My shell EA code

[SteveHopwood]

Excellent stuff Mahmut. I have posted a link in post 1 here, to make sure others see your work.

Cheers

[gaheitman]

Hi Steve, George, Mike and others who might be interested,

While working on an update to the swingONE EA I decided to build what I feel is an easier framework for building EAs that I think may interest you as well.

This is primarily based on the shell EA and I have incorporated other parts such as the DFTD and lock trading thread from Mike's CTC EA.

My primary motivations for this were:
- Easier to update all EAs based on this framework when there is a new bug fix or feature.
- The framework does not need to be edited for each EA based on it but instead it is intended to grow to accommodate all bug fixes and improvements.
- The framework does not include any trading logic at all - this logic is stored separately.
- Have everything stored under a source control system so it is easy for all to see differences between versions
- Even though the code is stored as multiple files a single .mq4 file should be created for easier installation.

I setup the following repositories to host this code:
https://bitbucket.org/mahmut/ct4-eabase
https://bitbucket.org/mahmut/ct4-eas

I have done this primarily for the EAs I will work on personally however if you like the idea I don't see why it cannot be used by all of us.

I have not documented it well yet so if there is anything not obvious I will be happy to clarify.

Please have a look if you have the time and I would very much appreciate your feedback. Also please keep in mind this is still a work in progress.

Kind Regards,
Mahmut

I'd actually prefer code I've written NOT be subject to GPL.

Thanks,
George

[markft]

I'd actually prefer code I've written NOT be subject to GPL.

Thanks,
George

Hi George,

I don't have a strong preference for licences. My intention was others should be able to use the finished code for their own use and make changes but not sell commercially (some would just ignore anyway but still). Do you find GPL too restrictive or too open?

I understand the portion of code developed by one individual continues to belong to the original author regardless of additonal licences that may be granted to other parties e.g. GPL. So authors of code are themselves not limited by GPL in any way.

In this case I have used the libGMT and libOrderReliable from Matt (also from this forum); and these are GPLed. This requires the derivative product (the complete EA e.g. swingONE.mq4) including this code become GPL as well. The only way to make it non-GPL is to either remove this code, ask Matt for another licence or reimplement it. I don't mind the GPL restriction so I didn't do this.

However this only applies to the derivative product (i.e. the complete EA); other libraries used by it can be non-GPL'ed. For example I am not sure what is your intended licence for DFTD; but it is not itself GPL'ed. It has whichever licence you wish it to have.

Perhaps this can be clarified by saying at the beginning of the EA that the EA uses GPL'ed libraries and therefore is available as GPL; and other libraries or code used in the EA are copyrighted by individual authors and are not GPL?

To be honest I didn't think much of it until now; just trying to get it working well. Are you OK with using your code such as the DFTD in the swingONE EA and others I may develop using this framework? I thought it was OK since everyone has been resuing each others' code in the forum.

Your feedback and comments are most welcome; and please let me know if you would like me to remove code developed by you from this framework.

Cheers,
Mahmut

[mbkennel]

If anybody wants to use libGMT and libOrderReliable with another license, PM me.

I just wanted it to be GPL as it transits "out there in the void".

[gaheitman]

I'd actually prefer code I've written NOT be subject to GPL.

Thanks,
George

To be honest I didn't think much of it until now; just trying to get it working well. Are you OK with using your code such as the DFTD in the swingONE EA and others I may develop using this framework? I thought it was OK since everyone has been resuing each others' code in the forum.

Your feedback and comments are most welcome; and please let me know if you would like me to remove code developed by you from this framework.

Cheers,
Mahmut

Of course you can continue to use any code I've written in swingONE and any other EAs you are writing. The stuff I put up here is free for any one to use any way they want. I'm just not a fan of "free...but" licensing schemes, specifically their viral nature. As long as the license relates to the compilation/framework and the individual parts remain here for anyone to grab, then I don't have an issue.

Thanks for taking the time to put it all together.

George

[markft]

Of course you can continue to use any code I've written in swingONE and any other EAs you are writing. The stuff I put up here is free for any one to use any way they want. I'm just not a fan of "free...but" licensing schemes, specifically their viral nature. As long as the license relates to the compilation/framework and the individual parts remain here for anyone to grab, then I don't have an issue.

Thanks for taking the time to put it all together.

George

Thank you George. Full steam ahead then. Also will make sure there is a proper copyright for each section to indicate ownership just in case.

[markft]

If anybody wants to use libGMT and libOrderReliable with another license, PM me.

I just wanted it to be GPL as it transits "out there in the void".

Hi Matt,

Thank you for this. I do not mind GPL for the framework I am working on; and it is still "out there" so no need from me at least.

Cheers

[SteveHopwood]

Latest update in post 1.

Stealth could cause the Breakeven function to repeatedly try to move a stop to BE, even though BE has already been achieved.

Also, I received this pm this morning:
"Hi Steve,

There is a bug in the hanover filter of your Shell EA.

Does not work with a slope above 0
if (SlopeCandles[cc] > 0 && prevstrength1 >= prevstrength2) return(false);

Modified
if (SlopeCandles[cc] > 0 && prevstrength1 >= strength1) return(false);

Many Thanks
Andy"

Many thanks to Andy for spotting this.

[SteveHopwood]

Another update in post 1. This one is in response to a pm from Mahmut (markft):

"Hi Steve,

I think there is a small typo in the latest version.

Shouldn't the line "if (SlopeCandles[cc] > 0 && prevstrength1 >= prevstrength1) return(false);" be updated to "if (SlopeCandles[cc] > 0 && prevstrength1 >= strength1) return(false);" as you mentioned in the post? Same for the OP_SELL version of course.

Cheers,
Mahmut"

Thanks Mahmut. Great piece of bloop-spotting.

Keep the bloop-spots coming guys. If you were not a part of the original Hanover module development back at FF, you will not know how difficult I found it to code, so you will not understand how easy it was for errors to creep in. Believe me, LUC started to write out party invitations long before I published any actual code.

[mobthehop]

Steve, as you know I am quite enamoured with the strength indicator xmeter....

Not sure whether you visited this part of your (excellent) site already http://viewtopic.php?f=20&t=48.

Would you please consider to modulize the xmeter calculation as an alternative to the hanover module?

And possibly consider to add it to the PB EA?

Please....

Mob