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

Add links to grok debugger


Adds some links that point to the grok debugger so users will know that the tool exists.

Not sure why we were pointing to current in the index, but I think it’s better to point to the branch (we actually can’t point to current because grok debugger docs won’t exist in current until the code is released).

Updated 26/06/2017 23:18

New docs page: Extending Bootstrap


Right now this is a placeholder PR to stir some conversation around how to extend Bootstrap. This might not be the right list to start with, but I wanted to open the PR to gather more feedback before jumping into it full-steam. Some of this might make more sense in our Getting Started pages, but let it rip if you have ideas no matter the place just yet.

Will aim to tackle this after Beta 1.

Updated 26/06/2017 21:08

5.5.0 Task: Documentation build is not working correctly

% grep 'Fetch of' /tmp/x
Fetch of failed: #<Net::HTTPNotFound:0x0055f2a224be78>
Fetch of failed: #<Net::HTTPNotFound:0x0055f2a26a9dd0>
Fetch of failed: #<Net::HTTPNotFound:0x0055f2a25e9e90>
Fetch of failed: #<Net::HTTPNotFound:0x0055f2a2376d48>
Fetch of failed: #<Net::HTTPNotFound:0x0055f2a2375fd8>

At least 5 plugins are targeted for Logstash 5.5..0 but they have no docs/index.asciidoc which probably means either:

  • maybe: They got docs migrated (docs/index.asciidoc) but did not get released
  • or, maybe: they got released, but 5.5.0 cannot install them due to dependency conflicts
  • or maybe: something else.
Updated 27/06/2017 00:09 15 Comments

Add missing descriptions to plugin docs


The following plugins are missing descriptions. To get the plugin docs building without errors, we are removing the empty “Description” sections from the plugin docs.

However, we need to add descriptions to the docs source for the following plugins as soon as possible.

codecs/edn.asciidoc codecs/edn_lines.asciidoc codecs/msgpack.asciidoc inputs/log4j.asciidoc outputs/circonus.asciidoc outputs/datadog.asciidoc outputs/email.asciidoc outputs/http.asciidoc outputs/librato.asciidoc outputs/lumberjack.asciidoc outputs/stomp.asciidoc codecs/edn.asciidoc

It’s possible that some of the descriptions might have been lost because they were in the wrong location prior to the doc extract. It would be good to check the old asciidoc comments in the ruby source to see if this happened.

We might find additional descriptions missing when we build the plugin docs for 5.5 with all the files included.

This fixes one of the issues captured in

Updated 26/06/2017 16:40

Fix documentation to not use multiple types


There are still some places in the documentation that should be fixed to either use the new parent join facilities or should be fixed in general to not use multiple types:

  • reference/aggregations/bucket/children-aggregation.asciidoc
  • reference/docs/index_.asciidoc
  • reference/indices/put-mapping.asciidoc
  • reference/mapping/dynamic/default-mapping.asciidoc
  • reference/mapping/fields/parent-field.asciidoc
  • reference/mapping/fields/type-field.asciidoc
  • reference/mapping.asciidoc
  • reference/query-dsl/parent-id-query.asciidoc

some of them might even be removed. Relates to #24961

Updated 26/06/2017 16:20

[DOCS] clrDropdownMenuOpen and toggleDropdown() method missing from dropdown docs



Select one … (check one with “x”) [x] bug [ ] feature request [ ] enhancement

clrDropdownMenuOpen and toggleDropdown() method missing from dropdown docs. Add them to the dropdown Summary of Options table.

Reproduction of behavior


Environment details

  • Angular version: NA

  • Clarity version: NA

  • OS and version: NA

  • Browser: [all]

Updated 26/06/2017 14:12

Nginx config client_max_body_size


Nginx defaults to a max body size of 1mb, which isn’t great for image and theme uploads.

I have been adding client_max_body_size 50m; as part of the install process for nginx before running ghost-cli.

On a fresh nginx install, I do this by adding the line after types_hash_max_size 2048; in /etc/nginx/nginx.conf. Ansible has a lovely tool called “lineinfile” that I wonder if we can reproduce?

I wonder if we should either document this, or get ghost-cli to attempt to set it?

Updated 26/06/2017 17:56 3 Comments

Mention the recursor's SNMP support in


Short description

<!– Write a small description of what this Pull Request fixes or provides, including the issue #s –> Carbon is documented but I forgot to add mention of SNMP to that page.


<!– please indicate if any of these things are done/included with this Pull Request. Not all boxes need to be checked for the Pull Request to be accepted –> I have: - [x] read the document - [ ] compiled and tested this code - [x] included documentation (including possible behaviour changes) - [ ] documented the code - [ ] added or modified regression test(s) - [ ] added or modified unit test(s)

Updated 26/06/2017 12:43

the yellow icons for "Needs statistical model" in SpaCy 101


Hi there, I am a tech writer and I volunteer to work on the Chinese translation for this manual. Here’s my page I am stuck here becuase I don’t understand what the yellow icon means here. Does it mean “it depends”? Surely green means “yes, it requires model” Red means “no, it doesn’t need to install a model”

Am I getting a wrong understanding? Thanks!

Updated 26/06/2017 11:27 1 Comments

website clean up


I am checking, and there are a few places need to be cleaned up.

  • [ ] API quickstart is under both getting_started.html and examples.html
  • [ ] Markdown Header within notebook should be ## and the title being ##, otherwise they appear the same on the right panel (e.g., see Sampler statistics and Multiple samplers)
  • [ ] redundent notebook and
Updated 26/06/2017 11:09 3 Comments

Documentation gh-pages throwing error on drawer link click


Problem description

Hey, just wanted to let you know that the documentation pages for the latest version (v0.18.4) is throwing when you click on every Link in the drawer.

Uncaught TypeError: Cannot read property 'query' of undefined
    at Object.y [as push] (app.js:15)
    at Object.r.handleChangeList [as onChange] (app.js:25)
    at n.r.handleItemTouchTap (app.js:9)
    at Object.onTouchTap (app.js:9)
    at Object.r.handleTouchTap [as onTouchTap] (app.js:6)
    at r.handleTouchTap (app.js:2)
    at Object.r (app.js:11)
    at a (app.js:7)
    at Object.s [as executeDispatchesInOrder] (app.js:7)
    at f (app.js:4)
    at m (app.js:4)
Updated 26/06/2017 13:44 1 Comments

Make sure you don't have race conditions or at least document it if you do [session locking]


Just found your fork. Original version is naive about concurrency and doesn’t implement any locking. I haven’t looked at your code but from what I understand you implemented batch updates. Batch updates significantly increase race condition windows so I thought it would be good idea to mention it here.

Updated 26/06/2017 19:36 5 Comments

Getting started mislead


I was following the tutorial “Getting Started” and found a slight issue in 6.1. Perhaps it was just worded in a strange way. First it mentioned about the new addition and one should examine the ‘db/scema.rb’ file, and showed an example output. I assumed this file would be identical to mine, however the ‘create_table “comments”’ section didn’t exist (yet). Next it said to do ‘db:migrate’. Only after that did the scema file appear like the example.

If I were writing this, I’d say ‘db:migrate’ first then show the output. This was misleading and confusing. So I would’ve put them around the other way. Do you agree?

Updated 26/06/2017 16:29 2 Comments

pyboards docs: Put hardware/datasheet info under "General information about pyboard"


As mentioned and shown at, pyboard docs inconsistent wrt to other port docs in that they put 3 headings of this doc: as top-level chapters. That leads to pretty messy ToC. Instead, that information should go under “general information about board” like other ports do (well, like esp8266, the freshest, does).

If approved, should this be a PR or can be committed directly?

Updated 26/06/2017 08:14 2 Comments

[Docs]: React- State


A State doc already exists for this project’s doc book, but is currently an empty place holder until its contents are filled in.

This doc should be updated to include all things state related for react. Including things like: - What is component state? - Why is managing state important? - Why you can’t and shouldn’t directly mutate state? - How to set initial state - How to update a component’s state - How to update a component’s state on an event action (preferably with code example) - How re-rendering works when state changes

Updated 23/06/2017 22:28

[Docs]: React - Proptypes


A Proptypes doc already exists for this project’s doc book, but is currently an empty place holder until its contents are filled in.

This doc should be updated to include all things proptype related for react. Including things like: - What is typechecking? - What are PropTypes? - Why use PropTypes? - How to use PropTypes

Also this doc should make sure we are referencing PropTypes usage from the recent React 15.5 or later. This is because this release moved PropTypes out of the main React node module and into a separate module of its own.

Code example should be used when possible! 🙂

Updated 23/06/2017 22:25

Update plugin docs ascidoc


Currently our plugins use the wrong defined ids at the top of the asciidoc.


And it should be


Note the s after type

Updated 23/06/2017 20:04 1 Comments

Color palette documentation: Need two small corrections


Select one … (check one with “x”) [X] bug [ ] feature request [ ] enhancement

Need two corrections in color palette documentation:

1) Title for Primary swatch #F52F22 (red) is shown with white text. Text should be black #000

2) Color swatch ##0065AB (blue) shows both a black and a white “A”. I verified that only white passes accessibility for this blue. Please remove the black “A”.



Updated 23/06/2017 18:10

@cerebral/http documentation is confusing


Hello! I’m new to cerebral and documentation for @cerebral/http makes me so dejected.

According to doc, I can use it like this: ``` import {set} from ‘cerebral/operators’ import {httpGet} from ‘@cerebral/http/operators’ import {state, props} from ‘cerebral/tags’

export default [ httpGet(/items/${propsitemKey}), set(stateapp.currentItem, propsresult) ] ```

But if I do so, I’ll get the following error: Uncaught TypeError: Cannot read property 'get' of undefined at Object.httpGet [as function] (httpGet.js:17) Which says nothing to me.

So there’re two problems: 1. Unclear error message If use http operators without adding HttpProvider 2. Gap in documentation, since there’s no even word about I should instantiate it first in controller section.

Updated 25/06/2017 08:29 1 Comments

Move playground prereq info out of doc and include tutorial link in the playground


If someone is trying to use the playground with a browser that is likely to cause massive issues, we should warn them directly in the UI. It would also be useful to include a link to the tutorial.


Even though anyone can get to the online playground directly, and there is a “Try it online” button on the first page of the Hyperledger Composer site, there is prereq information in the documentation, along with suggestions on what to do next. Having an ‘install’ guide to “Try Online Playground” is pretty odd to begin with, and the method of launching the playground in that topic is an unfriendly link…

The Online Playground can be accessed here:

Expected Behavior

I should be able to go directly to the playground without requiring information from the documentation.

Actual Behavior

The documentation contains a topic about ‘installing’ the online playground, which mentions ‘modern browsers’, requirement for a github account, and links to tutorials.

Possible Fix

  • [ ] Add basic checking to the playground to warn someone if they are likely to experience problems due to an outdated browser. (Ideally based on the functionality required, not specific browsers!)

  • [ ] Add link to tutorial and any other useful information to the splash screen

  • [ ] Remove redundant information from documentation

Updated 26/06/2017 07:25 1 Comments

Permissions system


Permissions system

I am writing this to myself just to create an outline and to discuss ideas. I want to go for a simple permission system with 3 tiers:

  • 3: Admin Can change FredBoat configuration and edit permissions

  • 2: DJ Can skip, can pause/unpause, can change repeat mode, can change shuffle mode, etc

  • 1: User Can add tracks to the queue

TBD: Who controls voteskipping?

Special notes

  • Administrators and owners implicitly have all perms
  • Permissions are for the entire Guild
  • Higher ranks have lower rank permissions
  • Permissions can be granted to users or roles

Please give me ideas for better names for these

Proposed commands

;;admin list ;;admin add <name> ;;admin remove|del <name>

Note that admin can be replaced by dj and user.


  • [x] Set up JPA entities
  • [x] Create PermsUtil. Named like this to avoid confusion with JDA’s PermissionsUtil
  • [x] Create these commands
  • [x] Implement the perms system into the other commands
  • [x] Do I18n and upload to Crowdin
  • [ ] Write documentation (help wanted)
  • [ ] Announce and deploy
Updated 26/06/2017 22:11 21 Comments

A Section on Operational requirements to configure Composer to work with Hyperledger Fabric V1 are required


With our one click installs and dev server setups we create an easy development environment not requiring any work to get up and running using a real hyperledger fabric v1.0 system. We have information about what is required to ensure hyperledger composer is able to work with Hyperledger Fabric 1.0 but it isn’t consolidated into a single place. So If you have a pre-configured environment what do you need to do in order to be use hyperledger composer with it.

  1. A Connection Profile - Point to the details about V1 connection profiles
  2. Importing Of Peer and Channel Admin credentials - point to the import identity command

Provide a concrete example based on the fabric-dev-server setup

Updated 26/06/2017 09:58

Remove 3 logos from the Symbols module


The 3 brand specific logos currently in the Symbols module are causing confusion and adding additional work when updating the GUI. They are… .symbol .symbol-logo-alt .symbol .symbol-logo-phone .symbol .symbol-logo-tablet These logos were originally designed to make life easier for developers when switching brands as they all use the exact same dimensions. Over time we have seen a decline in their usage. Moving forward I’d like to create several multi-brand, responsive header components which use the icons in the icon module. Removing the icons should mean the Symbols module becomes a global module so we only need to update 1 instance.

Updated 27/06/2017 01:04 1 Comments

Updates to the Primary Colour palette


This is big and scary.

Complementary A and Complementary B are causing confusion when it comes to designing multi-brand. These colours need to be removed from the primary palette and put into the additional colours. We also need to add a colour to the Primary palette for headings. Doing this will mean that every colour in the primary palette has a purpose and will map correctly when switching brands.

Updated 23/06/2017 02:16

Documentation - QuickStart Guide - Step 8. Customize robust theme


The first change that we have to make is to use a different default image instead of the one used in the theme. The theme’s default image used in both the list and single view page resides inside themes/hugo_theme_robust/static/images/default.jpg. We can easily override it by creating a simple directory structure inside the repository’s static directory.

Create an images directory inside the bookshelf/static directory and copy an image with name default.jpg inside it.

However, if you copy the Hand/Book image, name it default.jpg and place in the Hugo\Sites\\static directory as stated above, the page will not update with this image.

You should clarify that in this example, you must place the Hand/Book image named default.jpg in the: \Hugo\Sites\\themes\hugo_theme_robust\static\images, then refresh the page to see the changes you made.

Updated 25/06/2017 21:03 2 Comments

Cleanup doc repos to remove duplicated content


Currently, the doc build is pulling files from both the logstash and logstash-doc repos:

For 5.0 and later, the build pulls:

  • index files and static doc source files from the logstash repo
  • plugin source files from the logstash-docs repo

The files in logstash-docs under /docs/static are no longer used by the build.

For 1.5 through 2.4, the build pulls: - index files from the logstash repo. - static docs and plugin docs from the logstash-docs repo.

Eventually, we want to clean up branches 1.5 through 2.4 so that we follow the same build strategy that we follow for 5.0+.

TO DO list:

  • [ ] For version 1.5 through 2.4, make sure the docs/static folder in the logstash repo contains the authoritative version of the files. Right now there are a few doc changes that never got into the logstash-docs. I need to confirm that the opposite has not happened with the logstash-docs repo. It should not have changes that are not also in the logstash repo.
  • [ ] Update the build to pull the static doc source files from the logstash repo instead of logstash-docs.
  • [ ] Delete the /docs/static folder from all branches of the logstash-docs repo starting with version 1.5
  • [ ] Delete all index files from the logstash-docs repo.
Updated 22/06/2017 19:59

Larger visualization in doc/example notebooks?


In doc/example notebooks, I think it would be nice to have larger visualizations.

However, doing so might cause other problems, or be difficult to achieve. Or maybe other people don’t agree it would be better in the first place.

Current approach

I think the plots just come out at the backend’s default size, maybe?

Proposal 1: per notebook setting of output size

One option would be to make all the visualizations bigger by adding output size=150 or something at the start of every doc/example notebook. That way, people would be aware of the option and could copy/change it in their own work. But seems a shame to have a distraction like that at the start of every notebook.

Proposal 2: set output size for nbpublisher

Could put output_size=150 or whatever in the nbpublisher jupyter profile instead of adding to individual notebooks. Might be problematic for notebooks where the visualizations are already large. Wouldn’t help users to have larger visualizations themselves. Could make things difficult to debug: doc/example nbs will look different when published vs when local.

Proposal 3: do something automatic?

For instance, could decide to make any single plot take 80% of the space, 2 plots to add up to 80% of the space, 3 to add up to 100% of the space, 4 be like 3 but to split across lines, etc…

Any other ideas?

Updated 22/06/2017 19:33

{{first}} helper does not recognize array


Using the provided example of {{first "['a', 'b', 'c', 'd', 'e']" 2}} returns ['. Seems like its being interpreted as a string instead of an array.

Here is the code I am using based on the gulp example. Its possible I am missing something or Handlebars just does not like an inline array.

var $ = require('jquery');
var helpers = require('handlebars-helpers')({
  handlebars: Handlebars
$(function() {

  function render() {
    var tmpl = "<h2>{{first \"['a', 'b', 'c', 'd', 'e']\" 2}}</h2>";
    var fn = Handlebars.compile(tmpl);
Updated 22/06/2017 18:54 1 Comments

Tutorial template draft: needs new examples


This snippet is a stripped down version of the old tutorial, with examples removed. Would probably be good to develop some more complex examples (to demonstrate the value proposition better). (This could be an effort post-initial release.) (template is provided in raw markdown here and in attached text file because GH doesn’t support attaching markdown files :-( ) TUTORIAL_template.txt


The ksonnet readme shows you how to create a default deployment.json file that lets you deploy an nginx container to an existing Kubernetes cluster. This tutorial shows you how to:

  • Modify the deployment using ksonnet definitions
  • Define other Kubernetes objects
  • Work with mixins to develop complex configurations


This tutorial assumes that you have performed the following tasks. For details, see the readme.

  • Installed Jsonnet
  • Cloned the ksonnet repository locally
  • Installed and configured the VisualStudio Code extension (optional)
  • Created a test Kubernetes cluster

NOTE: All import paths are relative to the root of the *ksonnet** repository.

Modify the default deployment

ksonnet lets you configure or modify any Kubernetes object.

For example, to customize the default <> deployment, you can write:

// TODO: provide ksonnet example

Save the file as customDeploy.libsonnet and run:

jsonnet customDeploy.libsonnet
kubectl apply -f deployment.json

(here we seem to be looking at YAML we get from running kubectl get deployment <appName> -o yaml but in other examples we’re looking at JSON generated from ksonnet. Figure out, fix.) And the generated YAML looks like this. You can see the new <> and <> fields:

# TODO: provide example

Define other Kubernetes objects

ksonnet lets you define any Kubernetes object. For example, you can define a container:

// TODO: update ksonnet example; container is probably a good one
// consider whether nginx is the right one for deployment 
// (issue also in previous example)

Save this snippet as container.libsonnet, and run:

jsonnet container.libsonnet

The JSON output looks like this:

TODO: fix up to be consistent with YAML vs JSON across tutorial examples


You can include a liveness probe with your container:

// TODO: ksonnet example. Is liveness probe appropriate here?

Save the file again – container.libsonnet – and run:

jsonnet container.libsonnet

(As above, YAML or JSON?) The JSON output now looks like this:


Now you can define a pod that runs this container:

// TODO: ksonnet example

Save the file as pod.libsonnet, and run the following commands to deploy the pod to your cluster:

jsonnet pod.libsonnet // create pod.json
kubectl apply -f pod.json // apply pod definition to cluster

Work with mixins

You’ve seen how to modify the default deployment by writing mixins to add custom fields. As the Jsonnet tutorial explains in more detail, mixins provide dynamic inheritance, at runtime instead of compile time. This approach means that different team members can define the Kubernetes objects that they need. You can then mix them into your master definition without having to copy all the details.

For example, you could write the following code to define a container for your application:

// TODO: ksonnet example

And your teammate could write the following code to define the VolumeMounts:

//TODO: ksonnet example

Then, you write a deployment definition for the container that adds the VolumeMounts for logging using mixins:

//TODO: ksonnet example with mixins
Updated 24/06/2017 03:40 1 Comments

Complete Doc Comments


Detailed Description

Most of the public APIs have documentation already. But Visual Studio still has ~370 XML doc comment related warnings.

Acceptance Criteria

All XML Doc Comment related build warnings are resolved and all public and protected APIs and members have complete documentation.

Updated 22/06/2017 05:39


  • LaravelAuth Version: 2.2
  • Laravel Version: 5.4
  • PHP Version: 7.1.4


Hi arcanedev-maroc. This package looks awesome.

I followed the installation steps successfully. So it seems to have installed correctly. I’m just not clear on the usage. Should i build the forms and views myself? If there’s any examples out there I’d love to see them.

Thanks for any guidance.

Updated 22/06/2017 09:24 1 Comments

Adding @jonparrott's sphinx-docstring-typing module.


This allows usage of Python 3.5+ typing module, which is very helpful.

Note that the changes to source docstrings are likely bugs in sphinx-docstring-typing, so that may be addressed on that project.

@jonparrott This is marked “don’t merge” right now because I want to check with you what the issues actually are with the source docstrings.

Updated 22/06/2017 19:14 2 Comments

[Proposal] Add a guide on how to use Overlay



Add a guide on how to use Overlay, like there is one for theming and flexbox.

What is the use-case or motivation for changing an existing behavior?

I’ve been using overlay lately for a couple components, and it’s a bit hard to use due to the lack of documentation, but it’s easy once you understand what’s going on. Thankfully I was able to understand some of it with the MdDialog and MdTooltip, as well as reading the Overlay code, but it was a tough learning curve which I feel I haven’t conquered yet.

Some things that might be useful for such guide are:

  • How to override the overlay container so I put things in a known DOM element and make it easier for testing.
  • How to get the component instance or overlayRef in a test without having to keep a property in the object.
  • How to use position and scrolling strategies and how to test them.
  • How to create a new strategy.
Updated 23/06/2017 02:05 1 Comments

Versions in docs


What problem does this feature solve?

Every language in the docs show version 2.0 and I think it should be the last updated version. Right now if a user wants to learn about a new feature (let’s say, implemented in v2.6) there is no way of knowing if the docs in the desired language are up to date

What does the proposed API look like?

    2.0 - English (Up to v2.6)
    2.0 - French (Up to v2.5)
    2.0 - Japanese (Up to v2.4)
    2.0 - 中文 (Up to v2.3)
    2.0 - German (Up to v2.2)
    2.0 - Русский (Up to v2.1)
    2.0 - 한국어(Korean) (Up to v2.0)
    2.0 - Español (Up to v2.1)
    0.7 Docs


    2.6 - English
    2.5 - French
    2.4 - Japanese
    2.3 - 中文
    2.2 - German
    2.1 - Русский
    2.0 - 한국어(Korean)
    2.1 - Español
    0.7 Docs

<!– generated by vue-issues. DO NOT REMOVE –>

Updated 22/06/2017 06:11

Wheels.RouteNotFound error (during Beginner Tutorial: Hello Database)


I followed all the steps to create the Beginner Tutorial: Hello Database, but I got a bug on the page after I submitted the little form with name , email, and password.


Could not find a route that matched this request.

Suggested action

Make sure there is a route configured in your config/routes.cfm file that matches the users/create request.

Tag context

Error thrown on line 111 in wheels\dispatch\functions.cfm - called from line 177 in wheels\dispatch\functions.cfm - called from line 143 in wheels\dispatch\functions.cfm - called from line 5 in wheels\index.cfm - called from line 2 in index.cfm - called from line 5 in wheels\events\onrequest.cfm

Updated 25/06/2017 18:17 3 Comments

Documentation needed on the return values of the class methods


Taking an example, I want to make a bot that behaves differently for subscribed users, followers etc…

to get subscriptions, I should use :

client = twitch.TwitchClient(client_id, oauth)

channel =

subscribes = channel.get_subscriptions(args) #offset moving 100 at a time in a loop

so now, what exactly is in subscribes, it’s a list, but how are users represented in python ?

twitch dev gives : ``` {

"_total": 4,

"subscriptions": [


        "_id": "e5e2ddc37e74aa9636625e8d2cc2e54648a30418",

        "created_at": "2016-04-06T04:44:31Z",

        "sub_plan": "1000",

        "sub_plan_name": "Channel Subscription (mr_woodchuck)",

        "user": {

            "_id": "89614178",

            "bio": "Twitch staff member who is a heimerdinger main on the road to diamond.",

            "created_at": "2015-04-26T18:45:34Z",

            "display_name": "Mr_Woodchuck",

            "logo": "",

            "name": "mr_woodchuck",

            "type": "staff",

            "updated_at": "2017-04-06T00:14:13Z"



. . .


} `` so to access the id of the first user, should I use subscribes[0].user.id_ ` ?

Also I think you’re missing the token revoke endpoint.

I also think you could add some image stuff to the library to parse images from the links, and have them accessible as python objects (using pillow maybe)

Also, sub_plan refers to sub tiers ? so 1000 is 4.99$, 2000 9.99$ and 6000(or 3000, read somewhere that 25$ subs count for 6 so just interpreting the data) 24.99$ ?

Updated 22/06/2017 14:08 1 Comments

Counterintuitive date rounding behavior in date filters


Elasticsearch version: 5.4.1

Plugins installed: [mapper-murmur3, repository-s3]

JVM version (java -version): Java 1.8.0_60-b27

OS version (uname -a if on a Unix-like system): Centos 7

Description of the problem including expected versus actual behavior:

Date range rounding behavior is counterintuitive, imo, and inconsistent with what I think users would expect, based on experiences with other database/IR systems. Consider the following query:

GET /_search
  "size": 0,
  "query": {
    "bool": {
      "filter": [
          "range": {
            "@timestamp": {
              "gte": "now-1d/d",
              "lte": "now/d"
  "aggs": {
    "MaxDT": {
      "max": {
        "field": "@timestamp"
    "MinDT": {
      "min": {
        "field": "@timestamp"

This yields

  "aggregations": {
    "MinDT": {
      "value": 1497916800044,
      "value_as_string": "2017-06-20T00:00:00.044Z"
    "MaxDT": {
      "value": 1498067981073,
      "value_as_string": "2017-06-21T17:59:41.073Z"  <--- note this

Consider instead the following query:

GET /_search
  "size": 0,
  "query": {
    "bool": {
      "filter": [
          "range": {
            "@timestamp": {
              "gte": "now-1d/d",
              "lt": "now/d"
  "aggs": {
    "MaxDT": {
      "max": {
        "field": "@timestamp"
    "MinDT": {
      "min": {
        "field": "@timestamp"

This yields

  "aggregations": {
    "MinDT": {
      "value": 1497916800044,
      "value_as_string": "2017-06-20T00:00:00.044Z"
    "MaxDT": {
      "value": 1498003199988,
      "value_as_string": "2017-06-20T23:59:59.988Z" <--- note this

The later query is correct for “Show me what happened yesterday.” The former produces an unexpected date range in that the values of @timestamp themselves appear to be rounded down and compared against now/d, which is not what I would expect based in the wording in the existing documentation.

I am still looking for the date rounding handling in the source code, but my gut feeling is that @timestamp is being rounded down and compared against now/d. This may or may not be what you intended and there may or may not be good reasons to do it this way, if you are. But it is not the behavior I would expect based on the docs and based on experience with other databases. If this behavior is intentional, would it be possible to document this behavior as a gotcha?

Updated 22/06/2017 08:39 1 Comments

Beats doc updates in 6.0


Work in progress….just getting this started….

Beta 1

  • [ ] Document new command line with sub-commands. See

GA - [ ] Change doc structure to accommodate prospector types. See - [ ] Update dashboards for 6.0

Updated 21/06/2017 16:47

Form examples display submission values


Issue Description

<!– Describe the issue as best you can. We love screenshots! –> In all of our form component examples, it would be useful for consumers to be able to view the submission values. The example at the bottom of the form page demonstrates this: We should do the same for all of the form components so we can test the expected behavior: - [ ] date picker - [ ] time input - [ ] select - [ ] search field

Issue Type

<!– Is this a new feature request, enhancement, bug report, other? –> - [ ] New Feature - [x] Enhancement - [ ] Bug - [ ] Other

Updated 22/06/2017 18:18

Azure builder device login failes in AzureGermanCloud


With packer 1.0.0 we see the following error, when running device login version of packer in Azure Germany:

==> azuregerman: Running builder ...
Build 'azuregerman' errored: packer-azure application not set up for Azure environment "AzureGermanCloud"

Reason seems to be which is missing an entry for Azure Germany.

Updated 21/06/2017 18:43 8 Comments

Brokend Documentation links


The below documentation link is broken

Any future plan for ‘plug n play’ samples?

Updated 23/06/2017 10:23 2 Comments

Docs build without gammapy-extra


Currently building the Gammapy docs fails with a not very clear error message (a different one for @wegenmat and me) if GAMMAPY_EXTRA isn’t set or doesn’t point to a checkout of gammapy-extra. @woodmd also mentioned this on Slack.

This is a reminder issue for me to: - [ ] check if we can make the docs build work without GAMMAPY_EXTRA with limited effort - [ ] if yes, do it - [ ] if no, improve the error message - [ ] in any case, improve the Gammapy dev docs explaining this.

Updated 21/06/2017 11:21

List of elm-test packages?


Do we want a list/table of packages that provide additional functionality on top of elm-test? Right now, we’re mentioning elm-html-test and elm-testable under “Application-specific techniques”, but there are two more I think should be mentioned.

Updated 25/06/2017 21:09 3 Comments

Documentation is not fully responsive


On my mobile (Huawei P9 lite) and also on Chrome for 320px images page is not properly responsive, content is overflowing for about 2px. Probably because container has 15px margin, but bd-example bd-example-images has 1rem, and it is not consistent. Will it be fixed?

Updated 25/06/2017 11:21 1 Comments

Better docs

  • [ ] Make it more readable
  • [ ] More examples
  • [ ] XML config examples
  • [ ] Show happy path
  • [ ] Consider putting docs onto separate directory
  • [ ] GitHub Pages?
Updated 26/06/2017 05:34

Fork me on GitHub