From: cross@spitfire.i.gajendra.net   
      
   In article ,   
   Dave Froble wrote:   
   >On 4/14/2025 1:06 PM, Simon Clubley wrote:   
   >> On 2025-04-14, Dave Froble wrote:   
   >>>   
   >>> An example of a style I favor:   
   >>>   
   >>> Stat% = SYS$QIOW( , ! Event flag &   
   >>> ListenCh% By Value, ! VMS channel &   
   >>> IO$_SETCHAR By Value, ! Operation &   
   >>> IOSB::Stat%, ! I/O status   
   block &   
   >>> , ! AST routine &   
   >>> , ! AST parameter &   
   >>> ListenOpt::Protocol%, ! P1 &   
   >>> , ! P2 &   
   >>> ServerItemLst::Len%, ! P3 - local   
   socket na^   
   >>> BackLog% By Value, ! P4 -   
   connection back^   
   >>> SockOptItemList::Len%, ! P5 - socket   
   options &   
   >>> ) ! P6   
   >>>   
   >>> Ok, how many lines of code?   
   >>>   
   >>   
   >> Exactly one line of code.   
   >   
   >Indeed! However, some utilities used to count lines of code might count more   
   >than one. So, the entire exercise should be considered subject to errors.   
      
   There are utilities that are designed to count lines of code   
   that account for that sort of thing. Whether those can handle   
   VMS BASIC (let alone Macro-32) is, of course, a question, but   
   most people trying to analyze large codebases are aware that   
   they need special tools to do it.   
      
   >Perhaps looking at the size of object files, or executables, might be a better   
   >measure?   
      
   That's usually not a great measure, as lots of things can go   
   into this that do not directly reflect code: consider unrolled   
   loops that might artificially increase the size of program text,   
   large but mostly static tables of data, etc.   
      
    - Dan C.   
      
   --- SoupGate-DOS v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   
|