Page 2 of 8

Re: Potential replacement of FOCS with Python

Posted: Mon Oct 24, 2016 1:10 pm
by Geoff the Medio
em3 wrote:My 2c: Value could be made of type with __add__ and __radd__ methods overloaded...
Note that the current implementation has arbitrary parameters for what value = is set to... it can be just Value, or Value + 5, or Max(Value, 5), or Value*3 + 1, or sin(Value^log(Value)), or various other complicated expressions... Point being, don't just think of needing to overload the add methods due to that being the most common case.

Re: Potential replacement of FOCS with Python

Posted: Mon Oct 24, 2016 1:55 pm
by em3
Geoff the Medio wrote:Note that the current implementation has arbitrary parameters for what value = is set to... it can be just Value, or Value + 5, or Max(Value, 5), or Value*3 + 1, or sin(Value^log(Value)), or various other complicated expressions... Point being, don't just think of needing to overload the add methods due to that being the most common case.
I just wanted to note, that example created by adrian_bother could be discouraging to some people because of the syntax:

Code: Select all

Set("Detection", Add(Value, 10))
, and that it can be simplified - actual arithmetic operation could be used instead of the lengthy "functions", like Add, Multiply etc. Granted, some of these functions would have to stay (Max, Sin, Log, ...).

I understand, that designing everything about potential foss replacement in python is a bit too soon, as far as this discussion is concerned. I just wanted to make it sound like a better idea to people that would be bothered by the implied complexity.

Re: Potential replacement of FOCS with Python

Posted: Tue Oct 25, 2016 12:07 am
by spikethehobbit
Effects groups could be made callable. That would eliminate any special handling or typing requirements.

Code: Select all

Special(
    name = "WORLDTREE_SPECIAL",
    description = "WORLDTREE_SPECIAL_DESC",
    stealth = 0,
    spawnrate = 1.0,
    spawnlimit = 1,
    location = lambda planet: (
        isinstance(planet, Planet)
        and (planet.planet_type not in [PlanetType.Asteroids, PlanetType.GasGiant, PlanetType.Inferno])
        and (planet.size not in [PlanetSize.TINY, PlanetSize.HUGE])
        and not (within_starlane_jumps(planet, jumps=2, condition=lambda system: (
            isinstance(system, System)
            and not [planet2 for planet2 in system.planets if planet2.owner]
            )))),
)

Re: Potential replacement of FOCS with Python

Posted: Tue Oct 25, 2016 10:07 am
by em3
spikethehobbit wrote:Effects groups could be made callable. That would eliminate any special handling or typing requirements.

Code: Select all

[...]
I think the whole point is for effects to not be executed in python, just constructed from python scripts (compiled into already existing engine structures).

Re: Potential replacement of FOCS with Python

Posted: Wed Oct 26, 2016 2:35 am
by spikethehobbit
em3 wrote:
spikethehobbit wrote:Effects groups could be made callable. That would eliminate any special handling or typing requirements.

Code: Select all

[...]
I think the whole point is for effects to not be executed in python, just constructed from python scripts (compiled into already existing engine structures).
Fair enough.

Re: Potential replacement of FOCS with Python

Posted: Sun May 05, 2019 9:20 am
by o01eg
em3 wrote: Tue Oct 25, 2016 10:07 am
spikethehobbit wrote:Effects groups could be made callable. That would eliminate any special handling or typing requirements.

Code: Select all

[...]
I think the whole point is for effects to not be executed in python, just constructed from python scripts (compiled into already existing engine structures).
It's a little necroposting but there are AST modules for Python 2 and Python 3: https://docs.python.org/3/library/ast.html https://docs.python.org/2/library/ast.html

Re: Potential replacement of FOCS with Python

Posted: Sun May 05, 2019 9:48 am
by Oberlus
o01eg wrote: Sun May 05, 2019 9:20 amIt's a little necroposting
No really. It's been mentioned recently the common desire to substitute FOCS with Python for several reasons (compilation time and requirements, many more people know Python than FOCS, ...). So good to resurrect this.

Re: Potential replacement of FOCS with Python

Posted: Sun May 05, 2019 6:30 pm
by Ophiuchus
o01eg wrote: Sun May 05, 2019 9:20 am
em3 wrote: Tue Oct 25, 2016 10:07 am
spikethehobbit wrote:Effects groups could be made callable. That would eliminate any special handling or typing requirements.

Code: Select all

[...]
I think the whole point is for effects to not be executed in python, just constructed from python scripts (compiled into already existing engine structures).
It's a little necroposting but there are AST modules for Python 2 and Python 3: https://docs.python.org/3/library/ast.html https://docs.python.org/2/library/ast.html
As far as i get it these modules are for accessing the python a.syntax tree. So usuable for tools and metaprogramming python. I dont see how these help in generating/metaprogramming FOCS (?)

Re: Potential replacement of FOCS with Python

Posted: Sun May 05, 2019 6:36 pm
by o01eg
Ophiuchus wrote: Sun May 05, 2019 6:30 pm
o01eg wrote: Sun May 05, 2019 9:20 am
em3 wrote: Tue Oct 25, 2016 10:07 am
I think the whole point is for effects to not be executed in python, just constructed from python scripts (compiled into already existing engine structures).
It's a little necroposting but there are AST modules for Python 2 and Python 3: https://docs.python.org/3/library/ast.html https://docs.python.org/2/library/ast.html
As far as i get it these modules are for accessing the python a.syntax tree. So usuable for tools and metaprogramming python. I dont see how these help in generating/metaprogramming FOCS (?)
I suppose this is the entire thread about replace FOCS with Python.

Re: Potential replacement of FOCS with Python

Posted: Sat May 11, 2019 9:08 am
by Ophiuchus
o01eg wrote: Sun May 05, 2019 6:36 pm
Ophiuchus wrote: Sun May 05, 2019 6:30 pm
o01eg wrote: Sun May 05, 2019 9:20 am

It's a little necroposting but there are AST modules for Python 2 and Python 3: https://docs.python.org/3/library/ast.html https://docs.python.org/2/library/ast.html
As far as i get it these modules are for accessing the python a.syntax tree. So usuable for tools and metaprogramming python. I dont see how these help in generating/metaprogramming FOCS (?)
I suppose this is the entire thread about replace FOCS with Python.
Yes, but you do not answer my question/the topic.


The OP is about replacing FOCS as a text format. (For me extending the capabilities would actually be enough).
Python would be used to generate definition structure which we could freeorion content [abstract syntax] tree (FOCT/FOCAST)?

So we are talking about FOCAST not python AST.

I will rephrase:

As far as i get it these modules are for accessing the python a.syntax tree. So usuable for tools and metaprogramming python. I dont see how these help in generating/metaprogramming freeorion content.

Re: Potential replacement of FOCS with Python

Posted: Sun May 31, 2020 8:06 am
by Ophiuchus
A bump here as oleg mentioned and adrian is working on something related.

Given I have the content specified in python and I have the ability to understand what has to be done (by some binding or analysing the python file using python-ast)
  • (how) can I create all game structures in the game via python?
  • (how) can I start python in the backend?

Re: Potential replacement of FOCS with Python

Posted: Sun May 31, 2020 10:14 am
by Geoff the Medio
Ophiuchus wrote: Sun May 31, 2020 8:06 amGiven I have the content specified in python and I have the ability to understand what has to be done (by some binding or analysing the python file using python-ast)
  • (how) can I create all game structures in the game via python?
  • (how) can I start python in the backend?
Your questions are unclear. Do you mean:
-How can you use Python right now to define stuff like building types or species? You can't. Such a system doesn't exist.
-How could you use Python if FOCS was replaced with a Python parser to defined stuff like buildings or species? Don't know; such a system doesn't exist.
-How can you implement support for such a system? Not sure; it would be a bit complicated to set up.

Python is already run on the server for turn events, and for AI clients. You can check how that's done in the C++ code by searching... Perhaps that answers part of your questions?

Re: Potential replacement of FOCS with Python

Posted: Sun May 31, 2020 1:53 pm
by o01eg
Ophiuchus wrote: Sun May 31, 2020 8:06 am A bump here as oleg mentioned and adrian is working on something related.

Given I have the content specified in python and I have the ability to understand what has to be done (by some binding or analysing the python file using python-ast)
  • (how) can I create all game structures in the game via python?
  • (how) can I start python in the backend?
  • I suppose you should feed file to ast.parse. Then convert python object to AST object with PyAST_obj2mod. And then somehow process resulting structure.
  • Maybe you mean background?
    I suppose it could be done by using sub-interpreters in non-main thread.

Re: Potential replacement of FOCS with Python

Posted: Sun May 31, 2020 2:11 pm
by Ophiuchus
Geoff the Medio wrote: Sun May 31, 2020 10:14 am
Ophiuchus wrote: Sun May 31, 2020 8:06 amGiven I have the content specified in python and I have the ability to understand what has to be done (by some binding or analysing the python file using python-ast)
  • (how) can I create all game structures in the game via python?
  • (how) can I start python in the backend?
Your questions are unclear. Do you mean:
-How can you use Python right now to define stuff like building types or species? You can't. Such a system doesn't exist.
-How could you use Python if FOCS was replaced with a Python parser to defined stuff like buildings or species? Don't know; such a system doesn't exist.
-How can you implement support for such a system? Not sure; it would be a bit complicated to set up.

Python is already run on the server for turn events, and for AI clients. You can check how that's done in the C++ code by searching... Perhaps that answers part of your questions?
Yes, thank you. I am interested in creating the c++ building types/species... via python. So it seems that is currently not possible.

Does any of you pythonistas have experience with creating c data structures from python? Or would i need to provide an api like we have and expose that?

Re: Potential replacement of FOCS with Python

Posted: Sun May 31, 2020 2:41 pm
by Geoff the Medio
Ophiuchus wrote: Sun May 31, 2020 2:11 pmI am interested in creating the c++ building types/species... via python. So it seems that is currently not possible.

Does any of you pythonistas have experience with creating c data structures from python? Or would i need to provide an api like we have and expose that?
Even if there was a way to create the C++ data structures in Python, similar to how they are created when parsing FOCS files, you'd need to then pass them into the server and clients, either sending them from server to clients or ensuring they are created the same way on both. You'd need some sort of API exposed to do that regardless.