Comments are closed

cJass on the Google Code.
Our BugTracker (recommend to post bugreports and features request there).
Beta tester notes.

all times are GMT +03:00
Latest posts

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 21 22 23 24 25 26 27

posted at 26/01/10 12:47
ADOLF

1.4.1.29
New feature - anonymous functions
<span class="code">ForGroup(gr, <b>lambda</b> void() { KillUnit(GetEnumUnit()) } )</span>
And yeah, we're back.
~VD

posted at 26/01/10 05:35
Nestharus

Ok, so rather than doing Lua preprocessing etc, allow for Lua global blocks so we don't run into syntax errors and take the preprocessing a step further.
Use the Euphoria interpreter to process and let people write Euphoria code. Also include a lexical object so that people can build and manage code more easily.
Just check to see if Euphoria is installed on the machine, and if it's not, download it and run the exe or give a link or something : ).
So what should people be able to do with Euphoria? Generate code ofc. The euphoria preprocessor would run before the cJASS and vJASS stuff. It'd run before anything really.
Anywho, do what I was going to do with Lua. I'm thinking of doing the Euphoria thing, but I think you guys might be up for it. You could call it Euphorium : P, the ultimate preprocessor : D.
Furthermore, people could also obviously use Euphoria exes or C exes or w/e, lol.
Oh well : p. You know, dynamic code generation is now becoming a big thing. I hear the new VS 2010 has a full preprocessing language that generates code. From there, that generated code is compiled.
Get to it : o.

posted at 26/01/10 02:33
TheLifelessOne

@Van Damm
Nope. About new versions. :D

posted at 25/01/10 10:21
Van Damm

@TheLifelessOne: updates on what? Your questions? You can find answers to them by reading cJass Manual - everything you need is written there.

posted at 25/01/10 09:17
TheLifelessOne

None? D:

posted at 24/01/10 02:51
TheLifelessOne

Any updates?

posted at 20/01/10 15:37
Dark Dragon

i see, do u know maybe when he is supposed to come back? 1 month, 1 week?
anyway if not then np! ill just wait and until then ill have to code with as i did till now xD
Greets!
~DD

posted at 20/01/10 12:33
Van Damm

Unfortunately, ADOLF is away so no one will be able to make fixes untill he's back :(

posted at 20/01/10 12:32
Dark Dragon

Hi VD!
well i am not sure did you already read my last bug report but here it is:
when i use #include which contains #guard too much times cJass crashes...
and as request anonymous funcs are all that i really want...
ForGroup(g, nothing() { KillUnit(GetEnumUnit()) })
i know as u said ADOLF does not have time now, but laiter if possible i would really like this feature...
Regards!
~DD

posted at 20/01/10 12:10
Van Damm

You don't need to write "globals" at all
Answers to all questions you ask can be found in cjass manual cjass.xgm.ru/manual-en

posted at 20/01/10 08:33
TheLifelessOne

Bug report!
globals {
}
doesn't work.

posted at 20/01/10 07:22
TheLifelessOne

I'm trying to do something like:
#define CREATE name = func()
CREATE would call a function, and name would be the arguments.

posted at 20/01/10 07:17
TheLifelessOne

Hey, is there any way to do defines with arguments?

posted at 19/01/10 11:36
Anonymous

I made a spelling error. Sorry, I've gotta point it out.
"Two with have the"...
What I had meant to say was this: "Two would have the".

posted at 19/01/10 11:32
Anonymous

Damn, one more thing (using the request I just made).
Initializing arrays (parsed into an initialization function which runs during start). I believe this isn't a current feature at the moment.
Example:
int[] counter = 1, 2, 3, 4, 5,
Each index, 0-4 would be set to those values using an intialization function.
Also:
string[] str = 2 = "Hello", 5 = "World"
Two with have the value "Hello" and five would have the value "World". All other indexs would be blanks (feel free to change to syntax on this one, mine seems just a little... Messy).
Thanks for reading.

posted at 19/01/10 11:27
Anonymous

Just realized I should be more specifc:
int i[] // array.
int[] i // array, standard (the suggested way).
[]int i // not array, throw error.
[int i] // not array, throw error (it'd be interesting though).
And for includes, the .j extension isn't needed. It either automatically adds the .j extension, or includes regardless of the extension.
Thanks for reading.

posted at 19/01/10 11:18
Anonymous

Requests:
Can you add the ability to create arrays using "[]".
Example: int[] // is an array now
And add the ability to do includes using "<>".
Example: #include <filename.j>
Thanks for reading.

posted at 19/01/10 00:08
MasterofRa

I have a request, can you make it so that a macro like this would replace the variable in the text with the variables contents, so it would display what the condition is rather than saying condition?
assert(condition) = {
if not (condition) then
debug call BJDebugMsg("Assertion Failure: `condition`")
debug MACRO_CRASHER = MACRO_CRASHER //thread crash
endif
}

posted at 16/01/10 15:18
Dark Dragon

ohh hi VD! long time no see :)
i hope ADOLF will find time to release some more cJass updates!
Greets!
~DD

posted at 16/01/10 03:37
Nestharus

Ok, after some more thought, I think there needs to be a central exe for running the lua and passing lua into the various exes that they might be using. This will allow for one global lua block so that the lua can communicate with each other. It'll also let lua change from exe to exe (ObjectMerger to FileMerger etc) and it'll also allow for scripts that use no exe.
Each given block of lua would be properly scoped. For example, if you had local variables within a block, another lua block wouldn't be able to access them. This just means put each lua block passed into the central exe inside of do blocks, lol.
Now, I don't know who is going to do this portion, but I think it should be included in the standard jass newgen pack. You guys could develop the exe to manage all of this and let MindWorX know about it. What I do know is that this core is just really important so that lua can communicate with lua in other sections of that map : o.
All of this JASS generation stuff would also be better in C (in the core) and called from Lua.
From here, should also include-

posted at 15/01/10 14:09
Nestharus

Oh yea, final notes on Lua.
If you look, the Lua interpreter is in the NewGen directory =).
I'm sure you guys can plug Lua script directly in to it instead of going through externalblocks =), so should support both both ^_-.
Ok, going to bed now : O.
And yea, if you guys support a general framework for generating JASS files like the one I'm doing, that'd be so epic awesome ^_^.
One amazing req would be a var to get the file name rather than having the user make a text macro like I did, lol.
From there you can use my code to do the framework, or you could write it in C and distribute it with cJASS ;o, or hell, you could even write it in assembly, lol.
Oh well, hopefully with everyone letting you know of all of the major issues with cJASS, it can get fixed up as soon as possible so it's usable.
I know I stopped using cJASS now because if I had to pick between Lua and cJASS, I def 100% pick Lua ;o. Lua has way more preprocessing power than cJASS does ><.
So yea, find ADOLF so cJASS can get these direly needed updates : O.

posted at 15/01/10 13:13
Nestharus

Lua JASS Framework specializing in preprocessing for generation of JASS Code o-o.

posted at 15/01/10 11:01
Nestharus

Can't because it's a single character

posted at 15/01/10 10:43
TheLifelessOne

Try making a seperate library for it, and using private defines to change it? (if possible)

posted at 15/01/10 05:56
Nestharus

Bug..
Your macro replacement totally screwed up all of external blocks for lua..
//! externalblock extension=lua ObjectMerger $FILENAME$
needs to be fixed asap as this is a critical component of jasshelper -.-.

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 21 22 23 24 25 26 27

[back to top]