Re: [vi-mode] widgets for case manipulation: `gU` and `U` in visual mode

[Filipe Silva <filipe.silva@xxxxxxxxx>]

Oliver,


"If this is the problem, it has also been fixed in newer versions by
making it continue waiting for further keys if the keys typed so far,
such as y, correspond to a widget, such as vi-yank, that needs to wait
for a movement. This also allows tricks like
  bindkey -a -s 'cw' 'dwi'
to work much as the equivalent vi map."

What recent versions would those be? since 5.2?

cheers,

Filipe

On Wed, Jun 22, 2016 at 12:25 PM, Filipe Silva <filipe.silva@xxxxxxxxx>
wrote:

> Hum, I had this as the first lines of my plugin
>
> # no delays for mode switching.
> export KEYTIMEOUT=1
>
>
> https://github.com/ninrod/nin-vi-mode/blob/master/nin-vi-mode.plugin.zsh#L1-L2
>
> I guess I would have to type really fast for `ys` to work with this
> configuration.
>
>
> On Wed, Jun 22, 2016 at 12:09 PM, Oliver Kiddle <okiddle@xxxxxxxxxxx>
> wrote:
>
>> Filipe Silva wrote:
>> > for instance, the `ys` chord is not being picked up.
>> >
>> > supose i have a word: `abcde` and the cursor is in `c`. if I'm in
>> > normal mode and type `ys` zle thinks that I've hit `s` and deletes the
>> > `c` character and puts me in insert mode.
>> >
>> > Maybe it's a bug?
>>
>> I think this occurs if you don't type ys fast enough. What have you got
>> $KEYTIMEOUT set to? Try typing the ys especially fast. And perhaps try
>> binding ys to something that will be obvious like kill-whole-line to see
>> if it is surround that is failing or the key binding.
>>
>> If this is the problem, it has also been fixed in newer versions by
>> making it continue waiting for further keys if the keys typed so far,
>> such as y, correspond to a widget, such as vi-yank, that needs to wait
>> for a movement. This also allows tricks like
>>   bindkey -a -s 'cw' 'dwi'
>> to work much as the equivalent vi map.
>>
>> Similarly, I can define
>>   bindkey -s -a "gUU" "gUgU"
>> but because vi-uppercase is not a builtin widget, I need to hit the keys
>> fast even in 5.2.
>>
>> > On Wed, Jun 22, 2016 at 10:19 AM, Filipe Silva <filipe.silva@xxxxxxxxx>
>> > wrote:
>> > > I've fixed the bug including the following condition:
>>
>> Thanks for that!
>>
>> Oliver
>>
>
>