I have fixed these four issues.

IS that the line before or after that context? What comes after "ICE"?

I would appreciate that, thank you.

Huh, now that I look at it again, looks like this was before the issue. It seems you don't support the multiplcation sign, which is an x symbol on the calculator

If that was indeed 0xA9 as the error message implied, that should now be fixed.

I sent you the image by PM, as it contains code for my contest entry.

I have received the image. Oddly enough, not only are some of the colons on the second line, but some of the colons seem to be entirely absent. This is a CodeMirror problem rather than a SourceCoder 3 problem specifically, but if I get a chance I'll try to poke around and see if I can get to the bottom of it. Since it's a relatively minor aesthetic detail, I probably won't get too worried if I can't figure it out.

I'm writing a program using SourceCoder 3 in Axe. The project saves fine, but when I open it again, the tokenization/syntax highlighting goes back to Ti Basic. Every time. I have tried using the Axe program description as well:


Code:


to no avail.

Screenshot:
https://www.dropbox.com/s/69gp523c4wiysyj/Capture.PNG

What is the saved project name? I'll see if I can get to the bottom of that. It might have something to do with my changes to the "RawSave" format that SourceCoder uses internally.

The project is called MusicTheory

As mentioned on IRC, there is a token issue with 10^(.
In Sourcecoder, it reads as
Code:

because the token is 10^ instead of 10^(.
When the code is transferred to jsTIfied, the code reads as
Code:

It adds the opening parenthesis.

This can be confusing as the formula shown in Sourcecoder will not match the result on the calculator.

Thanks for reporting this! It has been repaired.

I have no idea if this is a bug, but sometimes the code will have a random "^^" in it before a character. Ex: "^^o" I've seen a lot. Can anyone explain that? Is it a bug?

Not a bug, it's the simplest way to show that that is the degree sign, since people will not want to have to dig around in character map to gain access to all the characters not on the keyboard.

As tifreak says, there are a number of equivalents like that: ^^2 is square, ^^3 is cubed, ^^o is the degree sign, ^^r is the radian sign, and so on. You can search through the equivalents using the Token Pane at the right side of the editor.

this image: http://img.ourl.ca/00-1.png
gets converted to this sprite: http://img.ourl.ca/animations.gif (ofc slime only)

I tried doing the leveling in GIMP and got something like this. Is that more what you wanted?

There's still one pixel off
And anyways, I noticed that for some weird reason i took the image, converted to jpeg, and then back to png. So now i too straight away the png (wich was only 4 shades of gray for real) and it worked

Ok this image doesn't get converted at all
http://img.ourl.ca/reuben2asmtitle.png

EDIT: NVM, the image was scaled up x2, and thus too large

This image:


gets converted to this 16x16 4-lv gray mess:
front
[1B0016800A7C1A08][2004434E3DA81258][1248081017E82BD4][2DB41FF811880E70]
back
[1B001F800FFC1FF8][3FFC7FFE3DB81258][124808101FF82C34][2E741A581FF80E70]
If I display the first 8*8 sprite I get this

A few other remarks:
1) I have no idea what the format is here. Is it top left to bottom right?
2) Another test program indicated that the front and back buffers for Axe are swapped.

The format used here is hex, that's all, it doesn't split your image into 8x8 sprites, it just converts each sequence of 8 pixels into a byte in hex format. Which means that your image displays fine with this code (tested, works).

Code:


If you want to have separate sprites, you'll have to (un)do some interlacing.

Speaking as someone who does not regularly write Axe, is there a more programmer-friendly format I could make SourceCoder spit out Axe sprites in? Or is the current format reasonable for general-purpose use?