Add <img> decoding="" attribute #3221

vmpstr · 2017-11-10T22:33:38Z

This pull introduces a decoding attribute to the img tag. The attribute indicates a decoding hint to
the user agent. That is, this hint aids the user agent in deciding how to process and decode the image
before rasterizing it. Possible values are as follows:

sync - prefer to decode the image synchronously for atomic presentation with other content
async - prefer to decode the image asynchronously to reduce delay in presenting other content
auto - default mode, which indicates no preference for the decoding mode.

Closes #1920

/embedded-content.html ( diff )
/images.html ( diff )
/index.html ( diff )
/indices.html ( diff )

vmpstr · 2017-11-10T22:36:43Z

/cc: @chrishtr @smfr @domenic @annevk

This attribute has three states (sync/async/auto) as per discussion on #1920. Also, the name of the attribute is decodingMode to eliminate amiguity with the decode() function.

chrishtr · 2017-11-10T23:42:32Z

source

@@ -25473,6 +25475,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  cannot process images or who have image loading disabled (i.e. it is the <code>img</code>
  element's <span>fallback content</span>).</p>

+  <p>Image data can appear in encoded form in order to reduce file size.
+  <dfn data-x="img-decoding-process">Decoding</dfn> is a process which converts image's media data


"converts an image's"

"is the process"

chrishtr · 2017-11-10T23:43:25Z

source

+   <tbody>
+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-sync">sync</code></dfn>
+     <td>Inidicates a preference to <span data-x="img-decoding-process">decode</span> this image


sp: indicates

chrishtr · 2017-11-10T23:43:45Z

source

+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-sync">sync</code></dfn>
+     <td>Inidicates a preference to <span data-x="img-decoding-process">decode</span> this image
+     synchronosly for atomic presentation with other content.


sp: synchronous

This doesn't say what "synchronous" is relative to. I think the spec needs to say something to the effect of "synchronous" meaning that the image will get painted immediately after the image data is available (i.e. the load event firing).

Yes, any ideas for how to make this clearer would be appreciated, perhaps by integrating with the img processing model. (Although that seems to have very little about actually presenting on the screen.)

I've added a section on Decoding images (as @domenic recommended). I appreciate any suggestions on how to phrase that section better.

chrishtr · 2017-11-10T23:44:14Z

source

+     asynchronously to avoid delaying presentation of other content.
+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-auto">auto</code></dfn>
+     <td>Indicates no preference in decoding mode, which also servces as the default value if the


smfr · 2017-11-11T00:15:26Z

source

@@ -25473,6 +25475,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  cannot process images or who have image loading disabled (i.e. it is the <code>img</code>
  element's <span>fallback content</span>).</p>

+  <p>Image data can appear in encoded form in order to reduce file size.


"can appear" is weird (it sounds like it's about putting something on the screen). Maybe "Image data is usually encoded to reduce file size".

smfr · 2017-11-11T00:15:46Z

source

@@ -25473,6 +25475,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  cannot process images or who have image loading disabled (i.e. it is the <code>img</code>
  element's <span>fallback content</span>).</p>

+  <p>Image data can appear in encoded form in order to reduce file size.
+  <dfn data-x="img-decoding-process">Decoding</dfn> is a process which converts image's media data


"is the process"

smfr · 2017-11-11T00:17:32Z

source

+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-sync">sync</code></dfn>
+     <td>Inidicates a preference to <span data-x="img-decoding-process">decode</span> this image
+     synchronosly for atomic presentation with other content.


This doesn't say what "synchronous" is relative to. I think the spec needs to say something to the effect of "synchronous" meaning that the image will get painted immediately after the image data is available (i.e. the load event firing).

smfr · 2017-11-11T00:19:25Z

source

   <dt><var>image</var> . <code subdfn data-x="dom-img-decode">decode</code>()</dt>

   <dd>

-    <p>Images usually exist in some encoded form; user agents need to decode them into raw pixels
+    <p>Images usually exist in some encoded form; user agents need to


Odd that this wording is similar to, but not exactly the same as the wording above.

I've simplified this part and referenced the decoding images section.

domenic

Great start :)

domenic · 2017-11-12T00:24:09Z

source

@@ -25511,6 +25517,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  <span>referrer policy attribute</span>. Its purpose is to set the <span>referrer policy</span>
  used when <span data-x="concept-fetch">fetching</span> the image. <ref spec=REFERRERPOLICY></p>

+  <p>The <dfn data-x="attr-img-decodingmode"><code>decodingmode</code></dfn> attribute which, when
+  set from a list of possible <span>image decoding mode hints</span>, indicates the preferred
+  decoding mode to use for this image.</p>


This needs to match other enumerated content attribute definitions more closely. In particular, it needs to state the states, keywords, invalid value default, and missing value default. See for example https://html.spec.whatwg.org/#the-dir-attribute or https://html.spec.whatwg.org/#attr-track-kind. You have some of this below, but it belongs with the content attribute, not the IDL attribute.

Gotcha. Done.

domenic · 2017-11-12T00:26:13Z

source

@@ -25473,6 +25475,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  cannot process images or who have image loading disabled (i.e. it is the <code>img</code>
  element's <span>fallback content</span>).</p>

+  <p>Image data can appear in encoded form in order to reduce file size.


As noted below, this paragraph should be inside the image processing model section.

domenic · 2017-11-12T00:28:33Z

source

+  <code data-x="attr-img-decodingmode-sync">sync</code> or
+  <code data-x="attr-img-decodingmode-async">async</code>. If the value specified is
+  <code data-x="attr-img-decodingmode-auto">auto</code>, then the user-agent is free to choose any
+  decoding behavior.</p>


This should not be located here; instead it should be in the normative section on the image processing model, along with the definition of decoding. I would create a new section, "Decoding images", before "Updating the image data", which explains that decoding is usually part of how the user agent presents the image, but can happen at various times depending on user agent heuristics, the value of the decodingmode attribute, and the decode() method.

Good idea, done! I wasn't sure if I should move the decode() description there as well listing all of the steps that take place when decode is invoked. WDYT?

domenic · 2017-11-12T00:29:29Z

source

+
+   <dd>
+
+     <p>Returns the image's preferred <span data-x="dom-img-decodingMode">decoding mode</span>.</p>


dom-img-decodingMode is just an IDL attribute. You should instead link to "image decoding mode hint".

domenic · 2017-11-12T00:30:18Z

source

+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-sync">sync</code></dfn>
+     <td>Inidicates a preference to <span data-x="img-decoding-process">decode</span> this image
+     synchronosly for atomic presentation with other content.


Yes, any ideas for how to make this clearer would be appreciated, perhaps by integrating with the img processing model. (Although that seems to have very little about actually presenting on the screen.)

domenic · 2017-11-12T00:31:00Z

source

+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-async">async</code></dfn>
+     <td>Indicates a preference to <span data-x="img-decoding-process">decode</span> this image
+     asynchronously to avoid delaying presentation of other content.


Using "asynchronously" (and "synchronously") is kind of begging the question, here. Can we do better? Either describing the implementation strategy, or the user-facing effects, or both?

Added some description in a separate section.

domenic · 2017-11-12T00:31:55Z

source

+     <th> <code data-x="">decodingmode</code>
+     <td> <code data-x="attr-img-decodingmode">img</code>;
+     <td> Decoding mode hint to use when processing this image for presentation.
+     <td> Comma-separated list of <span data-x="image decoding mode hints">image decoding mode hints</span>


I don't think it's a comma-separated list...

Following the example of other enumerated attributes, this should be a list of strings each of which link to the keyword definition. See e.g. the dir attribute in this same table.

domenic · 2017-11-12T00:33:25Z

source

@@ -117474,6 +117524,11 @@ interface <dfn>External</dfn> {
          <span>valid non-negative integer</span>, or
          <span>valid duration string</span>
    <tr>
+     <th> <code data-x="">decodingmode</code>
+     <td> <code data-x="attr-img-decodingmode">img</code>;
+     <td> Decoding mode hint to use when processing this image for presentation.


No period. Probably something like "Decoding mode hint for the image" where "Decoding mode hint" links to the image decoding mode hints concept.

vmpstr

I've renamed the attribute to "decoding" as per conversation in "decode" attribute on <img> #1920. I think it makes sense, but that's an easy thing to change if we change our mind.
I've also added Decoding images section as a place to describe some of the processes that happen in the user agent and to define what synchronous and asynchronous mean.

vmpstr · 2017-11-13T19:28:18Z

source

@@ -25473,6 +25475,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  cannot process images or who have image loading disabled (i.e. it is the <code>img</code>
  element's <span>fallback content</span>).</p>

+  <p>Image data can appear in encoded form in order to reduce file size.
+  <dfn data-x="img-decoding-process">Decoding</dfn> is a process which converts image's media data


vmpstr · 2017-11-13T19:28:38Z

source

@@ -25511,6 +25517,10 @@ interface <dfn>HTMLImageElement</dfn> : <span>HTMLElement</span> {
  <span>referrer policy attribute</span>. Its purpose is to set the <span>referrer policy</span>
  used when <span data-x="concept-fetch">fetching</span> the image. <ref spec=REFERRERPOLICY></p>

+  <p>The <dfn data-x="attr-img-decodingmode"><code>decodingmode</code></dfn> attribute which, when
+  set from a list of possible <span>image decoding mode hints</span>, indicates the preferred
+  decoding mode to use for this image.</p>


Gotcha. Done.

vmpstr · 2017-11-13T19:28:44Z

source

+   <tbody>
+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-sync">sync</code></dfn>
+     <td>Inidicates a preference to <span data-x="img-decoding-process">decode</span> this image


vmpstr · 2017-11-13T19:29:36Z

source

+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-sync">sync</code></dfn>
+     <td>Inidicates a preference to <span data-x="img-decoding-process">decode</span> this image
+     synchronosly for atomic presentation with other content.


I've added a section on Decoding images (as @domenic recommended). I appreciate any suggestions on how to phrase that section better.

vmpstr · 2017-11-13T19:29:53Z

source

+    <tr>
+     <td><dfn><code data-x="attr-img-decodingmode-async">async</code></dfn>
+     <td>Indicates a preference to <span data-x="img-decoding-process">decode</span> this image
+     asynchronously to avoid delaying presentation of other content.


Added some description in a separate section.

vmpstr · 2017-11-13T19:30:05Z

source

+  <code data-x="attr-img-decodingmode-sync">sync</code> or
+  <code data-x="attr-img-decodingmode-async">async</code>. If the value specified is
+  <code data-x="attr-img-decodingmode-auto">auto</code>, then the user-agent is free to choose any
+  decoding behavior.</p>


Good idea, done! I wasn't sure if I should move the decode() description there as well listing all of the steps that take place when decode is invoked. WDYT?

vmpstr · 2017-11-13T19:36:32Z

source

+
+   <dd>
+
+     <p>Returns the image's preferred <span data-x="dom-img-decodingMode">decoding mode</span>.</p>


vmpstr · 2017-11-13T19:37:07Z

source

@@ -117474,6 +117524,11 @@ interface <dfn>External</dfn> {
          <span>valid non-negative integer</span>, or
          <span>valid duration string</span>
    <tr>
+     <th> <code data-x="">decodingmode</code>
+     <td> <code data-x="attr-img-decodingmode">img</code>;
+     <td> Decoding mode hint to use when processing this image for presentation.


vmpstr · 2017-11-13T19:37:33Z

source

   <dt><var>image</var> . <code subdfn data-x="dom-img-decode">decode</code>()</dt>

   <dd>

-    <p>Images usually exist in some encoded form; user agents need to decode them into raw pixels
+    <p>Images usually exist in some encoded form; user agents need to


I've simplified this part and referenced the decoding images section.

vmpstr · 2017-11-13T19:37:40Z

source

+     <th> <code data-x="">decodingmode</code>
+     <td> <code data-x="attr-img-decodingmode">img</code>;
+     <td> Decoding mode hint to use when processing this image for presentation.
+     <td> Comma-separated list of <span data-x="image decoding mode hints">image decoding mode hints</span>


chrishtr · 2017-11-14T19:50:22Z

Latest patchset LGTM.

domenic · 2017-11-16T23:16:46Z

source

+     attribute is not specified.
+  </table>
+
+  <p class="note">The user agent <!--non-normative-->should respect the value of the


This doesn't seem like a non-normative note. You know something's going wrong if you're overriding the linter...

I'll fix.

domenic · 2017-11-16T23:22:11Z

source

+  <p class="note">If the <code data-x="attr-img-decoding">decoding</code> attribute is missing, its
+  value is not specified, or the value specified is not one of the above listed values, then the
+  effect is equivalent to specifying the <code data-x="attr-img-decoding">decoding</code>
+  attribute with <code data-x="attr-img-decoding-auto">auto</code> value.</p>


You need to normatively specify this using invalid and missing value defaults. I will fix.

domenic · 2017-11-16T23:34:09Z

I pushed a commit with a few more tweaks. This now looks good to me :).

As for tests, the only ones I can think of are updating https://github.com/w3c/web-platform-tests/blob/11abef74e5f06c23176ab2f492c71c9a33250742/html/dom/elements-embedded.js#L3 . If we can get those updated, either as a WPT PR or via Chromium's auto-export, I'm ready to merge!

domenic · 2017-11-16T23:36:01Z

Hmm I suppose we should get Apple's sign-off on the name. Let's have that discussion back in #1920, I guess? Marking "do not merge yet" just in case.

annevk

It seems there's again some unrelated commits in this PR. (I suspect they might all be related to the little incident we had on master yesterday...)

annevk · 2017-11-17T13:12:31Z

source

+   <tbody>
+    <tr>
+     <td><dfn><code data-x="attr-img-decoding-sync">sync</code></dfn>
+     <td><dfn data-x="attr-img-decoding-sync-state">Sync</dfn>


Should we spell these out as "Synchronous" and "Asynchronous"? We typically don't abbreviate states I think.

I find the whole states vs. keywords distinction already a bit hard to deal with, so I try for new stuff to keep them as close as possible. But I don't have a principled suggestion going forward, so happy to change it if you prefer.

annevk

As far as I can tell you haven't updated the Index for the img element.

Other review comments.

domenic · 2017-11-17T16:31:57Z

Fixed the extra commit; that was my bad. All of yesterday I had an extra commit floating around on master messing up all my rebases. Also updated the element index. (It'd be so nice if those were generated...)

annevk · 2017-11-19T13:40:52Z

source

@@ -115885,6 +115963,7 @@ interface <dfn>External</dfn> {
         <code data-x="attr-img-ismap">ismap</code>;
         <code data-x="attr-dim-width">width</code>;
         <code data-x="attr-dim-height">height</code>;
+         <code data-x="attr-img-decoding">decoding</code>;


Nit: this should come after referrerpolicy if we want to retain the order we started in the img element section.

annevk · 2017-11-19T14:03:40Z

Adding "do not merge yet" back as the discussion in the corresponding issue hasn't quite completed. Note that this is also still blocked on new web-platform-tests tests.

vmpstr · 2017-11-22T19:00:54Z

Adding "do not merge yet" back as the discussion in the corresponding issue hasn't quite completed. Note that this is also still blocked on new web-platform-tests tests.

As an update, I think the discussion on #1920 has concluded with the consensus that we should name the attribute decoding (as is in the latest patchset here). I'm adding web platform tests in here: web-platform-tests/wpt#8382 which is currently pending review.

domenic · 2017-11-27T17:39:40Z

Tests are arriving via https://chromium-review.googlesource.com/c/chromium/src/+/770106.

@annevk, @dbaron, do you feel that the concerns raised that you linked to have been addressed or no? They seem addressed to me, but I want to be sure before merging...

annevk · 2017-12-05T10:23:53Z

@vmpstr would you be willing to file browser bugs? I'm guessing that nobody but Chrome has this recorded thus far.

vmpstr · 2017-12-05T19:03:58Z

@vmpstr would you be willing to file browser bugs? I'm guessing that nobody but Chrome has this recorded thus far.

Yes, I'll file some bugs.

smfr · 2017-12-05T19:19:28Z

WebKit has https://bugs.webkit.org/show_bug.cgi?id=179432

vmpstr · 2017-12-05T19:32:42Z

Firefox: https://bugzilla.mozilla.org/show_bug.cgi?id=1423318

vmpstr · 2017-12-05T19:39:46Z

Edge: https://developer.microsoft.com/en-us/microsoft-edge/platform/issues/14965992/

vmpstr · 2017-12-05T19:40:36Z

Chrome: crbug.com/772470

annevk · 2017-12-06T07:48:50Z

Thanks! Kudos on getting this done. (By the way, are you mentioned in the Acknowledgments section already?)

vmpstr · 2017-12-06T18:18:11Z

By the way, are you mentioned in the Acknowledgments section already?

Yeah I'm there :) Thanks!

sideshowbarker · 2017-12-06T23:17:42Z

Added https://wpdev.uservoice.com/forums/257854-microsoft-edge-developer/suggestions/32541025-img-decoding-attribute

https://bugs.webkit.org/show_bug.cgi?id=179432 Patch by Said Abou-Hallawa <[email protected]> on 2017-12-06 Reviewed by Darin Adler. Source/WebCore: Support the HTMLElementImage decoding attribute to allow controlling the decoding of an image as per: whatwg/html#3221 https://whatpr.org/html/3221/images.html#decoding-images Tests: fast/images/decode-decoding-attribute-async-large-image.html fast/images/decoding-attribute-async-small-image.html fast/images/decoding-attribute-dynamic-async-small-image.html fast/images/decoding-attribute-sync-large-image.html * html/HTMLAttributeNames.in: * html/HTMLImageElement.cpp: (WebCore::HTMLImageElement::decodingMode const): * html/HTMLImageElement.h: * html/HTMLImageElement.idl: * platform/graphics/BitmapImage.cpp: (WebCore::BitmapImage::draw): (WebCore::BitmapImage::internalStartAnimation): (WebCore::BitmapImage::advanceAnimation): (WebCore::BitmapImage::internalAdvanceAnimation): * platform/graphics/DecodingOptions.h: (WebCore::DecodingOptions::DecodingOptions): (WebCore::DecodingOptions::isAuto const): (WebCore::DecodingOptions::isAsynchronousCompatibleWith const): (WebCore::DecodingOptions::isNone const): Deleted. * platform/graphics/ImageDecoder.h: * platform/graphics/ImageFrame.cpp: (WebCore::ImageFrame::clearImage): * platform/graphics/ImageSource.cpp: (WebCore::ImageSource::frameAtIndexCacheIfNeeded): * platform/graphics/avfoundation/objc/ImageDecoderAVFObjC.h: * platform/graphics/cg/ImageDecoderCG.h: * platform/graphics/win/ImageDecoderDirect2D.h: * platform/image-decoders/ScalableImageDecoder.h: * rendering/RenderBoxModelObject.cpp: (WebCore::RenderBoxModelObject::decodingModeForImageDraw const): The element attributes and the document settings should be checked before checking our heuristics. And since the "decoding" attribute is per an image and isLargeImageAsyncDecodingEnabledForTesting() is a global setting, the decoding attribute should be checked first. LayoutTests: * fast/images/async-attribute-with-small-image-expected.html: Removed. * fast/images/async-attribute-with-small-image.html: Removed. * fast/images/decode-decoding-attribute-async-large-image-expected.html: Added. * fast/images/decode-decoding-attribute-async-large-image.html: Added. * fast/images/decoding-attribute-async-small-image-expected.html: Added. * fast/images/decoding-attribute-async-small-image.html: Added. * fast/images/decoding-attribute-dynamic-async-small-image-expected.html: Added. * fast/images/decoding-attribute-dynamic-async-small-image.html: Added. * fast/images/decoding-attribute-sync-large-image-expected.html: Added. * fast/images/decoding-attribute-sync-large-image.html: Added. git-svn-id: http://svn.webkit.org/repository/webkit/trunk@225616 268f45cc-cd09-0410-ab3c-d52691b4dbfc

foolip · 2017-12-14T13:50:21Z

@tobie, https://whatpr.org/html/3221/8b055c0...3f6a0fe/images.html is b0rked, can you investigate?

tobie · 2017-12-14T14:03:35Z

@foolip it's an issue with the <pattern> element which isn't recognized by Tidy on htmldiff. I emailed @dontcallmedom to see if he could special-case it on his side.

tobie · 2017-12-14T14:37:10Z

Fixed on the htmldiff side by @dontcallmedom. Now restarting the build and cleaning the caches on the pr-preview side.

tobie · 2017-12-14T14:40:59Z

And now fixed.

foolip · 2017-12-14T16:35:25Z

Awesome, thanks!

…z,tnikkel This adds support for HTMLImageElement's decoding attribute, as described by: whatwg/html#3221 https://whatpr.org/html/3221/images.html#decoding-images It also exposes the same attribute on SVGImageElement, just as Blink has chosen to do so.

This adds support for HTMLImageElement's decoding attribute, as described by: whatwg/html#3221 https://whatpr.org/html/3221/images.html#decoding-images

…z,tnikkel This adds support for HTMLImageElement's decoding attribute, as described by: whatwg/html#3221 https://whatpr.org/html/3221/images.html#decoding-images It also exposes the same attribute on SVGImageElement, just as Blink has chosen to do so. UltraBlame original commit: 630049a9ac3b24ec0a141669937df372373a5d6f

chrishtr reviewed Nov 10, 2017

View reviewed changes

smfr reviewed Nov 11, 2017

View reviewed changes

domenic requested changes Nov 12, 2017

View reviewed changes

vmpstr commented Nov 13, 2017

View reviewed changes

vmpstr changed the title ~~Add img.decodingMode attribute~~ Add img.decoding attribute Nov 14, 2017

domenic reviewed Nov 16, 2017

View reviewed changes

domenic approved these changes Nov 16, 2017

View reviewed changes

domenic added addition/proposal New features or enhancements topic: img needs tests Moving the issue forward requires someone to write tests labels Nov 16, 2017

domenic added the do not merge yet Pull request must not be merged per rationale in comment label Nov 16, 2017

domenic mentioned this pull request Nov 16, 2017

"decode" attribute on <img> #1920

Closed

domenic removed the do not merge yet Pull request must not be merged per rationale in comment label Nov 16, 2017

annevk reviewed Nov 17, 2017

View reviewed changes

annevk requested changes Nov 17, 2017

View reviewed changes

vmpstr and others added 4 commits November 17, 2017 11:28

add decodingmode img attr

da51fd2

Added image decoding section.

d66f6ce

Other review comments.

Minor tweaks

a7dc0f8

Update element index with new attribute

3f6a0fe

annevk approved these changes Nov 19, 2017

View reviewed changes

annevk added the do not merge yet Pull request must not be merged per rationale in comment label Nov 19, 2017

chrishtr mentioned this pull request Nov 20, 2017

Use spec-compliant async decoding API ampproject/amphtml#12131

Closed

vmpstr mentioned this pull request Dec 1, 2017

Img decoding attribute w3ctag/design-reviews#220

Closed

3 tasks

annevk merged commit 05525c8 into whatwg:master Dec 5, 2017

domenic changed the title ~~Add img.decoding attribute~~ Add <img> decoding="" attribute Dec 5, 2017

fathyb mentioned this pull request Mar 22, 2018

Add the decoding property to Image microsoft/TypeScript-DOM-lib-generator#392

Closed

xxczaki mentioned this pull request Jul 23, 2019

Add decoding prop to gatsby-image img element gatsbyjs/gatsby#16005

Closed


		<dd>

		<p>Returns the image's preferred <span data-x="dom-img-decodingMode">decoding mode</span>.</p>

Add <img> decoding="" attribute #3221

Add <img> decoding="" attribute #3221

Conversation

vmpstr commented Nov 10, 2017 • edited by pr-preview bot Loading

vmpstr commented Nov 10, 2017

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

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

domenic 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

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

vmpstr 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

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

chrishtr commented Nov 14, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

domenic commented Nov 16, 2017

domenic commented Nov 16, 2017

annevk left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

annevk left a comment

Choose a reason for hiding this comment

domenic commented Nov 17, 2017

Choose a reason for hiding this comment

annevk commented Nov 19, 2017

vmpstr commented Nov 22, 2017

domenic commented Nov 27, 2017 • edited Loading

annevk commented Dec 5, 2017

vmpstr commented Dec 5, 2017

smfr commented Dec 5, 2017

vmpstr commented Dec 5, 2017

vmpstr commented Dec 5, 2017

vmpstr commented Dec 5, 2017

annevk commented Dec 6, 2017

vmpstr commented Dec 6, 2017

sideshowbarker commented Dec 6, 2017

foolip commented Dec 14, 2017

tobie commented Dec 14, 2017

tobie commented Dec 14, 2017

tobie commented Dec 14, 2017

foolip commented Dec 14, 2017

vmpstr commented Nov 10, 2017 •

edited by pr-preview bot

Loading

domenic commented Nov 27, 2017 •

edited

Loading