Re: pango patch -- fix some warnings, at least one bug

[Owen Taylor <otaylor redhat com>]

Darin Adler <darin bentspoon com> writes:

> I fixed a bunch of warnings. In one case, I had to do a bit of
> research, since the warning indicated a missing non-obvious line of
> code.
> 
> OK to commit?

Yes, looks good (with two suggestions below).

Thanks,
                                        Owen
 
> Index: modules/arabic/arabic-x.c
> ===================================================================
> RCS file: /cvs/gnome/pango/modules/arabic/arabic-x.c,v
> retrieving revision 1.21
> diff -p -u -r1.21 arabic-x.c
> --- modules/arabic/arabic-x.c	2001/06/25 15:09:39	1.21
> +++ modules/arabic/arabic-x.c	2001/07/11 23:39:32
> @@ -178,7 +178,7 @@ arabic_engine_shape (PangoFont        *f
>                        PangoGlyphString *glyphs)
>   {
>       PangoXSubfont   subfont;
> -    long            n_chars;
> +    int             n_chars;
>       int             i;
>       ArabicFontInfo *fs;
>       const char     *p;

I think it would be marginally better to change the argument to
arabic_reshape() instead.

> Index: modules/tamil/tamil-x.c
> ===================================================================
> RCS file: /cvs/gnome/pango/modules/tamil/tamil-x.c,v
> retrieving revision 1.13
> diff -p -u -r1.13 tamil-x.c
> --- modules/tamil/tamil-x.c	2001/06/14 20:38:22	1.13
> +++ modules/tamil/tamil-x.c	2001/07/11 23:39:33
> @@ -232,7 +232,7 @@ tamil_engine_shape (PangoFont        *fo
>     wc = (gunichar *)g_malloc (sizeof(gunichar) * n_chars * 2);
> 
>     p = text;
> -  prevchar = 0;complete = 1;/* One character look behind */
> +  prevchar = 0;currchar = 0;complete = 1;/* One character look behind */
>     n_glyph = 0;
>     cluster_start = text;

Could you do this as:

 gunichar currchar = 0;   /* Quiet compiler */

To indicate that the assignment is just a dummy to suppress a spurious
warning. (Initializing currchar to 0 doesn't actually make sense - 0
is not a legitimate value.)

> Index: pango/opentype/ftxgpos.c
> ===================================================================
> RCS file: /cvs/gnome/pango/pango/opentype/ftxgpos.c,v
> retrieving revision 1.1
> diff -p -u -r1.1 ftxgpos.c
> --- pango/opentype/ftxgpos.c	2000/12/20 04:41:36	1.1
> +++ pango/opentype/ftxgpos.c	2001/07/11 23:39:45
> @@ -873,6 +873,8 @@
> 
>   	if ( gpi->face->glyph->format != ft_glyph_format_outline )
>             return TTO_Err_Invalid_GPOS_SubTable;	
> +
> +	ap = an->af.af2.AnchorPoint;
>   	
>   	outline = gpi->face->glyph->outline;

Yep this looks right.