[Plplot-devel] Core width patch

[Alan W. I. <ir...@be...>]

On Fri, 22 Feb 2002, Olof Svensson wrote:

>   (Alexandre [Gobbo] writes) The line in plcore:
>
>     if (width != plsc->width && width > 0) {
>
>   avoids you from setting width =0. But a width =0 has
>   meaning in windows and unix, so I changed into
>
>         if (width != plsc->width && width >= 0) {
>
>   width=0 is a thin line, more efficiently drawn than width =1

This is a (minor) API change so I thought this should be discussed on the
list before I make the change (and also change the documentation
appropriately).  I checked the code and apparently the setting width = 0
should cause no problems because that is what it is by default.  (The
PLStream memory area is malloced, then memset to zero.  BTW, couldn't this
be done with calloc?)

I didn't check in detail for every driver, but it looks like drivers that
worry about width set the minimum possible appropriate width for that driver
when they encounter width=0 which is how I will document this change if we
all agree to it.

Alan