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

Making the "Stage" field mandatory


We’ve been asked by a number of recording schemes to make the “Stage” field (occAttr:106 in the iRecord General survey) mandatory, in order to reduce the number of records that arrive without a stage - for insect groups in particular this makes phenology analysis and verification checking much more difficult than it needs to be.

David and I have agreed that we should make this change. As far as I’m aware, to do this I will have to mark the Stage attribute as “Required” within the iRecord General survey settings. I intend to make equivalent changes to other relevant surveys, such as iRecord Moths.

@BirenRathod @johnvanbreda @JimBacon is there any reason why I shouldn’t go ahead and make this change? Is it right to make the change at the Survey level, rather than at the Global level?

@kazlauskis does this have any implications for the apps? (I’m assuming it doesn’t since they feed in to different surveys?)


Updated 27/03/2017 12:33 1 Comments

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

NginX custom default access log issues


Hi Folks

I’m banging my head on the desk here as I cannot get the Easy Engine default NginX log to format in goaccess. nginx.conf:

        # Log format Settings
        log_format rt_cache '$remote_addr $upstream_response_time $upstream_cache_status [$time_local] '
        '$http_host "$request" $status $body_bytes_sent '
        '"$http_referer" "$http_user_agent"';

Actual output: - - [27/Mar/2017:07:51:28 -0400] "GET /news/page/3/ HTTP/1.1" 301 178 "-" "Mozilla/5.0 (compatible; Googlebot/2.1; +" 0.618 - [27/Mar/2017:07:51:49 -0400] "GET /2012/04/30/leopard-films-ground-breaking-new-documentary-series-starts-today-monday-30th-april-11am-bbc1/ HTTP/1.1" 200 7978 "-" "Mozilla/5.0 (compatible; Googlebot/2.1; +"

None of the standard log readers work and my last attempt was: time-format %H:%M:%S date-format %d/%b/%Y log-format %h - %^ [%d:%t %^] “%r” %s %b “%R” “%u” %T “%^”

Which gives the error:

Token 'ET' doesn't match specifier '%s'
Token 'ET' doesn't match specifier '%s'
Token 'ET' doesn't match specifier '%s'
Token '[27/Mar/2017' doesn't match specifier '%d'
Token 'ET' doesn't match specifier '%s'
Token 'ET' doesn't match specifier '%s'
Token '[27/Mar/2017' doesn't match specifier '%d'
Token 'ET' doesn't match specifier '%s'
Token '[27/Mar/2017' doesn't match specifier '%d'
Token '[27/Mar/2017' doesn't match specifier '%d'

Format Errors - Verify your log/date/time format

Help, pretty please, cherry on top.

Is there an option for you guys to add EasyEngine NginX default logging in to the build?

Updated 27/03/2017 12:14 1 Comments

Possible bug in the DensityScanner


ssIterateOverTerrain is called on :update. In that function is determined how many mapSegments there are and based on it the callback is run. A bug can occur when during that update session the timescale is changed between >120 and <=120. The number of segments have to be determined at a different moment. Probably inside the job.

Updated 27/03/2017 12:15 5 Comments

Providing short description for the plots



I was wondering if you can provide some short description for each graph output. Like: - What the plot is showing - How the values are computed - How can we read the plot

Because, for some of them (e.g. Hadamard, similarity error, ..) I can’t explain what’s on the plolt. Thanks

Updated 27/03/2017 11:56

qc::image_cache_data Performance


With the latest release of qcode-tcl (6.69.2) installed I’m finding that loading up a web page which contains lots of images takes a long time (2mins+). I’ve noticed that the process convert appears to run a lot:


I think the problem is caused by a recent fix to qc::image_cache_data:

Here are the timings for a web page with lots of images with qcode-tcl-6.69.2 installed: image

Timings for the same web page with qcode-tcl-6.68.2 installed: image

Updated 27/03/2017 11:36 1 Comments

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

What does 'enumerable' mean?


‘Lists are enumerable’, What does enumerable mean?

Is the meaning of enumerable that you can refer to a value by numerical reference (by counting from left to right)?

Say I have a list: ["a", "b", "c"], the numerical reference/ index of a would be 0, b = 2, c = 3?

I’ve looked at dictionary definitions as well as this forum question: and can’t tell if it’s more complex than my original assumption.

Updated 27/03/2017 11:16 1 Comments

Make reusable and adaptable with a config file


After the prototype is complete we should incorporate the ability to supply a config file to 1. Allow the developer to supply custom field mappings 2. Enable/disable functionality (facets, type to search, is past event, etc.)

This would allow this to be reusable for other Core Competencies, and perhaps EBI Training and Events…

Updated 27/03/2017 11:02 1 Comments

Editorial interface for training resources


As an administrator I want to be able to create, update and delete training resources in BioExcel So that I can add new training resources and fix existing ones (e.g. moved links, expired courses)

A/C - [ ] I can link training resources to a competency in the competency profile - [ ] We can restrict who the administrators are for this knowledge and competency framework - [ ] It’s possible to deploy this administrative interface outside of the EBI, so that other people could use this. For example if we use Drupal lets make sure we’re not relying on EBI specific modules or functionality. - [ ] We document the process for somebody else to deploy this system

Updated 27/03/2017 10:42 1 Comments

bitwarden: App login screen only shown on newtab/aboutpages ?

  • Did you search for similar issues before submitting this one? Yes

  • Describe the issue you encountered: bitwarden: App login screen only shown on newtab/aboutpages ?

  • Platform (Win7, 8, 10? macOS? Linux distro?): Windows 10 x64

  • Brave Version (revision SHA): Brave 0.14.0 rev 61b6b6b

  • Steps to reproduce:

    1. Enable bitwarden from settings under security
    2. Open in a new tab, click on the bitwarden icon, widget login not shown
    3. Open any about page/newtab, click on the bitwarden icon, widget login is shown
  • Actual result: App login screen is not shown on webpage

  • Expected result: Widget login screen should not be dependent on any page

  • Will the steps above reproduce in a fresh profile? If not what other info can be added? Yes

  • Is this an issue in the currently released version? N/A

  • Can this issue be consistently reproduced? Yes

  • Extra QA steps:

  • Screenshot if needed: bitwardenwidget

  • Any related issues: cc: @kspearrin

Updated 27/03/2017 11:58 1 Comments

Moving configuration to dedicated file / environment


As I’m working on the database update script (#112), I obviously need to access the database, so I require the database credentials. The way it is now, all configs are saved directly in hook.php, i.e. database connection, custom commands paths, admins, command configs, etc. So the user would need to re-enter them in a database update script, instead of being able to reuse them.

I really think (as @lichtscheu has also suggested here) we should move these things to a central config file. This would also make it easier to set up different environments for development or testing.

As for the database schema updates, it would make it very easy to just include the config and run the update script.

Some ideas: - Use phpdotenv to use configs saved in environment variables. This is great because the config can easily be overridden with system environment variables. The downside is that it can’t execute arbitrary PHP code, which might be helpful for some instances, like setting up arrays or configs that require some logic in them.

  • Use a config.php file in the root folder, that sets up all necessary constants (e.g. TG_DB_HOST, TG_DB_USER, etc.) and any extra variables.

  • A very different approach would be to keep it the way it is now, but add a new method to the DB class called updateSchema, and then writing a script that doesn’t use the Telegram object to handle any requests, but just to initialise DB and call that method.

  • Or as a simple fix, have a folder with a bunch of update scripts that contain pure SQL.

I’m very open for other ideas to make this as usable and flexible as possible. Maintaining backwards compatibility might not work that easily, but it should be more or less doable by simply setting the same config values within the Telegram initialisation, if no “new” config method is found.

What are your thoughts?

Updated 27/03/2017 10:52 1 Comments

Basic knowledgebase update mechanism


As an administrator I want to be able to create/update/delete training resources and associate them with competencies using a spreadsheet So that I can make updates without having to ask the technical team to do it

A/C - [ ] A semi-automated way of updating the knowledgebase. We’re ok if this can’t do everything, but at least lets the training team and us make the most common updates in the least painful way

Updated 27/03/2017 10:33 2 Comments

Card details required for email notifications using AWS SES


In order to send email notifications we need to use a service that enables us to do this. dwyl advocates using Amazon Web Services Simple Email Service (“AWS SES”) for reliably sending email messages.

The key reasons we choose AWS SES are because it has good documentation, excellent “deliverability” and no minimum spend (10 cents per 1000 emails sent)! There’s also a generous “Free Tier” of 65k emails per month if you’re new to AWS.

You can learn more about it on our README:

In order to register for an AWS SES account you must provide some card details (regardless of whether you reach a quantity of emails that incurs a charge). @ellemindwave please can you discuss this with the client and provide us with the necessary details?

Updated 27/03/2017 10:13

Service scoring data


The process - The user is shown nine quotes one-by-one from people who have recently been diagnosed with Parkinson’s - For each quote the user picks whether they relate to that quote or not - Each quote represents one service (primary) and up to two other services (secondary) - From the quotes chosen, the relevant services are displayed as recommendations to the user

The rules - Only three services will be recommended to the user - If more than three quotes have been chosen by the user, any quote where a service appears as a secondary AND a primary, the secondary instance will be overwritten by the primary quote - Four services can be shown in the rare occurrence of no service appearing more than once, either as a primary or a secondary - If a user is under 50 then whatever their recommendations, they will also be recommended the early onset page - If a user only chooses one quote then all services (both primary and secondary associated with that quote) are shown to the user as recommendations - If the third service spot is empty after the above criteria are fulfilled then ‘Parkinson’s nurse’ should be shown to fill the gap if it isn’t already one of the recommended services

Updated 27/03/2017 10:12

Changing the calculation of BestReviews


In the previous semester the calculation of the best n reviews worked as follows: assume the number of pinned reviews is x. if n <= x then we calculate is by average of the x reviews. if n > x then we calculate the rating by average the pinned reviews and the best n-x unpinned reviews.

Do you think we should change this calculation and instead of using average to consider the activity and reliability of the user? (i.e to add each user some level tells how much we trust his review based on his activity in our application) @KaplanAlexander @yaelAmitay @ArthurSap @Kolikant

Updated 27/03/2017 09:10

Replace instead of add property



cf push Change value of e.g. host or services in manifest.yml cf push

What occurred

The url is not replaced, but rather there are now two urls for my app now. Same with services.

Is this behavior really intended?

What you expected to occur

I expect that the values are replaced and not added.

CLI Version

cf --version cf version 6.25.0+787326d.2017-02-28

CC API Endpoint Version

API version: 2.75.0

Platform & Shell Details

Ubuntu 16.10

Updated 27/03/2017 12:27 4 Comments

This project is not covered by Drupal’s security advisory policy.


Use case: - There is notification message when accessing link for “ATM Plugin” on “” page

Question: - Should it be fixed to pass “Drupal’s security advisory policy” validation?

Please refer to attachment for more details: <img width=“653” alt=“screen shot 2017-03-27 at 11 27 55” src=“”>

Updated 27/03/2017 08:38

[BusinessEntity] Transform properties


I have an object that use vichmage to save a picture, but to display an image i must use a specific twig function to get the path and display the image and i cant set the properties imageable. Can we have a way to automatic transform a properties to an “*able” vic propertie or define a custom vicTransformer. Ex: ```php

 * @var string
 * @ORM\Column(name="logo", type="string", length=255, nullable=true)
 * @VIC\BusinessProperty("imageable")
 * @Vic\VicTransformer("VichToImage")
private $logo;

 * @Vich\UploadableField(mapping="product_logo", fileNameProperty="logo")
 * @Assert\Image(
 *      minWidth = 300,
 *      minHeight = 92,
 * )
 * @var File
private $logoFile;


Updated 27/03/2017 09:07

Rewrite Article Options


At this moment, we put a lot of options in the Article Options. I suggest to break this up in to

  • Article options – paddings – mobile full width – advertorial – divider

  • Image options – paddings – image above title

  • Social Share Options – fb share – whatsapp share – twitter share – whatsapp share – mail share – google plus share

Not sure to put TITLE/SUBTITLE/CONTENT options in Article options or just create options for each one of them

Updated 27/03/2017 10:29

Trying to compile the UWP library but I get a fault


I am just familiarizing myself with this very promising looking library and was wanting to run it under UWP. In Visual Studio 15 with Brutile (Installed using NuGet) I get the following errors:-

Error The package BruTile with version 0.20.0 could not be found in C:\Users\paul.nuget\packages. Run a NuGet package restore to download the package. Mapsui.Samples.Uwp
Error The package Microsoft.CSharp with version 4.0.1 could not be found in C:\Users\paul.nuget\packages. Run a NuGet package restore to download the package. Mapsui.UI.Uwp

Paul, is there a path adjustment I must make to make the download find the appropriate Module? I had a look around and I could not work out where to make the changes. I would be most grateful for your input here.

Updated 27/03/2017 08:27 2 Comments

What would you like to see?


Use this Issue thread to post requests for various Python frameworks and libraries you would like to see a demo for. Include the topic (Networking, Math, Encryption, GUI) and a link to the documentation in your comment. Once a site admin sees and reviews your request a separate Issue will be created for it and you notified of its (hopeful) construction soon!

Updated 27/03/2017 07:28

Update link to angular2-swing


Hello @gajus,

I was wondering if it is possible to update the link of the angular2-swing repo, since the owner of the current base repo has become unresponsive ( I am willing to take on the issues and missing features in my fork (fork: (discussion: if the link gets updated, publish a new npm package linked to this new fork.


Updated 27/03/2017 12:07 7 Comments

Reg: Dynamic Date Selection in Reports


Dear Support Team,

We are using hosted erpnext.

In Reports, we can select date range like, last month etc.. But these date ranges are not dynamic.

If we generate auto email alert for these reports, it will show old same results.

Kindly let us know, how hosted user can make reports with dynamic date selection filters/option.

Updated 27/03/2017 10:15

No Active Cache Available


I followed all the steps for setting browscap and caching the Full_PHP_BrowscapINI but I’m still getting the error “Uncaught exception ‘BrowscapPHP\Excption’ with message ‘there s no active cache available, please run the update command’ in composer/vendor/browscap/browscap-php/src/Browscap.php. I checked and that file doesn’t have a function for update the BrowscapUpdater.php file does. I ran the command on both but files just to be sure. I got the errors:

.Browscap.php: line 1: ?php: No such file or directory
.Browscap.php: line 2: /bin: is a directory
.Browscap.php: line 3: syntax error near  unexpected token `('
.Browscap.php: line 3: ` * Copyright (c) 1998-2015 Browser Compatibilities Project'

I got the same with BrowscapUpdater.php.

The steps I followed were: 1. Recommended Setup part a. 2. Setting up a Memcached 3. Taking the user agent from the global $_SERVER variable

Can anyone tell me what I might have done wrong?

Updated 27/03/2017 09:59 3 Comments

공간요청 사용자, 요청자 정의 재확인


<현재 상황> 1. 공간요청 업무 1) 요청서 - 요청자 : 로그인한 사번 - 사용자 : 로그인한 사번

2) 접수화면(담당자 요청서 작성시) - 요청자 : 로그인한 사번(권혁상대리님) - 사용자 : 사용부서의 사용자 사번

<문의사항> - 공간요청 업무에서 요청자와 사용자를 분리여부 확인(권혁상대리님) EX) 접수화면에서 담당자(권혁상대리님)이 대리요청서를 작성하는 경우, 요청자, 사용자의 의미 확인

Updated 27/03/2017 06:15

Higher breakpoints for higher resolutions


I’ve been making web pages lately without thinking of high resolution monitors (Quad HD, 4k, Retina, what'not). So after buying a 1440p monitor I notice my web pages also should be optimized for this resolution.

Can predefined breakpoints for higher resolutions be bundled with sass-mq in the future?

$mq-breakpoints: (
    mobile:  320px,
    tablet:  740px,
    desktop: 980px,
    wide:    1300px

I know I can do this myself, but thought this would be a great idea anyway. And by the way: what resolutions would you recommend for these breakpoints and a “hardcode case”?

Cheers. Love sass-mq and your contributions! :dancer:

Updated 27/03/2017 06:50 1 Comments

question: - should we use it?


@keith9820 I saw that you added to this project. I am wondering whether or not we need it given that GitHub added project support:

What are the advantages of using over this? I’m fine either way, I was just wondering. I’ve never used it. :)

Updated 27/03/2017 04:59

info command


Should the info command be used once a file system is open, or should we pass the file system as a parameter?

Any error messages? YES! Specified under the ‘close’ command requirement in the document

Assume: Info should be used once a file system is open. Info will read the information about the open file system.

Updated 27/03/2017 04:50

Binaural and UHJ not working under ReaJS VST (Windows)


I’m trying to get the ATK plugins to work in Max 7 (Windows 10) using ReaJS VST. Most of them seem to work except for binaural and UHJ decoding. I also tested by loading ReaJS VST inside Reaper and the result was the same as in Max: there’s no audio output and the ear images are not refreshed. Is this probably due to a path issue that works differently in ReaJS VST? In Reaper, using the standard JS loading (no ReaJS VST) these plugins work fine.

Updated 27/03/2017 08:40 1 Comments

Name of Auto-Generated 3rd Pivot Table in Many-to-Many Relation


Hi, I just want to ask, what is the pattern of auto-generated 3rd pivot table name? Example: In News Model: ‘tags’ => array( ‘has-many’ => array(‘\Model\Tag’,‘news’), ) $table = ‘tags’; In Tag Model: ‘news’ => array( ‘has-many’ => array(‘\Model\News’,‘tags’), ) $table = ‘news’; Is it the 3rd pivot table will be ‘news_tags’ or ‘tags_news’ or something else?

I am asking about this because I want to use a 3rd parameter (name of the pivot table) in the field configuration and I want to use a prefix in the table.


Updated 27/03/2017 09:44 1 Comments

Some Advice


Howdy @bahmutov,

I’m looking at refactoring semantic-release-gitlab to remove package publishing from semantic-release-gitlab, and move that functionality into a separate package.

The act of publishing an npm package to the npm registry would fall into a job separate from the job that publishes a tag and release notes to GitLab.

A detailed explanation of the goal, and reasoning, is available at the link above.

It looks like ci-publish can handle most of my requirements, such as setting up the CI’s .npmrc configuration file, and calling npm publish.

The one remaining requirement, because of the way semantic-release-gitlab generates a new version based on the commits since the last tag, is the need to write a version number to the package.json.

If publishing to the npm registry happens in a separate job from the job that runs semantic-release-gitlab, the value written to package.json by semantic-release-gitlab won’t be available during the npm publish.

Therefore, I was wondering whether I should bake ci-publish into another tool that also writes the version number to a package.json file, or whether that functionality can, or should, exist in a project like ci-publish?

Updated 27/03/2017 04:17 1 Comments



非登录状态下,点击绘图,会直接进入编辑状态,存盘按钮也是激活的。 可是点击后没任何反应。 此处可考虑: 1.非作者不能打开编辑(权限控制) 2.允许编辑的话,提示另存(提示用户会保存在自己的网站、绘画中,未登陆先提示登陆)

Updated 27/03/2017 03:17 1 Comments

301/302 redirect page


Says we currently have an API redirects user to another page:

Request --> API --> PageA

restc send a request from frontend with fetch instead of jumping page which breaks the process, currently I have to disable restc on prod env.

<img width=“285” alt=“2017-03-27 10 52 25” src=“”>

thanks in advance

Updated 27/03/2017 03:04

Question about custom function parameters



First of all, super cool project! We have a current project that uses DotLiquid. While nice, the fact that they don’t have a proper parser is holding them back. I’m looking into switching to Sciban and was wondering if I could do something like this:

{{ myfunction foo bar }}

In this example, myfunction would be a function that I would define in the C# project (similar to drops and tags in DotLiquid) while foo and bar would be values that I would receive in the C# function. In the C# code I would actually need to see values foo and bar (not treat them as variables). Is something like this allowed or would I have to string-ify them? Like:

{{ myfunction "foo" "bar" }}


Updated 27/03/2017 12:15 2 Comments

Integrate with React?


The front-end has so-far been implemented with javascript and jQuery, however it can also be perfectly integrated with advanced JS frameworks. React, in particular, is designed to be only the View part of an MVC architecture, so it could fit perfectly within the PoP architecture.

It could be used for:

  • Give bigger autonomy to modules. Eg: right now, only blocks are able to fetch data from the webserver, and modules are not independent from the block; using React, modules could communicate with the webserver under a different URL from that of the block, eg: “Recommend” button count
Updated 27/03/2017 02:35

Idee zu System für unsere Mapper


Regeln: 1. Beim Mappen im TS sein auch wenn gemuted (geht darum das ihr euch sehen könnt und absprechen) 2. In dieser Issue ein Kommentar verfassen, wenn man beginnt zu mappen und es wieder beendet 3. Vor dem Mappen diese Issue auf neue Kommentare überprüfen.

Beispiel folgt unten drunter :P

Diese Regeln sind nur für Bugfixes an der Map gedacht, d.h. sollte neuer Content eingefügt werden wird dies durch Mergen der Maps getan und somit steht sich keiner im Weg. Sollte allerdings an der bestehenden Map Änderungen vorgenommen werden sind die Regeln zu befolgen um Fehler zu vermeiden. Ebenfalls immer die Mission syncen in GitHub!

Updated 27/03/2017 11:40 4 Comments

Normalization lost when executing method refresh



Working with a SPA application, For some reason the normalization of fields is lost when executing the refresh method, so the result of the filter is not displayed.

I added a call to the normalization method whenever loadData is executed, and it worked.

My alternate implementation to work around this problem:


  let self = this      
  let obj
  let normalizedFields = []
  this.fields.forEach(function(field, i) {
    if (typeof (field) === 'string') {
      obj = {
        name: field,
        title: self.setTitle(field),
        titleClass: '',
        dataClass: '',
        callback: null,
        visible: true,
    } else {
      obj = {
        title: (field.title === undefined) ? self.setTitle( : field.title,
        sortField: field.sortField,
        titleClass: (field.titleClass === undefined) ? '' : field.titleClass,
        dataClass: (field.dataClass === undefined) ? '' : field.dataClass,
        callback: (field.callback === undefined) ? '' : field.callback,
        visible: (field.visible === undefined) ? true : field.visible,


  this.tableFields = normalizedFields


Is this really necessary, or am I wrong somewhere?

Translated by google ;)

Updated 27/03/2017 05:55 1 Comments

counts not showing in status bar on macOS


<img width=“1920” alt=“screen shot 2017-03-26 at 6 17 21 pm” src=“”>

I cannot get the status bar counts to load. Opening dev console, I see no errors. Opening the same project on Linux works. Unknown how I can debug this. Can I easily turn on linter debugging?


Updated 27/03/2017 01:47 1 Comments

Getting Error "https://localhost:44358/home/error?errorId=08e36a48955f31ea67525c7d3b0a1792"



I have used below Configuration in identity server 4

public class Config { // scopes define the resources in your system public static IEnumerable<IdentityResource> GetIdentityResources() { return new List<IdentityResource> { new IdentityResources.OpenId(), new IdentityResources.Profile(), new IdentityResources.Email(), new IdentityResource { Name = “role”, UserClaims = new List<string> {“role”} } }; }

    public static IEnumerable<ApiResource> GetApiResources()
        return new List<ApiResource>
            new ApiResource
                Name = "customAPI",
                DisplayName = "Custom API",
                Description = "Custom API Access",
                UserClaims = new List<string> {"role"},
                ApiSecrets = new List<Secret> {new Secret("scopeSecret".Sha256())},
                Scopes = new List<Scope>
                    new Scope(""),
                    new Scope("customAPI.write")

    // clients want to access resources (aka scopes)
    public static IEnumerable<Client> GetClients()
        // client credentials client

        return new List<Client>
            new Client
                ClientId = "oauthClient",
                ClientName = "Example Client Credentials Client Application",
                AllowedGrantTypes = GrantTypes.ClientCredentials,
                ClientSecrets = new List<Secret>
                    new Secret("superSecretPassword".Sha256())
                AllowedScopes = new List<string> {""}
            new Client
                ClientId = "openIdConnectClient",
                ClientName = "Example Implicit Client Application",
                AllowedGrantTypes = GrantTypes.Implicit,
                AllowedScopes = new List<string>
                RedirectUris = new List<string> {"https://localhost:44330/signin-oidc"},
                PostLogoutRedirectUris = new List<string> { "https://localhost:44330" }

and i am using below code for MVC Client

app.UseCookieAuthentication(new CookieAuthenticationOptions { AuthenticationType = “Cookies” });

        app.UseOpenIdConnectAuthentication(new OpenIdConnectAuthenticationOptions
            ClientId = "openIdConnectClient",
            Authority = "https://localhost:44358",                              
            SignInAsAuthenticationType = "Cookies",
            ResponseType = "id_token",
            Scope = "openid profile",                               

Please help me

<log goes here>
Updated 27/03/2017 09:48 3 Comments

2.x: Uncaught errors fail silently in junit tests.


I was writing a utility that involved decorating observers and catching uncaught errors from the delegate observer’s onError handling when I noticed in testing that tests always passed. There would be log output, but tests would pass none the less.

Demo of the issue can be found here, but the gist of it is that the following test passes in standard JUnit:

public void blerg() {
    Observable.error(new RuntimeException("This should fail the test"))

I’m trying to understand why errors are passed directly to the handler rather than just thrown, and also curious to get thoughts on making this behavior configurable (if for no other reason than to disable in junit tests). I can’t find any details about why this behavior happens in JUnit either.

I think this has worrisome implications. In our codebase alone, I dropped in an error watching rule to track these and discovered over 150 tests with uncaught exceptions that are being marked as passing otherwise.

Updated 27/03/2017 11:11 4 Comments

Validation error for classname collisions?


Before emitting stylesheets, the elm-css binary could detect when you’ve made two class declarations that share a classname, and raise a validation error.

So this would become an error:

.foo {
  display: inline-block;
  font-weight: bold;

.foo {
  margin: 10px;

It seems like if you ever have this situation, it would have been clearer to combine the styles into one declaration—for example, using a mixin. It seems like if you ever did this, it would have been unintentional, a namespacing mistake.

The underlying question I’m trying to answer is: is it ever a good idea to have two different declarations in your stylesheets which style the same classname?

NOTE: I’m aware that this could be configurable; please don’t respond saying that it ought to be. 🙂

Updated 27/03/2017 00:11

Two dropdowns not working in Angular


I am working on a page that has 2 dropdowns on it using angular. Here is the sample code:

<StackLayout class="input-field">
    <Label text="Location" class="label"></Label>
    <DropDown #dd1 backroundColor="blue" [items]="locs" [selectedIndex]="selectedIndexLoc" (selectedIndexChanged)="onchangeLoc($event)"
        (opened)="onopenLoc()" >
    <StackLayout class="hr-light"></StackLayout>
<StackLayout class="input-field">
    <Label text="Department" class="label"></Label>
    <DropDown #dd2 [items]="depts" [selectedIndex]="selectedIndexDept" (selectedIndexChanged)="onchangeDept($event)"
        (opened)="onopenDept()" hint="Select a department">
    <StackLayout class="hr-light"></StackLayout>

The problem is it only works for one dropdown, eiher dd1 (if dd2 commented out) or dd2 (if dd1 commented out).

What am I missing here? Thank you for your help.


Updated 27/03/2017 07:21 1 Comments

Reload watcher when webpack.mix.js file is udpdated


Not sure if this is a relative issue, but there isn’t so much of a change when the webpack.mix.js file is updated while the run watch <package name> is executing (changes in the webpack file are of course in the scope of the requested package!). So you have to stop the execution of that command, make the changes to webpack file and run it again to see the changes.

Again, not sure if this is an appropriate question or the webpack is technically not meant to change or update during watch operation.

Updated 27/03/2017 06:27 1 Comments

Bootstrapping the compiler


Basic Problem

Building previous versions of the compiler from source using Alpaca and the rebar3 plugin in their current states in order to build the current compiler from source will recurse into the past. I don’t know what all the different versions of Alpaca coexisting will do to each other or if this is even a real concern. Projects that use Alpaca (one of which is Alpaca) need a way to typecheck and compile Alpaca code. Danger: optimizing Alpaca to build Alpaca, see this post.

Questions, Possibilities

How are we going to bootstrap the pieces of Alpaca that require Alpaca themselves? One idea would be to build escript-ized Alpaca releases or sets of .beam files for each release that can be used directly to bootstrap later builds. I don’t see much of a way around needing the previous version to build the current one. Current issue: v0.2.8 is going to need v0.2.7 in order to build itself and v0.2.9 will need the one previous so is it turtles all the way down?

Updated 26/03/2017 23:08

vegindex.R future handling of 'returnHCR' arg


Stil needed after raster integration of HyperSpecRaster handling in vegindex()?

  • Input HyperSpecRaster. Output (to disk) -> RasterLayer/Brick

  • Input HyperSpecRaster. Output (memory) -> HyperSpecRaster (maybe change to Raster* object?)

  • Input Speclib. Output data.frame

When input == Speclib, is there any case when an “HyperSpecRaster” output is desired? -> missing CRS etc.

Updated 26/03/2017 22:32

Hybrid offset?


I was reading the source code of computeOffsetPixels.js and now understand how it works. I would like to raise an discussion on a possible expansion for the “hybrid offset”, e.g. allowing 100% - 50px:

I found one use case in my project where I use react-waypoint as a scrollspy agent. I have a fixed-height <nav> at the top of the page. And I want to trigger something when the user scroll down to a point where a <div> is right ‘at flush’ to the bottom of the <nav>.

Since I want the above behavior when scrolling down, I need to use bottomOffset, not topOffset (right?). So basically I need 100% - 50px-ish value for bottomOffset, which is not supported. Such functionality will be handy, since _handleScroll() gets the bound from _getBounds() fresh, and that takes care of resizing issues.

Updated 27/03/2017 00:54 11 Comments

Is it possible to only load components css if it is mounted?


I have an app thats allows the user to choose from a few different themes/templates which all have their own components, so I only want to load the components css if the user has selected it. i.e there might be 8 Navbar components but only one can be active so I don’t want to load the other 7.

from my brief testing with the demo project, it looks like it bundles all the css like a standard meteor app. Is there a way around this?

Perhaps #114 will help with this?

Updated 27/03/2017 10:04 1 Comments

[Question]: Handling CloudKit network errors


I am trying to handle network errors within my CloudKit & ProcedureKit based app. All networking is currently handled by various CloudKitProcedure wrapped CloudKit operations. I want to be able to detect when any of these fail due to network conditions.

To start with, I have overridden the default errorHandlers on all : ``` someProcedure.set(errorHandlerForCode: .networkUnavailable) { _, error, log, suggested in log.fatal(message: “Exiting due to CloudKit Networking Error: (error)”) return nil }

someProcedure.set(errorHandlerForCode: .networkFailure) { _, error, log, suggested in log.fatal(message: “Exiting due to CloudKit Networking Error: (error)”) return nil } ```

While using the Network Link Conditioner, I have been unable fire these errors using very Bad Network or 100% Loss. They don’t fire when Airplane Mode on either.

What I do get, is a timeout error after 30 seconds: ProcedureKit.ProcedureKitError.Context.timedOut(for 30.0 seconds)

This length is set within the convenience init of CloudKitProcedure (file:CloudKit.swift). The CKOperation.timeoutIntervalForRequest default is 60 seconds. I could understand that the mismatch in the 2 timeouts would mean that I don’t receive a .networkFailure error. The CloudKitProcedure could simply not be waiting long enough to receive this error. But, surely the . networkUnavailable error would be fired while on airplane mode…?

Looking inside ProcedureKit, it appears the these errors are passed directly on from CloudKit. So, this may be a issue with CloudKit and not ProcedureKit. Can anyone point me in the best direction to handle network errors?

Updated 27/03/2017 10:46 1 Comments

Event types?


The designs show filterable event types like Official ACLU Event, Muslim Ban Action, etc. But in the events data file ( #21 ) all the events appear to have the same type: event_type: "Action".

Do we have a list of the event types?

And given an event object like the following, how do we determine its event type if not by its event_type field?

    "city": "Brooklyn", 
    "start_datetime": "2017-04-04T19:00:00", 
    "group": "Follow-Up Organizational Meeting", 
    "event_type": "Action", 
    "zip": "11216", 
    "title": "Follow-Up Organizational Meeting", 
    "url": "", 
    "supergroup": "PeoplePower", 
    "venue": "AN APARTMENT", 
    "state": "NY", 
    "lat": 30.6797, 
    "lng": -74.9303, 
    "starts_at_utc": "2017-04-04T23:00:00", 
    "categories": null
Updated 26/03/2017 21:57 1 Comments

Bug using Docker builder and Ansible provisioner



Seems Docker Builder and Ansible Provisioner doesn’t work together:

packer build template.json

docker output will be in this color.

1 error(s) occurred:

* exit status 250

Packer version : Packer v0.12.3 Host Platform : Mac OSX 10.11.6 Gist :

Doesn’t found anything in documentation who discourage the use of those two together in the same build

Updated 27/03/2017 06:40 2 Comments

Do not import base namespaces


Some schemas import XSD namespaces, like and

What is the right way to avoid import these URLs? Per domain? Any URL in namespaces module? Only these two hard-coded?

    schemaLocation="" />
    schemaLocation="" />

As a side note, xsd2py generate broken code with them.

Updated 26/03/2017 21:05

Problems with years before Christ (negative years)



I’m trying to make a timeline and in this timeline I have some ranges with negative years .

How could I use negative years with vis.js?

{id: 83, content: '<img src="" class="imagen_listado" alt="Biografia de Arquimedes">287 212 a.C. Arquímedes, el hombre que movió el mundo.<a href="" target="_blank"> <span class="icohishuma-headphones"></span></a>', start: 'UTC(-287)', end: 'UTC(-212)'},

Thank you very much, Héctor

Updated 27/03/2017 06:38

Fork me on GitHub