Trizbort v1.5.9.6 - Minor Update released

Wanted to let everyone here know first that I released this morning v1.5.9.6 of Trizbort. This is a relatively minor release, but I wanted to get all the changes I’ve made out there since it was August I last released something. Probably the nicest new feature in this version, is I’ve now included a Find. This is still somewhat limited and only searches room names, sub-titles objects and descriptions. But I’ve found it really handy on large maps.

You can find everything on GitHub. I will be updating the trizbort.com site later and perhaps putting out a notification on the other forum.

Please let me know if you find any issues, have further suggestions or anything. I hope to ramp up some more work on Trizbort as I’ve had some requests come in over the last few months.

Enjoy

2 Likes

A quick little bug fix release that fixed one issue introduced in the automapper in the last release and an older bug that caused issues exporting to I7 code for older maps.

2 Likes

Yesterday, I tried Trizbort’s automapping with a transcript produced by zarf’s Lectrote interpreter. The good news is that Lectrote updates the transcript each turn so that isn’t an issue. The bad news is that I noticed my rooms were occasionally getting funky names. I’m guessing it’s likely due to the fact that Lectrote saves its text files with unix encoding.

I asked zarf about the possibility of having the Windows port save with pc encoding. He didn’t know offhand where in the stack he’d have to make the change and didn’t foresee himself having time to look at it in the near future.

I don’t know what kind of hassle it would be to add non-pc-encoded file compatibility to Trizbort itself, but if it wouldn’t be much of a problem, that’d probably be a good thing to add.

What’s a file look like? If you could attach a few, I could open them and poke around.

A google for C# Unix Text File turns up this: http://stackoverflow.com/questions/10500930/read-unix-encoded-file-with-c-sharp

At the very least it seems like there can/should be a toggle in the automap features for Windows/Unix (or whatever code page the text is) input. I don’t suppose we can hit them all, but we can nail down the main ones.

Here’s the one that brought the issue to my attention: https://drive.google.com/file/d/0B_4ZXs4Z_yoWMDNzTEdydFVtWVU/view?usp=sharing

I zipped the text file just to avoid Google Drive doing some kind of automatic conversion.

Thanks! Looking at it I don’t think the problem is the encoding. Rather, Trizbort sees “On the old” as a potential room, which is wrong. Maybe there could be a way to ignore certain phrases? Or sharpen up what Trizbort sees as a sentence?

If I replace “On the old, wooden table is a blue book.” with “You see a blue book on the old, wooden table” then it is okay.

If I replace 0xa with 0xd 0xa, we still have the room-misreading problems.

So this looks like a word parsing issue and not a character set thing, which is a relief.

One odd thing though is that Lectrote sends out a bunch of null characters after each command. This isn’t critical, but it doesn’t look right in Notepad++. I might report a (low-priority/impact) issue with the file you attached, if you haven’t.

Ah, good to know it’s not the encoding. To be honest, I don’t use the auto-mapper all that often because it gets so easily thrown off, but I had already been using it to automap a different game but had to switch interpreters because of interpreter-bug issues and was just testing Lectrote before I picked up where I left off.

In general, the automapper needs rewritten I think. Since I started maintaining it, it seems somewhat flacky and difficult to enhance without breaking stuff. I’m sure Genstein created it just fine, but it’s been modified enough over the years and we’ve added new features that the automapper is not taking advantage of.

Someday, perhaps soon, I will rip the whole thing out and start over.

Even better, the bug I saw with the 0x00 characters is fixed. I hadn’t used Lectrote much but I might use it more now. It’s neat to see how it progressed.

Yeah, you can only do so much planning for possible future features. It’s good, though, that people are finding new ways to beat it up.

Do you mean this bug is fixed? If so can you please close the issue? :slight_smile:

@curiousdannii yes! Thanks for poking me here. I could’ve sworn I closed it before then, but I had some weirdness where my browser crashed because I opened too many tabs.

This topic has gotten me using Lectrote more, though, so that’s a good thing. Maybe I can pitch in by testing some of my more intricate games which I’m planning to re-release anyway. If nothing else I can say “game X works on Lectrote.”