[vim/vim] Vim9: first argument of "cursor()" cannot be of type "any" even when it's a number (#8610)

5 views
Skip to first unread message

lacygoill

unread,
Jul 22, 2021, 3:46:42 PM7/22/21
to vim/vim, Subscribed

Describe the bug

In Vim9 script, the first argument of cursor() cannot be of type any even when it's a number.

To Reproduce

Run this shell command:

vim -Nu NONE -S <(cat <<'EOF'
    vim9script
    g:lnum = 1
    def Func()
        cursor(g:lnum, 1)
    enddef
    defcompile
EOF
)

This error is raised:

E1013: Argument 1: type mismatch, expected number but got any

Expected behavior

No error is raised.

Environment

  • Vim version: 8.2 Included patches: 1-3203
  • OS: Ubuntu 20.04.2 LTS
  • Terminal: XTerm(353)

Additional context

Regression introduced in 8.2.3188.

At first, I wasn't sure it was a bug (see here for the rationale).

But since then, I've found out that the patch breaks too many functions in my config. I think any should be accepted at compile time.

If we use a global variable for the column index, rather than the line index, no error is raised:

vim9script
g:col = 1
def Func()
    cursor(1, g:col)
enddef
defcompile
echo 'no error'
no error

Which is inconsistent. Why an error for the first argument, but not the second one? They have the same type.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.

Yegappan Lakshmanan

unread,
Jul 23, 2021, 1:07:38 AM7/23/21
to vim_dev, reply+ACY5DGE64SWAR2TWLQ...@reply.github.com, vim/vim, Subscribed
Hi,

I have created PR 8611 with a fix for this issue along with other changes.

Regards,
Yegappan

vim-dev ML

unread,
Jul 23, 2021, 1:07:58 AM7/23/21
to vim/vim, vim-dev ML, Your activity

Hi,

On Thu, Jul 22, 2021 at 12:46 PM lacygoill ***@***.***>
wrote:

> *Describe the bug*

>
> In Vim9 script, the first argument of cursor() cannot be of type any even
> when it's a number.
>
> *To Reproduce*

>
> Run this shell command:
>
> vim -Nu NONE -S <(cat <<'EOF'
> vim9script
> g:lnum = 1
> def Func()
> cursor(g:lnum, 1)
> enddef
> defcompile
> EOF
> )
>
> This error is raised:
>
> E1013: Argument 1: type mismatch, expected number but got any
>
> *Expected behavior*
>
> No error is raised.
>
> *Environment*
>
> - Vim version: 8.2 Included patches: 1-3203
> - OS: Ubuntu 20.04.2 LTS
> - Terminal: XTerm(353)
>
> *Additional context*
>
> Regression introduced in 8.2.3188.
> ------------------------------

>
> At first, I wasn't sure it was a bug (see here for the rationale
> <https://github.com/vim/vim/issues/8602#issuecomment-885145463>).

>
> But since then, I've found out that the patch breaks too many functions in
> my config. I think any should be accepted at compile time.
> ------------------------------

>
> If we use a global variable for the column index, rather than the line
> index, no error is raised:
>
> vim9scriptg:col = 1def Func()
> cursor(1, g:col)enddefdefcompileecho 'no error'

>
> no error
>
> Which is inconsistent. Why an error for the first argument, but not the
> second one? They have the same type.
>
>
> I have created PR 8611 with a fix for this issue along with other changes.

Regards,
Yegappan

lacygoill

unread,
Jul 23, 2021, 2:44:04 PM7/23/21
to vim/vim, vim-dev ML, Comment

Probably fixed in 8.2.3206

—
You are receiving this because you commented.

lacygoill

unread,
Jul 23, 2021, 2:44:04 PM7/23/21
to vim/vim, vim-dev ML, Comment

Closed #8610.

—
You are receiving this because you commented.

Reply all
Reply to author
Forward
0 new messages