Partial patch for block caret display

[John Ehresman]

Hi,

I want a block caret that displays over the character at the end of a
selection like the cursor in vi does. I've attached a proof of concept
patch. Two questions: is this the best way to do this? And, should I
add a new caret style for this behavior or just modify how the existing
CARETSTYLE_BLOCK works?

Thanks,

John

[Neil Hodgson]

John Ehresman:

> I want a block caret that displays over the character at the end of a selection like the cursor in vi does. I've attached a proof of concept patch. Two questions: is this the best way to do this?

It looks reasonable to me. The drag state looks to be taken care of. Inverted range (caret before anchor) should be OK.

Well, maybe a comment: // Show block caret on last character of non-empty selection

> And, should I add a new caret style for this behavior or just modify how the existing CARETSTYLE_BLOCK works?

That will depend on whether the other users of CARETSTYLE_BLOCK always want this behaviour.

Neil

[Nicholai Benalal]

On 20 December 2016 at 23:23:52 +01:00, Neil Hodgson <nyama...@me.com> wrote:

And, should I add a new caret style for this behavior or just modify how the existing CARETSTYLE_BLOCK works?

That will depend on whether the other users of CARETSTYLE_BLOCK always want this behaviour.

This patch looks very good and in my opinion, there's no need for an extra caret style as I think the modified behaviour is what most people would expect from a caret mode. Actually, I raised this back in 2011 but failed to get much traction ;-). Possibly because I failed in explaining the issue properly.

https://groups.google.com/forum/#!topic/scintilla-interest/MH46tzfw5qw

[John Ehresman]

I can't think of a good rationale for the old behavior, but it is a
change and not 100% backwards compatible.

Thanks,

John

[Neil Hodgson]

The patch to display the block caret on the last character of the selection was committed as
https://sourceforge.net/p/scintilla/code/ci/d29f1dad85ac3ec2a57c9341c5bd62a05c07d1e0/
If any one wants the previous behaviour to remain possible with this as an option then they should say so.

Neil

[Matthew Brush]

Why should the block caret behave different from the I-beam caret? If
it's not just a different presentation, it should have a separate
setting to make it behave differently rather than breaking
normal/existing behaviour, IMO.

Related:
https://sourceforge.net/p/scintilla/bugs/1924/
https://github.com/geany/geany/issues/1442

Regards,
Matthew Brush

[Semyon Filippov]

I definitely prefer the old behavior. Not only it was uniform with non-block caret style and some other text editors (where shift+arrow moves the cursor), but also it was easier to see that the last character is selected. Perhaps, the latter is a matter of habit, but I this it's a matter of user experience as well.

Semyon

[Neil Hodgson]

Matthew Brush:

> Why should the block caret behave different from the I-beam caret? If it's not just a different presentation, it should have a separate setting to make it behave differently rather than breaking normal/existing behaviour, IMO.

It would be reasonable to implement another choice for SCI_SETCARETSTYLE if any of the interested parties want to work on this. Something like CARETSTYLE_BLOCKATEND.

Neil

[Semyon Filippov]

On Saturday, March 25, 2017 at 5:29:53 AM UTC+7, Neil Hodgson wrote:

    It would be reasonable to implement another choice for SCI_SETCARETSTYLE if any of the interested parties want to work on this. Something like CARETSTYLE_BLOCKATEND.

   Neil

I'm interested in such choice, but I'm not skilled at C++, sorry. I would be very thankful if someone else would implement this )

Semyon