[mapserver-users] Runaway process and other wierdness (fwd)

[Tim Sutton]

Hi

Just an update on the image caching problem and the runaway thread
problem. I found a copy of the nighly build from 30/08/2001 and the
runaway thread problem does not occur with this version. I applied the
patch provided below and was able to properly test if it resolves the
problem. It seems that the problem of random cached images being
displayed in the map window has _not_ been resolved.

The problem can be seen on the gmap demo I have running on my web
server :

http://www.giug.net/~sutton_t/gmap/htdocs/gmap75.phtml

If you zoom in a bit and pan around a bit, the images will not display
consistently.

Any help to resolve this will be greatly appreciated. The 3.5. nightly
mentioned above works fine otherwise (for my needs anyway), I just
need to resolve the image caching problem.

Many thanks

Tim Sutton

On Mon, 29 Oct 2001, Daniel Morissette wrote:

> Bruno Gendron wrote:
> >
> > I will keep you informed with my experiments,
> > particularly on the unique filename issue!
> >
>
> Brunu, Tim,
>
> About the random map image problem, can you please try the following
> change and tell me if that makes a difference?
>
> --- php_mapscript.c     2001/10/17 21:40:50     1.59
> +++ php_mapscript.c     2001/10/29 15:23:15
> @@ -351,7 +351,7 @@
>  static int le_mslegend;
>
>  static char tmpId[128]; /* big enough for time + pid */
> -static int  tmpCount;
> +static int  tmpCount = 0;
>
>  /* -------------------------------------------------------------------- */
>  /*      class entries.                                                  */
> @@ -736,7 +736,7 @@
>
>      /* We'll use tmpId and tmpCount to generate unique filenames */
>      sprintf(PHPMS_GLOBAL(tmpId), "%ld%d",(long)time(NULL),(int)getpid());
> -    tmpCount = 0;
> +/*    tmpCount = 0; */
>
>  #ifdef PHP4
>      INIT_CLASS_ENTRY(tmp_class_entry, "map", php_map_class_functions);
>
>
>
>
>