Zugg Software :: View topic - 2.17

Apprentice Joined: 03 Nov 2007 Posts: 132

This could be pretty major but I'll leave that to someone else to determine. In 2.17 I am finding when I have a lot of settings with ansi codes firing, the screen tends to not repaint properly and those %ansi codes can bleed over to other lines or just blank out huge parts of lines, or what have you. It's not pretty! I'll try to isolate it more but something is up with %ansi.

Wizard Joined: 25 Mar 2003 Posts: 1113 Location: USA

This has been reported in other threads for a while now. I noticed it happening in 2.17 myself and submitted feedback with the raw log and a screenshot, so hopefully that helps narrow it down. It seems to only bleed background colors (for me), which I think may be the key to tracking it down and fixing it.

Apprentice Joined: 03 Nov 2007 Posts: 132

For me it bleeds background colors as well as modifiers such as bold, blink, etc. It was a problem a few versions back but it had stopped, for me anyway, for last couple.

Posted: Fri Dec 14, 2007 5:27 pm

Do you have scripts that are firing and changing the screen? For example, triggers that output their own messages? I need to know if the ANSI display coming directly from the MUD is a problem, or just when you have your scripts enabled. In Larkin's case, he has some scripts enabled that are causing a complex interaction with the screen.

I'd really love to fix this if someone can come up with a way to reproduce it. If there is a simple script that shows the color problem using the #SHOW command, or if there is a MUD I can log into that shows the problem. But unless I get a way to reproduce this today, I might need to let this one slip through into the public version. I've just run out of time to fix this stuff in the current version. I'm doing a quick fix this morning for the #ALL problem because it's a serious parser issue. But without a way to test the Ansi problem, I'm stuck with not being able to do anything about it.

Wizard Joined: 25 Mar 2003 Posts: 1113 Location: USA

I pasted one of my display aliases that causes the problem and a raw log into the error report, trying to keep it all together with the screenshot. If you want, I can give you my package file and even a test server address to try and re-create it exactly as I see it.

Posted: Fri Dec 14, 2007 5:41 pm

Larkin, that would be great. If you can get that sent to me this morning, then I might have a chance to fix it. Just give me the exact instructions in the email message that I would need to log in and test it. And post here when you send the email so that I can be on the lookout for it. Email everything to the sales@zuggsoft.com account.

Wizard Joined: 25 Mar 2003 Posts: 1113 Location: USA

Okay. I sent it out now, with basic instructions on how to use it. The subject will be "[2.17 Pro] ANSI colors problem" then. Let me know if you need anything else!

Posted: Fri Dec 14, 2007 8:12 pm

Got the files. I'll email you with questions.

Posted: Fri Dec 14, 2007 8:56 pm

Test the new 2.18 version and let me know if it helps with the ANSI bleeding problems. I logged onto Larkin's MUD and didn't see any obvious problems. And I played with Malach's stuff from previous posts. When I first tried some of Malach's old ANSI color posts in 2.17, I also reproduced a problem with the ANSI color. I fixed that specific problem in 2.18.

Basically, the #SAY and #SAYP commands were still not properly restoring the previous color value in some special cases. The problem required doing a #SAYP to add something to the prompt from within a prompt trigger.

While I was working on this, I also found a bug where the prompt trigger could be called more than one in certain weird situations. Again, it only happened if text was added to the screen from within another prompt trigger.

So, let me know if I broke anything in your prompt triggers, and whether the color issue is fixed. Although at this point I'm not expecting any other versions before the holidays, so you might need to live with any problems that still exist.

Wizard Joined: 25 Mar 2003 Posts: 1113 Location: USA

Well, regardless of whether or not my settings and test server helped find and fix this, it certainly appears to be fixed in 2.18. Thanks, Zugg!

Posted: Fri Dec 14, 2007 9:02 pm

Great news! Thanks very much for the quick response.

Apprentice Joined: 03 Nov 2007 Posts: 132

Unfortunately, it's not fixed with me. Well, correction, the %ansi portion of this is fixed but I am noticing a similar problem with my usage of #CW. Basically I have a prompt trigger and dependent on certain statuses, the end portion of the line is colored differently. So kind of like this:

4900h, 4000m, 3700e, 10p, 22400en, 17896w elrxk<>-

With the elrxk<>- being the colored portion with the actual command to color it being nothing more than a switch statement with #CW high,yellow, #CW high,red, etc. as the commands. Sometimes what happens is that everything in front of the elrxk<>- is blanked out entirely for several lines. This includes stopping me from seeing any other lines received from the mud during that time though they trigger fine so cmud is seeing them. It won't stop until some other block of text other than the prompt shows up.

This is hardly major as I can just ignore it or stop coloring it for the time being but the issue seems to be with #CW.