TextualSummaries: part of client side rendering with React. #3420

martinpovolny · 2018-02-16T20:30:05Z

The goal is to implement a top level component for client-side rendering of textual summaries. Such component can then consume a JSON endpoint and later the API.

Extracted PRs

TODO

~~tests (snapshot testing)~~
~~links~~
~~images in general (see e.g. http://localhost:3000/ems_cluster/show/10000000000032 missing img)~~
(will be fixed in a separate PR) ~~PDF rendering: replace with full-screen display with browser print?~~
(dropped) ~~preprocess inlined images (needed in PDF rendering, so we may drop it)~~
~~TextualCustom: 2x in physical server:~~ ~~textual_firmware_table~~, ~~textual_network_adapter_table~~
~~TextualCustom: in app/helpers/vm_helper/textual_summary.rb: textual_normal_operating_ranges~~
~~TextualListview: in ops/ vmdb tables~~
TextualMultilabel -- seems to be a superset of TextualTable with sorting and value expanding
(expanding used in app/helpers/container_build_helper/textual_summary.rb)
TextualMultilink
app/helpers/ems_cluster_helper/textual_summary.rb: ret.blank? ? nil :
TextualMultilink.new(("OpenStack Status"), :items => ret)
app/helpers/host_helper/textual_summary.rb: TextualMultilink.new(("OpenStack Service Status"),
:items => textual_generate_openstack_status) (FIXME: specs)
~~TextualTable (andible playbok job plays, security group firewall rules) (scurity group, network router)~~
(dropped) ~~:spinner => true used only once in app/helpers/vm_helper/textual_summary.rb in textual_genealogy can we drop it?~~
~~:no_value => true -- only used in GO, can we drop it? (introduced in Follow-up for 2846 #2852 I don't understand that change)~~

Next (separate PRs)

~~share the click behavior using HOC https://reactjs.org/docs/higher-order-components.html~~
use patternfly-react icon?
use patternfly-react tables?
~~move to https://github.com/karelhala/react-ui-components (or move the generic parts)~~
fix printing (already broken by the quadicon component)

ohadlevy

Thanks @martinpovolny - I just did a really quick code review - more inline:

ohadlevy · 2018-02-19T09:52:29Z

app/javascript/packs/application-common.js

+
+$(function () {
+  const summary = $('#textual_summary').data('summary');
+  ReactDOM.render(<TextualSummary summary={summary} />, document.getElementById('textual_summary'));


Please use the mounter helper instead ?

yes, will do that

cannot be done right now, unless the component uses only data:

https://github.com/ManageIQ/manageiq-ui-classic/blob/master/app/javascript/react/componentRegistry.js#L43

will wait with this change until that is fixed

Should be resolved in #3228, reviewing tomorrow..

ohadlevy · 2018-02-19T09:53:04Z

app/javascript/react/textual_summary/generic_group.jsx

@@ -0,0 +1,30 @@
+import * as React from 'react';


as a conversion, please just use .js, no need for .jsx extension imho.

Well. I am not sure if this is a good idea. @himdel, @karelhala, @skateman, @Hyperkid123: what do you think?

To be honest I like to use .jsx for react specific code, that way you can immediately see that it's react component and nothing else.

I would even recommend using this ${cmp_name}.component.jsx for components, ${action_name}.action.js for redux actions and ${reducer_name}.reducer.js for reducers, but that is my preference and no need to do it that way.

Same here, better explicit than implicit :)

I disagree :) but thats OK.. there is an ongoing discussion in the foreman
community about how to structure this (and redux etc) at
theforeman/foreman#5240

If we were using just React i would be fine with .js, but i think it would be better to use .jsx in our case.

ohadlevy · 2018-02-19T09:58:34Z

app/javascript/react/textual_summary/generic_group.jsx

+import PropTypes from 'prop-types';
+import GenericTableRow from './generic_table_row';
+
+export default class GenericGroup extends React.Component {


this can probably be written as a const, e.g.:

const GenericGroup = ({ title, items }) => ( <table className="table table-bordered table-hover table-striped table-summary-screen"> <thead> <tr> <th colSpan="2" align="left"> {title} </th> </tr> </thead> <tbody> {items.map((item, i) => <GenericTableRow key={i} item={item} />)}</tbody> </table> ); export default GenericGroup;

and probably consider using patternfly-react table component.

this can probably be written as a const,

yes, the ones where there'll be just the render and no other code I'll convert to functions. (I guess you mean "use function rather than class").

and probably consider using patternfly-react table component.

not in the repo at this point, but yes, I will take a look if it's usable in this context

ohadlevy · 2018-02-19T09:59:29Z

app/javascript/react/textual_summary/icon_or_image.jsx

+
+  renderIcon() {
+    // %i{:class => item[:icon], :title => item[:title]}
+    return <i className={this.props.icon} title={this.props.title} />;


please use patternfly-react icon component.

Not now, later when we include patternfly-react in manageiq, we can do it.

TBH I think we should just add it, no point in writing the code twice...

"writing the code twice..." -- you mean the single line ;-).

I will take a look at a later point. It's not a 5 minutes job. After 5 minutes of searching I found the component in patternfly-react. Figured it's not in the storybook https://rawgit.com/patternfly/patternfly-react/gh-pages/index.html. Have to figure what that means, hopefully it does not mean it should not be used.

Then I figured it's actually 2 components: https://github.com/patternfly/patternfly-react/blob/master/src/components/Icon/Icon.js and the resulting markup seems different. So have to check if it renders that same.

All that work is to prevent writing <i className={this.props.icon} title={this.props.title} />; Probably not. That does not match my sense of effective work.

In this PR I am trying to start this task by moving the rendering to the client side while making sure that the textual summaries render as they did before. W/o extra changes if possible.

Later we can change the markup and/or styling. In a separate PR or at least a separate commit.

I've added patternfly/patternfly-react#234 to track the missing storybook.

the two components are simply because we support two type of components (patternfly icons and font-awesome)
using i assume its only font-awesome?

to use it you can simply do

import { Icon } from 'patternfly-react'; ... export default = () => ( <Icon type='fa' name='react' /> );

ohadlevy · 2018-02-19T10:44:39Z

app/javascript/react/textual_summary/icon_or_image.jsx

+  image: null,
+  icon: null,
+};
+


can be written as:

import * as React from 'react'; import PropTypes from 'prop-types'; /** * Render icon or an image with a title */ const IconOrImage = ({ icon, title, image }) => { const renderIcon = () => <i className={icon} />; const renderImage = () => <img src={image} alt={title} title={title} />; return image ? renderIcon() : renderImage(); }; IconOrImage.propTypes = { title: PropTypes.string, image: PropTypes.string, icon: PropTypes.string, }; IconOrImage.defaultProps = { title: null, image: null, icon: null, }; export default IconOrImage;

yes, I will move the methods inside the class (I did that in some places already).

But the line return image ? renderIcon() : renderImage(); is not the same as my original code. This simplification would not work.

sorry i did not test the code, what did not work? how about

return (image === null ? renderIcon() : renderImage());

no ;-)

in case of both not present I need an empty string. And nested ternary is discouraged so I have that if

ohadlevy · 2018-02-19T10:45:23Z

package.json

@@ -54,6 +54,8 @@
    "babel-polyfill": "^6.23.0",
    "babel-preset-env": "~1.4.0",
    "babel-preset-react": "^6.24.1",
+    "babel-preset-es2015": "^6.24.1",
+    "babel-preset-stage-1": "^6.24.1",


are those required?

const renderIcon = () => <i className={icon} />;

Yes. W/o those 2 lines (and the change to .babelrc) this code would not compile ^^^

what was the error? stage-1 should normally not be required.

at transpile (/home/martin/Projects/manageiq-ui-classic/node_modules/babel-loader/lib/index.js:63:13) at Object.module.exports (/home/martin/Projects/manageiq-ui-classic/node_modules/babel-loader/lib/index.js:163:20) @ ./app/javascript/react/textual_summary/textual_group.jsx 20:17-39 @ ./app/javascript/react/textual_summary/textual_row.jsx @ ./app/javascript/react/textual_summary/index.jsx @ ./app/javascript/packs/application-common.js ERROR in ./app/javascript/react/textual_summary/icon_or_image.jsx Module build failed: SyntaxError: Unexpected token (12:13) 10 | } 11 | > 12 | renderIcon = () => <i className={this.props.icon} title={this.props.title} />; | ^

diff --git a/.babelrc b/.babelrc index 9d3ea8b..936c5c8 100644 --- a/.babelrc +++ b/.babelrc @@ -1,5 +1,6 @@ { - "presets": [["env", { "modules": false }], "react", "es2015", "stage-1"], + //"presets": [["env", { "modules": false }], "react", "es2015", "stage-1"], + "presets": [["env", { "modules": false }], "react", "es2015"], "env": { "test": { "plugins": ["transform-es2015-modules-commonjs"]

The problem is it's inside the class, so I need class fields. I can change the code to use methods instead or make it a function, not a class.

The missing feature is called class fields: https://github.com/tc39/proposal-class-fields

class Foo { renderIcon = 5; }

Currently at stage3 .. so maybe we just need to bump es2015 to a newer version..

Looks like maybe we should just be using env (and react) with some settings to target IE9+,
and maybe include this particual feature..

We need to figure the best way ;-)

Figuring the best way, I have some notes:

I like some of the poweful expressivity of the functional notation. But there are some parts that are really ugly inside JSX:

the contional inclusion of markup: (condition && (<your_crap_here>))

{(index === 0) && ( <td rowSpan={item.value.length} className="label">{item.label}</td> )}

case construct (hash + lookup):

{{ 'shared/summary/textual': ( <GenericGroup items={props.group.items} title={props.group.title} /> ), 'shared/summary/textual_tags': ( <TagGroup items={props.group.items} title={props.group.title} /> ), 'shared/summary/textual_table': ( <SimpleTable items={props.group.items} title={props.group.title} /> ), }[props.group.partial]}

I am glad for the <React.Fragment>. That helps.

Overall the JSX seems to be just another way to write markup and not particularly better than HAML. How many do we need?

ohadlevy · 2018-02-19T15:23:22Z

On Mon, Feb 19, 2018 at 5:20 PM, Karel Hala ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In app/javascript/react/textual_summary/generic_group.jsx <#3420 (comment)> : > @@ -0,0 +1,30 @@ +import * as React from 'react'; To be honest I like to use .jsx for react specific code, that way you can immediately see that it's react component and nothing else. I would even recommend using this ${cmp_name}.component.jsx for components, ${action_name}.action.js for redux actions and ${reducer_name}.reducer.js for reducers, but that is my preference and no need to do it that way. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#3420 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AABOx9SE7Y9TdYvMAVsNC0QU3AwKo8_qks5tWZFBgaJpZM4SI4mV> .

I disagree :) but thats OK.. there is an ongoing discussion in the foreman community about how to structure this (and redux etc) at theforeman/foreman#5240

martinpovolny · 2018-02-28T09:57:15Z

@AparnaKarve : can you, please, tell me, why the :no_value key was added in
#2852 ? What was the problem before that PR? Thx!

AparnaKarve · 2018-03-01T00:18:43Z

no_value is to render a table row with a single cell with no associated column or value.
So far it has been used only in the Generic Objects context.

In the image below, notice the Attributes, Associations and Methods table, that do not have a value.
All we need is a single cell that displays - No Attributes defined

martinpovolny · 2018-03-01T08:04:48Z

no_value is to render a table row with a single cell with no associated column or value.
So far it has been used only in the Generic Objects context.

Thanks, @AparnaKarve. I saw that much it the code ;-) But I don't understand why we need a separate attribute :no_value. Why don't we just check :value for emptyness. Can you explain that, please?

Thanks!

martinpovolny · 2018-03-01T08:34:38Z

@AparnaKarve : to be more explicit: What was the motivation behind the PR that introduced the new attribute? Why testing just :value was not enough?

I'd like to remove this feature if it's not really needed, but there's no test that would fail if I do nor do I understand it when reading the PR.

Thx!

himdel · 2018-03-01T12:24:15Z

@martinpovolny the babel stuff ..

diff --git a/.babelrc b/.babelrc
index 9d3ea8b30..83a481bc7 100644
--- a/.babelrc
+++ b/.babelrc
@@ -1,5 +1,6 @@
 {
-  "presets": [["env", { "modules": false }], "react", "es2015", "stage-1"],
+  "presets": [["env", { "modules": false }], "react"],
+  "plugins": ["transform-class-properties", "transform-object-rest-spread"],
   "env": {
     "test": {
       "plugins": ["transform-es2015-modules-commonjs"]
diff --git a/package.json b/package.json
index 394df1c49..77e5ed098 100644
--- a/package.json
+++ b/package.json
@@ -54,12 +54,11 @@
     "babel-jest": "^22.0.0",
     "babel-loader": "~7.0",
     "babel-plugin-syntax-dynamic-import": "^6.18.0",
-    "babel-plugin-transform-class-properties": "^6.24.1",
+    "babel-plugin-transform-class-properties": "~6.24.1",
+    "babel-plugin-transform-object-rest-spread": "~6.26.0",
     "babel-polyfill": "^6.23.0",
-    "babel-preset-env": "~1.4.0",
-    "babel-preset-es2015": "^6.24.1",
+    "babel-preset-env": "^1.6.1",
     "babel-preset-react": "^6.24.1",
-    "babel-preset-stage-1": "^6.24.1",
     "coffee-loader": "~0.7.3",
     "coffee-script": "~1.12.5",
     "compression-webpack-plugin": "~0.4.0",

(and making a TODO for myself to update to babel7)

AparnaKarve · 2018-03-02T00:16:18Z

Why don't we just check :value for emptyness

That's what I thought too and even adjusted the code to check for existence of :value
5e2b5a6 (PR #2846)

But turns out that a nil value or a blank value could be a valid value in certain cases - in which case we need to render the blank value cell.
(I think the Foreman configured systems summary screen has such examples)

martinpovolny · 2018-03-02T07:14:04Z

@AparnaKarve : Can we have a nil value render nothing and an empty string to be rendered as empty string?

The 5e2b5a6 that you referenced is what turned out not to work? If so then it is pretty counter-intuitive that it does not work. I would suggest having it documented in a test or at least add a source code comment. Can you remember what the problem was?

Taking it from the other side: why cannot the GO behave the same way, why is GO the only entity that needs a special behavior? Would it be a problem if the behavior was consistent?

I mean before you changes we already had 2 behaviors for handling empty values:

empty groups are not rendered at all so instead of a header "Associations" and text "No Associations defined" there would simply be nothing. That would be consistent with all the other summary pages.
The helper functions that render links to association also can handle a link to an empty list.

For some reason, it seems that you implemented a 3rd way:

  def textual_group_associations                                                                    
    if @record.property_associations.count > 0                                                      
      TextualGroup.new(_("Associations"), associations)                                             
    else                                                                                            
      TextualGroup.new(_("Associations"), %i(associations_none))                                    
    end                                                                                             
  end

So this is completely inconsistend with all the other textual summaries. It's a new type of behavior.

If I accept that then looking at the code path that one has to go to display:

+----------------------------------------+
| Associations...........................|
+----------------------------------------+
| No Associations defined................|
+----------------------------------------+

Is pretty obscure. I mean the functionality to display this is added to the most complex path in the TextualSummary rendering.

So what next?

I suggest that:

I remove the functionality.
You then help me understand what where the requirements or reasons behind the different behavior of GO compared to all the other places.
If it really cannot behave the same as the other places, I get it working again.
or you show me the places that get broken (somewhere under Foreman)

This way we should end up with less code, more tests and more consistend UX.

martinpovolny · 2018-03-02T07:20:55Z

@AparnaKarve : I also need you help with this: #3492

It's another area where the behavior is different to the other summaries. It's hard for me to fix that, because I have no data for the physical infra. Given you reviewed it, can you show me how you tested/debugged? I mean do you have a gist or script or database, that I could use? I don't see those two special partials covered in any tests so I need some help.

miq-bot · 2018-03-02T13:29:33Z

This pull request is not mergeable. Please rebase and repush.

AparnaKarve · 2018-03-03T00:31:40Z

I remove the functionality.
You then help me understand what where the requirements or reasons behind the different behavior of GO compared to all the other places.

GO instances need to display Attributes, Associations and Methods.
If these don't exist, then we need to explicitly say so.

Not showing data because it does not exist can be confusing for the user - I think that's what the BZ was about.
The No Attributes defined text clearly conveys that there are no Attributes defined for the GO instance.

or you show me the places that get broken (somewhere under Foreman)

With 5e2b5a6, certain tables in Foreman Configured Systems summary page looked like this without the value column, which is incorrect.

(Bad)

Compare the above with this -

(Good)

because I have no data for the physical infra

6_0_dc_MBU_demo_01202018.backup from Dan's folder should have all the data for physical infra.
Look up record# 55 (/physical_server/show/55)

martinpovolny · 2018-03-08T11:58:13Z

@AparnaKarve : I have redone the empty groups a bit. I created a trivial component to display those, rather than having that behavior mixed onto the complex table.

I found that the behavior is different in GO vs GO definitions. A am not sure that is desired. In my eyes it's yet another inconsistency.

I did some extra fixing:

I fixed the broken i18n, (but meanwhile @mzazrivec fixed that in master so I effectively removed those changes when rebasing).
I removed some methods, 2 where broken anyway (returning identity where each was used in place of map).

Overall it should work better that before and your review is welcome.

AparnaKarve · 2018-03-09T01:23:11Z

I created a trivial component to display those

Thanks. I was going to suggest that as well.

the behavior is different in GO vs GO definitions

Non-admin users cannot view a GO definition.
Hence for them, information like "No Attributes/Associations/Methods defined" can be quite useful.

Admins, on the other hand, can view the GO definitions.

However, for the sake of consistency, we could also add "No Attributes/Associations/Methods defined" for GO definitions.

martinpovolny · 2018-03-12T16:48:54Z

@AparnaKarve : I figured one of the Physical summary files is unused and the 2nd can be trivially replaced with a generic one: Please, check the 2 that mention "physical" close to the end.
p.s. It should also fix the i18n in the summary.

AparnaKarve · 2018-03-13T00:24:17Z

The TextualTable implementation looks nice and clean.

The unused one is a result of this PR #3398.
The network adapters (aka guest devices) now have a view of their own, which means they don't need to be displayed in a table in the Physical Server Summary screen anymore.

miq-bot · 2018-03-13T07:07:44Z

This pull request is not mergeable. Please rebase and repush.

miq-bot · 2018-05-16T09:27:21Z

This pull request is not mergeable. Please rebase and repush.

skateman · 2018-05-24T12:35:43Z

When you open an openstack computenode's (infra -> hosts) summary screen, the boolean value isn't properly displayed for the introspected field:
Before:

After:

skateman · 2018-05-24T13:39:14Z

With configuration profiles this one:

skateman · 2018-05-24T13:41:21Z

And configured systems:

miq-bot · 2018-05-28T11:48:34Z

Checked commits martinpovolny/manageiq-ui-classic@89fcb55~...f90b7f3 with ruby 2.3.3, rubocop 0.52.1, haml-lint 0.20.0, and yamllint 1.10.0
54 files checked, 1 offense detected

**

💣 💥 🔥 🚒 - Linter/Haml - Linter::Haml STDERR:

warning: parser/current is loading parser/ruby23, which recognizes
warning: 2.3.6-compliant syntax, but you are running 2.3.3.
warning: please see https://github.com/whitequark/parser#compatibility-with-ruby-mri.

skateman

martinpovolny · 2018-05-28T13:54:01Z

ping @mzazrivec, @himdel : please, merge

(this is going to break already broken printing even more, but that must be fixed separately)

skateman · 2018-05-28T13:56:08Z

I'm working on the printing right now, let's get this in so I can test it properly 😉

martinpovolny added wip gaprindashvili/no textual summaries labels Feb 16, 2018

ohadlevy suggested changes Feb 19, 2018

View reviewed changes

martinpovolny mentioned this pull request Feb 28, 2018

Physical infrastructure: problematic partials #3492

Closed

martinpovolny force-pushed the reactive_textual branch from 9e57467 to ac72d55 Compare February 28, 2018 09:26

martinpovolny force-pushed the reactive_textual branch from 2293a81 to d61a798 Compare February 28, 2018 09:58

miq-bot added the unmergeable label Mar 2, 2018

himdel mentioned this pull request Mar 2, 2018

FormButtons component, with redux #3509

Merged

martinpovolny force-pushed the reactive_textual branch from 08336da to d24f12d Compare March 8, 2018 11:53

martinpovolny force-pushed the reactive_textual branch 2 times, most recently from 91ae5f6 to ae83ad6 Compare March 12, 2018 08:55

miq-bot removed the unmergeable label Mar 12, 2018

miq-bot added the unmergeable label Mar 13, 2018

miq-bot removed the wip label May 9, 2018

martinpovolny force-pushed the reactive_textual branch from ea0564f to e722632 Compare May 9, 2018 10:03

miq-bot removed the unmergeable label May 9, 2018

martinpovolny force-pushed the reactive_textual branch 3 times, most recently from d6444f9 to 4ef09df Compare May 15, 2018 14:22

miq-bot added the unmergeable label May 16, 2018

martinpovolny force-pushed the reactive_textual branch from 4ef09df to d972b41 Compare May 23, 2018 10:59

miq-bot removed the unmergeable label May 23, 2018

martinpovolny force-pushed the reactive_textual branch from 239399e to c23cc30 Compare May 24, 2018 07:33

martinpovolny force-pushed the reactive_textual branch from c23cc30 to a055b97 Compare May 25, 2018 11:52

martinpovolny added 3 commits May 28, 2018 10:53

TextualSummaries: client side rendering with React.

89fcb55

Make Hakiri and Codeclimate happy.

ec9b93c

TextualSummary: style fixes.

d89f23c

martinpovolny force-pushed the reactive_textual branch from a055b97 to f3390cd Compare May 28, 2018 08:59

TextualSummaries: fix rendering of individual parts.

f90b7f3

martinpovolny force-pushed the reactive_textual branch from f3390cd to f90b7f3 Compare May 28, 2018 11:42

skateman approved these changes May 28, 2018

View reviewed changes

mzazrivec added this to the Sprint 87 Ending Jun 4, 2018 milestone May 28, 2018

mzazrivec merged commit e4e6149 into ManageIQ:master May 28, 2018

skateman mentioned this pull request May 28, 2018

Use browser's native printing dialog when exporting summaries to PDF #3998

Merged

AparnaKarve mentioned this pull request Jun 2, 2018

Broken infra mapping button ManageIQ/manageiq-v2v#363

Closed

ZitaNemeckova mentioned this pull request Sep 14, 2018

Add React textual summary for PXE images #4662

Merged

TextualSummaries: part of client side rendering with React. #3420

TextualSummaries: part of client side rendering with React. #3420

Conversation

martinpovolny commented Feb 16, 2018 • edited Loading

Extracted PRs

TODO

Next (separate PRs)

ohadlevy left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

himdel Feb 19, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Hyperkid123 Feb 19, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

martinpovolny Feb 19, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

martinpovolny Feb 19, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

himdel Feb 19, 2018 • edited Loading

Choose a reason for hiding this comment

martinpovolny Feb 19, 2018 • edited Loading

Choose a reason for hiding this comment

ohadlevy commented Feb 19, 2018 via email

martinpovolny commented Feb 28, 2018 • edited Loading

AparnaKarve commented Mar 1, 2018

martinpovolny commented Mar 1, 2018

martinpovolny commented Mar 1, 2018

himdel commented Mar 1, 2018 • edited Loading

AparnaKarve commented Mar 2, 2018 • edited Loading

martinpovolny commented Mar 2, 2018 • edited Loading

So what next?

martinpovolny commented Mar 2, 2018 • edited Loading

miq-bot commented Mar 2, 2018

AparnaKarve commented Mar 3, 2018

martinpovolny commented Mar 8, 2018

AparnaKarve commented Mar 9, 2018

martinpovolny commented Mar 12, 2018 • edited Loading

AparnaKarve commented Mar 13, 2018 • edited Loading

miq-bot commented Mar 13, 2018

miq-bot commented May 16, 2018

skateman commented May 24, 2018

skateman commented May 24, 2018

skateman commented May 24, 2018

miq-bot commented May 28, 2018

skateman left a comment

Choose a reason for hiding this comment

martinpovolny commented May 28, 2018

skateman commented May 28, 2018

martinpovolny commented Feb 16, 2018 •

edited

Loading

himdel Feb 19, 2018 •

edited

Loading

Hyperkid123 Feb 19, 2018 •

edited

Loading

martinpovolny Feb 19, 2018 •

edited

Loading

martinpovolny Feb 19, 2018 •

edited

Loading

himdel Feb 19, 2018 •

edited

Loading

martinpovolny Feb 19, 2018 •

edited

Loading

martinpovolny commented Feb 28, 2018 •

edited

Loading

himdel commented Mar 1, 2018 •

edited

Loading

AparnaKarve commented Mar 2, 2018 •

edited

Loading

martinpovolny commented Mar 2, 2018 •

edited

Loading

martinpovolny commented Mar 2, 2018 •

edited

Loading

martinpovolny commented Mar 12, 2018 •

edited

Loading

AparnaKarve commented Mar 13, 2018 •

edited

Loading