job thus far btw. This truly is an amazing tool!

Updated 30/09/2017 08:37 2 Comments

Support Lazy Loading (Dynamic Import)


Just an issue to talk about supporting lazy loading mainly for webpack users 🎉

Seems rollup is not intending on supporting lazy loading after all:

One day! 🎉

Mentioned here:

Updated 20/09/2017 06:32 2 Comments

Improve Performance


I was looking at the benchmark and why is that Hyperapp near last (for the 5 frameworks I selected) for so many operations? Elm is the worst. Wondering when optimizations they have made?

Note: I also dont know how common these operations are versus other operations where Hyperapp shines. I think that’s the issue with these benchmarks :)


<img width=“989” alt=“append-rows” src=“”>


<img width=“950” alt=“remove-row” src=“”>


<img width=“985” alt=“swap-row” src=“”>


<img width=“968” alt=“select-row” src=“”>


<img width=“982” alt=“partial-update” src=“”>

Updated 20/09/2017 08:34 6 Comments

Always name fields after API request/response fields; remove all cases of weird abbreviation


Especially near the beginning of this library’s life we engaged in a lot of weird abbreviate. For example a source is a Src and a transaction is a Tx. This most often obfuscates things unnecessarily when users are comparing our API reference to what they see in the library.

We haven’t been abbreviated newly added fields for some time, so at some point let’s go back and and strip out all the pre-existing ones.

Updated 19/09/2017 16:53 2 Comments

Cleanly separate all returned resources from parameters


Some while back we accidentally introduced a pattern where a single struct would be shared as parameters for API requests and also appear in response structures.

This usually works okay, but it can cause problems where there is some asymmetry between request and response. In the worst cases this asymmetry will necessitate a custom AppendTo implementation for no good reason. Take IdentityDocument for example:

// IdentityDocument is the structure for an identity document.
type IdentityDocument struct {
    Created int64  `json:"created" form:"-"`
    ID      string `json:"id" form:"-"` // See custom AppendTo implementation
    Size    int64  `json:"size" form:"-"`

func (d *IdentityDocument) AppendTo(body *form.Values, keyParts []string) {
    body.Add(form.FormatKey(keyParts), d.ID)

Most of these shared structs are in account.go, but you can find all of them with a search like ag -C 5 'json:.form:.'.

Given that it’s the convention used in 95% of the rest of the library already anyway, we should just copy all these shared structs, suffix the new one with Params, and then remove any fields in either that aren’t relevant to the request or response. This is a breaking change, so let’s try to do them all at once.

Updated 17/09/2017 13:56

App State


Curious if there are any plans to implement some form of state in stripe-mock. I tried to switch over to using stripe-mock in our test suite, but had to revert once I realized creating tokens then retrieving them would return different data, i.e. card tokens turned into bank accounts when fetched.

I’d be happy to contribute, but I figure keeping state probably requires some discussion about the design and whether or not it aligns with your roadmap.

Updated 12/10/2017 22:21 2 Comments

Crate Tagging Historian


While the mongo tagging historian can service a crate tagging service, it should not be necessary. This will allow us to create a tagging historian for an existing crate database.

Updated 13/09/2017 23:06

Deploy multiple projects


A feature that the maven plugin has, is to be able to deploy multiple projects at once.

Our circuit breaker booster is a good example.

not sure how it would work, but it would be nice

Updated 13/09/2017 20:09

Typescript object causes error when calling charges.create()


Node version: 8.4.0 Stripe version: 4.23.1

I created a model representing a Stripe Charge object in typescript so we can ensure type safety when building requests. However, when using that Stripe Charge object when calling charges.create, I get the error:

TypeError: Path must be a string. Received undefined

Relevant Code:

// This causes the mentioned error
const paymentRequest = new StripeCharge();
paymentRequest.amount = req.amount;
paymentRequest.currency = req.currency;
paymentRequest.customer = req.customerId;
paymentRequest.source = req.token;

const response: StripeCharge = await this.stripe_client.charges.create(paymentRequest);
// This does not cause the error
const test: any = {
    amount: req.amount,
    currency: req.currency,
    customer: req.customerId,
    source: req.token

const response: StripeCharge = await this.stripe_client.charges.create(test);

By debugging the stripe library I was able to narrow it down to isPlainObject(args[0]) returning false in getDataFromArgs inside util.js

getDataFromArgs: function(args) {
  if (args.length > 0) {
    if (isPlainObject(args[0]) && !utils.isOptionsHash(args[0])) {
      return args.shift();
  return {};

I am able to work around this temporarily by importing toPlainObject from lodash, but we have no other use for lodash so I’d much rather not include the unneeded dependency.

Updated 25/09/2017 23:13 4 Comments

Feature Request: Toogle command assigned to keyboard shortcut


Hi (love the app on Linux gnome ubuntu 17.04) I can’t seem to add a command to toggle Tusk window when it is opened to a keyboard shortcut. Similar to how I have CopyQ a clipboard manager command ‘copyq toggle’ assigned to a keyboard shortcut. Is there a command which works like clicking Systray icon to toggle the window? I couldn’t find anything in the readme section. Thanks

Updated 21/09/2017 11:10

Schedule Component


Feature Request

A Material Scheduler, calendar with events would be a nice addition to the library.

What is the expected behavior?

Date Range event spanning Calendar Navigation Something like Primeng but with Material.

Updated 11/10/2017 18:25

Coast phases


UPFG in the current implementation assumes that all the burns are executed immediately one after another, but in reality some time is needed between stages to let them separate and ignite. This time is user-input and thus predictable. Lack of handling those small coasts isn’t usually a big issue (any error that it generates is eventually accounted for in the next stage), but in some cases this might cause weird behavior. Example of such case is the Saturn V, where the final stage (S-IVb) ignites only for a few seconds to circularize, but before it does, the vehicle is in ballistic flight for a while. Since PEGAS isn’t aware of that coast, it doesn’t prepare for it, so there are several seconds during which the vehicle is coasting but all the calculations were done assuming it would be thrusting instead. This is the potential cause to the resulting problem*: after the stage ignites, attitude rapidly changes and the vehicle makes two sharp pitches.

In the original implementation, UPFG is able to handle at least one coast stage (period after jettisoning the Shuttle’s External Tank, before the OMS burn). The thrust integrals potentially could handle more than that, and implementing it would make PEGAS more robust to coasts, potentially even allowing longer coast periods.

Marking this as future enhancement, since 1) it is not a dire issue, 2) it might be difficult to implement, requiring alpha-test in a prototype environment.

* - this problem has at least one alternative cause: the convergence check could be still imperfect and allow PEGAS to take steering even though the UPFG isn’t actually ready yet.

Updated 12/09/2017 10:33

future: alpha shiplist

  • [x] fix i18n #311
  • [x] put i18n into a class @ladjs/i18n #310
  • [ ] convert jobs/email and jobs/locales into @ladjs/email and @ladjs/locales #309
  • [x] convert _404-handler into a component #308
  • [ ] move dynamic helpers into 3 packages (@ladjs/state-ctx, @ladjs/state-flash, @ladjs/state-locals) #307
  • [ ] move policies into a class @ladjs/policies #306
  • [ ] write a basic test for API and web to hit all public endpoints to get 200 code w/supertest/superagent or something #305
  • [ ] koa-better-views integration #304
  • [ ] move context-helpers into a class #313
  • [x] fix env overriding
    • [x]
    • [x]
    • [x]
  • [x] fix koa-response-time header not getting added to request in web.js and api.js
  • [x] koa-better-error-handler (need to use this.state.meta.title and this.state.meta.description

    koa-better-error-handler set this.state.title to Not Found +0ms koa-better-error-handler set this.state.desc to Not Found +0ms

Updated 18/09/2017 03:52

zohlednit rušení rozhodnutí NS Ústavním soudem


Víme o tom, na webu na to i upozorňujeme, přesto by to ale chtělo urychleně řešit.

Nemálo nemeritorních rozhodnutí NS je následně zrušeno Ústavákem, což zkresluje statistiky advokáta (má pak například minus bod u NS, plus bod u ÚS a následně plus bod u NS).

Je tedy třeba: - z webu NS vycrawlovat další metadata, a to ta, která se nachází v tabulce obsažené v hlavičce (někdy tam je, někdy není) - viz například 30 Cdo 3463/2014 - pokud je v tabulce “výsledek” roven “zrušeno” (měly by tam být 3 stavy - odmítnuto, zamítnuto, zrušeno - ale až bude vycrawlováno, chtělo by to samozřejmě ověřit), zrušit přiřazení daného rozhodnutí advokátu, ale ponechat jej v databázi (- původně jsem navrhovala zobrazování hlášky na stránce případu o tom, že toto rozhodnutí jsme nepřiřadili, protože bylo zrušeno, to by asi bylo také dobré, ale v prvotní fázi to nechme stranou)

Updated 12/10/2017 20:04 2 Comments

Support for error tokens, webhooks and mocked checkout


I work as ruby programmer and I really like the concept of this project. It is great that an official solution like this has emerged. Also, it is fantastic, that one line - changing Stripe.api_base - is enough to run test suite in ruby with the mock server.

However, there are some aspects that refrain me from using it for tests.

The first one is lack of support for error tokens. Maybe I didn’t find the proper configuration? Without proper error responses it is difficult to think of using the library in real project for tests, while with VCR it is possible to record Stripe responses in a variety of error cases simply catching tokens with debugger.

The second thing are client-side scripts. By default they call the real Stripe servers. Having a test version of js that called local mock instance would make it possible to write functional tests with Capybara, testing standard payment integration in end-to-end manner. Similar solution is implemented by fake_stripe.

Lastly, support for webhooks would make it possible to test more complicated integrations in functional tests. It would be possible to e.g. test a whole subscription flow with first payment if the mock server sent back a request with event to the app server. I know this would either be limited to only subset of cases or bring relatively big complexity to the library. Still, it seems to me that having only the subset of the cases like creating subscription or charge would make the library more usable. Another option would be to manually trigger webhooks in a similar manner as it is done in stripe-ruby-mock.

I am looking forward for this project to grow as, imho, there is no best option for mocking Stripe integration at the moment. Keep up the good work! :)

Updated 03/10/2017 17:37 2 Comments

feature request: Wiki Style Links


If evernote is missing one thing, it’s wiki style links: I would love to just open a search box and then search for the note I want to link (or make a new note if the link doesn’t exist) Then I want to just click through that link (and have a back button to return to where I came from). This would be amazing.

Updated 10/09/2017 19:57

V4L2: 1280p@30fps not available by default any more


after the merge of #9479, 1280p@30fps is no longer available by default. One is now required to do: py cap.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*"MJPG")) to be able to select this mode.

Used camera Logitech C615. v4l2-ctl --list-formats-ext:

    Index       : 0
    Type        : Video Capture
    Pixel Format: 'YUYV'
    Name        : YUYV 4:2:2
        Size: Discrete 1280x720
            Interval: Discrete 0.100s (10.000 fps)
            Interval: Discrete 0.133s (7.500 fps)
            Interval: Discrete 0.200s (5.000 fps)
    Index       : 1
    Type        : Video Capture
    Pixel Format: 'MJPG' (compressed)
    Name        : Motion-JPEG
        Size: Discrete 1280x720
            Interval: Discrete 0.033s (30.000 fps)
            Interval: Discrete 0.042s (24.000 fps)
            Interval: Discrete 0.050s (20.000 fps)
            Interval: Discrete 0.067s (15.000 fps)
            Interval: Discrete 0.100s (10.000 fps)
            Interval: Discrete 0.133s (7.500 fps)
            Interval: Discrete 0.200s (5.000 fps)
Updated 05/09/2017 15:07 2 Comments

feature: datetime picker (basic functionalities)


What is the motivation ?

Considering that the material team will most likely start the develop of this feature in the next year, I think it would be a great functionality to add into covalent (in a basic functional form).

There are definitely a lot of developers out there who would love to see covalent offering this.

Expected behavior

Create a new component that adds to the current material datepicker the capability to manipulate time instance variables of a DateTime class (moment, Date or others depending on the supplied DateAdapter implementation).

Other information

The issue tracker for material datetime picker can be found here

Examples of implementations of this can be found in primeng what I would personally suggest as starting point and ngx-ui (personally dislike the overlay part of it). Both of this implementations dont follow the material spec but hey, its at least something! 😄

Updated 11/10/2017 18:16 5 Comments

Fork me on GitHub