(no subject)

[Frank Heckenbach]

Oldham, Adam wrote:

> Well how do you guys debug GPC when you are testing new features you have
> written?

Only speaking for myself, not Peter, I usually write a series of test
programs (in the test suite in the GPC sources) to test every conceivable
use or misuse of the new feature and whether GPC handles them right (i.e.,
accepts all legitimate uses and rejects all invalid ones, possibly
different with different dialect or other options). Of course, I also
test manually the cases in my real code (or others' code, posted to the
list) which caused the need for the changes.

After each change, I let the test suite run (under Linux and DJGPP), and
if any problems are found, I fix them (which then is usually not too
hard because I know which changes to GPC caused them).

I use gdb on gpc1 rather rarely, and mostly only to get the code position
of a crash or something.

Frank

-- 
Frank Heckenbach, frank at g-n-u.de, http://fjf.gnu.de/
GPC To-Do list, latest features, fixed bugs:
http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html