SVCD::write [NOT READY] #39
Conversation
| resolve_table(L, "TIMEOUT"); | ||
| int timeout = (int) lua_tonumber(L, -1); | ||
|
|
||
| resolve_table(L, "handlers"); // 7 |
There was a problem hiding this comment.
While it is a good abstraction for development, this may add extra overhead to the embedded system since SVCD need to be load/unload every time this function is called, but in reality, it only need to be load once.
Untitled
There was a problem hiding this comment.
I agree with Jack that this adds unnecessary overhead. The point of these C functions is that although they are ugly to write, you spend the time trying to optimise them. Development and prototyping would be done in Lua, in C you should be trying to be as efficient as possible
There was a problem hiding this comment.
This is experimental code that we forgot to remove.
There was a problem hiding this comment.
Can you please remove it before your final pull request due in 1 hour and 20 minutes?
|
I am not convinced that this code was tested, how does anon_function get visible in the global scope? Can you detail how this was tested? |
| storm.os.invokeLater(timeout_ms*storm.os.MILLISECOND, function() | ||
| if SVCD.handlers[ivkid] ~= nil then | ||
| SVCD.handlers[ivkid](SVCD.TIMEOUT) | ||
| SVCD.handlers[ivkid] = nil | ||
| end | ||
| end) | ||
| storm.net.sendto(SVCD.wcsock, storm.mp.pack({svcid, attrid, ivkid, payload}), targetip, 2526) | ||
| end |
There was a problem hiding this comment.
Not really sure what is going on with this bit of Lua left over. Do you still require it?
There was a problem hiding this comment.
Yea, we still require the invoke later. We were trying to switch it over to C and we got stuck.
|
|
||
| /* SEND TO */ | ||
| lua_pushlightfunction(L, libstorm_net_sendto); | ||
| // resolve_table(L, "print_sendto"); |
There was a problem hiding this comment.
You can remove this line; It was a special debugging statement. Was resolving a lua function SVCD.print_sendto...
This is the same SCVD write pull request, except with a cleaned commit log. Please comment on either.
#36