I would suggest adding the possibility to work with more people on one project/program.

This existed in SourceCoder 2, but because no one used the feature, I didn't make the effort to port it to SourceCoder 3. I have considered adding some sort of GitHub/BitBucket integration, though.

Wouldn't it be easier to allow to zip-pack a project (with encryption) and have a page for downloading those exported, packed and encrypted projects within 24h. So that everyone can download such a project but only members knowing the password may use it and open it in SourceCoder... This will take some space if heavily used but will be no work for coding... (At least I think so)

I think the point was to prevent having to export it and unzip it every time you want to trade code with whoever else you`re working with. If you were to implement such functionality, might as well just pm download links to eachother ;

I meant something like that... Not an actual network system for coding, just a way to let people download stuff others wrote. Maybe not a page but some sort of special PM that let's you open the project in SourceCoder directly, I like this idea even more

I recommend to color all the commands in the Catalog.
Because now it looks like this:

PT_: Clearly I missed a few commands when I put the highlighter together. Thanks for the bug report.

Nik, PT_: I'll keep looking into good ways to share projects.

An issue? Or just not implemeted code? If you load a program from your calculator that has a Label, Lbl L1, in SourceCoder 3 it will get parsed to text as Lbl L1. However, when loading it back onto your calc it will be back-parsed as Lbl List1 (the list L1).

Also, I havent tried this yet, but I am sure that if putting a Token into a string, let's say "Test and anothertest->Str1 while the "and" was a token, it will probably be back-parsed as a typed out "and".

My suggestion is marking tokens the following way:

Code:


This will make an easy and effective solution.
Or another solution would be to automatically tokenize as much as possible of strings and automatically ignore tokenizing after commands like Lbl and so on, where a token is impossible.

(I once had to edit a whole program to exchange all those Goto's and Lbl's... This took me half an hour, despite the find & replace.)

Someone has probably suggested this before, but it would be nice if comments were possible in SourceCoder. Make it so that you can put either // or # or whatever at the end of a line of code so that everything after it gets ignored and doesn't actually get put into the program. This would definitely help me a bit, especially when coming back to some of my projects after a while.

This is already a feature.

Oh it is? Well I'm confused because I have actually seen the text change color when you do // like it is a comment, but it still get's put into the actual code when i transfer it to jsTIfied. What am I doing wrong?

Can you provide a sample program that displays this behavior?

I believe you have to put at least one space before the start of the comment

I never really use comments because I don't like having my code cluttered up with a bunch of things I probably won't read, but another question about comments, what if someone would try to have a string with all the components of a comment, would it then be removed from the program or is there some check to see if it is in a string?
(something like this)

Code:

This is correct. Space, tab, or any other whitespace will do the trick.
In the final program, that would be an empty string, followed by the real() command on the next line.

*bump*



I support this. It would help, especially when me, PT_, and NoahK worked on our Tic-Tac-Toe program. Instead of just having each person have to post/PM the code, we would have been able to see each other's work. And that one feature would have simplified the process of creating it, and would make it much easier to work together.

No, it may not be used a lot... but just like the ability to call 911, it's there when you need it.

It would be immensely helpful to extend parenthesis matching to match close parentheses with parentheses that are part of tokens. For example, in the following:


Code:


the tan( should be highlighted when my cursor is on the third close-paren from the right.

Most of my close-parens, in fact, are used to close functions, so this would be useful.

Suggestion: Add an HP PPL ICON statement generator for image conversion.

The ICON statement consists of a hex-encoded PNG with a specific format. I can try and track down the PNG specs that it uses and get something to create them.

lirtosiast: Yes, I'd like to implement that, and I'll do my best to find the time to do it.
AHelper: If you could share a spec, I'd be more than happy to implement it.