I'm wondering if there is an online IDE for the TI-89 Titanium and basically all other 68k calculators like SC3 here on Cemetech. If there wasn't, I was thinking of making one. I wanted to use the TIVars lib by Adriweb, but that only supported Z80 models. Is there any library that makes the files readable or do I have to find a way of detokenizing them myself? If there isn't, I will start making my own. I've been wanting to get into the web technologies, so any help is appreciated. It'd be great. Thanks in advance.

It'd be really cool to see an online 68k (BASIC, I presume?) IDE!

If you dig through the various archives on ticalc.org you may be able to find some 68k format converters, although I'm reasonably confident though that there's no library in common usage for detokenizing 68k BASIC. There are things out there that can read and write the actual file formats (e.g. 89p/v2p/9xp) like libtifiles2, though.

This guide has a lot of information about the formats. It's mirrored online here by Merth; you may be aware of it already but I figured I should mention it. It's an awesome resource.

Yeah, it'd be BASIC. I don't do assembly too much. I think assemblers like BRASS and SPASM are better at that than I would be implementing that online. Too bad about detokenizing those files. Thanks for those links though. Never really looked through the "calcinfo" archives. I'll look through these, plan on this and start working on it probably tomorrow. Thanks again for those links.

Back in the day I started trying to add TI-89 support to SourceCoder 2. I remember that the untokenized format was straightforward (ie, just text), but the tokenized format was seriously problematic. Certainly tokens had meanings that would change how the remaining text (or a subset of the remaining text) was detokenized; at the time I had a strictly table-based detokenizer, and it was not possible to cleanly fit the detokenization algorithm into that table format. For SourceCoder 3, the sad reality is that there just aren't enough 68k TI-BASIC programmers still working to make it worth my time, I think. Of course, if someone were to help write some PHP to tokenize and detokenize TI-89 files and/or just port something from existing C/C++/whatever-ish code, I'd be more than willing to add support to SourceCoder.

Did you ever add it? If so, how did you read the files and convert them to a human-readable format? Did you just go across the file and convert it right there? If you encountered such tokens that affect a subset of text, how did you have handled those, and you knew when the subset of special text stopped?

There had been some work on the TI-Planet side as well, based on Tokens89, which underlies Daisuke Basic Edit and TI-Edit (which contains an improved fork). Tokens89 is terrible code with zero portability (MS VB), but it remains the reference implementation of a TI-68k tokenizer / detokenizer.
The work didn't come out in a releasable state, plus we didn't have an online IDE anyway. But now we do have a flexible one, whose source code is in the process of being published under the GPLv3

That's great. Mind if you tell me when the newer one is released?

I'll try to remember about this topic. At least, I've subscribed to it

Yes, SourceCoder 2's primary purpose was to convert .8** files into human-readable text and back. My technique was to include a special type of tagged function pointer in my token array for these special tokens, and call that function pointer with the current state of the parser. I must say that I don't remember exactly how I knew when I reached the end of the special text section.

Thanks for the info Kerm. I've been working on the GUI of the IDE. I haven't made it pretty yet, but that'll be later. After I have been able to get proper knowledge on the whole file format of the 68k calcs, I will start adding that. First though is the GUI, with editing, new file creation, and compilation menus. I will start off by reading and editing .txt files to see if I can properly read it and then edit it.

I haven't been working much on the file tokenization and detokenization, but instead the website that it will work on. Little hard making it though, with positioning, CSS, and menus. I don't really like CSS . I have been doing fine on all else though. Without having much access to a Windows computer most of the time, I can't really examine the files of some already made projects that have been made for Windows. So, I will finish the site this week probably, and then try my best to find the processes of the 68k calcs. I think I might e-mail TI about this project, and see what they say and if they can help.

I totally screwed up. I finished the website. It looked awesome. It had all that good stuff, like a file upload screen, a pretty nice layout, and a pretty sweet everything else if I do say so myself. I made backups every once in a while. They had the same name with a "(n)" appended to it, with n being a number. The latest one had the biggest number in it. I thought I didn't need those backups anymore since I finished it and wouldn't be making big changes to it. I pressed Ctrl + A and selected every file and clicked Delete. I realized I just deleted the ACTUAL WEBSITE FILE. I turned on the setting to automatically delete everything after I sent it to the recycle bin. That backfired and now I have to start from scratch. I will get one done this time on Monday or Tuesday probably, but with testing going on, I might not have much time. Just wonderful, huh?

Ugh, that's horribly annoying; sorry that happened. Good luck rebuilding what you had made from your memory. Did you have the backend completed yet, or is that future work after you're happy with how the frontend looks?

If you haven't made too many changes to files on the disk, you may be able to recover it. "deleting" a file doesn't actually remove the data, it just marks it a free.

I've already recovered files after they were deleted, multiple times, from both HDDs (without formatting) and other persons' USB MSDs (after formatting).
Of course, the more one waits / the more disk output operations one triggers after the erase, the least the chances of recovering files are.