headers
Ralph Versteegen | 1 Jul 2007 02:26
Picon

Re: SVN: teeemcee/1172 Got rid of EXXERROR, for those of us who can't remember such things

On 7/1/07, Mike Caron <caron.mike <at> gmail.com> wrote:
> On 6/30/07, James Paige <Bob <at> hamsterrepublic.com> wrote:
> > On Sat, Jun 30, 2007 at 02:23:25AM -0700, subversion <at> HamsterRepublic.com wrote:
> > > teeemcee
> > > 2007-06-30 02:23:25 -0700 (Sat, 30 Jun 2007)
> > > 168
> > > Got rid of EXXERROR, for those of us who can't remember such things
> > >
> >
> > Wait! that was there for a reason!
> >
> > -exx turns on array bound checking, whichis good, but it also causes
> > file-not-found errors to result in a crash rather than an error code.
> >
> > This change might have broken some stuff. I'll do some testing to see
> > whether or not it needs to be added back in.
> >

Why would it? It was only used in one place, which I rewrote.

>
> If I may, anything that breaks due to stronger error checking should
> probably be fixed, not ignored.

Your application of logic does not apply to FreeBASIC's illogical
design. Why on earth didn't they have an option to just enable array
bounds checking, one of the strengths of BASIC, without requiring all
that ON ERROR nonsense?? Like a call to readmouse killing the program
when the mouse is offscreen!
(Continue reading)

Permalink | Reply | 3 comments |
headers
Ralph Versteegen | 1 Jul 2007 02:29
Picon

Re: [Bug 188] segmentation violation when compiled using -exx -s console

On 7/1/07, James Paige <Bob <at> hamsterrepublic.com> wrote:
> On Sat, Jun 30, 2007 at 10:05:34PM +1200, Ralph Versteegen wrote:
> > On 6/26/07, Ralph Versteegen <teeemcee <at> gmail.com> wrote:
> > > On 6/26/07, Bob-bugzilla <at> hamsterrepublic.com
> > > <Bob-bugzilla <at> hamsterrepublic.com> wrote:
> > > > http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/show_bug.cgi?id=188
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > ------- Comment #4 from Bob <at> HamsterRepublic.com  2007-06-25 14:14 -------
> > > > One note about this-- The -exx compile option means that GAME and CUSTOM will
> > > > crash immediately whenever an array-bounds error occurs (rather than silently
> > > > allowing it)
> > > >
> > > > When running from a console in Linux, you get a nice display of the error
> > > > message, with  file, function, and line number where the error occured... but
> > > > on Windows you get nothing. the program just exits instantly.
> > > >
> > > > For this reason, I will not enable -exx in the batch file for building on
> > > > Windows, but I will be enabling it in the shell scripts for compiling on Linux.
> > > >
> > >
> > > Horray! This was really annoying me, and I had started working on it,
> > > before exams intervened.
> > >
> > > Now that my exams are over, I can stop letting you have all the fun.
> > > Looks like Ubersetzung is moving up quickly!
> > >
(Continue reading)

Permalink | Reply | 2 comments |
headers
subversion | 1 Jul 2007 04:33
Favicon

SVN: teeemcee/1179 Bug 401: more whitespace

teeemcee
2007-06-30 19:33:23 -0700 (Sat, 30 Jun 2007)
24
Bug 401: more whitespace
---
U   wip/hspeak.exw
Permalink | Reply | 0 comments |
headers
Bob-bugzilla | 1 Jul 2007 04:33
Favicon

[Bug 401] Hspeak is strict about whitepace around string assignment

http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/show_bug.cgi?id=401

teeemcee <at> gmail.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

------- Comment #4 from teeemcee <at> gmail.com  2007-06-30 19:33 -------
Fixed.

--

-- 
Configure bugmail: http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
Permalink | Reply | 0 comments |
headers
subversion | 1 Jul 2007 05:55
Favicon

SVN: pkmnfrk/1180 Fixing an array out of bounds access in bam2mid.bas

pkmnfrk
2007-06-30 20:55:00 -0700 (Sat, 30 Jun 2007)
267
Fixing an array out of bounds access in bam2mid.bas

But, the thing about this that weirds me out is that it was being triggered by a BAM with 16 channels... but,
such a thing doesn't exist!

It was song #8 in Wandering Hamster (the mountain theme) that caused it...
---
U   wip/bam2mid.bas
Permalink | Reply | 1 comment |
headers
Bob-bugzilla | 1 Jul 2007 10:35
Favicon

[Bug 233] OHR freezes on startup

http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/show_bug.cgi?id=233

------- Comment #9 from teeemcee <at> gmail.com  2007-07-01 01:35 -------
I was just about to say that I haven't seen it since, but then it happened with
music_sdl a few times. I then did alot of testing with music_native and got it
several times. I guess it doesn't show up with -exx or FB0.17, explaining why I
hadn't seen it any of the hundred previous runs of game.

A second place it freezes alot (and always has, but I don't seem to have
distinguished) is on the Loading... screen.

This is not a music backend bug, my suspect is gfx_fb.

--

-- 
Configure bugmail: http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
Permalink | Reply | 0 comments |
headers
Bob-bugzilla | 2 Jul 2007 00:51
Favicon

[Bug 405] Hero Stays In Weak Pose after being healed

http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/show_bug.cgi?id=405

Bob <at> HamsterRepublic.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|backend:gfx_fb              |Battles

--

-- 
Configure bugmail: http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
Permalink | Reply | 0 comments |
headers
James Paige | 2 Jul 2007 01:15
Favicon

Re: SVN: teeemcee/1172 Got rid of EXXERROR, for those of us who can't remember such things

> > >
> > > Wait! that was there for a reason!
> > >
> > > -exx turns on array bound checking, whichis good, but it also causes
> > > file-not-found errors to result in a crash rather than an error code.
> > >
> > > This change might have broken some stuff. I'll do some testing to see
> > > whether or not it needs to be added back in.
> > >
> 
> Why would it? It was only used in one place, which I rewrote.

*reads the code*
Huh. Interesting. I have never seen the syntax you used before:

  if open(f$ for binary access read as #fh) = 2 then

I had no idea there was a way to get a retun value from "open"
That is actually a lot cleaner than my hacky code. I like it. 

> >
> > If I may, anything that breaks due to stronger error checking should
> > probably be fixed, not ignored.
> 
> Your application of logic does not apply to FreeBASIC's illogical
> design. Why on earth didn't they have an option to just enable array
> bounds checking, one of the strengths of BASIC, without requiring all
> that ON ERROR nonsense?? Like a call to readmouse killing the program
> when the mouse is offscreen!
(Continue reading)

Permalink | Reply | 2 comments |
headers
James Paige | 2 Jul 2007 01:18
Favicon

Re: [Bug 188] segmentation violation when compiled using -exx -s console

On Sun, Jul 01, 2007 at 12:29:35PM +1200, Ralph Versteegen wrote:
> > > - it's just going to cause file not
> > > found crashes all over the place,
> >
> > ... which is why I added the EXXERROR hack... :P
> >
> > > and the error messages aren't much use.
> >
> > I can't even *find* the error messages on windows, but on Linux, I have
> > found them highly helpful. I see file, function, and line number.
> >
> 
> Oh, that's nice. I have to compile with -s console -exx -g and run
> with gdb to able to see and trace the error messages. All I get is
> "File not found on line 3275"

Weird. I wonder why the error doesn't get printed on the console when 
you compile with -s console

---
James
Permalink | Reply | 1 comment |
headers
James Paige | 2 Jul 2007 01:28
Favicon

Re: SVN: pkmnfrk/1180 Fixing an array out of bounds access in bam2mid.bas

On Sat, Jun 30, 2007 at 08:55:00PM -0700, subversion <at> HamsterRepublic.com wrote:
> pkmnfrk
> 2007-06-30 20:55:00 -0700 (Sat, 30 Jun 2007)
> 267
> Fixing an array out of bounds access in bam2mid.bas
> 
> But, the thing about this that weirds me out is that it was being triggered by a BAM with 16 channels... but,
such a thing doesn't exist!

The BAM format has always allowed 16 voices, but no implementation of it 
could ever encode or play more than 9

> It was song #8 in Wandering Hamster (the mountain theme) that caused it...

Huh. That is a *really* old one. One of the BAM's I composed when I was 
first writing notate. I must have done something flukey.

---
James Paige
Permalink | Reply | 0 comments |

Gmane