Linux (recommended):

sudo apt-get install python-pip python-wxgtk2.8
pip install M30W

Run using “M30W” from terminal or Alt+F2.

Windows:
We have windows installers here. However, there's a bug with python x64 installers that requires you to manually modify your registry. (instructions)

Mac:
install pip, then follow the linux instructions.

If you're not sure, you can get the code from our SF svn repo at

svn checkout https://svn.code.sf.net/p/m30w/svn/

You can run it using python M30W.py
If you have some trouble installing M30W, here is an excellent guide by Gravitation.

M30W needs:
- python 2.7 (3.x support is NOT planned, nor 2.6 compatibility)
- kurt
- PIL
- wxPython version 2.8 or higher (Version 2.9 no more regularly tested)

Yes, of course! Just post your sourceforge name, grab the code, and start working!
You can also help by reporting bugs on this page

• roijac
  
• davidkt

We currently only support .sb files (old 1.x format)
We are aware of this problem, but our current focus is on the execution of scripts (And we're doing well)

We use the sourceforge ticketing system at our sourceforge page.

scimonster wrote:

Doesn't work on Fedora. I can't find the python-wxgtk2.8 package, even with RPMFusion. Also, the command is python-pip install M30W, but it's obviously missing a dependency.

sudo yum install wxPython

roijac wrote:

Who are current developers?

• roijac
  
• davidkt

davidkt wrote:

roijac wrote:

Who are current developers?

• roijac
  
• davidkt

Really? What happened to Magnie, slinger and aion5?

davidkt wrote:

Good. I had to reinstall Windows a few days ago(because I accidentally deleted a windows theme file) so I'll have to reinstall m30w.
P.S. I don't want to make the cmd-line version- I'm horrible with command-line stuff.

roijac wrote:

davidkt wrote:

roijac wrote:

Who are current developers?

• roijac
  
• davidkt

Really? What happened to Magnie, slinger and aion5?

They don't really… develop the program, you know? They're pretty much just interested in the project. Not offending or something, but they didn't write a single line of code (nor did you, but in you I still have faith )

roijac wrote:

davidkt wrote:

roijac wrote:

Who are current developers?

• roijac
  
• davidkt

Really? What happened to Magnie, slinger and aion5?

They don't really… develop the program, you know? They're pretty much just interested in the project. Not offending or something, but they didn't write a single line of code (nor did you, but in you I still have faith )

davidkt wrote:

I could do is remove that annoying “v” from the end of lists.

#blockspecs.py
block_plugin_inserts = dict((insert, fmt.replace(" v", "")) for (insert, fmt) in block_plugin_inserts.items())

davidkt wrote:

I see. So should I update the wiki page?

davidkt wrote:

I know what I could do! I could modify Kurt for the M30W standards! And one of the first things I could do is remove that annoying “v” from the end of lists. I could give M30W a completely different syntax altogether…

roijac wrote:

I REALLY hate the scratchblocks syntax and how it's so messy and badly defined, but I also want to keep the nice feature of copy-paste into the forums. I should probably talk to ST, or whoever does it in S2.0.

How would you design the syntax?

move item n + 1 of amounts steps
set player 2 score to 1 + 3 / 2

move (item ((n) + (1)) of [amounts v]) steps
set [player 2 score v] to ((1) + ((3) / (2)))

nXIII wrote:

How would you design the syntax?

I made a parser a while ago that was a lot more logo-esque, i.e., it rarely used parentheses/brackets for grouping and relied on block information to parse code.

blob8108 wrote:

nXIII wrote:

How would you design the syntax?

I made a parser a while ago that was a lot more logo-esque, i.e., it rarely used parentheses/brackets for grouping and relied on block information to parse code.

Ah yes, you said. And then you refused to show me the code.

I definitely agree with you there: if one can do it, then your parser has a much better syntax. I still can't quite figure out how to combine TDOP with mixfix blocks. Did you use TDOP or recursive decent or a parser generator or some other custom homemade algorithm?

Maybe I should even rewrite scratchblocks2's parser in the same way… though the existing block plugins both render exactly what you type, and don't actually validate the code in any way, so it would be a slight change in philosophy.

nXIII wrote:

blob8108 wrote:

nXIII wrote:

How would you design the syntax?

I made a parser a while ago that was a lot more logo-esque, i.e., it rarely used parentheses/brackets for grouping and relied on block information to parse code.

Ah yes, you said. And then you refused to show me the code.

Oops! Here: http://dl.dropboxusercontent.com/u/10715865/tb/index.php (I apologize for the slightly messy coded quality. It also still doesn't parse a lot of stuff [e.g., my demo] correctly yet)

I definitely agree with you there: if one can do it, then your parser has a much better syntax. I still can't quite figure out how to combine TDOP with mixfix blocks. Did you use TDOP or recursive decent or a parser generator or some other custom homemade algorithm?

It's a recursive descent parser for the operators with a lot of homemade stuff for parsing the blocks.

Maybe I should even rewrite scratchblocks2's parser in the same way… though the existing block plugins both render exactly what you type, and don't actually validate the code in any way, so it would be a slight change in philosophy.

I think the current scratchblocks is good the way it is—you get exactly what you write, it's fast, and it's simple. I wrote my parser both as a test and and for large personal projects so I could write them quickly in a real text editor; I think it's less suited to everyday use for small scripts in the forums.

nXIII wrote:

Here:

Maybe I should even rewrite scratchblocks2's parser in the same way…