[Webkit-unassigned] [Bug 202033] New: [GTK] Granularize Undo stack

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=202033

            Bug ID: 202033
           Summary: [GTK] Granularize Undo stack
           Product: WebKit
           Version: Other
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: WebKitGTK
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: mcrha at redhat.com
                CC: bugs-noreply at webkitgtk.org

With a move to JSC API I'm also considering depending on WebKit's Undo stack, instead of having my own undo/redo manager, which adds maintenance burden and which probably wastes memory, do to having two undo managers in parallel. It has its pros and cons, obviously. I've been trying this in the MiniBrowser of the 2.26.0 release and there are some obvious issues with it (I've been clicking the undo/redo buttons at the toolbar). For the initial issues:

a) there should be added granularity, because when I type several paragraphs, then the Undo deletes all of them at once, like KB-s of text, which is useless; it should rather remember by a letter and after space/enter it can join the sequence into one word;

b) backspace is not handled properly - when I type some text, then press backspace twice, then it does not undelete the text, it removes the text I wrote before pressing the backspace instead;

c) probably related to b), when I type some text, then Ctrl+A followed by Delete, then type a new text, then press Undo, then I end up not with a blank page, but in a page where the previously deleted text is shown and it is selected (like after Ctrl+A and before the Delete press).

I can give this some extensive testing, if you'd like to, even for any proposed fixes, but I do not know whether you've time to look on this. Neither I like lengthy bug reports with semi-mixed things, which this one could end up in easily, thus I'll rather wait for you on this.

The bug #88595 would be also quite interesting to have, but, well, it's an old one, which may make this one being such as well.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20190920/45a1d248/attachment.html>