Advice on removing code redundancy

[Saeed Ehteshamifar]

>
> Use paramterized CPP (C pre-processor) macros with string
> concatenation via ##, something like

Thanks Gedare. That was really helpful.
Now the source code is much more readable.

Kind Regards,
SAeeD


On Wed, Aug 19, 2015 at 8:47 PM, Gedare Bloom <gedare at rtems.org> wrote:

> Ah, and then usage is:
>
> TEST_BODY(tc_abs, 1, TC_abs)
>
> Or something like that.
>
> On Wed, Aug 19, 2015 at 2:46 PM, Gedare Bloom <gedare at rtems.org> wrote:
> > Use paramterized CPP (C pre-processor) macros with string
> > concatenation via ##, something like
> >
> > #define TEST_BODY(test_name, test_num, test_tag) \
> > void __run_##test_name##_##test_num##() \
> > { \
> > ... \
> >   rtems_task_start (tc_task_id, ___##test_tag##_##test_num##, 0);
> > ... \
> > } \
> > rtems_task Init (rtems_task_argument) \
> > { \
> >   tc_finished = false; \
> >   ___run_##test_name##_##test_num##(); \
> > ...
> > } \
> >
> > Something like that, the syntax may be a bit different but that's the
> idea.
> >
> > On Wed, Aug 19, 2015 at 10:26 AM, Saeed Ehteshamifar
> > <salpha.2004 at gmail.com> wrote:
> >> Hi,
> >>
> >> Please take a look at following snippet:
> >>
> >> testcase_executer.cpp:
> >> 1.  #include "___TC_abs_1.h"
> >> 2.
> >> 3.  void ___run_tc_abs_1 ()
> >> 4.  {
> >> 5.    rtems_id tc_task_id;
> >> 6.    rtems_name tc_task_name = rtems_build_name( 'T', 'C', ' ', ' ');
> >> 7.    rtems_task_create(
> >> 8.    tc_task_name, TC_TASK_PRIORITY, RTEMS_MINIMUM_STACK_SIZE,
> >> 9.      RTEMS_DEFAULT_MODES, RTEMS_DEFAULT_ATTRIBUTES, &tc_task_id);
> >> 10.
> >> 11.   rtems_task_start (tc_task_id, ___TC_abs_1, 0);
> >> 12.
> >> 13.   for (int i=0; i<TC_TIMEOUT_DIVISION_FACTOR; i++) {
> >> 14.     status = rtems_task_wake_after(RTEMS_MICROSECONDS_TO_TICKS(
> >> 15.       TC_TIMEOUT/TC_TIMEOUT_DIVISION_FACTOR));
> >> 16.     if (tc_finished)
> >> 17.       break;
> >> 18.   }
> >> 19. }
> >> 20.
> >> 21. rtems_task Init (rtems_task_argument)
> >> 22. {
> >> 23.   tc_finished = false;
> >> 24.   ___run_tc_wait_1 ();
> >> 25.   if (tc_finished)
> >> 26.     cout << "___TC_wait_1: " << "PASS" << endl;
> >> 27.   else
> >> 28.     cout << "___TC_wait_1: " << "RESTART" << endl;
> >> 29.   cout << endl;
> >> 30.
> >> 31.   cout << "finito!" << endl;
> >> 32.   exit (0);
> >> 33. }
> >>
> >> It's going to be thousands of test cases, and I want to serially run
> each
> >> test case in a task other than Init.
> >> In the code above, I've marked symbols that are per each test case,
> with 3
> >> leading underscores (___).
> >> This is the basic duplication structure I've come up with. Since RTEMS
> >> doesn't support dynamic linking, I'm going to duplicate lines 3 to 19
> >> (___run_tc_abs_1) and 23 to 29 for each test case which leads to having
> a
> >> very big source file for testcase_executer.cpp.
> >> Is there any better solution that avoids code redundancy that I'm not
> aware
> >> of? Do you have advices on how to change this structure to minimize code
> >> redundancy? Any help is much appreciated.
> >> Note: ___TC_abs_1 is already defined in a separate CPP file.
> >>
> >> Thanks for your time.
> >>
> >> Kind Regards,
> >> SAeeD
> >>
> >> _______________________________________________
> >> users mailing list
> >> users at rtems.org
> >> http://lists.rtems.org/mailman/listinfo/users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20150820/b3268d63/attachment-0002.html>