POV-Ray : Newsgroups : povray.documentation.inbuilt : Remaining documentation question for FS#226 Server Time: 18 Jun 2019 16:39:19 GMT
  Remaining documentation question for FS#226 (Message 1 to 3 of 3)  
From: William F Pokorny
Subject: Remaining documentation question for FS#226
Date: 19 Oct 2016 11:56:43
Message: <58075f7b$1@news.povray.org>
The remaining question for the still open FlySpray issue 214 was whether 
to better document which of the old X11 flags no longer functioned due 
the use of the Simple DirectMedia Library for display.

The last comment was:

Comment by Grimbert Jérôme (Le_Forgeron) - Friday, 09 December 2011, 
14:28 GMT+5

 From my point of view, the original issue is closed (as now with #5460, 
the offending line would be dumped, the detection of forgotten alias or 
magic macro should be trivial).

The part about the Wiki of Unix options is still open: 3.7 use the SDL 
library, which does not intercept the X11 setting. so -display, -icon 
and so on are not supported. Supported specific options are in the unix 
specific source file (from memory: -help, -version; and their shortcuts).

-display and al. were supported at least in 3.1, and that was before SDL 
library. I do not know when the change occurred (3.5, 3.6, or 3.7)

----------

See: 
http://wiki.povray.org/content/Documentation:Unix_Section_6#Understanding_POV-Ray_Options

In my testing of the listed special unix options do the following:

-display <display_name>  --> Will fail with a command line parse error 
so long as a display name is given. +-display otherwise acts like +-D.

-geometry [WIDTHxHEIGHT][+XOFF+YOFF] --> Fails with a command line parse 
error.

-help --> Works fine.

-icon --> Fails due the -i being seen as specifying a scene file name 
con, unless the scene file comes after -icon in which case things run 
without complaint.

-title <title> --> Fails with a command line parse error.

-visual <visual_type> --> Fails with a command line parse error if a 
visual_type is specified and runs without complaint otherwise.

I've closed the issue and the information above can be added to the 
documentation - or not. My personal opinion is we probably should add at 
least a sentence saying explicitly not all the listed X11 supported 
options will work as expected - something the known bugs link implies 
but doesn't say outright.

I'll be closing the issue with a comment pointing to this thread.

Bill P.


Post a reply to this message

From: clipka
Subject: Re: Remaining documentation question for FS#226
Date: 19 Oct 2016 17:11:47
Message: <5807a953$1@news.povray.org>
Am 19.10.2016 um 13:56 schrieb William F Pokorny:
> The remaining question for the still open FlySpray issue 214 was whether
> to better document which of the old X11 flags no longer functioned due
> the use of the Simple DirectMedia Library for display.
...

> I've closed the issue and the information above can be added to the
> documentation - or not. My personal opinion is we probably should add at
> least a sentence saying explicitly not all the listed X11 supported
> options will work as expected - something the known bugs link implies
> but doesn't say outright.

In my personal opinion, I think we should properly amputate any
remaining code related to X11 options (if still present) -- either by
deletion or via "#ifdef", depending on whether we want to retain the
option to ever revive the code -- and then strip all reference to the
options from the docs.


Post a reply to this message

From: William F Pokorny
Subject: Re: Remaining documentation question for FS#226
Date: 20 Oct 2016 13:37:37
Message: <5808c8a1$1@news.povray.org>
On 10/19/2016 01:11 PM, clipka wrote:
> Am 19.10.2016 um 13:56 schrieb William F Pokorny:
>> The remaining question for the still open FlySpray issue 214 was whether
>> to better document which of the old X11 flags no longer functioned due
>> the use of the Simple DirectMedia Library for display.
> ...
>
>> I've closed the issue and the information above can be added to the
>> documentation - or not. My personal opinion is we probably should add at
>> least a sentence saying explicitly not all the listed X11 supported
>> options will work as expected - something the known bugs link implies
>> but doesn't say outright.
>
> In my personal opinion, I think we should properly amputate any
> remaining code related to X11 options (if still present) -- either by
> deletion or via "#ifdef", depending on whether we want to retain the
> option to ever revive the code -- and then strip all reference to the
> options from the docs.
>
I'm not up on X11 history, but given the length of time we have now been 
using the Simple DirectMedia Layer (SDL) alternative, I lean this way too.

I hedge some because I still occasionally see flaky behavior with SDL 
V1.2. Some of that I suspect is due the V1.2 SDL library and some of I 
think is probably due our current code not closing down SDL as cleanly 
as it could(1).

I've been reading up on and playing with SDL V2.0 on the side. My plan 
is to get my own daily version of povray onto SDL V2.0 to take it for a 
test drive, but we'll see. If I get the V2 library going, I'll open up a 
related github issue.

Bill P.

(1) - One reason for pull request #109 is that I have been for years 
using cntl-c on unix to quit renders early. Slowly, I came to suspect it 
was occasionally after these aborts that I would find my windowing 
messed up in not updating properly or at all. The 'q' key is a more 
organized way to quit, but I didn't know it was active during a 
previewed render by the current unix console messages. I'm now trying to 
use 'q' only to quit to see if I see the window update issue less, but 
I'm finding my cntl-c quit-fingering-memory habit hard to break.


Post a reply to this message

Copyright 2003-2008 Persistence of Vision Raytracer Pty. Ltd.