Hey,
Can anyone tell me which files of the SDK (2.60) take part in the parsing of the script file before loading a map?
I need them for ETScript, want to include this functionality…
Till now I have found some bits in g_script.c and mainly in g_script_actions.c but I don’t think that’s ALL there is, or is it?
I see…
Where can I find the code for these COM_Parse calls?
I’m sorry, I have absolutely NO experience in C…
I had a feeling the actuall parsing was done in the COM_Parse sections… but couldn’t find them.
In MS Visual Studio you can search all project file with STRG+SHIFT+F (most editors have such features). When you enter “COM_Parse” as search phrase it will list all occurences in a window. You only need to double click a find to open the according source file at the correct position. In the source file you can right click a function name etc. (to open the context menu) and choose “Go To Declaration” or “Go To Definition”.
I just checked it with MS Visual Studio Beta 2 (its currently freely available: http://msdn2.microsoft.com/en-us/evalcenter/bb655864.aspx ). Just double click the wolf.sln-project file and VS will open up. Click through the dialogs to convert the project file to the new version. You can even build the game (don’t forget to set $(WOLFDIR)).
But imho (when you have absolutely no C experience) you should go another way: Use something like yacc+lex to mimic script parsing and execution. This way its much easier to support different versions (e.g. ETPro vs. ETMain) or to apply changes/updates. Yacc- resp. lex-like tools should be available for most programming languages.
I have VS2005 so that’s no problem.
What’s “STRG” though, lol?
I will try some things out…
I will see what yacc and lex is… are they plugins or something?
Thanks for the help.
Oops, STRG (Steuerung) is German for the English CTRL (Control).
Lex creates a scanner and Yacc is a compiler-compiler.
So lex creates the program source for a scanner, which parses a file into tokens. It takes only a (relativly) simple syntax file based on regular expressions. And yacc creates the source for a compiler (syntax usually given in BNF), so you can create a program which runs scripts.
Yacc and lex work for C, but there are similar tools available for nearly all programming languages.
Look in q_shared.c
I typically use grep to find things like this.
If you have no experience in C, the parsing code will probably take you quite a bit of work to fully understand.
The Q3 parsing stuff doesn’t come from lex and yacc
edit: I realize Ragnar_40k wasn’t suggesting that it was, but you should be aware that et script syntax might have quirks specific to the way it was implemented.
Afaics WET just looks at the script code character by character and accepts a string or rewinds to an earlier position and tries again (resp. returns an error). Lex has the lookahead-operator, which does basically the same thing, so you can mimic this 1:1 if needed. But I guess lexical analysis isn’t the problem.
A problem might be to implement script execution correctly (including bugs). But here you can start with an approach which works correctly in 99.9% of the cases and add handling for special quirks later on if needed (which should be relativly easy when you use yacc or something similar).