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

Strange behavior with `prunner` in sourced test script.


From: In testing prunner, I have ended up with the following scenario:

  • prunner -c “gzip -fk” *.out - works as expected in test.bash and interactively.
  • find . -maxdepth 1 -name “*.out” | prunner -c echo -t 6 - works interactively, but not in test.bash.

Testing was performed on Ubuntu 16.04 with Bash 4.3 and on Mac OS X with Bash 4.4.

What appears to be happening with the latter in test.bash is that getopts is refusing to process -c, and thus prunner will try to directly execute the argument without the prefix command it was given. The strange part is that I am able to observe it accepting the -t option, so getopts is at least partially working. Bash debugging with set -x has not been able to shed any light on why this is happening for me.

Can anyone please help me to determine why -c options are not working correctly for prunner in test.bash when lines are piped to stdin?

Updated 22/03/2018 23:39

Switch to headless Chrome as the driver for feature specs


Our current driver for browser tests is Poltergeist, which uses the now-unmaintained PhantomJS to run headless browser tests.

We can likely replace this with headless Chrome, which has been available for some time now. There will likely be some trial-and-error getting the configuration worked out properly; in preliminary exploration, I was able to get some tests to pass but ran into trouble related to URL blacklists.

Updated 22/03/2018 23:34

Find a better way to classify logs


I want a new log system that auto adds to the JSON whether it’s a database error, scan error, connection error, etc. Maybe simply by declaring a “category” in each file and automatically adding it when logging using Object.Merge()?

Updated 22/03/2018 23:30

New elements and repos ideas


This is a list of ideas of elements that could be implemented in this org: - <integrated-terminal> - A terminal that could be fully customized greatly styled (Huge Project) - <screen-capture> - A custom element that allows the user to select parts of the screen to capture

Other Repos that might be implemented: - @electron-elements/utils repo - A utility modules that could provide basic utilities like TemplateManager and AttributeManger in send-feedback and much more

Updated 22/03/2018 23:25 1 Comments

Armor defense stat range


Is there any interest in including the range in the defense stat that the armor pieces have? Like the non-upgraded base defense value and the fully upgraded end value of them?

I could really use the info to show the result of the mixed set on my set builder/sharer.

Updated 22/03/2018 23:35 1 Comments

FederationResolver QML Component


Class to resolve federations, maybe nice to think about what would be more easy to use.

I suggest a component with a property that when it is set with a federation address, a signal and another property is filled with the response.

A variable with the current status could be useful, to distinguish between error, pending, ready.

Updated 22/03/2018 23:10

General discussion about the new QML API


The plan is to build an optional higher level API to interact with horizon from QML. Using a new .pri subproject, the plan is to add a defined “QML_STELLAR_API” to conditional compile Q_DECLARE_METATYPE/qmlRegisterType/interface classes etc.. as required.

I realized that the below classes could be nice to have in any Stellar application and would be great to implement a QML declarative version of them: - StellarConfiguration, singleton to setup common settings (use public or test network, horizon server, etc.. ) and to access Stellar current variables. API version number, minimal fee, base reserve, etc… - FederationResolver, class to resolve federations - QQuickImageProvider to resolve assets icons. URI could be something like stellar://icon?issuer=issuer_public_key&code=asset_code - Repeater/Loader like requests classes. For each request class, a declarative version to request data to horizon.

You can feel free to comment them or suggest some more ideas.

Updated 22/03/2018 23:00

Improve TOML processing


I tried to use an existing library, but it’s not working under android and may have problems in other platforms. As there are not many options to process, it would be great to integrate other permissive licence libraries or implement one or a function to deal with these type of files.

Actually it is used on federation answers, but it will be required to deal with assets icons and new features coming to stellar.

Updated 22/03/2018 22:53

Create a TCP reordering module


In some cases, TCP packets get scrambled on the network and don’t arrive to the other side in the same order that they were sent. The purpose of this module is to re-order the in the same way they were sent, meaning according to the TCP stream order. This sounds a little like TCP reassembly, but the difference is that this module should only re-order packets, nothing more. It should be simpler and more lightweight than TCP reassembly

Updated 22/03/2018 22:50

Filter: prepend


A preprend filter would add a given number of a character to a sprite. This would be useful to shift sprites without having to change the original sprite. The arguments would be the character you want to prepend and the number of times you want it to be prepended. As an example: "Filter": ["0", 64] would add the character 0 64 times before the start of a sprite.

Updated 22/03/2018 22:40

Seems to be draining battery and cellular data


We have built an Ionic app for both Android and iOS and most of our testing is done on iOS, it seems as though the geolocation settings are killing battery life and cellular data is sky high.

Here are the configuration settings that we are using:

          desiredAccuracy: 0,
          stationaryRadius: 25,
          distanceFilter: 0,
          locationUpdateInterval: 10000,
          fastestLocationUpdateInterval: 10000,
          stopOnTerminate: false,
          startOnBoot: true,
          preventSuspend: false,

All we want to do in the app is simply track users incase they pass by a certain location (a retail store), then it will ping them with a push notification, so it doesn’t need to track them to heavily. Thanks.

Updated 22/03/2018 23:04 1 Comments

Create a new endpoint in ServerConfigurationController which will handle a form for submitting both baseUrl and port


We need to create a new endpoint in the ServerConfigurationController which will be responsible for validating a form which contains both the baseUrl and port. This endpoint will be used in the front-end to “bulk” update the url and port. Please make sure you follow the How-To-Write-Controllers guide.

Validation rules: * baseUrl should not be blank or null * port should be an integer between 1 and 65535

The endpoint will receive the following json:

    "baseUrl": "",
    "port": 1234
Updated 22/03/2018 21:55

The future of sonar-golang


SonarSource is going to provide its own open source sonarqube plugin for the go language called SonarGo next month.

This is a good news in our opinion since now go will be a language officially supported by SonarQube.

The approach taken by SonarGo seems different from our plugin: instead of retrieving the issues/data from external tools as we do, they are analyzing with their own rules the files.

The advantage is to be able to support the same set of rules across all languages.

The disadvantage may be to loose some rules you are used to with current go linters.

In the long term, it also means a tighter integration with SonarQube.

As such, the future of this plugin is not clear once SonarSource plugin will be available.

If there is no need for this plugin anymore, we are happy to move to SonarGo.

Some people may however prefer to rely on current linters, thus continue to use this plugin. In that case we would be happy to continue maintaining it.

Please feel free to comment in this issue to tell us if you think there is still a need for this plugin.

Updated 22/03/2018 21:47

REPL appears not to start


A command invocation such as bloop console kaleidoscope/core appears to work right up until it starts the REPL. The last lines I see are (including a blank one), ``` Starting scala interpreter…

The invocation does not appear to terminate, and the cursor is left "hanging".

There is no output at all on the server until I hit `Ctrl+C`, then:

Client in <path> has disconnected with a ‘SocketError’ event. Cancelling tasks… ```

This is the beaviour for the first five invocations of bloop console. On the sixth attempt, no output is shown on the frontend at all.

Bloop-frontend: 1.0.0-M7 Zinc: 1.1.0+9-57ca9e0f Scala: 2.12.4

Updated 22/03/2018 22:48 1 Comments

Misleading exception report in CLI mode


When running fx-sig-verify locally, if the C parts of M2Crypto could not be installed, the program exits with a SigVerifyBadSignature. That raises the blood pressure, and doesn’t really provide any hints as to issue. (Fortunately, running print-pe-certs program gives a much more informative message: Catching ('%s requires M2Crypto, which is not available', ...)

If it can’t be fixed, at least document the possibility to reduce chance of heart attacks on the job.

Updated 22/03/2018 21:05

Histogram plot type


It’s tedious to have to bucket & count data yourself, when jp could just do it for you. We should have a histogram plot type with some default number of buckets (or strategy for auto-choosing one) and a way to specify other (numbers of) buckets.

Updated 22/03/2018 20:48

100 continue documentation is wrong


I’m pretty certain that ATS support the Expect header and 100 continue. I’ve tested this in the past, and I just looked through the code and there is support in there.

Someone internally claimed it wasn’t supported because of a FAQ note in the 5.3 documentation. In fact it is in the current documentation

It seems like we should update the docs or remove some code.

Updated 22/03/2018 20:18

Data Table does not render row data entered through this.props


The datatable component will not render the data when it is passed through props, but is rendering when I pass the data directly from a json from another file. Please note the data passed directly through with a json file is passed through at the very start upon rendering of the whole page, while the data passed through props is through user generated input, so happens during app usage.

Also not no error pops up in console.

After digging through source code, we think it may have something to do with this code block below:

It looks like the table will only render newly imported rows when you are sorting the header keys or sorting the direction. However in our case, we are injecting new rows into the rows prop of <DataTable>, see code below:

        rows={this.props.investors} //why wont this render???
        filterRows={(rows) => {
          console.log('rows adam', rows)
          return =>
          // selectedRows,
        }) => {
          console.log('inside2', rows, headers)

          return (
            <TableContainer title='DataTable with batch actions'>
                {/* make sure to apply getBatchActionProps so that the bar renders */}
                <TableBatchActions {...getBatchActionProps()}>
                  {/* inside of you batch actinos, you can include selectedRows */}
                  <TableBatchAction >
                    Remove Investor
                  <TableBatchAction >
                    Modify Restriction Dates
                <TableToolbarSearch onChange={onInputChange} />
                  // onClick={action('TableToolbarAction - Download')}
                  // onClick={action('TableToolbarAction - Edit')}
                  // onClick={action('TableToolbarAction - Settings')}
                    // modalProps={{ onBlur: { onBlur() }, onClick: { onClick() }, onFocus: { onFocus() }, …}}
                    buttonTriggerText='Add New'
                    // modalLabel='Add New Investor'
                    modalHeading='Add New Investor'
                    <p className='bx--modal-content__text'>
                      Please enter the information below to add a single investor.
                    <br />
                    <InvestorForm onSubmit={this.handleInvestorSubmit} />
                    <TableSelectAll {...getSelectionProps()} />
                    { => (
                      <TableHeader {...getHeaderProps({ header })}>
                  { => (
                    <TableRow key={}>
                      <TableSelectRow {...getSelectionProps({ row })} />
                      { => (
                        <TableCell key={}>{cell.value}</TableCell>

So we think it may be that an injection of new items does not trigger updating of the internal state of the component.

Please also note I am using the pagination component to get the length of the this.props.investors, and it is correctly reading it and rendering it, so I am definitely not passing through an empty rows prop. It really seems like something is happening through the internal sorting/filtering of the rows prop, because this.props.investor has real data and then when it renders inside datatable it is just picking up a blank array.

Any help or direction is greatly appreciated!

Is this a feature request (new component, new icon), a bug, or a general issue?

appears to be a bug

What browser are you working in?


What version of the Carbon Design System are you using?


Updated 22/03/2018 20:32 1 Comments

Dimension of lorenz system is inaccurate


So the Capacity dimension of the lorenz system is very close to 2.0.

Yet, check this out: ```julia using ChaosTools using PyPlot

lor = Systems.lorenz() tr_lor = trajectory(lor, 10000.0; dt = 0.05);

sizes = logspace(-1, 2, 10) es = estimate_boxsizes(tr_lor) ds = genentropy.(0, sizes, tr_lor)

figure(figsize=(6,4)) plot_linear_regions(-log.(sizes), ds) D = linear_region(-log.(sizes), ds)[2] 1.8 ``` Try whatever combination of parameters you want, different boxsizes, different initial conditions.

The result is just “wrong”, assuming the literature is correct.

The problem is that the linear_regions function is behaving perfectly fine… :( And I cannot possibly believe that the entropy function is wrong, since it gives correct results in other cases….

I really can’t understand where the problem is, pliz halp!

Updated 22/03/2018 19:42

View initialization when class member


I’m experiencing a similar issue to #1254, but I can’t seem to fix the issue with the advice given.

I have a class that has three Kokkos::Views as class members (no nested Views): ``` typedef Kokkos::HostSpace MemorySpace; typedef Kokkos::OpenMP AccExecSpace; typedef Kokkos::OpenMP ExecutionSpace;

Kokkos::View<Kokkos::vector<double>, MemorySpace> pdf_; Kokkos::View<lbm::NodeStatus, MemorySpace> node_status; Kokkos::View<Kokkos::vector<double>***, MemorySpace> node_solid_and_velocity; ``` I’m compiling for OpenMP, but get the following runtime error:

terminate called after throwing an instance of ‘std::runtime_error’ what(): Constructing View and initializing data with uninitialized execution space Traceback functionality not available

In the class constructor, I’m using the default constructor for the View members in the initializer list: Lattice(...) : ... pdf_(), node_status_(), node_solid_and_velocity_(), ... Then I resize and initialize them in a separate function: ``` Kokkos::resize(pdf, nodes_x, nodes_y, nodes_z); Kokkos::resize(node_status, nodes_x, nodes_y, nodes_z); Kokkos::resize(node_solid_and_velocity, nodes_x, nodes_y, nodes_z);

// Set initial node status to active fluid Kokkos::parallel_for(“init_node_status”, Kokkos::RangePolicy<AccExecSpace>(0, nodes_x * nodes_y * nodes_z), KOKKOS_LAMBDA(const lbm::LatUint& index) { const lbm::LatUint z_index = index / (nodes_x * nodes_y); const lbm::LatUint y_index = (index % (nodes_x * nodes_y)) / nodes_x; const lbm::LatUint x_index = (index % (nodes_x * nodes_y)) % nodes_x; node_status(x_index, y_index, z_index) = lbm::NodeStatus::ACTIVE_FLUID; });

// Allocate memory for discrete velocities in pdf Kokkos::parallel_for(“init_pdfs”, Kokkos::RangePolicy<ExecutionSpace>(0, nodes_x * nodes_y * nodes_z), KOKKOS_LAMBDA(const lbm::LatUint& index) { const lbm::LatUint z_index = index / (nodes_x * nodes_y); const lbm::LatUint y_index = (index % (nodes_x * nodes_y)) / nodes_x; const lbm::LatUint x_index = (index % (nodes_x * nodes_y)) % nodes_x; pdf_(x_index, y_index, z_index) = Kokkos::vector<double>(27, 0.0); }); ```

Based on my understanding, this should work–Kokkos::OpenMP execution space should have access to Kokkos::HostSpace memory space. Am I missing something obvious about why the execution space is uninitialized? Sorry to bother you with simple implementation questions…

Updated 22/03/2018 23:39 6 Comments

Standarize the format of vulnerabilities on core and npm databases


There are few discrepancies in core and npm databases: - -

For example dates are inconsistent across different files. We should document the format and add CI to validate PRs. @vdeturckheim I think you did some work on this with Joi validators.

Updated 22/03/2018 19:33 1 Comments

Sum of transactions does NOT include internal transactions


This is a duplicate from the old repo Issue#47

Problem: Fetching the Data through etherscan API get txlist only lists external transactions. Internal transactions seem to be ignored.

Tried to find a solution for this for weeks now. I have visited the Internet all over only to see that seemingly everybody (except etherscan) is having this problem. The question is maybe if the etherscan API also allows other requests, like list internal transactions that we could just add to the board (I should add that we use botth the usual API as well as the websockets API which is even more limited)

Possible Solutions: 1 - We find some undocumented(?) etherscan API secret 2 - We make our own scraper (maybe based on @maxskant take on the leaderboard backend or @sponnet’s Token Scraper) and listen for the donation event emitted from the multisig for example 3 - We wait until a better API comes along, maybe from an Open Block Explorer

Updated 22/03/2018 19:16

signature_algorithms_cert backwards compatibility


update existing scripts to advertise full range of signature_algorithms_cert (as far as RSA is concerned) in all test cases that verify support for signature_algorithms.

As server should not send certificates that don’t match our advertised sig algs, and we’re trying to coerce the server to use specific signature algorithm on the SKE (or CV in TLS1.3) message, we don’t want to confuse it with cert selection (where we will accept anything it sends, as long it’s RSA).

Updated 22/03/2018 19:48

[Preview] stop showing immutable `entries` field


The debugger has a really nice feature for detecting immutable objects and showing its JS representation. Unfortunately, we show a top level entries field when we should just show the fields


  1. go to
  2. click “stuff”
  3. hover on map and list


diff --git a/src/components/Editor/Preview/Popup.js b/src/components/Editor/Preview/Popup.js
index 5abd115..3c59a26 100644
--- a/src/components/Editor/Preview/Popup.js
+++ b/src/components/Editor/Preview/Popup.js
@@ -171,6 +171,8 @@ export class Popup extends Component<Props> {

+    // this is where we show entries... not sure how to prevent it...
+    // maybe we need to avoid using `createNode` and instead create two nodes...
     const roots = [
       createNode(null, "entries", "entries", { value: immutable.entries })


Screen Shot 2018-03-22 at 3.09.27 PM.png Screen Shot 2018-03-22 at 3.09.22 PM.png

Updated 22/03/2018 19:12

TLS 1.3 version negotiation


Verify that if the Client Hello versions extension doesn’t include TLS 1.3 version (but does include, say, TLS 1.4, or past draft versions, with all the necessary extensions) TLS 1.3 is not negotiated.

verify that supported_versions were not sent in ServerHello when negotiating TLS 1.2

check if supported_versions can be used to negotiate TLS 1.1 even if TLS 1.2 is enabled on the server.

Updated 22/03/2018 19:46

concept drawings


@phase4space/careful-cots Who would be willing to draw or find someone to draw a set of concept drawings for a high-level proposal review? So far, we have two votes for “6U with ion drive” from team participants as a preferred format.

Updated 22/03/2018 19:01

TLS1.3 integrity tests


Verify that server performs integrity checks on all encrypted messages (initial handshake messages and actual application data).

Check if modifying any bit of the AEAD tag causes the message to be rejected.

Check if server is not “lenient” – will not accept messages that use slightly off sequence number (few values off in either direction) or encrypted with wrong keys (server’s vs client’s).

Updated 22/03/2018 18:52

Cleanup sfinae machinery


Our sfinae machinery is old. It uses the old method of doing stuff. We should find a way to cleanup the mess that it is right now and implement the traits in a way that is both compatible with visual studio 2015 and readable for mortals.

Here’s the current plan:

  1. Find a way to expand an index sequence without using private member function
  2. Have clear semantics on what is checked and what is assumed
  3. Fix bug we deal today (lack of check with supplied services, visual studio’s intellisense)
  4. Remove duplicated highly duplicated code in sfinae checks.
Updated 22/03/2018 18:37

Add support for ECDSA certificates


Add support for ECDSA certificates, both on client and server side.

Since behaviour of the signature algorithms extension is different for TLS 1.2 and TLS 1.3 – in TLS 1.3 the curve is bound to a hash algorithm – it’s possible for the implementation to be lazy and use the same negotiation mechanism for both, possibly causing issues in either of those protocols.

resolves part of #55, depends on

Updated 22/03/2018 18:36

Design a logo


I don’t have a logo for use on website, Facebook, Twitter, and newsletter.

I’m picturing a large focus on OBSERVED with a small ‘city’ up the side.

I’ll have a go at this but I’m no designer.

Comment here or email: with the issue in the subject line.

Updated 22/03/2018 20:59

Find a way to remove the Safari reader mode hack


The <nav> element is required to have Safari reader mode render the page correctly. But semantically this shouldn’t be needed.

This only affects the first <h2> element (About header) after the title / nav.

I found if there is another element after the <h1> (like a <h2> or <p>) it also resolves the issue.

Not sure if I’m doing something semantically incorrect or the Safari reader implementation is buggy.

Tested on Safari 11.0.3 (MacOS) and Safari 11 (iOS 11.2).

Updated 22/03/2018 18:24 1 Comments

Can you help me get more subscribers?


I need to build subscribers and get the word out. I’ve launched the first newsletter and online support areas – so what are the next steps to maximise reach?

This is an email for Birmingham UK so I appreciate this is a geographically limited project but maybe you could Subscribe if interested in the project, or tweet out the Subscribe link to anyone you know in Birmingham or the Midlands area of the UK?

Thank you!

Updated 22/03/2018 18:45

Can I get some feedback on my launch issue, please?


I picked a great week to launch a data privacy email (Cambridge Analytica undercover investigation)! It was going to just be a draft but I decided to go for it in a basic form that I can refine in the next issue.

What do you think? General feedback/advice would be great – reply here or email ( The web version is online at [Observed City Tinyletter] (

There’s also a new website with sign up link and intro/about: [Observed.City] (

And social media areas for discussion: Twitter account [@observedcity] ( and a [Facebook closed group] (

Updated 22/03/2018 18:37

Possible readme update


As a first time Pheonix user I followed the Step 1 in the readme and created the app, but when checking to see if I had done it properly I received the following error message numerous times, filling my terminal. [error] Postgrex.Protocol (#PID<0.363.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name): database "chat_dev" does not exist I understand that the site works correctly as shown in the demo and that this error occurred as I had not created my database yet. However, I wasn’t expecting to see a terminal full of errors and I thought it may put some newer users off.

Perhaps you could consider moving Step 6 Create/Configure Database into Step 1 Or noting that when starting the server there will be errors and the reason for them at this time.

Updated 22/03/2018 18:10 1 Comments

Allow builds to tag event



In a pipeline events page GIT SHA and commit message is what uniquely identifies an event. It would be helpful for users to quickly identify an event using a key->value pair which a build can set to identify a specific event rather than going through the SHA and commit message.

For example a build in a release pipeline can set value RELEASE_VERSION=1.5.1 to identify an event which built Version 1.5.1 and use that to quickly identify the event. In a roll back scenario this helps users to quickly identify the event from which they need to invoke roll back.


  1. Expose an sd command to tag an event.
  2. UI should display the tag for the event.
Updated 22/03/2018 18:04

Document Parcel v1.7.0 features


Parcel 1.7.0 is coming out soon, and it’s gonna be a really huge update with tons of very exciting features and changes! 🚀 😃

The only problem is… none of it is documented yet 😟

I think we should all work together on getting the major changes documented in depth: - [ ] New Resolver (parcel-bundler/parcel#850) - Explain tilde/absolute paths, aliases, etc. - [ ] Auto Install Feature (parcel-bundler/parcel#805) - There’s not much to explain, but it’s still new concept so we’re gonna need to document what it and why it’s useful - [ ] New File Naming Strategy (parcel-bundler/parcel#1025) - We should add something to the advanced documentation with that table that @jamiebuilds made in the PR comments

Lmk if I missed anything else that needs to be documented.

Updated 22/03/2018 18:54 1 Comments

Fork me on GitHub