Re: string_exec()

new topic     » goto parent     » topic index » view thread      » older message » newer message
petelomax said...

Please try to provide specific examples to back up any comments or claims.

I have no clue to the structure of compiled OE, but what if the line containing the string_eval() caused a Non-Fatal interrupt, the line was evaluated to be a valid string_eval(), it (or the file it mentions) is compiled and then inlined (rather more of a local gosub in machine code) in place of the line which caused the interrupt, and then program flow resumes normally from the point of interrupt?

Again, from a clueless perspective, this smells like it will cause the least headaches in var/procedure scopes, as well as goto and name scopes (include balh.e as balh). No new environments would need to be built, no massive copying of var tables, permission slips, colors, or battery voltages. If the string crashes because it assumes something blah blah blah, then it's a normal blah blah blah program crash.

A working code example:

     .timer 1 %defflooddelay %temptext 

Kat

new topic     » goto parent     » topic index » view thread      » older message » newer message

Search



Quick Links

User menu

Not signed in.

Misc Menu