Re: [code] [SCINTERM] API documentation mistake and Scintilla patch

From: Mitchell <m.att.foicica.com>
Date: Tue, 3 Mar 2015 09:22:42 -0500 (EST)

Hi Robin,

On Tue, 3 Mar 2015, Robin Haberkorn wrote:

> Hello,
>
> the Scinterm documentation contains a severe mistake. It is said that
>
> "scintilla_delete (sci)
>
> Deletes the given Scintilla window. This function does not delete the
> curses WINDOW associated with it. You will have to delete the WINDOW
> manually."
>
> This is wrong. Scinterm does call delwin() on the WINDOW. But for some
> strange reason it never seems to do any harm on NCurses. I had this
> wrong (deleted the WINDOW before calling scintilla_delete()) in SciTECO
> for a long time - probably double-freeing the Curses windows thousands
> of times. But when I tried compiling on MinGW/PDCurses again, it
> crashed immediately.

Yes, you are right. I think in an earlier version of Scinterm the
documentation was accurate, but certainly not now. Thanks for catching
this. I'll update the documentation.

> Also I fixed a bug in Scintilla that's related to Scinterm. You may
> have found that hard tabs are laid out wrong in Scintilla/Scinterm. I
> painstakingly traced this down in Scintilla (see attachment). It has to
> be discussed with the Scintilla maintainers, as the padding applied in
> the calculation may have a use in graphical UIs. Couldn't find any
> explanation for it in the Scintilla commit history. But for the time
> being you can apply it as a patch on the Scintilla code tree like I do.

Thanks for taking the time to track this down. I know how difficult it can
be sometimes. I'll try to transform your patch into something acceptable
upstream and let you know what happens.

Cheers,
Mitchell

-- 
You are subscribed to code.att.foicica.com.
To change subscription settings, send an e-mail to code+help.att.foicica.com.
To unsubscribe, send an e-mail to code+unsubscribe.att.foicica.com.
Received on Tue 03 Mar 2015 - 09:22:42 EST

This archive was generated by hypermail 2.2.0 : Wed 04 Mar 2015 - 06:45:58 EST