Global variables g:db_ui_* VS g:dbui_* are misleading

[bew]

Hello,
I think that the use of variables g:db_ui_* (for config) & g:dbui_* (in the plugin) are misleading, what I mean is that when I look at the source code of the plugin and find a g: variable, and I try to override it in my config, it does nothing.

For example, I wanted to disable to execution on write. Since I didn't see the option documented in the top readme (and didn't thought at checking the :h pages at that time, and went to check the source code and found

vim-dadbod-ui/autoload/db_ui/query.vim

Lines 184 to 186 in ed73c98

	if g:dbui_execute_on_save && is_sql
	autocmd BufWritePost <buffer> nested call s:method('execute_query')
	endif

So I figured that setting g:dbui_execute_on_save to v:false (or 0) would disable this behavior.
Unfortunately, that variable is overwritten by

vim-dadbod-ui/plugin/db_ui.vim

Line 19 in a0045ae

let g:dbui_execute_on_save = get(g:, 'db_ui_execute_on_save', 1)

which takes its value from g:db_ui_execute_on_save instead.

Why did you do this distinction? Would it be possible to keep a single source of truth, and maybe change all g:dbui_* to g:db_ui_* with default if they are not already set?

[kristijanhusak]

g:db_ui vars are for the user to change, where g:dbui is for internal purposes. I think I had some issues with tests or something, can't really remember at this moment.
I agree it's a bit misleading, but didn't really think about it since I presume that who checks the source code, can figure it out :)

Yeah, it could be unified, why not.

[kristijanhusak]

I updated this. Names are now unified. Thanks for reporting.