hull@hao.UUCP (Howard Hull) (03/27/86)
I purchased the Lattice TMU (Text Management Utilities) package, and decided
I would test it out on the corrections to the mandV2.01 sources that went to
the net in January. The TMU diff output below is for both CBM Lattice 3.02
and for Lattice 3.03 (under which RJ supposedly checked to see that the code
would compile). Mand was probably developed on yet another compiler at Amiga
(Greenhills/Sun?). Of course just checking a program to be sure it compiles
doesn't do much for the bugs; only serious testing will find those. And then
there is forethoughtless malice by the mailers...
I don't represent the list below as being complete, but it does address all
the bug fixes issued by various contributors to the net, and as such is a
summary of their work. I thank them for their mercy, without which I have
no particular capacity to fix the broken. The Lattice TMU diff gave quite
a bit of context, so in most cases I edited the output down to the essentials.
I usually appreciate more context than the Unix diff gives; however some of
what the Lattice diff put out may have to do with the way Amiga ED mangles
tab characters. I have not yet had time to study the Lattice diff options
(there may be one similar to Unix's -b option).
In the stuff below, anything prefixed by Lattice diff as a_ is the net version,
anything prefixed as b_ is for the 3.02 compiler, and c_ is for the 3.03 one.
TO TRANSFORM a_mand.c INTO b_mand.c ...
*** CHANGE 246 IN a_mand.c TO [246,247] IN b_mand.c ***
< fwrite(&start_r,sizeof(start_r),1,fp);
---------------
> fwrite(&start_r,sizeof(start_r),1,fp);
> fwrite(&end_r,sizeof(end_r),1,fp);
*** CHANGE 574 IN a_mand.c TO [575,580] IN b_mand.c ***
<
---------------
>
>fl_assign (var, val)
>float *var, val;
>{
>*var = val;
>}
*** CHANGE [657,661] IN a_mand.c TO [663,667] IN b_mand.c ***
< start_r.f = -2.85;
< end_r.f = 2.85;
< start_i.f = -2.05;
< end_i.f = 2.05;
< max_count = 15;
---------------
> fl_assign (&start_r.f, -2.85);
> fl_assign (&end_r.f, 2.85);
> fl_assign (&start_i.f, -2.05);
> fl_assign (&end_i.f, 2.05);
> max_count = 15;
*** CHANGE [666,669] IN a_mand.c TO [672,675] IN b_mand.c ***
< start_r.f = 1.703418;
< end_r.f = 2.016930;
< start_i.f = -0.169450;
< end_i.f = 0.169450;
---------------
> fl_assign(&start_r.f, 1.703418);
> fl_assign(&end_r.f, 2.016930);
> fl_assign(&start_i.f, -0.169450);
> fl_assign(&end_i.f, 0.169450);
*** CHANGE [674,677] IN a_mand.c TO [680,683] IN b_mand.c ***
< start_r.f = 1.937821;
< end_r.f = 1.945044;
< start_i.f = -0.00259;
< end_i.f = 0.00259;
---------------
> fl_assign(&start_r.f, 1.937821);
> fl_assign(&end_r.f, 1.945044);
> fl_assign(&start_i.f, -0.00259);
> fl_assign(&end_i.f, 0.00259);
*** CHANGE [682,685] IN a_mand.c TO [688,691] IN b_mand.c ***
< start_r.f = -0.294473;
< end_r.f = -0.288444;
< start_i.f = 0.012677;
< end_i.f = 0.014839;
---------------
> fl_assign(&start_r.f, -0.294473);
> fl_assign(&end_r.f, -0.288444);
> fl_assign(&start_i.f, 0.012677);
> fl_assign(&end_i.f, 0.014839);
*** CHANGE [691,694] IN a_mand.c TO [697,700] IN b_mand.c ***
< start_r.f = 0.225;
< end_r.f = 0.275;
< start_i.f = 0.8425;
< end_i.f = 0.8525;
---------------
> fl_assign(&start_r.f, 0.225);
> fl_assign(&end_r.f, 0.275);
> fl_assign(&start_i.f, 0.8425);
> fl_assign(&end_i.f, 0.8525);
*** CHANGE [700,703] IN a_mand.c TO [706,709] IN b_mand.c ***
< start_r.f = 0.115206;
< end_r.f = 0.168190;
< start_i.f = -1.036059;
< end_i.f = -0.985386;
---------------
> fl_assign(&start_r.f, 0.115206);
> fl_assign(&end_r.f, 0.168190);
> fl_assign(&start_i.f, -1.036059);
> fl_assign(&end_i.f, -0.985386);
/* mand.c 729 Warning 61: Undefined Structure Tag "Iconbase" (Harmless) */
TO TRANSFORM a_mand3.c INTO b_mand3.c ...
*** CHANGE 139 IN a_mand3.c TO [139,140] IN b_mand3.c ***
< ns.Height = 400;
---------------
> ns.Height = 400;
> ns.ViewModes |= INTERLACE;
TO TRANSFORM a_mand7.c INTO b_mand7.c ...
*** CHANGE [245,246] IN a_mand7.c TO [246,247] IN b_mand7.c ***
< if (PutDiskObject(filename, &ProjectObject) == FALSE)
< goto WRITE_ERROR;
---------------
>/* if (PutDiskObject(filename, &ProjectObject) == FALSE)
> goto WRITE_ERROR; */
[This next one isn't part of the code sent over the net. It is a tweaking
of the Lattice 3.02 file include/workbench/workbench.h to get it to deal more
rationally with enums and bit fields, a matter that had been fixed by 3.03.]
TO TRANSFORM a_workbench.h INTO b_workbench.h ...
*** CHANGE [33,35] IN a_workbench.h TO [33,40] IN b_workbench.h ***
<enum WBObjectType {
< WBDISK = 1, WBDRAWER, WBTOOL, WBPROJECT, WBGARBAGE, WBDEVICE, WBKICK
<};
---------------
>/*enum*/ typedef int WBObjectType;
>#define WBDISK 1
>#define WBDRAWER 2
>#define WBTOOL 3
>#define WBPROJECT 4
>#define WBGARBAGE 5
>#define WBDEVICE 6
>#define WBKICK 7
*** CHANGE [69] IN a_workbench.h TO [74] IN b_workbench.h ***
< enum WBObjectType do_Type;
---------------
> /*enum*/ WBObjectType do_Type;
*** CHANGE [94] IN a_workbench.h TO [99,100] IN b_workbench.h ***
<
---------------
> struct WBObject * wo_Parent;
> UBYTE WB_JUNK /*Place holder for Lattice 3.02, see correction below */
>/* Lattice 3.02 Doesn't like the bit fields below */
*** CHANGE [96,101] IN a_workbench.h TO [102,107] IN b_workbench.h ***
< UWORD wo_IconDisp:1; /* icon is currently in a window */
< UWORD wo_DrawerOpen:1;/* we're a drawer, and it is open */
< UWORD wo_Selected:1; /* our icon is selected */
< UWORD wo_Background:1;/* set if icon is in background */
<
< enum WBObjectType wo_Type; /* what flavor object is this? */
---------------
>/* UWORD wo_IconDisp:1; icon is currently in a window */
>/* UWORD wo_DrawerOpen:1; we're a drawer, and it is open */
>/* UWORD wo_Selected:1; our icon is selected */
>/* UWORD wo_Background:1; set if icon is in background */
>
> /*enum*/ WBObjectType wo_Type; /* what flavor object is this? */
3.02 Compiled Executable is 66280 bytes.
-------------------------------------------------------------------------------
TO TRANSFORM a_mand.h INTO c_mand.h ...
*** CHANGE [213,221] IN a_mand.h TO [213] IN c_mand.h ***
<
</* ======================================================================== */
<
</* This constant is for Intuition V1.1. If you are getting "redefinition"
< * warnings when you compile, it's because of this redeclaration.
< * Is harmless, and when we all get to V1.1 then we all can delete this.
< */
<#define INTUITICKS 0x00400000
<
---------------
>
TO TRANSFORM a_mand.c INTO c_mand.c ...
*** CHANGE [246,248] IN a_mand.c TO [246,249] IN c_mand.c ***
< fwrite(&start_r,sizeof(start_r),1,fp);
< fwrite(&start_i,sizeof(start_i),1,fp);
< fwrite(&end_i,sizeof(end_i),1,fp);
---------------
> fwrite(&start_r,sizeof(start_r),1,fp);
> fwrite(&end_r,sizeof(end_r),1,fp);
> fwrite(&start_i,sizeof(start_i),1,fp);
> fwrite(&end_i,sizeof(end_i),1,fp);
TO TRANSFORM a_mand3.c INTO c_mand3.c ...
*** CHANGE 139 IN a_mand3.c TO [139,140] IN c_mand3.c ***
< ns.Height = 400;
---------------
> ns.Height = 400;
> ns.ViewModes |= INTERLACE;
3.03 Compiled Executeable is 67920 bytes.
The following problems are still to be fixed:
- when executing Analize, one gets stuck and Exit from that screen does not
work. One has to reboot.
- when creating a 640x400 high-res image (mode 7), the image is created
correctly, but when one tries to invoke a menu the machine crashes with an
"out of Memory error" and Guru Meditation #04010003.4C500000
Marco Papa
-when recovering an image (saved by the "SA <filename>" command after closing
the display window) with the "L <filename>" command, issuing the "D" command
to re-display it will leave you stuck in the display window, wits end fashion.
Every attempt to leave writes a new dp iff format file and returns you to the
display. You have to reboot to get out.
-if you use the icon editor to make an icon for the mand executable (say, by
copying CLI.info to Mand.info) and then invoke the Mand program using the
icon, the Amiga hangs after the Q (quit) command, and you have to reboot.
-Picking up a SAved image does not reset the color parameters; they should be
saved with the image (I guess this is a requested correction for mandV2.02
rather than a bug report).
Best Regards, Howard Hull
[If yet unproven concepts are outlawed in the range of discussion...
...Then only the deranged will discuss yet unproven concepts]
{ucbvax!hplabs | allegra!nbires | harpo!seismo } !hao!hull