I'm not sure if this really belongs here, but it has to do with the
current
POV-Ray implementation so...

Just flame me if I should move to another group! ;-)

If you use the write-command to save some string/vector/whatever
to disk, and stop the trace, it happened pretty often that the
vector it just wrote wasn't complete, or the string broke off
middle-sentence...

So I'm thinking, is there a way to have POV-Ray at least finish
the write-command before it stops?

I think it would help some POV-Scripters, but in this case
especially me... :-)

I hope this is not a feature request...

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

On Fri, 21 Jun 2002 02:28:56 +0200, Tim Nikias <tim### [at] gmxde> wrote:
> If you use the write-command to save some string/vector/whatever
> to disk, and stop the trace, it happened pretty often that the
> vector it just wrote wasn't complete, or the string broke off
> middle-sentence...

with or without #fclose ?

> Just flame me if I should move to another group! ;-)

Not necessary to another group since there could be really some problem - but
hard to say without proper bug report. Have you noticed main parts of typical
bug report ?

ABX

Post a reply to this message

> with or without #fclose ?
>

With and without...

>
> > Just flame me if I should move to another group! ;-)
>
> Not necessary to another group since there could be really some problem - but
> hard to say without proper bug report. Have you noticed main parts of typical
> bug report ?

I'm not too sure if this is really a bug, and if it really happens that often.
I just do know that it occured to me several times, but I'm trying to
figure if it was Windows which crashed on me, or if POV-Ray did
something wrong.

If any of the programmers could tell me if POV-Ray does finish
a command like #write when being stopped manually (not by system
crash, which never results in expected behaviour...;-), I'd probably
just shut up, because there are probably some issues involved if write
may not be finished...


--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

On Fri, 21 Jun 2002 11:09:17 +0200, Tim Nikias <tim### [at] gmxde> wrote:
> If any of the programmers could tell me if POV-Ray does finish
> a command like #write when being stopped manually

IMO The reason why beta-test group exist is that usually reports first are
verified by the users and _after_ confirmations programmers tries to fix it.
As date of linux release shows programmers have enough tasks with RL. It is
hard to help them and confirm and investigate anything without a little bit of
wrongly worked script and comparision of system parameters. The excpected
behaviour is IMO that after exiting if something is not written then buffer is
moved to file and then file is closed. But 6.2.3.2 is not precise about it.

ABX

Post a reply to this message

> IMO The reason why beta-test group exist is that usually reports first are
> verified by the users and _after_ confirmations programmers tries to fix it.
> As date of linux release shows programmers have enough tasks with RL. It is
> hard to help them and confirm and investigate anything without a little bit of
> wrongly worked script and comparision of system parameters. The excpected
> behaviour is IMO that after exiting if something is not written then buffer is
> moved to file and then file is closed. But 6.2.3.2 is not precise about it.
>
> ABX

Hm, okay. We'll leave it at that, and I'll just expect POV to finish
the #write-command while scripting. I'm not too sure about that
buffer you're talking, I would be expecting POV to write to
disk as soon as it sees #write, but internally close the file, making
output to that file impossible, when using #fclose.
IMHO #fclose is needed when you want to make sure that
your system doesn't interact with another one which uses the same
file-handles. So, one system might open a file, naming it _O (I always
do), and when its finished, the system should #fclose _O, so in
case another system open another file with _O as handle, it
doesn't issue a warning and doesn't think we're using the same
file...

I think I'll stay settled with #write is done when its called, unless
some POV-Team member tells me different...

And regarding the bug-issue etc you're mentioning at the beginning:
I told you, I was not sure if it belonged here... :-)

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

On Fri, 21 Jun 2002 12:18:07 +0200, Tim Nikias <tim### [at] gmxde> wrote:
> I think I'll stay settled with #write is done when its called, unless
> some POV-Team member tells me different...

You are not precise - do you think it is a bug or not ? If yes ...

> And regarding the bug-issue etc you're mentioning at the beginning:
> I told you, I was not sure if it belonged here... :-)

... when you post to beta-test you could accept beg-reporting guidelines since
they are clear. Since you have problems with making sample script take a look
at below:

// START
  #fopen _O "test.txt" write
  #write (_O,"What's the problem?")
  #fclose _O
  #error "Please, stop it!"
// END

works fine with POV 3.5 RC6 icl on WinNT Sp 6 PII 233 with 128 MB

ABX

Post a reply to this message

Okay, okay. I admit, I've been a little stupid... I'll go for advanced-users
next time.


> ...  beg-reporting guidelines ...
>

Beg-reporting? :-)

"Hey, Thorsten, there's someone begging for POV-Ray 4.0, what
shall I do with him?" :-)

Okay. Now that's off-topic. We'll end it here, or move to off-topic,
okay?

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message