W3C home > Mailing lists > Public > public-editing-tf@w3.org > February 2016

Re: CTRL+BACKSPACE or its equivalent behavior when deleting block elements

From: Johannes Wilm <mail@johanneswilm.org>
Date: Wed, 17 Feb 2016 04:17:14 +0100
Message-ID: <CABkgm-S+hMhmNKMHaBdaGpDTOoHSRLW_iKfh33mb3dyMGusLYQ@mail.gmail.com>
To: Grisha Lyukshin <glyuk@microsoft.com>
Cc: Chaals McCathie Nevile <chaals@yandex-team.ru>, "public-editing-tf@w3.org" <public-editing-tf@w3.org>
Delete by itself is an area of interest by itself, I think. Also, it fairly
large, because it means you will have to deal with cutting and merging
block elements (if the selection goes from block element to block element).
If you guys are up for it, I can go around to the JS projects and collect
everyone's most urgent delete issues.

On Tue, Feb 16, 2016 at 7:46 PM, Grisha Lyukshin <glyuk@microsoft.com>
wrote:

> I think it makes sense to have at the very least, very similar behavior
> before we even spec it.
>
>
> *I am open to discuss these behaviors now.*
>
>
> @Chrome/FF/Safari/Yandex what do you guys think about that? Would you be
> interested in cleaning up Delete behavior at the very least?
>
>
>
> Also, Johannes, if there are other interop issues that deal with
> *deletion* that could be revived, perhaps we could tackle them as well.
>
>
> As we discussed before, ExecCommands are complicated and there seems to
> be a split of opinions on whether they even need to be specced. And if we
> do decide to spec, I think, we will have few thoughts on how we can do.
>
>
> --grisha
>
>
>
>
>
> *From:* johanneswilm@gmail.com [mailto:johanneswilm@gmail.com] *On Behalf
> Of *Johannes Wilm
> *Sent:* Thursday, February 11, 2016 4:44 PM
> *To:* Grisha Lyukshin <glyuk@microsoft.com>
> *Cc:* Chaals McCathie Nevile <chaals@yandex-team.ru>;
> public-editing-tf@w3.org
>
> *Subject:* Re: CTRL+BACKSPACE or its equivalent behavior when deleting
> block elements
>
>
>
>
>
>
>
> On Thu, Feb 11, 2016 at 10:02 PM, Grisha Lyukshin <glyuk@microsoft.com>
> wrote:
>
> @Johannes: This came  up because we are revisiting  DELETE and FORWARD
> DELETE commands behavior in Edge. So I felt like bringing the findings to
> the editing community. I am not sure whether we should call this an attempt
> to standardize CE or not. At this point this is just pointing
> out non-interop and possibly, "buggy" behavior in browsers for different
> editing use cases.
>
>
>
> Ok, got it. Yes, feel free to send in pull requests for the execCommand
> spec where a lot of this will currently be described.
>
>
>
> My personal background on some of this: I and some other JS people filed
> bugs with Firefox and Chrome, some of which has to do with deletion of
> content, others with caret movement, etc. . At some stage we received the
> answer that these bugs would not be fixed as long as there was not a
> standard to work against.  That's how we ended up getting involved here. So
> I think several of us will be very interested in you guys making certain
> things (such as deletion in various contexts) behave the same in all the
> browsers even if we don't have a full spec. So let us know if this is now a
> general policy that either Edge or all the browsers are following.
>
>
>
> @Chaals: In Edge, all of it is removed in 6 steps:
>
>
>
> 1st CTRL+BACKSPACE moves the caret at the end of the line above
>
> 2nd CTRL+BACKSPACE  removes the image
>
> 3rd CTRL+BACKSPACE does nothing
>
> 4th CTRL+BACKSPACE moves the caret at the end of the line above
>
> 5th CTRL+BACKSPACE removes *sdsdsd*
>
> 6th CTRL+BACKSPACE removes *asdsad*
>
>
>
> The experience that I got in Chrome and FF on windows machine seems to
> differ from the one you described though:
>
>
>
> Firefox(43.0.4):
>
> 1st CTRL+BACKSPACE removes the image and visually places the caret at the
> beginning of the very first line.
>
> 2nd CTRL+BACKSPACE  removes *sdsdsd* and places caret in front of *asdsad*
>
> 3rd CTRL+BACKSPACE removes *asdsad*
>
>
>
> Chrome(48.0.2564.109 m)/Yandex:
>
> 1st CTRL+BACKSPACE removes the image and *sdsdsd*.
>
> 2nd CTRL+BACKSPACE  removes *asdsad*
>
>
>
> --grisha
>
>
>
> Sent from Outlook <http://aka.ms/weboutlook>
> ------------------------------
>
> *From:* johanneswilm@gmail.com <johanneswilm@gmail.com> on behalf of
> Johannes Wilm <johannes@fiduswriter.org>
> *Sent:* Thursday, February 11, 2016 5:13:42 AM
> *To:* Chaals McCathie Nevile
> *Cc:* public-editing-tf@w3.org
> *Subject:* Re: CTRL+BACKSPACE or its equivalent behavior when deleting
> block elements
>
>
>
> Hey,
>
> this sounds like a contenteditable=true issue. We originally moved on to
> work on other parts because the number of issues and inconsistencies of
> cE=true were so numerous that it did not seem solvable in the next decade.
>
>
>
> But now it sounds as if there is some interest in trying to standardize
> it. Is that correct?
>
>
>
> On Thu, Feb 11, 2016 at 12:01 PM, Chaals McCathie Nevile <
> chaals@yandex-team.ru> wrote:
>
> Hi,
>
> On Thu, 11 Feb 2016 03:38:33 +0100, Grisha Lyukshin <glyuk@microsoft.com>
> wrote:
>
> Hello All,
>
> There is a special behavior that is currently not interoperable between
> browsers for a specific use case below:
>
> I think we should only delete one block element at a time when
> CTRL+BACKSPACE is pressed.
>
> Chrome, FF and Edge all behave differently (sorry, I don't have Mac to
> test how Safari behaves).
>
> Repro steps:
>
> jsfiddle<https://jsfiddle.net/svn9Lztd/>
>
> Place the caret in front of "def" and press CTRL+BACKSPACE
>
>
> Opening this in Yandex.browser (blink-based), Safari and Firefox on mac,
> and pressing command-backspace, Firefox deleted everything in 4 steps, the
> other two in 5. (ctrl-backspace did nothing in FF and a normal backspace in
> the other two).
>
> It seems Firefox deletes the paragraph when it removes the last thing in
> it, whereas the others remove the text node in a separate operation. What
> happened on windows? (I don't have a windows machine here to test…)
>
> Cheers
>
> Chaals
>
> --
> Charles McCathie Nevile - web standards - CTO Office, Yandex
>  chaals@yandex-team.ru - - - Find more at http://yandex.com
>
>
>
>
>
> --
>
> Johannes Wilm
>
> Fidus Writer
>
> http://www.fiduswriter.org
>
>
>
>
>
> --
>
> Johannes Wilm
> http://www.johanneswilm.org
> tel: +1 (520) 399 8880
>



-- 
Johannes Wilm
http://www.johanneswilm.org
tel: +1 (520) 399 8880
Received on Wednesday, 17 February 2016 03:17:43 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 17 February 2016 03:17:44 UTC