Differences

This shows you the differences between two versions of the page.

Link to this comparison view

luabot_vs_hbot [2010/03/21 16:40]
hadrien created
luabot_vs_hbot [2010/03/21 16:42] (current)
hadrien
Line 5: Line 5:
 ==== Evolution of luabot ==== ==== Evolution of luabot ====
  
-At the very beginning, I used lua as a replacement for the configuration files I had, meaning that only the controller, servos, sensors, etc. where described by luabot. At this point I had a C++ main file that runs the configuration script, then instantiate the various algorithms (map generation, discovery, etc.) and connected the signals and slots in a deterministic way. This was really limitating, so I started writing lua bindings for these algorithms and components, and for the signal/slot mechanism (i.e. the "​connect"​ call). I had to enhance many of the hbot internals to support the dynamicity of a lua script (introducing device repositories and device factories).+At the very beginning, I used lua as a replacement for the configuration files I had, meaning that only the controller, servos, sensors, etc. where described by luabot. At this point I had a C++ main file that run the configuration script, then instantiate the various algorithms (map generation, discovery, etc.) and connected the signals and slots in a deterministic way. This was really limitating, so I started writing lua bindings for these algorithms and components, and for the signal/slot mechanism (i.e. the "​connect"​ call). I had to enhance many of the hbot internals to support the dynamicity of a lua script (introducing device repositories and device factories).
  
-This was a real progress because I could have just one executable file (the luabot interpreter),​ and have everything allocated and connected to each other at runtime (and not at compile time as it was the case before). The problem is that I could only connect signals and slots, not interact on the objects, algorithms, components of the hbot world from the luabot world. I expanded the bindings to add that, and gradually most of the high-level components and their methods were available in lua.+This was a real progress because I could have just one compiled program ​(the luabot interpreter),​ and have everything allocated and connected to each other at runtime (and not at compile time as it was the case before). The problem is that I could only connect signals and slots, not interact on the objects, algorithms, components of the hbot world from the luabot world. I expanded the bindings to add that, and gradually most of the high-level components and their methods were available in lua.
  
 The last step is to "​finish the work" and make the last bindings with the low-level elements (servos, sensors, bumpers, etc.). This is mostly done now, some methods bindings are still to be developped, but we are getting closer. Now the big challenge is to keep consistency between the hbot world and the luabot world. And not lose simplicity (that already suffered a bit from this exchanges back-and-forth). This challenge today has several aspects: The last step is to "​finish the work" and make the last bindings with the low-level elements (servos, sensors, bumpers, etc.). This is mostly done now, some methods bindings are still to be developped, but we are getting closer. Now the big challenge is to keep consistency between the hbot world and the luabot world. And not lose simplicity (that already suffered a bit from this exchanges back-and-forth). This challenge today has several aspects:
 
luabot_vs_hbot.txt ยท Last modified: 2010/03/21 16:42 by hadrien
Recent changes RSS feed Creative Commons License Donate Powered by PHP Valid XHTML 1.0 Valid CSS La rache Driven by DokuWiki