On 11/17/24 2:21 PM, Harald Anlauf wrote:
Dear all,
the attached patch fixes a rejects-valid / rejects-potentially-valid code issue
for ALLOCATE of CHARACTER with type-spec, and add character length checking
with -fcheck=bounds for the case at hand. It also improves checking of
character function declarations and references slightly, using the diagnostics
of NAG as a guidance.
Some testcases popped up during regtesting, suggesting that one needs to be
careful not to generate too many false positives, so I decided to not spend
to much time on the FIXME's therein. (Character length might be expressions
in an explicit interface and the actual declaration, where we don't have a
reliable way to compare.)
Regtested on x86_64-pc-linux-gnu. OK for mainline?
Thanks,
Harald
Looks good, OK for mainline.
Jerry
