so someone else fixed up the wreck we know as HuC and it now uses ANSI function declarations (death to small C!) as well as structs. oh yeah, it is also 64 bit compatible.
the source is on github: https://github.com/uli/huc
new version of HuC!
Re: new version of HuC!
Do you if some binary releases are planed? Otherwise maybe someone should take over the task and propose windows/linux/mac builds (provided he has the machines to build them).
Re: new version of HuC!
pff! it uses like nil external libs. if you dont already have a compiler, you dont need this compiler!MooZ wrote:Do you if some binary releases are planed? Otherwise maybe someone should take over the task and propose windows/linux/mac builds (provided he has the machines to build them).
Re: new version of HuC!
Sure but I know from experience that whole bunch of people will still use the old version because it comes as a packaged binary.
-
tomaitheous
- Posts: 88
- Joined: Mon Jun 23, 2008 1:58 pm
Re: new version of HuC!
Did some of the PCEAS bugs get fixed as well? Specifically, the bug where if an opcode of more than 1byte, crosses a 8k page boundary.
Re: new version of HuC!
His pceas version is based on nop's fork (https://github.com/nop00/pceas).
I don't think this bug was fixed. What happened when the instruction crosses the bank boundary? PCEAS vomits an error? The generated bin is buggy?
If it's the old "bank overflow" error I think it's fixed as the repo was started from your latest release.
I don't think this bug was fixed. What happened when the instruction crosses the bank boundary? PCEAS vomits an error? The generated bin is buggy?
If it's the old "bank overflow" error I think it's fixed as the repo was started from your latest release.
-
tomaitheous
- Posts: 88
- Joined: Mon Jun 23, 2008 1:58 pm
Re: new version of HuC!
Ahh ok, I see. It is. But, there's a possible incompatibility with old PCEAS generated code here; local label names used in high lever expressions. I added the ability to do .- and .+ (.+, .++, .---, .----, etc) for local labels. And you can use this just fine in expressions.. *IF* you use a space. The release can came with 3.21 Denki allowed you to do .local+#$33. But you can no longer do this in this new version. You have to use spaces in between the operator symbols and label names, else it thinks ".local+" is the label. So, .local + #$33 works fine, in the new build. But not without spaces, if that makes sense.
I fixed one error handling routine (didn't have an return/exit). I added .dwl and .dwh, which are define byte directives, but get the high byte or low byte of a word. The bank opcode in the middle of two banks now just increments the bank, but I didn't test all internal address calculations into the second bank (if there's a bug there before relative to that, then it's still there). I added "." character for macro names as well, i.e "MOVE.b" or "MOVE.w".
I'm coding some macros now, and there needs to be a new define for indirect addressing. Both [zp] and [zp],y return the same test value (4). Makes it a pain for making all purpose pseudo opcodes.
HuC also needs a form of assigning a bank/address to a higher level function. And possible a segment layout that's mapped on the function call (something the user can define). CC65 has this, and this is how it gets around the memory near/far/ram/etc layout problems of a banked system (this would also allow more optimized dev on SGX and CD projects that need to setup and access more than 8k of ram). If I can get this compiled on windows, I'll see if I can add this (I added a few things to HuC source in the past (C++ style comments, single inline asm directive, etc - just never made them public).
I fixed one error handling routine (didn't have an return/exit). I added .dwl and .dwh, which are define byte directives, but get the high byte or low byte of a word. The bank opcode in the middle of two banks now just increments the bank, but I didn't test all internal address calculations into the second bank (if there's a bug there before relative to that, then it's still there). I added "." character for macro names as well, i.e "MOVE.b" or "MOVE.w".
I'm coding some macros now, and there needs to be a new define for indirect addressing. Both [zp] and [zp],y return the same test value (4). Makes it a pain for making all purpose pseudo opcodes.
HuC also needs a form of assigning a bank/address to a higher level function. And possible a segment layout that's mapped on the function call (something the user can define). CC65 has this, and this is how it gets around the memory near/far/ram/etc layout problems of a banked system (this would also allow more optimized dev on SGX and CD projects that need to setup and access more than 8k of ram). If I can get this compiled on windows, I'll see if I can add this (I added a few things to HuC source in the past (C++ style comments, single inline asm directive, etc - just never made them public).
-
HyperFighting
- Posts: 6
- Joined: Thu Jun 26, 2014 12:58 am
Re: new version of HuC!
I'm with Mooz on this one.
A binary release would make it easier for new comers to get started. I have never compiled a compiler before but the ability to use C++ style comments and structs would be awesome among other things.
It boils me up when people turn their nose up and suggest "If you don't know how to do it don't bother it's not for you" this type of attitude sucks. We are people at different skill levels having fun with what we like. If you don't want to help then don't but on the other hand only contribute to the discussion if it is positive. I made up a little rap it goes like this... life is about F-U-N if you ain't about that F-U then. Sorry I know that last part wasn't very positive but I I'm so tired of high and mighty comments over bull crap.
A binary release would make it easier for new comers to get started. I have never compiled a compiler before but the ability to use C++ style comments and structs would be awesome among other things.
It boils me up when people turn their nose up and suggest "If you don't know how to do it don't bother it's not for you" this type of attitude sucks. We are people at different skill levels having fun with what we like. If you don't want to help then don't but on the other hand only contribute to the discussion if it is positive. I made up a little rap it goes like this... life is about F-U-N if you ain't about that F-U then. Sorry I know that last part wasn't very positive but I I'm so tired of high and mighty comments over bull crap.
Re: new version of HuC!
Trying to mash this into a small Haiku format as possible.
Well Hyperfighting, that just seems the attitude of Gravis. I have suggested many ideas and I get verbal abuse. Sad part he actually likes Zelda ( or at least liked ). In fact I am like three college grade classes behind from getting my bachelors in Computer Science. If I had not failed the last class* I would have been doing "Circuitry and logic", and the other Math classes.
* That was because for the last three years I decided to take care of my senile family member while I was on break from college. Very stressful to venture from one place with barely any electronics and your deciding what to do after stopping from college ( I have over 200+ usable credits to branch off from ). Being in new places without any idea to do with my time.
In fact being/visiting that crowd. The Sci-fi and Anime club. I do not want to be apart of that crowd anymore. Because nobody is open or wants to get along. That is how I feel in general. People want me as a tool. I purchased a copy of CV-X for the PCE. I tested it out. Somebody on the bigger PCE forum made a new account just to ask me if it was real or not. Instead of supporting me they hide behind another user name. If I was Japanese I would probably be doing animations or a comic right now. Even a filthy flash game. Apparently you have to grow a Vagina in order to be respected in the west as an artist, as nobody wants to fund men
In short I would rather move to Flash as a programming tool for non-programmers. Because the community is big.
Well Hyperfighting, that just seems the attitude of Gravis. I have suggested many ideas and I get verbal abuse. Sad part he actually likes Zelda ( or at least liked ). In fact I am like three college grade classes behind from getting my bachelors in Computer Science. If I had not failed the last class* I would have been doing "Circuitry and logic", and the other Math classes.
* That was because for the last three years I decided to take care of my senile family member while I was on break from college. Very stressful to venture from one place with barely any electronics and your deciding what to do after stopping from college ( I have over 200+ usable credits to branch off from ). Being in new places without any idea to do with my time.
In fact being/visiting that crowd. The Sci-fi and Anime club. I do not want to be apart of that crowd anymore. Because nobody is open or wants to get along. That is how I feel in general. People want me as a tool. I purchased a copy of CV-X for the PCE. I tested it out. Somebody on the bigger PCE forum made a new account just to ask me if it was real or not. Instead of supporting me they hide behind another user name. If I was Japanese I would probably be doing animations or a comic right now. Even a filthy flash game. Apparently you have to grow a Vagina in order to be respected in the west as an artist, as nobody wants to fund men
In short I would rather move to Flash as a programming tool for non-programmers. Because the community is big.